def test_plantcv_dilate():
img = cv2.imread(os.path.join(TEST_DATA, TEST_INPUT_BINARY), -1)
device, dilate_img = pcv.dilate(img=img, kernel=5, i=1, device=0, debug=None)
# Assert that the output image has the dimensions of the input image
if all([i == j] for i, j in zip(np.shape(dilate_img), TEST_BINARY_DIM)):
# Assert that the image is binary
if all([i == j] for i, j in zip(np.unique(dilate_img), [0, 255])):
assert 1
else:
assert 0
else:
assert 0
def test_plantcv_dilate():
img = cv2.imread(os.path.join(TEST_DATA, TEST_INPUT_BINARY), -1)
device, dilate_img = pcv.dilate(img=img,
kernel=5,
i=1,
device=0,
debug=None)
# Assert that the output image has the dimensions of the input image
if all([i == j] for i, j in zip(np.shape(dilate_img), TEST_BINARY_DIM)):
# Assert that the image is binary
if all([i == j] for i, j in zip(np.unique(dilate_img), [0, 255])):
assert 1
else:
assert 0
else:
assert 0
def main():
# Get options
args = options()
# Read image
img, path, filename = pcv.readimage(args.image)
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(img, 's', device, args.debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, 'light', device, args.debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 0, device, args.debug)
device, s_cnt = pcv.median_blur(s_thresh, 0, device, args.debug)
# Fill small objects
#device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(img, 'b', device, args.debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, 'light', device, args.debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, 'light', device, args.debug)
# Fill small objects
#device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, args.debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(img, bs, 'white', device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, 'a', device, args.debug)
device, masked_b = pcv.rgb2gray_lab(masked, 'b', device, args.debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, 'dark', device, args.debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, 'light', device, args.debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, args.debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, args.debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 2, device, args.debug)
# Dilate to join small objects with larger ones
device, ab_cnt1=pcv.dilate(ab_fill1, 3, 2, device, args.debug)
device, ab_cnt2=pcv.dilate(ab_fill1, 3, 2, device, args.debug)
# Fill dilated image mask
device, ab_cnt3=pcv.fill(ab_cnt2,ab_cnt1,150,device,args.debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, 'white', device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, 'a', device, args.debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, 'b', device, args.debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, 'dark', device, args.debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255, 'light', device, args.debug)
device, ab_fill = pcv.logical_or(masked2a_thresh, masked2b_thresh, device, args.debug)
# Identify objects
device, id_objects,obj_hierarchy = pcv.find_objects(masked2, ab_fill, device, args.debug)
# Define ROI
device, roi1, roi_hierarchy= pcv.define_roi(masked2,'rectangle', device, None, 'default', args.debug,True, 525, 0,-490,-150)
# Decide which objects to keep
device,roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(img,'partial',roi1,roi_hierarchy,id_objects,obj_hierarchy,device, args.debug)
# Object combine kept objects
device, obj, mask = pcv.object_composition(img, roi_objects, hierarchy3, device, args.debug)
############## VIS Analysis ################
outfile=False
if args.writeimg==True:
outfile=args.outdir+"/"+filename
# Find shape properties, output shape image (optional)
device, shape_header,shape_data,shape_img = pcv.analyze_object(img, args.image, obj, mask, device,args.debug,outfile)
# Shape properties relative to user boundary line (optional)
device, boundary_header,boundary_data, boundary_img1= pcv.analyze_bound(img, args.image,obj, mask, 325, device,args.debug,outfile)
# Determine color properties: Histograms, Color Slices and Pseudocolored Images, output color analyzed images (optional)
device, color_header,color_data,color_img= pcv.analyze_color(img, args.image, mask, 256, device, args.debug,None,'v','img',300,outfile)
# Output shape and color data
result=open(args.result,"a")
result.write('\t'.join(map(str,shape_header)))
result.write("\n")
result.write('\t'.join(map(str,shape_data)))
result.write("\n")
for row in shape_img:
result.write('\t'.join(map(str,row)))
result.write("\n")
result.write('\t'.join(map(str,color_header)))
result.write("\n")
result.write('\t'.join(map(str,color_data)))
result.write("\n")
result.write('\t'.join(map(str,boundary_header)))
result.write("\n")
result.write('\t'.join(map(str,boundary_data)))
result.write("\n")
result.write('\t'.join(map(str,boundary_img1)))
result.write("\n")
for row in color_img:
result.write('\t'.join(map(str,row)))
result.write("\n")
result.close()
def main():
# Get options
args = options()
# Read image
img, path, filename = pcv.readimage(args.image)
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(img, 's', device, args.debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, 'light', device,
args.debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 0, device, args.debug)
device, s_cnt = pcv.median_blur(s_thresh, 0, device, args.debug)
# Fill small objects
#device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(img, 'b', device, args.debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, 'light', device,
args.debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, 'light', device,
args.debug)
# Fill small objects
#device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, args.debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(img, bs, 'white', device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, 'a', device, args.debug)
device, masked_b = pcv.rgb2gray_lab(masked, 'b', device, args.debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, 'dark',
device, args.debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, 'light',
device, args.debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device,
args.debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device,
args.debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 2, device, args.debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, args.debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, args.debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, args.debug)
img2 = np.copy(img)
device, masked2 = pcv.apply_mask(img2, ab_cnt3, 'white', device,
args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, 'a', device, args.debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, 'b', device, args.debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, 'dark',
device, args.debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255,
'light', device, args.debug)
device, ab_fill = pcv.logical_or(masked2a_thresh, masked2b_thresh, device,
args.debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(
masked2, ab_fill, device, args.debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(masked2, 'rectangle', device,
None, 'default', args.debug,
True, 550, 10, -600, -907)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(
img, 'partial', roi1, roi_hierarchy, id_objects, obj_hierarchy, device,
args.debug)
# Object combine kept objects
device, obj, mask = pcv.object_composition(img, roi_objects, hierarchy3,
device, args.debug)
############## Landmarks ################
device, points = pcv.acute_vertex(obj, 40, 40, 40, img, device, args.debug)
boundary_line = 900
# Use acute fxn to estimate tips
device, points_r, centroid_r, bline_r = pcv.scale_features(
obj, mask, points, boundary_line, device, args.debug)
# Get number of points
tips = len(points_r)
# Use turgor_proxy fxn to get distances
device, vert_ave_c, hori_ave_c, euc_ave_c, ang_ave_c, vert_ave_b, hori_ave_b, euc_ave_b, ang_ave_b = pcv.turgor_proxy(
points_r, centroid_r, bline_r, device, args.debug)
# Get pseudomarkers along the y-axis
device, left, right, center_h = pcv.y_axis_pseudolandmarks(
obj, mask, img, device, args.debug)
# Re-scale the points
device, left_r, left_cr, left_br = pcv.scale_features(
