def make_dataset(category=None,
dirlist=None,
height=32,
width=32,
channel=3,
extensions=None):
print("\n** Make " + category)
class_len = len(dirlist)
io_mode = "w"
label_number = 0
if (not (util.check_path(path=PACK_PATH + "/images/"))):
util.make_path(path=PACK_PATH + "/images/")
util.refresh_directory(PACK_PATH + "/images/dataset/")
channel = 1
for di in dirlist:
tmp_path = PACK_PATH + "/dataset/" + category + "/" + di
fi_list = util.get_filelist(directory=tmp_path, extensions=extensions)
cnt = 0
for fi in fi_list:
tmp_sub, tmp_file = util.get_dir_and_file_name(path=fi)
cnt += 1
image = cv2.imread(fi)
resized_image = cv2.resize(image, (width, height))
cvf.save_image(path=PACK_PATH + "/images/dataset/",
filename=str(label_number) + "_" + str(cnt) +
".png",
image=resized_image)
if (channel == 1):
resized_image = cv2.cvtColor(resized_image, cv2.COLOR_BGR2GRAY)
height, width = resized_image.shape
else:
height, width, channel = resized_image.shape
resized_image = resized_image.reshape((height * width * channel))
np.save(file=tmp_path + "/" + tmp_file, arr=resized_image)
label_number += 1
if (os.path.exists(PACK_PATH + "/" + category)): # management storage
shutil.rmtree(PACK_PATH + "/" + category)
f = open(PACK_PATH + "/dataset/format.txt", "w")
f.write(str(label_number))
f.write("\n")
f.write(str(height * width * channel))
f.write("\n")
f.write(str(height))
f.write("\n")
f.write(str(width))
f.write("\n")
f.write(str(channel))
f.close()
def main():
extensions = ["BMP", "bmp", "PNG", "png", "JPG", "jpg", "JPEG", "jpeg"]
util.refresh_directory(PACK_PATH+"/images")
print("Enter the path")
# usr_path = input(">> ")
usr_path = "/media/yeonghyeon/Toshiba/lung/datasets/20171204"
if(util.check_path(usr_path)):
files = util.get_filelist(directory=usr_path, extensions=extensions)
for fi in files:
print(fi)
tmp_sub, tmp_file = util.get_dir_and_file_name(path=fi)
if(not(util.check_path(path=PACK_PATH+"/images/"+str(tmp_file)+"/"))):
util.make_path(path=PACK_PATH+"/images/"+str(tmp_file)+"/")
image = cvf.load_image(path=fi)
if(image.shape[0] > image.shape[1]): # height > width
resized = cvf.resizing(image=image, width=int(500*(image.shape[1]/image.shape[0])), height=500)
else:
resized = cvf.resizing(image=image, width=500, height=int(500*(image.shape[0]/image.shape[1])))
zeropad = cvf.zero_padding(image=resized, height=500, width=500)
print(image.shape)
print(resized.shape)
print(zeropad.shape)
cvf.save_image(path=PACK_PATH+"/images/", filename=str(tmp_file)+".png", image=zeropad)
else:
print("Invalid path :"+usr_path)
def GetLungROI(filename):
PACK_PATH = "G:/Work/ChestRadioGraphy/tb/tb-voc/VOC2007/JPEGImages"
outRoidir = "G:/Work/ChestRadioGraphy/tb/tb-voc/VOC2007/BoxJPEGImages"
outpath = os.path.join(PACK_PATH, "ROI")
if not os.path.exists(outpath):
os.makedirs(outpath)
if not os.path.exists(outRoidir):
os.makedirs(outRoidir)
pyramid = [800, 500, 100]
boxes = []
if filename.split(".")[-1] == "jpg":
# 导入图像
img = cvf.load_image(path=os.path.join(PACK_PATH, filename))
height, width = img.shape[0], img.shape[1]
# 在不同下采样程度上的图像上计算ROI框
for imgwidth in pyramid:
boxespart = GetRoiBoxes(filename, img, imgwidth, outpath)
boxes += boxespart
# 抑制重叠的框，并选出最大的两个框
boxes = cvf.NMS(boxes, (width, height), 0.5)
# 对选出的两个框进行合理性判断以及 并框
boxes = cvf.merge(boxes, (width, height))
# 输出ROI区域
RoiFilePath = os.path.join(outRoidir, filename + "_ROI.txt")
with open(RoiFilePath, "w") as fp:
if boxes != None:
# 输出ROI区域
fp.writelines("%d %d %d %d" %
(boxes[0][0], boxes[0][1], boxes[0][0] +
boxes[0][2], boxes[0][1] + boxes[0][3]))
for box in boxes:
rx, ry, rw, rh = box
cv2.rectangle(img, (rx, ry), (rx + rw, ry + rh),
(0, 255, 0), 2)
else:
# 没有正确找到ROI区域
fp.writelines("None" + "\n")
cvf.save_image(True,
path=outpath,
filename="boxes_" + filename,
image=img)
def GetRoiBoxes(filename, img, imgwidth, outpath):
origin = img.copy()
# rgb 转 gray
try:
gray = cvf.rgb2gray(rgb=origin)
except: # if origin image is grayscale
gray = origin
# 原图图像尺寸
h, w = gray.shape
#尺寸调整
resized = cvf.resizing(image=gray, width=imgwidth)
# 尺寸调整后的尺寸
hr, wr = resized.shape
cvf.save_image(saveflag,
path=outpath,
filename="resize_" + str(imgwidth) + "_" + filename,
image=resized)
resized = cvf.bluring(binary_img=resized, k_size=11)
mulmul = resized.copy()
for i in range(20):
# 将图像 按[0.3*均值，255]范围二值化
ret, thresh = cv2.threshold(mulmul,
np.average(mulmul) * 0.3, 255,
cv2.THRESH_BINARY)
cvf.save_image(saveflag,
path=outpath,
filename="thresh_" + str(imgwidth) + "_" + str(i) +
"_" + filename,
image=thresh)
# 将图像 规范化到均值为127
mulmul = cvf.normalizing(binary_img=resized * (thresh / 255))
cvf.save_image(saveflag,
path=outpath,
filename="normal_" + str(imgwidth) + "_" + str(i) +
"_" + filename,
image=mulmul)
movavg = cvf.moving_avg_filter(binary_img=mulmul, k_size=10)
adap = cvf.adaptiveThresholding(binary_img=movavg,
neighbor=111,
blur=True,
blur_size=3)
cvf.save_image(saveflag,
path=outpath,
filename="adaptive_" + str(imgwidth) + "_" + filename,
image=255 - adap)
masking = resized * ((255 - adap) / 255)
cvf.save_image(saveflag,
path=outpath,
filename="mask_" + str(imgwidth) + "_" + filename,
image=masking)
movavg = cvf.moving_avg_filter(binary_img=masking, k_size=5)
cvf.save_image(saveflag,
path=outpath,
filename="movavg_" + str(imgwidth) + "_" + filename,
image=movavg)
ret, thresh = cv2.threshold(movavg,
np.average(movavg) * 0.5, 255,
cv2.THRESH_BINARY_INV)
cvf.save_image(saveflag,
path=outpath,
filename="thresh_" + str(imgwidth) + "_" + str(imgwidth) +
"_" + filename,
image=thresh)
contours = cvf.contouring(binary_img=thresh)
cv2.drawContours(resized, contours, -1, 0)
cvf.save_image(saveflag,
path=outpath,
filename="Contours_" + str(imgwidth) + "_" + str(imgwidth) +
"_" + filename,
image=resized)
boxes_tmp = cvf.contour2box(contours=contours, padding=20)
boxes = cvf.rid_repetition(boxes=boxes_tmp, binary_img=thresh)
newboxes = []
for box in boxes:
box[0] = int(float(w) / float(wr) * box[0])
box[1] = int(float(h) / float(hr) * box[1])
box[2] = int(float(w) / float(wr) * box[2])
box[3] = int(float(h) / float(hr) * box[3])
newboxes.append(box)
return newboxes
dicom_data = dicom.read_file(fi)
try:
dicom_numpy = dicom_data.pixel_array
except:
print("TypeError: No pixel data found in this dataset.")