obj, mask, left, boundary_line, device, args.debug)
device, right_r, right_cr, right_br = pcv.scale_features(
obj, mask, right, boundary_line, device, args.debug)
device, center_hr, center_hcr, center_hbr = pcv.scale_features(
obj, mask, center_h, boundary_line, device, args.debug)
# Get pseudomarkers along the x-axis
device, top, bottom, center_v = pcv.x_axis_pseudolandmarks(
obj, mask, img, device, args.debug)
# Re-scale the points
device, top_r, top_cr, top_br = pcv.scale_features(obj, mask, top,
boundary_line, device,
args.debug)
device, bottom_r, bottom_cr, bottom_br = pcv.scale_features(
obj, mask, bottom, boundary_line, device, args.debug)
device, center_vr, center_vcr, center_vbr = pcv.scale_features(
obj, mask, center_v, boundary_line, device, args.debug)
## Need to convert the points into a list of tuples format to match the scaled points
points = points.reshape(len(points), 2)
points = points.tolist()
temp_out = []
for p in points:
p = tuple(p)
temp_out.append(p)
points = temp_out
left = left.reshape(20, 2)
left = left.tolist()
temp_out = []
for l in left:
l = tuple(l)
temp_out.append(l)
left = temp_out
right = right.reshape(20, 2)
right = right.tolist()
temp_out = []
for r in right:
r = tuple(r)
temp_out.append(r)
right = temp_out
center_h = center_h.reshape(20, 2)
center_h = center_h.tolist()
temp_out = []
for ch in center_h:
ch = tuple(ch)
temp_out.append(ch)
center_h = temp_out
## Need to convert the points into a list of tuples format to match the scaled points
top = top.reshape(20, 2)
top = top.tolist()
temp_out = []
for t in top:
t = tuple(t)
temp_out.append(t)
top = temp_out
bottom = bottom.reshape(20, 2)
bottom = bottom.tolist()
temp_out = []
for b in bottom:
b = tuple(b)
temp_out.append(b)
bottom = temp_out
center_v = center_v.reshape(20, 2)
center_v = center_v.tolist()
temp_out = []
for cvr in center_v:
cvr = tuple(cvr)
temp_out.append(cvr)
center_v = temp_out
#Store Landmark Data
landmark_header = ('HEADER_LANDMARK', 'tip_points', 'tip_points_r',
'centroid_r', 'baseline_r', 'tip_number', 'vert_ave_c',
'hori_ave_c', 'euc_ave_c', 'ang_ave_c', 'vert_ave_b',
'hori_ave_b', 'euc_ave_b', 'ang_ave_b', 'left_lmk',
'right_lmk', 'center_h_lmk', 'left_lmk_r',
'right_lmk_r', 'center_h_lmk_r', 'top_lmk',
'bottom_lmk', 'center_v_lmk', 'top_lmk_r',
'bottom_lmk_r', 'center_v_lmk_r')
landmark_data = ('LANDMARK_DATA', points, points_r, centroid_r, bline_r,
tips, vert_ave_c, hori_ave_c, euc_ave_c, ang_ave_c,
vert_ave_b, hori_ave_b, euc_ave_b, ang_ave_b, left, right,
center_h, left_r, right_r, center_hr, top, bottom,
center_v, top_r, bottom_r, center_vr)
############## VIS Analysis ################
outfile = False
#if args.writeimg==True:
#outfile=args.outdir+"/"+filename
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(
img, args.image, obj, mask, device, args.debug, outfile)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(
img, args.image, obj, mask, 935, device, args.debug, outfile)
# Determine color properties: Histograms, Color Slices and Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(
img, args.image, mask, 256, device, args.debug, None, 'v', 'img', 300,
outfile)
# Output shape and color data
result = open(args.result, "a")
result.write('\t'.join(map(str, shape_header)))
result.write("\n")
result.write('\t'.join(map(str, shape_data)))
result.write("\n")
for row in shape_img:
result.write('\t'.join(map(str, row)))
result.write("\n")
result.write('\t'.join(map(str, color_header)))
result.write("\n")
result.write('\t'.join(map(str, color_data)))
result.write("\n")
result.write('\t'.join(map(str, boundary_header)))
result.write("\n")
result.write('\t'.join(map(str, boundary_data)))
result.write("\n")
result.write('\t'.join(map(str, boundary_img1)))
result.write("\n")
for row in color_img:
result.write('\t'.join(map(str, row)))
result.write("\n")
result.write('\t'.join(map(str, landmark_header)))
result.write("\n")
result.write('\t'.join(map(str, landmark_data)))
result.write("\n")
result.close()
def main():
# Initialize device
device = 0
# Parse command-line options
args = options()
# Read image
img, path, filename = pcv.readimage(filename=args.image, debug=args.debug)
# Convert RGB to LAB and extract the Blue-Yellow channel
device, blue_channel = pcv.rgb2gray_lab(img=img,
channel="b",
device=device,
debug=args.debug)
# Threshold the blue image using the triangle autothreshold method
device, blue_tri = pcv.triangle_auto_threshold(device=device,
img=blue_channel,
maxvalue=255,
object_type="light",
xstep=1,
debug=args.debug)
# Extract core plant region from the image to preserve delicate plant features during filtering
device += 1
plant_region = blue_tri[0:1750, 600:2080]
if args.debug is not None:
pcv.print_image(filename=str(device) + "_extract_plant_region.png",
img=plant_region)
# Use a Gaussian blur to disrupt the strong edge features in the cabinet
device, blur_gaussian = pcv.gaussian_blur(device=device,
img=blue_tri,
ksize=(3, 3),
sigmax=0,
sigmay=None,
debug=args.debug)
# Threshold the blurred image to remove features that were blurred
device, blur_thresholded = pcv.binary_threshold(img=blur_gaussian,
threshold=250,
maxValue=255,
object_type="light",
device=device,
debug=args.debug)
# Add the plant region back in to the filtered image
device += 1
blur_thresholded[0:1750, 600:2080] = plant_region
if args.debug is not None:
pcv.print_image(filename=str(device) + "_replace_plant_region.png",
img=blur_thresholded)
# Fill small noise
device, blue_fill_50 = pcv.fill(img=np.copy(blur_thresholded),
mask=np.copy(blur_thresholded),
size=50,
device=device,
debug=args.debug)
# Identify objects
device, contours, contour_hierarchy = pcv.find_objects(img=img,
mask=blue_fill_50,
device=device,
debug=args.debug)
# Define ROI
device, roi, roi_hierarchy = pcv.define_roi(img=img,
shape="rectangle",
device=device,
roi=None,
roi_input="default",
debug=args.debug,
adjust=True,
x_adj=565,
y_adj=0,
w_adj=-490,
h_adj=-250)
# Decide which objects to keep
device, roi_contours, roi_contour_hierarchy, _, _ = pcv.roi_objects(
img=img,
roi_type="partial",
roi_contour=roi,
roi_hierarchy=roi_hierarchy,
object_contour=contours,
obj_hierarchy=contour_hierarchy,
device=device,
debug=args.debug)
# If there are no contours left we cannot measure anything
if len(roi_contours) > 0:
# Object combine kept objects
device, plant_contour, plant_mask = pcv.object_composition(
img=img,
contours=roi_contours,
hierarchy=roi_contour_hierarchy,
device=device,
debug=args.debug)
outfile = False
if args.writeimg:
outfile = args.outdir + "/" + filename
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(
img=img,
imgname=args.image,
obj=plant_contour,
mask=plant_mask,
device=device,
debug=args.debug,
filename=outfile)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img = pcv.analyze_bound(
img=img,
imgname=args.image,
obj=plant_contour,
mask=plant_mask,
line_position=440,
device=device,
debug=args.debug,
filename=outfile)
# Determine color properties: Histograms, Color Slices and Pseudocolored Images,
# output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(
img=img,
imgname=args.image,
mask=plant_mask,
bins=256,
device=device,
debug=args.debug,
hist_plot_type=None,
pseudo_channel="v",
pseudo_bkg="img",
resolution=300,
filename=outfile)
# Output shape and color data
result = open(args.result, "a")
result.write('\t'.join(map(str, shape_header)) + "\n")
result.write('\t'.join(map(str, shape_data)) + "\n")
for row in shape_img:
result.write('\t'.join(map(str, row)) + "\n")
result.write('\t'.join(map(str, color_header)) + "\n")
result.write('\t'.join(map(str, color_data)) + "\n")
result.write('\t'.join(map(str, boundary_header)) + "\n")
result.write('\t'.join(map(str, boundary_data)) + "\n")
result.write('\t'.join(map(str, boundary_img)) + "\n")
for row in color_img:
result.write('\t'.join(map(str, row)) + "\n")
result.close()
# Find matching NIR image
device, nirpath = pcv.get_nir(path=path,
filename=filename,
device=device,
debug=args.debug)
nir_rgb, nir_path, nir_filename = pcv.readimage(nirpath)
nir_img = cv2.imread(nirpath, 0)
# Make mask glovelike in proportions via dilation
device, d_mask = pcv.dilate(plant_mask,
kernel=1,
i=0,
device=device,
debug=args.debug)
# Resize mask
prop2, prop1 = conv_ratio()
device, nmask = pcv.resize(img=d_mask,
resize_x=prop1,
resize_y=prop2,
device=device,
debug=args.debug)
# Convert the resized mask to a binary mask
device, bmask = pcv.binary_threshold(img=nmask,
threshold=0,
maxValue=255,
object_type="light",
device=device,
debug=args.debug)
device, crop_img = crop_sides_equally(mask=bmask,
nir=nir_img,
device=device,
debug=args.debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(img=nir_img,
mask=crop_img,
device=device,
x=34,
y=9,
v_pos="top",
h_pos="right",
debug=args.debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(img=nir_rgb,
mask=newmask,
device=device,
debug=args.debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(
img=nir_rgb,
contours=nir_objects,
hierarchy=nir_hierarchy,
device=device,
debug=args.debug)
if args.writeimg:
outfile = args.outdir + "/" + nir_filename
# Analyze NIR signal data
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(
img=nir_img,
rgbimg=nir_rgb,
mask=nir_combinedmask,
bins=256,
device=device,
histplot=False,
debug=args.debug,
filename=outfile)
# Analyze the shape of the plant contour from the NIR image
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(
img=nir_img,
imgname=nir_filename,
obj=nir_combined,
mask=nir_combinedmask,
device=device,
debug=args.debug,
filename=outfile)
# Write NIR data to co-results file
coresult = open(args.coresult, "a")
coresult.write('\t'.join(map(str, nhist_header)) + "\n")
coresult.write('\t'.join(map(str, nhist_data)) + "\n")
for row in nir_imgs:
coresult.write('\t'.join(map(str, row)) + "\n")
coresult.write('\t'.join(map(str, nshape_header)) + "\n")
coresult.write('\t'.join(map(str, nshape_data)) + "\n")
coresult.write('\t'.join(map(str, nir_shape)) + "\n")
coresult.close()
def main():
# Initialize device
device = 0
# Parse command-line options
args = options()
# Read image
img, path, filename = pcv.readimage(filename=args.image, debug=args.debug)
# Convert RGB to LAB and extract the Green-Magenta channel
device, green_channel = pcv.rgb2gray_lab(img=img,
channel="a",
device=device,
debug=args.debug)
# Invert the Green-Magenta image because the plant is dark green
device, green_inv = pcv.invert(img=green_channel,
device=device,
debug=args.debug)
# Threshold the inverted Green-Magenta image to mostly isolate green pixels
device, green_thresh = pcv.binary_threshold(img=green_inv,
threshold=134,
maxValue=255,
object_type="light",
device=device,
debug=args.debug)
# Extract core plant region from the image to preserve delicate plant features during filtering
device += 1
plant_region = green_thresh[100:2000, 250:2250]
if args.debug is not None:
pcv.print_image(filename=str(device) + "_extract_plant_region.png",
img=plant_region)
# Use a Gaussian blur to disrupt the strong edge features in the cabinet
device, blur_gaussian = pcv.gaussian_blur(device=device,
img=green_thresh,
ksize=(7, 7),
sigmax=0,
sigmay=None,
debug=args.debug)
# Threshold the blurred image to remove features that were blurred
device, blur_thresholded = pcv.binary_threshold(img=blur_gaussian,
threshold=250,
maxValue=255,
object_type="light",
device=device,
debug=args.debug)
# Add the plant region back in to the filtered image
device += 1
blur_thresholded[100:2000, 250:2250] = plant_region
if args.debug is not None:
pcv.print_image(filename=str(device) + "_replace_plant_region.png",
img=blur_thresholded)
# Use a median blur to breakup the horizontal and vertical lines caused by shadows from the track edges
device, med_blur = pcv.median_blur(img=blur_thresholded,
ksize=5,
device=device,
debug=args.debug)
# Fill in small contours
device, green_fill_50 = pcv.fill(img=np.copy(med_blur),
mask=np.copy(med_blur),
size=50,
device=device,
debug=args.debug)
# Define an ROI for the brass stopper
device, stopper_roi, stopper_hierarchy = pcv.define_roi(
img=img,
shape="rectangle",
device=device,
roi=None,
roi_input="default",
debug=args.debug,
adjust=True,
x_adj=1420,
y_adj=890,
w_adj=-920,
h_adj=-1040)
# Identify all remaining contours in the binary image
device, contours, hierarchy = pcv.find_objects(img=img,
mask=np.copy(green_fill_50),
device=device,
debug=args.debug)
# Remove contours completely contained within the stopper region of interest
device, remove_stopper_mask = remove_countors_roi(mask=green_fill_50,
contours=contours,
hierarchy=hierarchy,
roi=stopper_roi,
device=device,
debug=args.debug)
# Define an ROI for a screw hole
device, screw_roi, screw_hierarchy = pcv.define_roi(img=img,
shape="rectangle",
device=device,
roi=None,
roi_input="default",
debug=args.debug,
adjust=True,
x_adj=1870,
y_adj=1010,
w_adj=-485,
h_adj=-960)
# Remove contours completely contained within the screw region of interest
device, remove_screw_mask = remove_countors_roi(mask=remove_stopper_mask,
contours=contours,
hierarchy=hierarchy,
roi=screw_roi,
device=device,
debug=args.debug)
# Identify objects
device, contours, contour_hierarchy = pcv.find_objects(
img=img, mask=remove_screw_mask, device=device, debug=args.debug)
# Define ROI
device, roi, roi_hierarchy = pcv.define_roi(img=img,
shape="rectangle",
device=device,
roi=None,
roi_input="default",
debug=args.debug,
adjust=True,
x_adj=565,
y_adj=200,
w_adj=-490,
h_adj=-250)
# Decide which objects to keep
device, roi_contours, roi_contour_hierarchy, _, _ = pcv.roi_objects(
img=img,
roi_type="partial",
roi_contour=roi,
roi_hierarchy=roi_hierarchy,
object_contour=contours,
obj_hierarchy=contour_hierarchy,
device=device,
debug=args.debug)
# If there are no contours left we cannot measure anything
if len(roi_contours) > 0:
# Object combine kept objects
device, plant_contour, plant_mask = pcv.object_composition(
img=img,
contours=roi_contours,
hierarchy=roi_contour_hierarchy,
device=device,
debug=args.debug)
outfile = False
if args.writeimg:
outfile = args.outdir + "/" + filename
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(
img=img,
imgname=args.image,
obj=plant_contour,
mask=plant_mask,
device=device,
debug=args.debug,
filename=outfile)
# Determine color properties: Histograms, Color Slices and Pseudocolored Images,
# output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(
img=img,
imgname=args.image,
mask=plant_mask,
bins=256,
device=device,
debug=args.debug,
hist_plot_type=None,
pseudo_channel="v",
pseudo_bkg="img",
resolution=300,
filename=outfile)
# Output shape and color data
result = open(args.result, "a")
result.write('\t'.join(map(str, shape_header)) + "\n")
result.write('\t'.join(map(str, shape_data)) + "\n")
for row in shape_img:
result.write('\t'.join(map(str, row)) + "\n")
result.write('\t'.join(map(str, color_header)) + "\n")
result.write('\t'.join(map(str, color_data)) + "\n")
for row in color_img:
result.write('\t'.join(map(str, row)) + "\n")
result.close()
# Find matching NIR image
device, nirpath = pcv.get_nir(path=path,
filename=filename,
device=device,
debug=args.debug)
nir_rgb, nir_path, nir_filename = pcv.readimage(nirpath)
nir_img = cv2.imread(nirpath, 0)
# Make mask glovelike in proportions via dilation
device, d_mask = pcv.dilate(plant_mask,
kernel=1,
i=0,
device=device,
debug=args.debug)
# Resize mask
prop2, prop1 = conv_ratio()
device, nmask = pcv.resize(img=d_mask,
resize_x=prop1,
resize_y=prop2,
device=device,
debug=args.debug)
# Convert the resized mask to a binary mask
device, bmask = pcv.binary_threshold(img=nmask,
threshold=0,
maxValue=255,
object_type="light",
device=device,
debug=args.debug)
device, crop_img = crop_sides_equally(mask=bmask,
nir=nir_img,
device=device,
debug=args.debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(img=nir_img,
mask=crop_img,
device=device,
x=0,
y=1,
v_pos="bottom",
h_pos="right",
debug=args.debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(img=nir_rgb,
mask=newmask,
device=device,
debug=args.debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(
img=nir_rgb,
contours=nir_objects,
hierarchy=nir_hierarchy,
device=device,
debug=args.debug)
if args.writeimg:
outfile = args.outdir + "/" + nir_filename
# Analyze NIR signal data
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(
img=nir_img,
rgbimg=nir_rgb,
mask=nir_combinedmask,
bins=256,
device=device,
histplot=False,
debug=args.debug,
filename=outfile)