else:
sumx = np.sum(dicom_numpy) / (dicom_numpy.shape[0]*dicom_numpy.shape[1])
dicom_normal = (dicom_numpy / sumx) * (2**7-1)
# area1 = np.mean(dicom_normal[:int(dicom_numpy.shape[0]/4), :int(dicom_numpy.shape[1]/4)])
# area2 = np.mean(dicom_normal[int(dicom_numpy.shape[0]/4*3):, :int(dicom_numpy.shape[1]/4)])
# area3 = np.mean(dicom_normal[:int(dicom_numpy.shape[0]/4), int(dicom_numpy.shape[1]/4*3):])
# area4 = np.mean(dicom_normal[int(dicom_numpy.shape[0]/4*3):, int(dicom_numpy.shape[1]/4*3):])
#
# threshold = np.mean([area1, area2, area3, area4])
# if(threshold > 127):
# dicom_normal = 255 - dicom_normal
cvf.save_image(path=main_dir, filename=tmp_file+".bmp", image=dicom_normal)
dist = (2**16-1) / np.max(dicom_numpy)
dicom_normal = dicom_numpy * dist
dicom_normal, _ = image_histogram_equalization(image=dicom_numpy, number_bins=int(2**(16/1)))
print(np.max(dicom_normal))
tiff.imsave(main_dir+tmp_file+".tiff", (dicom_normal*(2**0)).astype(np.uint16))
tmp = tiff.imread(main_dir+tmp_file+".tiff")
print(np.max(tmp))
def extract_segments(filename):
tmp_sub, tmp_file = util.get_dir_and_file_name(path=filename)
if (not (util.check_path(path=PACK_PATH + "/images/" + str(tmp_file) +
"/"))):
util.make_path(path=PACK_PATH + "/images/" + str(tmp_file) + "/")
origin = cvf.load_image(path=filename)
gray = cvf.rgb2gray(rgb=origin)
resized = cvf.resizing(image=gray, width=500)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + ".png",
image=resized)
mulmul = resized.copy()
for i in range(20):
ret, thresh = cv2.threshold(mulmul,
np.average(mulmul) * 0.3, 255,
cv2.THRESH_BINARY)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_thresh1.png",
image=thresh)
mulmul = cvf.normalizing(binary_img=resized * (thresh / 255))
movavg = cvf.moving_avg_filter(binary_img=mulmul, k_size=10)
adap = cvf.adaptiveThresholding(binary_img=movavg,
neighbor=111,
blur=False,
blur_size=3)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_adap.png",
image=255 - adap)
result = resized * ((255 - adap) / 255)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_result1.png",
image=result)
movavg = cvf.moving_avg_filter(binary_img=result, k_size=10)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_result2.png",
image=movavg)
ret, thresh = cv2.threshold(movavg,
np.average(movavg) * 0.5, 255,
cv2.THRESH_BINARY_INV)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_thresh2.png",
image=thresh)
contours = cvf.contouring(binary_img=thresh)
boxes = cvf.contour2box(contours=contours, padding=20)
resized = cvf.resizing(image=gray, width=500)
cnt = 0
for box in boxes:
x, y, w, h = box
if ((x > 0) and (y > 0)):
if ((x + w < resized.shape[1]) and (y + h < resized.shape[0])):
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_0_" + str(cnt) + ".png",
image=thresh[y:y + h, x:x + w])
pad = cvf.zero_padding(image=thresh[y:y + h, x:x + w],
height=500,
width=500)
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_1_" + str(cnt) + ".png",
image=pad)
pad2 = cvf.remain_only_biggest(binary_img=pad)
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_2_" + str(cnt) + ".png",
image=pad2)
pad_res = cvf.zero_padding(image=resized[y:y + h, x:x + w],
height=500,