# Analyze the shape of the plant contour from the NIR image
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(
img=nir_img,
imgname=nir_filename,
obj=nir_combined,
mask=nir_combinedmask,
device=device,
debug=args.debug,
filename=outfile)
# Write NIR data to co-results file
coresult = open(args.coresult, "a")
coresult.write('\t'.join(map(str, nhist_header)) + "\n")
coresult.write('\t'.join(map(str, nhist_data)) + "\n")
for row in nir_imgs:
coresult.write('\t'.join(map(str, row)) + "\n")
coresult.write('\t'.join(map(str, nshape_header)) + "\n")
coresult.write('\t'.join(map(str, nshape_data)) + "\n")
coresult.write('\t'.join(map(str, nir_shape)) + "\n")
coresult.close()
def process_sv_images_core(vis_id, vis_img, nir_id, nir_rgb, nir_cv2, traits, debug=None):
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(vis_img, 's', device, debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, 'light', device, debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 5, device, debug)
device, s_cnt = pcv.median_blur(s_thresh, 5, device, debug)
# Fill small objects
# device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(vis_img, 'b', device, debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
# Fill small objects
# device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(vis_img, bs, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, 'a', device, debug)
device, masked_b = pcv.rgb2gray_lab(masked, 'b', device, debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, 'dark', device, debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, 'light', device, debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 200, device, debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, 'a', device, debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, 'b', device, debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, 'dark', device, debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255, 'light', device, debug)
device, masked2a_thresh_blur = pcv.median_blur(masked2a_thresh, 5, device, debug)
device, masked2b_thresh_blur = pcv.median_blur(masked2b_thresh, 13, device, debug)
device, ab_fill = pcv.logical_or(masked2a_thresh_blur, masked2b_thresh_blur, device, debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(masked2, ab_fill, device, debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(masked2, 'rectangle', device, None, 'default', debug, True, 700,
0, -600, -300)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(vis_img, 'partial', roi1, roi_hierarchy,
id_objects, obj_hierarchy, device,
debug)
# Object combine kept objects
device, obj, mask = pcv.object_composition(vis_img, roi_objects, hierarchy3, device, debug)
############## VIS Analysis ################
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(vis_img, vis_id, obj, mask, device, debug)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(vis_img, vis_id, obj, mask, 384, device,
debug)
# Determine color properties: Histograms, Color Slices and
# Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(vis_img, vis_id, mask, 256, device, debug,
None, 'v', 'img', 300)
# Output shape and color data
vis_traits = {}
for i in range(1, len(shape_header)):
vis_traits[shape_header[i]] = shape_data[i]
for i in range(1, len(boundary_header)):
vis_traits[boundary_header[i]] = boundary_data[i]
for i in range(2, len(color_header)):
vis_traits[color_header[i]] = serialize_color_data(color_data[i])
############################# Use VIS image mask for NIR image#########################
# Flip mask
device, f_mask = pcv.flip(mask, "vertical", device, debug)
# Reize mask
device, nmask = pcv.resize(f_mask, 0.1154905775, 0.1154905775, device, debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(nir_rgb, nmask, device, 30, 4, "top", "right", debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(nir_rgb, newmask, device, debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(nir_rgb, nir_objects, nir_hierarchy, device, debug)
####################################### Analysis #############################################
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(nir_cv2, nir_id, nir_combinedmask, 256,
device, False, debug)
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(nir_cv2, nir_id, nir_combined, nir_combinedmask,
device, debug)
nir_traits = {}
for i in range(1, len(nshape_header)):
nir_traits[nshape_header[i]] = nshape_data[i]
for i in range(2, len(nhist_header)):
nir_traits[nhist_header[i]] = serialize_color_data(nhist_data[i])
# Add data to traits table
traits['sv_area'].append(vis_traits['area'])
traits['hull_area'].append(vis_traits['hull-area'])
traits['solidity'].append(vis_traits['solidity'])
traits['height'].append(vis_traits['height_above_bound'])
traits['perimeter'].append(vis_traits['perimeter'])
return [vis_traits, nir_traits]
def main():
# Get options
args = options()
# Read image
img, path, filename = pcv.readimage(args.image)
# roi = cv2.imread(args.roi)
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(img, "s", device, args.debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, "light", device, args.debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 0, device, args.debug)
device, s_cnt = pcv.median_blur(s_thresh, 0, device, args.debug)
# Fill small objects
# device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(img, "b", device, args.debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, "light", device, args.debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, "light", device, args.debug)
# Fill small objects
# device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, args.debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(img, bs, "white", device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, "a", device, args.debug)
device, masked_b = pcv.rgb2gray_lab(masked, "b", device, args.debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, "dark", device, args.debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, "light", device, args.debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, args.debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, args.debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 2, device, args.debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, args.debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, args.debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, args.debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, "white", device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, "a", device, args.debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, "b", device, args.debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, "dark", device, args.debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255, "light", device, args.debug)
device, ab_fill = pcv.logical_or(masked2a_thresh, masked2b_thresh, device, args.debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(masked2, ab_fill, device, args.debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(
masked2, "rectangle", device, None, "default", args.debug, True, 550, 0, -600, -925
)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(
img, "partial", roi1, roi_hierarchy, id_objects, obj_hierarchy, device, args.debug
)
# Object combine kept objects
device, obj, mask = pcv.object_composition(img, roi_objects, hierarchy3, device, args.debug)
############### Analysis ################
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(
img, args.image, obj, mask, device, args.debug, args.outdir + "/" + filename
)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(
img, args.image, obj, mask, 900, device, args.debug, args.outdir + "/" + filename
)
# Determine color properties: Histograms, Color Slices and Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(
img, args.image, kept_mask, 256, device, args.debug, None, "v", "img", 300, args.outdir + "/" + filename
)
# Output shape and color data
pcv.print_results(args.image, shape_header, shape_data)
pcv.print_results(args.image, color_header, color_data)
pcv.print_results(args.image, boundary_header, boundary_data)
def process_sv_images(session, url, vis_id, nir_id, traits, debug=None):
"""Process side-view images from Clowder.
Inputs:
session = requests session object
url = Clowder URL
vis_id = The Clowder ID of an RGB image
nir_img = The Clowder ID of an NIR grayscale image
traits = traits table (dictionary)
debug = None, print, or plot. Print = save to file, Plot = print to screen
:param session: requests session object
:param url: str
:param vis_id: str
:param nir_id: str
:param traits: dict
:param debug: str
:return traits: dict
"""
# Read VIS image from Clowder
vis_r = session.get(posixpath.join(url, "api/files", vis_id), stream=True)
img_array = np.asarray(bytearray(vis_r.content), dtype="uint8")
img = cv2.imdecode(img_array, -1)
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(img, 's', device, debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, 'light', device, debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 5, device, debug)
device, s_cnt = pcv.median_blur(s_thresh, 5, device, debug)
# Fill small objects
# device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(img, 'b', device, debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
# Fill small objects
# device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(img, bs, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, 'a', device, debug)
device, masked_b = pcv.rgb2gray_lab(masked, 'b', device, debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, 'dark', device, debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, 'light', device, debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 200, device, debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, 'a', device, debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, 'b', device, debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, 'dark', device, debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255, 'light', device, debug)
device, masked2a_thresh_blur = pcv.median_blur(masked2a_thresh, 5, device, debug)
device, masked2b_thresh_blur = pcv.median_blur(masked2b_thresh, 13, device, debug)
device, ab_fill = pcv.logical_or(masked2a_thresh_blur, masked2b_thresh_blur, device, debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(masked2, ab_fill, device, debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(masked2, 'rectangle', device, None, 'default', debug, True, 700,
0, -600, -300)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(img, 'partial', roi1, roi_hierarchy,
id_objects, obj_hierarchy, device,
debug)
# Object combine kept objects
device, obj, mask = pcv.object_composition(img, roi_objects, hierarchy3, device, debug)
############## VIS Analysis ################
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(img, vis_id, obj, mask, device, debug)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(img, vis_id, obj, mask, 384, device,
debug)
# Determine color properties: Histograms, Color Slices and
# Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(img, vis_id, mask, 256, device, debug,
None, 'v', 'img', 300)
# Output shape and color data
vis_traits = {}
for i in range(1, len(shape_header)):
vis_traits[shape_header[i]] = shape_data[i]
for i in range(1, len(boundary_header)):
vis_traits[boundary_header[i]] = boundary_data[i]
for i in range(2, len(color_header)):
vis_traits[color_header[i]] = serialize_color_data(color_data[i])
#print(vis_traits)
add_plantcv_metadata(session, url, vis_id, vis_traits)
############################# Use VIS image mask for NIR image#########################
# Read NIR image from Clowder
nir_r = session.get(posixpath.join(url, "api/files", nir_id), stream=True)
nir_array = np.asarray(bytearray(nir_r.content), dtype="uint8")
nir = cv2.imdecode(nir_array, -1)
nir_rgb = cv2.cvtColor(nir, cv2.COLOR_GRAY2BGR)
# Flip mask
device, f_mask = pcv.flip(mask, "vertical", device, debug)
# Reize mask
device, nmask = pcv.resize(f_mask, 0.1154905775, 0.1154905775, device, debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(nir_rgb, nmask, device, 30, 4, "top", "right", debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(nir_rgb, newmask, device, debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(nir_rgb, nir_objects, nir_hierarchy, device, debug)
####################################### Analysis #############################################
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(nir, nir_id, nir_combinedmask, 256,
device, False, debug)
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(nir, nir_id, nir_combined, nir_combinedmask,
device, debug)
nir_traits = {}
for i in range(1, len(nshape_header)):
nir_traits[nshape_header[i]] = nshape_data[i]
for i in range(2, len(nhist_header)):
nir_traits[nhist_header[i]] = serialize_color_data(nhist_data[i])
#print(nir_traits)
add_plantcv_metadata(session, url, nir_id, nir_traits)
# Add data to traits table
traits['sv_area'].append(vis_traits['area'])
traits['hull_area'].append(vis_traits['hull-area'])
traits['solidity'].append(vis_traits['solidity'])
traits['height'].append(vis_traits['height_above_bound'])
traits['perimeter'].append(vis_traits['perimeter'])
return traits
def process_sv_images(vis_img, nir_img, debug=None):
"""Process side-view images.
Inputs:
vis_img = An RGB image.
nir_img = An NIR grayscale image.
debug = None, print, or plot. Print = save to file, Plot = print to screen.
:param vis_img: str
:param nir_img: str
:param debug: str
:return:
"""
# Read VIS image
img, path, filename = pcv.readimage(vis_img)
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(img, 's', device, debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, 'light', device, debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 5, device, debug)
device, s_cnt = pcv.median_blur(s_thresh, 5, device, debug)
# Fill small objects
# device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(img, 'b', device, debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
# Fill small objects
# device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(img, bs, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, 'a', device, debug)
device, masked_b = pcv.rgb2gray_lab(masked, 'b', device, debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, 'dark', device, debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, 'light', device, debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 200, device, debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, 'a', device, debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, 'b', device, debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, 'dark', device, debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255, 'light', device, debug)
device, masked2a_thresh_blur = pcv.median_blur(masked2a_thresh, 5, device, debug)
device, masked2b_thresh_blur = pcv.median_blur(masked2b_thresh, 13, device, debug)
device, ab_fill = pcv.logical_or(masked2a_thresh_blur, masked2b_thresh_blur, device, debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(masked2, ab_fill, device, debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(masked2, 'rectangle', device, None, 'default', debug, True, 700,
0, -600, -300)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(img, 'partial', roi1, roi_hierarchy,
id_objects, obj_hierarchy, device,
debug)
# Object combine kept objects
device, obj, mask = pcv.object_composition(img, roi_objects, hierarchy3, device, debug)
############## VIS Analysis ################
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(img, vis_img, obj, mask, device, debug)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(img, vis_img, obj, mask, 384, device,
debug)
# Determine color properties: Histograms, Color Slices and
# Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(img, vis_img, mask, 256, device, debug,
None, 'v', 'img', 300)
# Output shape and color data
print('\t'.join(map(str, shape_header)) + '\n')
print('\t'.join(map(str, shape_data)) + '\n')
for row in shape_img:
print('\t'.join(map(str, row)) + '\n')
print('\t'.join(map(str, color_header)) + '\n')
print('\t'.join(map(str, color_data)) + '\n')
print('\t'.join(map(str, boundary_header)) + '\n')
print('\t'.join(map(str, boundary_data)) + '\n')
print('\t'.join(map(str, boundary_img1)) + '\n')
for row in color_img:
print('\t'.join(map(str, row)) + '\n')
############################# Use VIS image mask for NIR image#########################
# Read NIR image
nir, path1, filename1 = pcv.readimage(nir_img)
nir2 = cv2.imread(nir_img, -1)
# Flip mask
device, f_mask = pcv.flip(mask, "vertical", device, debug)
# Reize mask
device, nmask = pcv.resize(f_mask, 0.1154905775, 0.1154905775, device, debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(nir, nmask, device, 30, 4, "top", "right", debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(nir, newmask, device, debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(nir, nir_objects, nir_hierarchy, device, debug)
####################################### Analysis #############################################
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(nir2, filename1, nir_combinedmask, 256,
device, False, debug)
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(nir2, filename1, nir_combined, nir_combinedmask,
device, debug)
print('\t'.join(map(str, nhist_header)) + '\n')
print('\t'.join(map(str, nhist_data)) + '\n')
for row in nir_imgs:
print('\t'.join(map(str, row)) + '\n')
print('\t'.join(map(str, nshape_header)) + '\n')
print('\t'.join(map(str, nshape_data)) + '\n')
print('\t'.join(map(str, nir_shape)) + '\n')
def process_sv_images_core(vis_id,
vis_img,
nir_id,
nir_rgb,
nir_cv2,
traits,
debug=None):
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(vis_img, 's', device, debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, 'light', device, debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 5, device, debug)
device, s_cnt = pcv.median_blur(s_thresh, 5, device, debug)
# Fill small objects
# device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(vis_img, 'b', device, debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, 'light', device,
debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, 'light', device, debug)
# Fill small objects
# device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(vis_img, bs, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, 'a', device, debug)
device, masked_b = pcv.rgb2gray_lab(masked, 'b', device, debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, 'dark',
device, debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, 'light',
device, debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device,
debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 200, device, debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, 'white', device, debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, 'a', device, debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, 'b', device, debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, 'dark',
device, debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255,
'light', device, debug)
device, masked2a_thresh_blur = pcv.median_blur(masked2a_thresh, 5, device,
debug)
device, masked2b_thresh_blur = pcv.median_blur(masked2b_thresh, 13, device,
debug)
device, ab_fill = pcv.logical_or(masked2a_thresh_blur,
masked2b_thresh_blur, device, debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(
masked2, ab_fill, device, debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(masked2, 'rectangle', device,
None, 'default', debug, True,
700, 0, -600, -300)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(
vis_img, 'partial', roi1, roi_hierarchy, id_objects, obj_hierarchy,
device, debug)
# Object combine kept objects
device, obj, mask = pcv.object_composition(vis_img, roi_objects,
hierarchy3, device, debug)
############## VIS Analysis ################
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(
vis_img, vis_id, obj, mask, device, debug)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(
vis_img, vis_id, obj, mask, 384, device, debug)
# Determine color properties: Histograms, Color Slices and
# Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(
vis_img, vis_id, mask, 256, device, debug, None, 'v', 'img', 300)
# Output shape and color data
vis_traits = {}
for i in range(1, len(shape_header)):
vis_traits[shape_header[i]] = shape_data[i]
for i in range(1, len(boundary_header)):
vis_traits[boundary_header[i]] = boundary_data[i]
for i in range(2, len(color_header)):
vis_traits[color_header[i]] = serialize_color_data(color_data[i])
############################# Use VIS image mask for NIR image#########################
# Flip mask
device, f_mask = pcv.flip(mask, "vertical", device, debug)
# Reize mask
device, nmask = pcv.resize(f_mask, 0.1154905775, 0.1154905775, device,
debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(nir_rgb, nmask, device, 30, 4,
"top", "right", debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(
nir_rgb, newmask, device, debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(
nir_rgb, nir_objects, nir_hierarchy, device, debug)
####################################### Analysis #############################################
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(
nir_cv2, nir_id, nir_combinedmask, 256, device, False, debug)
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(
nir_cv2, nir_id, nir_combined, nir_combinedmask, device, debug)
nir_traits = {}
for i in range(1, len(nshape_header)):
nir_traits[nshape_header[i]] = nshape_data[i]
for i in range(2, len(nhist_header)):
nir_traits[nhist_header[i]] = serialize_color_data(nhist_data[i])
# Add data to traits table
traits['sv_area'].append(vis_traits['area'])
traits['hull_area'].append(vis_traits['hull-area'])
traits['solidity'].append(vis_traits['solidity'])
traits['height'].append(vis_traits['height_above_bound'])
traits['perimeter'].append(vis_traits['perimeter'])
return [vis_traits, nir_traits]
def main():
# Get options
args = options()
# Read image
img, path, filename = pcv.readimage(args.image)
# Pipeline step
device = 0
# Convert RGB to HSV and extract the Saturation channel
device, s = pcv.rgb2gray_hsv(img, "s", device, args.debug)
# Threshold the Saturation image
device, s_thresh = pcv.binary_threshold(s, 36, 255, "light", device, args.debug)
# Median Filter
device, s_mblur = pcv.median_blur(s_thresh, 0, device, args.debug)
device, s_cnt = pcv.median_blur(s_thresh, 0, device, args.debug)
# Fill small objects
# device, s_fill = pcv.fill(s_mblur, s_cnt, 0, device, args.debug)
# Convert RGB to LAB and extract the Blue channel
device, b = pcv.rgb2gray_lab(img, "b", device, args.debug)
# Threshold the blue image
device, b_thresh = pcv.binary_threshold(b, 137, 255, "light", device, args.debug)
device, b_cnt = pcv.binary_threshold(b, 137, 255, "light", device, args.debug)
# Fill small objects
# device, b_fill = pcv.fill(b_thresh, b_cnt, 10, device, args.debug)
# Join the thresholded saturation and blue-yellow images
device, bs = pcv.logical_and(s_mblur, b_cnt, device, args.debug)
# Apply Mask (for vis images, mask_color=white)
device, masked = pcv.apply_mask(img, bs, "white", device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked_a = pcv.rgb2gray_lab(masked, "a", device, args.debug)
device, masked_b = pcv.rgb2gray_lab(masked, "b", device, args.debug)
# Threshold the green-magenta and blue images
device, maskeda_thresh = pcv.binary_threshold(masked_a, 127, 255, "dark", device, args.debug)
device, maskedb_thresh = pcv.binary_threshold(masked_b, 128, 255, "light", device, args.debug)
# Join the thresholded saturation and blue-yellow images (OR)
device, ab = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, args.debug)
device, ab_cnt = pcv.logical_or(maskeda_thresh, maskedb_thresh, device, args.debug)
# Fill small noise
device, ab_fill1 = pcv.fill(ab, ab_cnt, 2, device, args.debug)
# Dilate to join small objects with larger ones
device, ab_cnt1 = pcv.dilate(ab_fill1, 3, 2, device, args.debug)
device, ab_cnt2 = pcv.dilate(ab_fill1, 3, 2, device, args.debug)
# Fill dilated image mask
device, ab_cnt3 = pcv.fill(ab_cnt2, ab_cnt1, 150, device, args.debug)
device, masked2 = pcv.apply_mask(masked, ab_cnt3, "white", device, args.debug)
# Convert RGB to LAB and extract the Green-Magenta and Blue-Yellow channels
device, masked2_a = pcv.rgb2gray_lab(masked2, "a", device, args.debug)
device, masked2_b = pcv.rgb2gray_lab(masked2, "b", device, args.debug)
# Threshold the green-magenta and blue images
device, masked2a_thresh = pcv.binary_threshold(masked2_a, 127, 255, "dark", device, args.debug)
device, masked2b_thresh = pcv.binary_threshold(masked2_b, 128, 255, "light", device, args.debug)
device, ab_fill = pcv.logical_or(masked2a_thresh, masked2b_thresh, device, args.debug)
# Identify objects
device, id_objects, obj_hierarchy = pcv.find_objects(masked2, ab_fill, device, args.debug)
# Define ROI
device, roi1, roi_hierarchy = pcv.define_roi(
masked2, "rectangle", device, None, "default", args.debug, True, 500, 0, -600, -885
)
# Decide which objects to keep
device, roi_objects, hierarchy3, kept_mask, obj_area = pcv.roi_objects(
img, "partial", roi1, roi_hierarchy, id_objects, obj_hierarchy, device, args.debug
)
# Object combine kept objects
device, obj, mask = pcv.object_composition(img, roi_objects, hierarchy3, device, args.debug)
############## VIS Analysis ################
outfile = False
if args.writeimg == True:
outfile = args.outdir + "/" + filename
# Find shape properties, output shape image (optional)
device, shape_header, shape_data, shape_img = pcv.analyze_object(
img, args.image, obj, mask, device, args.debug, outfile
)
# Shape properties relative to user boundary line (optional)
device, boundary_header, boundary_data, boundary_img1 = pcv.analyze_bound(
img, args.image, obj, mask, 845, device, args.debug, outfile
)
# Determine color properties: Histograms, Color Slices and Pseudocolored Images, output color analyzed images (optional)
device, color_header, color_data, color_img = pcv.analyze_color(
img, args.image, mask, 256, device, args.debug, None, "v", "img", 300, outfile
)
# Output shape and color data
result = open(args.result, "a")
result.write("\t".join(map(str, shape_header)))
result.write("\n")
result.write("\t".join(map(str, shape_data)))
result.write("\n")
for row in shape_img:
result.write("\t".join(map(str, row)))
result.write("\n")
result.write("\t".join(map(str, color_header)))
result.write("\n")
result.write("\t".join(map(str, color_data)))
result.write("\n")
result.write("\t".join(map(str, boundary_header)))
result.write("\n")
result.write("\t".join(map(str, boundary_data)))
result.write("\n")
result.write("\t".join(map(str, boundary_img1)))
result.write("\n")
for row in color_img:
result.write("\t".join(map(str, row)))
result.write("\n")
result.close()
############################# Use VIS image mask for NIR image#########################
# Find matching NIR image
device, nirpath = pcv.get_nir(path, filename, device, args.debug)
nir, path1, filename1 = pcv.readimage(nirpath)
nir2 = cv2.imread(nirpath, -1)
# Flip mask
device, f_mask = pcv.flip(mask, "vertical", device, args.debug)
# Reize mask
device, nmask = pcv.resize(f_mask, 0.1304, 0.1304, device, args.debug)
# position, and crop mask
device, newmask = pcv.crop_position_mask(nir, nmask, device, 65, 0, "top", "left", args.debug)
# Identify objects
device, nir_objects, nir_hierarchy = pcv.find_objects(nir, newmask, device, args.debug)
# Object combine kept objects
device, nir_combined, nir_combinedmask = pcv.object_composition(nir, nir_objects, nir_hierarchy, device, args.debug)
####################################### Analysis #############################################
outfile1 = False
if args.writeimg == True:
outfile1 = args.outdir + "/" + filename1
device, nhist_header, nhist_data, nir_imgs = pcv.analyze_NIR_intensity(
nir2, filename1, nir_combinedmask, 256, device, False, args.debug, outfile1
)
device, nshape_header, nshape_data, nir_shape = pcv.analyze_object(
nir2, filename1, nir_combined, nir_combinedmask, device, args.debug, outfile1
)
coresult = open(args.coresult, "a")
coresult.write("\t".join(map(str, nhist_header)))
coresult.write("\n")
coresult.write("\t".join(map(str, nhist_data)))
coresult.write("\n")
for row in nir_imgs:
coresult.write("\t".join(map(str, row)))
coresult.write("\n")
coresult.write("\t".join(map(str, nshape_header)))
coresult.write("\n")
coresult.write("\t".join(map(str, nshape_data)))
coresult.write("\n")
coresult.write("\t".join(map(str, nir_shape)))
coresult.write("\n")
coresult.close()
def main():
# Get options
args = options()
if args.debug:
print("Analyzing your image dude...")