width=500)
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_3_" + str(cnt) + ".png",
image=pad_res * (pad2 / 255))
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_4_" + str(cnt) + ".png",
image=resized[y:y + h, x:x + w])
cnt += 1
for b in boxes:
x, y, w, h = b
if ((x > 0) and (y > 0)):
if ((x + w < resized.shape[1]) and (y + h < resized.shape[0])):
cv2.rectangle(resized, (x, y), (x + w, y + h), (255, 255, 255),
2)
cv2.rectangle(thresh, (x, y), (x + w, y + h), (255, 255, 255),
2)
# cvf.save_image(path=PACK_PATH+"/images/"+str(tmp_file)+"/", filename="opened.png", image=dilated)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="contour.png",
image=thresh)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="resized.png",
image=resized)
cvf.save_image(path=PACK_PATH + "/images/",
filename="resized" + str(tmp_file) + ".png",
image=resized)
def concatenate(image=None, boxes=None, ratio=1, file_name=None):
box_left = []
box_right = []
for box in boxes:
x, y, w, h, result, acc = box
rx, ry, rw, rh = x * ratio, y * ratio, w * ratio, h * ratio
if ((rx > 0) and (ry > 0)):
if ((rx + rw < image.shape[1]) and (ry + rh < image.shape[0])):
if (result == "lung_left"):
box_left.append([rx, ry, rw, rh, result, acc])
elif (result == "lung_right"):
box_right.append([rx, ry, rw, rh, result, acc])
cnt = 0
tmp_boxes = []
for box_r in box_right:
x_r, y_r, w_r, h_r, result_r, acc_r = box_r
for box_l in box_left:
x_l, y_l, w_l, h_l, result_l, acc_l = box_l
center_rx = x_r + (w_r / 2)
center_ry = y_r + (h_r / 2)
center_lx = x_l + (w_l / 2)
center_ly = y_l + (h_l / 2)
dist_limit = max(h_r, h_l) / 3 * 2
if (abs(center_ry - center_ly) >
dist_limit): # concat by y relation.
continue
else:
x_start = min(x_r, x_l)
y_start = min(y_r, y_l)
x_end = max(x_r + w_r, x_l + w_l)
y_end = max(y_r + h_r, y_l + h_l)
if ((x_start > 0) and (y_start > 0)):
if ((x_end < image.shape[1]) and (y_end < image.shape[0])):
tmp_boxes.append([
x_start, y_start, x_end - x_start, y_end - y_start,
"lung", (acc_r + acc_l) / 2
])
box_concat = []
try:
max_idx = 0
tmp_size = 0
for idx in range(len(tmp_boxes)):
x, y, w, h, result, acc = tmp_boxes[idx]
if ((w * h) > tmp_size):
tmp_size = w * h
max_idx = idx
x, y, w, h, result, acc = tmp_boxes[max_idx]
box_concat.append([x, y, w, h, result, acc])
cvf.save_image(path=PACK_PATH + "/results/" + str(file_name) + "/",
filename=str(file_name) + "_concat_" + str(cnt) + "_" +
str(int((acc_r + acc_l) / 2 * 100)) + ".png",
image=image[y:y + h, x:x + w])
cnt += 1
except:
pass
# try:
# max_idx = 0
# tmp_acc = 0
# for idx in range(len(tmp_boxes)):
# x, y, w, h, result, acc = tmp_boxes[idx]
#
# if(acc > tmp_acc):
# tmp_acc = acc
# max_idx = idx
#
# x, y, w, h, result, acc = tmp_boxes[max_idx]
# box_concat.append([x, y, w, h, result, acc])
# except:
# pass
return box_concat # return only one box
def extract_lung(usr_path,
extensions=None,
height=None,
width=None,
channel=None,
sess=None,
x_holder=None,
training=None,
prediction=None,
saver=None):
if (not (util.check_path(path=PACK_PATH + "/results/"))):
util.make_path(path=PACK_PATH + "/results/")
summf = open(PACK_PATH + "/results/summary.csv", "w")
summf.write("FILENAME")
summf.write(",")
summf.write("DETECT")
summf.write(",")
summf.write("IOU")
summf.write("\n")
files = util.get_filelist(directory=usr_path, extensions=extensions)
files.sort()
for filename in files:
print(filename)
if (util.check_file(filename=filename)):
tmp_sub, tmp_file = util.get_dir_and_file_name(path=filename)
if (not (util.check_path(path=PACK_PATH + "/results/" +
str(tmp_file) + "/"))):
util.make_path(path=PACK_PATH + "/results/" + str(tmp_file) +
"/")
origin = cvf.load_image(path=filename)
try:
gray = cvf.rgb2gray(rgb=origin)
except: # if origin image is grayscale
gray = origin
resized = cvf.resizing(image=gray, width=500)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_pre1_origin.png",
image=resized)
mulmul = resized.copy()
for i in range(20):
ret, thresh = cv2.threshold(mulmul,
np.average(mulmul) * 0.3, 255,
cv2.THRESH_BINARY)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) +
"/",
filename=str(tmp_file) + "_pre2_thresh.png",
image=thresh)
mulmul = cvf.normalizing(binary_img=resized * (thresh / 255))
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) +
"/",
filename=str(tmp_file) + "_pre3_normalize.png",
image=mulmul)
movavg = cvf.moving_avg_filter(binary_img=mulmul, k_size=10)
adap = cvf.adaptiveThresholding(binary_img=movavg,
neighbor=111,
blur=False,
blur_size=3)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_pre4_adaptrhesh.png",
image=255 - adap)
masking = resized * ((255 - adap) / 255)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_pre5_mask1.png",
image=masking)
movavg = cvf.moving_avg_filter(binary_img=masking, k_size=5)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_pre6_mask2.png",
image=movavg)
ret, thresh = cv2.threshold(movavg,
np.average(movavg) * 0.5, 255,
cv2.THRESH_BINARY_INV)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_pre7_thresh.png",
image=thresh)
contours = cvf.contouring(binary_img=thresh)
boxes_tmp = cvf.contour2box(contours=contours, padding=20)
boxes = cvf.rid_repetition(boxes=boxes_tmp, binary_img=thresh)
if (os.path.exists(PACK_PATH + "/checkpoint/checker.index")):
saver.restore(sess, PACK_PATH + "/checkpoint/checker")
f = open(PACK_PATH + "/dataset/labels.txt", 'r')
content = f.readlines()
f.close()
for idx in range(len(content)):
content[idx] = content[idx][:len(content[idx]) -
1] # rid \n
boxes_pred = []
cnt = 0
for b in boxes:
x, y, w, h = b
if ((x > 0) and (y > 0)):
if ((x + w < resized.shape[1])
and (y + h < resized.shape[0])):
pad = cvf.zero_padding(image=thresh[y:y + h,
x:x + w],
height=500,
width=500)
pad2 = cvf.remain_only_biggest(binary_img=pad)
pad_res = cvf.zero_padding(image=resized[y:y + h,
x:x + w],
height=500,
width=500)
xdata = pad_res * (pad2 / 255)
prob = sess.run(prediction,
feed_dict={
x_holder:
convert_image(image=xdata,
height=height,
width=width,
channel=channel),
training:
False
})
result = str(content[int(np.argmax(prob))])
acc = np.max(prob)
boxes_pred.append([x, y, w, h, result, acc])
# cvf.save_image(path=PACK_PATH+"/results/"+str(tmp_file)+"/", filename=str(tmp_file)+"_"+str(result)+"_"+str(int(round(acc, 2)*100))+"_"+str(cnt)+".png", image=xdata)
cnt += 1
boxes_pred = sorted(boxes_pred,
key=lambda l: l[4],