# Read image
img = cv2.imread(args.image, flags=0)
# if a region of interest is specified read it in
roi = cv2.imread(args.roi)
# Pipeline step
device = 0
# Start by examining the distribution of pixel intensity values
if args.debug:
pcv.plot_hist(img, 'hist_img')
# Will intensity transformation enhance your ability to isolate object of interest by thesholding?
device, he_img = pcv.HistEqualization(img, device, args.debug)
if args.debug:
pcv.plot_hist(he_img, 'hist_img_he')
# Laplace filtering (identify edges based on 2nd derivative)
device, lp_img = pcv.laplace_filter(img, 1, 1, device, args.debug)
if args.debug:
pcv.plot_hist(lp_img, 'hist_lp')
# Lapacian image sharpening, this step will enhance the darkness of the edges detected
device, lp_shrp_img = pcv.image_subtract(img, lp_img, device, args.debug)
if args.debug:
pcv.plot_hist(lp_shrp_img, 'hist_lp_shrp')
# Sobel filtering
# 1st derivative sobel filtering along horizontal axis, kernel = 1, unscaled)
device, sbx_img = pcv.sobel_filter(img, 1, 0, 1, 1, device, args.debug)
if args.debug:
pcv.plot_hist(sbx_img, 'hist_sbx')
# 1st derivative sobel filtering along vertical axis, kernel = 1, unscaled)
device, sby_img = pcv.sobel_filter(img, 0, 1, 1, 1, device, args.debug)
if args.debug:
pcv.plot_hist(sby_img, 'hist_sby')
# Combine the effects of both x and y filters through matrix addition
# This will capture edges identified within each plane and emphesize edges found in both images
device, sb_img = pcv.image_add(sbx_img, sby_img, device, args.debug)
if args.debug:
pcv.plot_hist(sb_img, 'hist_sb_comb_img')
# Use a lowpass (blurring) filter to smooth sobel image
device, mblur_img = pcv.median_blur(sb_img, 1, device, args.debug)
device, mblur_invert_img = pcv.invert(mblur_img, device, args.debug)
# combine the smoothed sobel image with the laplacian sharpened image
# combines the best features of both methods as described in "Digital Image Processing" by Gonzalez and Woods pg. 169
device, edge_shrp_img = pcv.image_add(mblur_invert_img, lp_shrp_img, device, args.debug)
if args.debug:
pcv.plot_hist(edge_shrp_img, 'hist_edge_shrp_img')
# Perform thresholding to generate a binary image
device, tr_es_img = pcv.binary_threshold(edge_shrp_img, 145, 255, 'dark', device, args.debug)
# Prepare a few small kernels for morphological filtering
kern = np.zeros((3,3), dtype=np.uint8)
kern1 = np.copy(kern)
kern1[1,1:3]=1
kern2 = np.copy(kern)
kern2[1,0:2]=1
kern3 = np.copy(kern)
kern3[0:2,1]=1
kern4 = np.copy(kern)
kern4[1:3,1]=1
# Prepare a larger kernel for dilation
kern[1,0:3]=1
kern[0:3,1]=1
# Perform erosion with 4 small kernels
device, e1_img = pcv.erode(tr_es_img, kern1, 1, device, args.debug)
device, e2_img = pcv.erode(tr_es_img, kern2, 1, device, args.debug)
device, e3_img = pcv.erode(tr_es_img, kern3, 1, device, args.debug)
device, e4_img = pcv.erode(tr_es_img, kern4, 1, device, args.debug)
# Combine eroded images
device, c12_img = pcv.logical_or(e1_img, e2_img, device, args.debug)
device, c123_img = pcv.logical_or(c12_img, e3_img, device, args.debug)
device, c1234_img = pcv.logical_or(c123_img, e4_img, device, args.debug)
# Perform dilation
device, dil_img = pcv.dilate(c1234_img, kern, 1, device, args.debug)
# Get masked image
# The dilated image may contain some pixels which are not plant
device, masked_erd = pcv.apply_mask(img, c1234_img, 'black', device, args.debug)
device, masked_erd_dil = pcv.apply_mask(img, dil_img, 'black', device, args.debug)
# Need to remove the edges of the image, we did that by generating a set of rectangles to mask the edges
# img is (254 X 320)
device, box1_img, rect_contour1, hierarchy1 = pcv.rectangle_mask(img, (1,1), (64,252), device, args.debug)
device, box2_img, rect_contour2, hierarchy2 = pcv.rectangle_mask(img, (256,1), (318,252), device, args.debug)
device, box3_img, rect_contour3, hierarchy3 = pcv.rectangle_mask(img, (1,184), (318,252), device, args.debug)
device, box4_img, rect_contour4, hierarchy4 = pcv.border_mask(img, (1,1), (318,252), device, args.debug)
# combine boxes to filter the edges and car out of the photo
device, bx12_img = pcv.logical_or(box1_img, box2_img, device, args.debug)
device, bx123_img = pcv.logical_or(bx12_img, box3_img, device, args.debug)
device, bx1234_img = pcv.logical_or(bx123_img, box4_img, device, args.debug)
device, inv_bx1234_img = pcv.invert(bx1234_img, device, args.debug)
# Apply the box mask to the image
device, masked_img = pcv.apply_mask(masked_erd_dil, inv_bx1234_img, 'black', device, args.debug)
# Generate a binary to send to the analysis function
device, mask = pcv.binary_threshold(masked_img, 1, 255, 'light', device, args.debug)
pcv.analyze_NIR_intensity(img, args.image, mask, 256, device, args.debug, 'example')
def main():
# Parse command-line options
args = options()
device = 0
# Open output file
out = open(args.outfile, "w")
# Open the image file
img, path, fname = pcv.readimage(filename=args.image, debug=args.debug)
# Classify healthy and unhealthy plant pixels
device, masks = pcv.naive_bayes_classifier(img=img,
pdf_file=args.pdfs,
device=device)
# Use the identified blue mesh area to build a mask for the pot area
# First errode the blue mesh region to remove background
device, mesh_errode = pcv.erode(img=masks["Background_Blue"],
kernel=9,
i=3,
device=device,
debug=args.debug)
# Define a region of interest for blue mesh contours
device, pot_roi, pot_hierarchy = pcv.define_roi(img=img,
shape='rectangle',
device=device,
roi=None,
roi_input='default',
debug=args.debug,
adjust=True,
x_adj=0,
y_adj=500,
w_adj=0,
h_adj=-650)
# Find blue mesh contours
device, mesh_objects, mesh_hierarchy = pcv.find_objects(img=img,
mask=mesh_errode,
device=device,
debug=args.debug)
# Keep blue mesh contours in the region of interest
device, kept_mesh_objs, kept_mesh_hierarchy, kept_mask_mesh, _ = pcv.roi_objects(
img=img,
roi_type='partial',
roi_contour=pot_roi,
roi_hierarchy=pot_hierarchy,
object_contour=mesh_objects,
obj_hierarchy=mesh_hierarchy,
device=device,
debug=args.debug)
# Flatten the blue mesh contours into a single object
device, mesh_flattened, mesh_mask = pcv.object_composition(
img=img,
contours=kept_mesh_objs,
hierarchy=kept_mesh_hierarchy,
device=device,
debug=args.debug)
# Initialize a pot mask
pot_mask = np.zeros(np.shape(masks["Background_Blue"]), dtype=np.uint8)