reverse=True) # sort by result
boxes_pred = sorted(boxes_pred,
key=lambda l: l[5],
reverse=True) # sort by acc
ratio = origin.shape[0] / resized.shape[0]
save_crops(image=resized,
boxes=boxes_pred,
ratio=1,
file_name=tmp_file)
concats = concatenate(image=resized,
boxes=boxes_pred,
ratio=1,
file_name=tmp_file)
iou, bbox = intersection_over_union(filename=filename,
boxes=concats,
ratio=ratio)
summf.write(str(filename))
summf.write(",")
summf.write(str(len(concats)))
summf.write(",")
summf.write(str(iou))
summf.write("\n")
origin_res1 = cvf.resizing(image=origin, width=500)
origin_res2 = origin_res1.copy()
origin_res3 = origin_res1.copy()
origin_res_lr = draw_boxes(image=origin_res1,
boxes=boxes_pred,
ratio=1,
file_name=tmp_file)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) +
"/",
filename=str(tmp_file) + "_origin_lr.png",
image=origin_res_lr)
origin_res_concat1 = draw_boxes(image=origin_res1,
boxes=concats,
ratio=1,
file_name=tmp_file)
cvf.save_image(
path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_origin_lr_and_concat.png",
image=origin_res_concat1)
origin_res_concat2 = draw_boxes(image=origin_res2,
boxes=concats,
ratio=1,
file_name=tmp_file)
cvf.save_image(path=PACK_PATH + "/results/" + str(tmp_file) +
"/",
filename=str(tmp_file) + "_origin_concat.png",
image=origin_res_concat2)
if (len(bbox) > 0):
origin_res_bbox = draw_boxes(image=origin_res3,
boxes=bbox,
ratio=1,
file_name=tmp_file)
cvf.save_image(path=PACK_PATH + "/results/" +
str(tmp_file) + "/",
filename=str(tmp_file) + "_origin_bbox.png",
image=origin_res_bbox)
origin_res_concat3 = draw_boxes(image=origin_res3,
boxes=concats,
ratio=1,
file_name=tmp_file)
cvf.save_image(
path=PACK_PATH + "/results/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_origin_concat_bbox.png",
image=origin_res_concat3)
else:
print("You must training first!")
else:
print("Invalid File: " + str(filename))
summf.close()
def tmp_main():
util.refresh_directory(PACK_PATH + "/images")
img = cvf.load_image(
path="/home/yeonghyeon/Desktop/total/pul edema_post.bmp")
print(img.shape)
gray = cvf.rgb2gray(rgb=img)
print(gray.shape)
res = cvf.resizing(image=gray, width=500)
print(res.shape)
print("AVG: " + str(np.average(res)))
print(np.average(res), np.average(res * 2))
cvf.save_image(path=PACK_PATH + "/images/",
filename="resx2.png",
image=res * res + res)
feed = cvf.feeding_outside_filter(binary_img=res, thresh=100)
cvf.save_image(path=PACK_PATH + "/images/",
filename="feed.png",
image=feed)
movavg = cvf.moving_avg_filter(binary_img=feed, k_size=10)
cvf.save_image(path=PACK_PATH + "/images/",
filename="aveage.png",
image=movavg)
ret, thresh = cv2.threshold(movavg,
np.average(movavg) * 0.8, 255,
cv2.THRESH_BINARY_INV)
cvf.save_image(path=PACK_PATH + "/images/",
filename="thresh.png",
image=thresh)
# movavg = cvf.moving_avg_filter(binary_img=thresh, k_size=3)
# cvf.save_image(path=PACK_PATH+"/images/", filename="aveage2.png", image=movavg)
contours = cvf.contouring(binary_img=thresh)
boxes = cvf.contour2box(contours=contours, padding=15)
res = cvf.resizing(image=gray, width=500)
cnt = 0
for b in boxes:
x, y, w, h = b
if ((x > 0) and (y > 0)):
if ((x + w < res.shape[1]) and (y + h < res.shape[0])):