# Find the minimum bounding rectangle for the blue mesh region
rect = cv2.minAreaRect(mesh_flattened)
# Create a contour for the minimum bounding box
box = cv2.boxPoints(rect)
box = np.int0(box)
# Create a mask from the bounding box contour
cv2.drawContours(pot_mask, [box], 0, (255), -1)
# If the bounding box area is too small then the plant has likely occluded too much of the pot for us to use this
# as a marker for the pot area
if np.sum(pot_mask) / 255 < 2900000:
print(np.sum(pot_mask) / 255)
# Create a new pot mask
pot_mask = np.zeros(np.shape(masks["Background_Blue"]), dtype=np.uint8)
# Set the mask area to the ROI area
box = np.array([[0, 500], [0, 2806], [2304, 2806], [2304, 500]])
cv2.drawContours(pot_mask, [box], 0, (255), -1)
# Dialate the blue mesh area to include the ridge of the pot
device, pot_mask_dilated = pcv.dilate(img=pot_mask,
kernel=3,
i=60,
device=device,
debug=args.debug)
# Mask the healthy mask
device, healthy_masked = pcv.apply_mask(img=cv2.merge(
[masks["Healthy"], masks["Healthy"], masks["Healthy"]]),
mask=pot_mask_dilated,
mask_color="black",
device=device,
debug=args.debug)
# Mask the unhealthy mask
device, unhealthy_masked = pcv.apply_mask(img=cv2.merge(
[masks["Unhealthy"], masks["Unhealthy"], masks["Unhealthy"]]),
mask=pot_mask_dilated,
mask_color="black",
device=device,
debug=args.debug)
# Convert the masks back to binary
healthy_masked, _, _ = cv2.split(healthy_masked)
unhealthy_masked, _, _ = cv2.split(unhealthy_masked)
# Fill small objects
device, fill_image_healthy = pcv.fill(img=np.copy(healthy_masked),
mask=np.copy(healthy_masked),
size=300,
device=device,
debug=args.debug)
device, fill_image_unhealthy = pcv.fill(img=np.copy(unhealthy_masked),
mask=np.copy(unhealthy_masked),
size=1000,
device=device,
debug=args.debug)
# Define a region of interest
device, roi1, roi_hierarchy = pcv.define_roi(img=img,
shape='rectangle',
device=device,
roi=None,
roi_input='default',
debug=args.debug,
adjust=True,
x_adj=450,
y_adj=1000,
w_adj=-400,
h_adj=-1000)
# Filter objects that overlap the ROI
device, id_objects, obj_hierarchy_healthy = pcv.find_objects(
img=img, mask=fill_image_healthy, device=device, debug=args.debug)
device, _, _, kept_mask_healthy, _ = pcv.roi_objects(
img=img,
roi_type='partial',
roi_contour=roi1,
roi_hierarchy=roi_hierarchy,
object_contour=id_objects,
obj_hierarchy=obj_hierarchy_healthy,
device=device,
debug=args.debug)
device, id_objects, obj_hierarchy_unhealthy = pcv.find_objects(
img=img, mask=fill_image_unhealthy, device=device, debug=args.debug)
device, _, _, kept_mask_unhealthy, _ = pcv.roi_objects(
img=img,
roi_type='partial',
roi_contour=roi1,
roi_hierarchy=roi_hierarchy,
object_contour=id_objects,
obj_hierarchy=obj_hierarchy_unhealthy,
device=device,
debug=args.debug)
# Combine the healthy and unhealthy mask
device, mask = pcv.logical_or(img1=kept_mask_healthy,
img2=kept_mask_unhealthy,
device=device,
debug=args.debug)
# Output a healthy/unhealthy image
classified_img = cv2.merge([
np.zeros(np.shape(mask), dtype=np.uint8), kept_mask_healthy,
kept_mask_unhealthy
])
pcv.print_image(img=classified_img,
filename=os.path.join(
args.outdir,
os.path.basename(args.image)[:-4] + ".classified.png"))
# Output a healthy/unhealthy image overlaid on the original image
overlayed = cv2.addWeighted(src1=np.copy(classified_img),
alpha=0.5,
src2=np.copy(img),
beta=0.5,
gamma=0)
pcv.print_image(img=overlayed,
filename=os.path.join(
args.outdir,
os.path.basename(args.image)[:-4] + ".overlaid.png"))
# Extract hue values from the image
device, h = pcv.rgb2gray_hsv(img=img,
channel="h",
device=device,
debug=args.debug)
# Extract the plant hue values
plant_hues = h[np.where(mask == 255)]
# Initialize hue histogram
hue_hist = {}
for i in range(0, 180):
hue_hist[i] = 0
# Store all hue values
hue_values = []
# Populate histogram
total_px = len(plant_hues)
for hue in plant_hues:
hue_hist[hue] += 1
hue_values.append(hue)
# Parse the filename
genotype, treatment, replicate, timepoint = os.path.basename(
args.image)[:-4].split("_")
replicate = replicate.replace("#", "")
if timepoint[-3:] == "dbi":
timepoint = -1
else:
timepoint = timepoint.replace("dpi", "")
# Output results
for i in range(0, 180):
out.write("\t".join(
map(str, [
genotype, treatment, timepoint, replicate, total_px, i,
hue_hist[i]
])) + "\n")
out.close()
# Calculate basic statistics
healthy_sum = int(np.sum(kept_mask_healthy))
unhealthy_sum = int(np.sum(kept_mask_unhealthy))
healthy_total_ratio = healthy_sum / float(healthy_sum + unhealthy_sum)
unhealthy_total_ratio = unhealthy_sum / float(healthy_sum + unhealthy_sum)
stats = open(args.outfile[:-4] + ".stats.txt", "w")
stats.write("%s, %f, %f, %f, %f" %
(os.path.basename(args.image), healthy_sum, unhealthy_sum,
healthy_total_ratio, unhealthy_total_ratio) + '\n')
stats.close()
# Fit a 3-component Gaussian Mixture Model
gmm = mixture.GaussianMixture(n_components=3,
covariance_type="full",
tol=0.001)
gmm.fit(np.expand_dims(hue_values, 1))
gmm3 = open(args.outfile[:-4] + ".gmm3.txt", "w")
gmm3.write("%s, %f, %f, %f, %f, %f, %f, %f, %f, %f" %
(os.path.basename(args.image), gmm.means_.ravel()[0],
gmm.means_.ravel()[1], gmm.means_.ravel()[2],
np.sqrt(gmm.covariances_.ravel()[0]),
np.sqrt(gmm.covariances_.ravel()[1]),
np.sqrt(gmm.covariances_.ravel()[2]), gmm.weights_.ravel()[0],
gmm.weights_.ravel()[1], gmm.weights_.ravel()[2]) + '\n')
gmm3.close()
# Fit a 2-component Gaussian Mixture Model
gmm = mixture.GaussianMixture(n_components=2,
covariance_type="full",
tol=0.001)
gmm.fit(np.expand_dims(hue_values, 1))
gmm2 = open(args.outfile[:-4] + ".gmm2.txt", "w")
gmm2.write("%s, %f, %f, %f, %f, %f, %f" %
(os.path.basename(args.image), gmm.means_.ravel()[0],
gmm.means_.ravel()[1], np.sqrt(gmm.covariances_.ravel()[0]),
np.sqrt(gmm.covariances_.ravel()[1]), gmm.weights_.ravel()[0],
gmm.weights_.ravel()[1]) + '\n')
gmm2.close()
# Fit a 1-component Gaussian Mixture Model
gmm = mixture.GaussianMixture(n_components=1,
covariance_type="full",
tol=0.001)
gmm.fit(np.expand_dims(hue_values, 1))
gmm1 = open(args.outfile[:-4] + ".gmm1.txt", "w")
gmm1.write(
"%s, %f, %f, %f" %
(os.path.basename(args.image), gmm.means_.ravel()[0],
np.sqrt(gmm.covariances_.ravel()[0]), gmm.weights_.ravel()[0]) + '\n')
gmm1.close()