cvf.save_image(path=PACK_PATH + "/images/",
filename="box_0_" + str(cnt) + ".png",
image=res[y:y + h, x:x + w])
cnt += 1
cnt = 0
for box1 in boxes:
x1, y1, w1, h1 = box1
for box2 in boxes:
x2, y2, w2, h2 = box2
x_crop = min(x1, x2)
y_crop = min(y1, y2)
w_crop = max(x1 + w1, x2 + w2)
h_crop = max(y1 + h1, y2 + h2)
if ((x_crop > 0) and (y_crop > 0)):
if ((x_crop + w_crop < res.shape[1])
and (y_crop + h_crop < res.shape[0])):
cvf.save_image(path=PACK_PATH + "/images/",
filename="box_1_" + str(cnt) + ".png",
image=res[y_crop:h_crop, x_crop:w_crop])
cnt += 1
for box1 in boxes:
x1, y1, w1, h1 = box1
for box2 in boxes:
x2, y2, w2, h2 = box2
x_crop = min(x1, x2)
y_crop = min(y1, y2)
w_crop = max(x1 + w1, x2 + w2)
h_crop = max(y1 + h1, y2 + h2)
if ((x_crop > 0) and (y_crop > 0)):
if ((x_crop + w_crop < res.shape[1])
and (y_crop + h_crop < res.shape[0])):
cv2.rectangle(res, (x_crop, y_crop),
(x_crop + w_crop, y_crop + h_crop),
(255, 255, 255), 2)
cv2.rectangle(thresh, (x_crop, y_crop),
(x_crop + w_crop, y_crop + h_crop),
(255, 255, 255), 2)
# for b in boxes:
# x, y, w, h = b
#
# if((x > 0) and (y > 0)):
# if((x+w < res.shape[1]) and (y+h < res.shape[0])):
# cv2.rectangle(res,(x,y),(x+w,y+h),(255, 255, 255),2)
# cv2.rectangle(thresh,(x,y),(x+w,y+h),(255, 255, 255),2)
cvf.save_image(path=PACK_PATH + "/images/",
filename="withbox.png",
image=res)
cvf.save_image(path=PACK_PATH + "/images/",
filename="withbox_thre.png",
image=thresh)
def extract_segments(filename):
tmp_sub, tmp_file = util.get_dir_and_file_name(path=filename)
if (not (util.check_path(path=PACK_PATH + "/images/" + str(tmp_file)))):
util.make_path(path=PACK_PATH + "/images/" + str(tmp_file))
origin = cvf.load_image(path=filename)
gray = cvf.rgb2gray(rgb=origin)
resized = cvf.resizing(image=gray, width=500)
avg = np.average(resized)
# feed = cvf.feeding_outside_filter(binary_img=resized, thresh=100)
# cvf.save_image(path=PACK_PATH+"/images/"+str(tmp_file)+"/", filename="feed.png", image=feed)
movavg = cvf.moving_avg_filter(binary_img=resized, k_size=10)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="movavg.png",
image=movavg)
ret, thresh = cv2.threshold(movavg,
np.average(movavg) * 0.5, 255,
cv2.THRESH_BINARY_INV)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="origin.png",
image=origin)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="thresh.png",
image=thresh)
contours = cvf.contouring(binary_img=thresh)
boxes = cvf.contour2box(contours=contours, padding=50)
resized = cvf.resizing(image=gray, width=500)
cnt = 0
for box in boxes:
x, y, w, h = box
if ((x > 0) and (y > 0)):
if ((x + w < resized.shape[1]) and (y + h < resized.shape[0])):
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_0_" + str(cnt) + ".png",
image=thresh[y:y + h, x:x + w])
pad = cvf.zero_padding(image=thresh[y:y + h, x:x + w],
height=500,
width=500)
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_1_" + str(cnt) + ".png",
image=pad)
cvf.save_image(
path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename=str(tmp_file) + "_2_" + str(cnt) + ".png",
image=resized[y:y + h, x:x + w])
cnt += 1
for b in boxes:
x, y, w, h = b
if ((x > 0) and (y > 0)):
if ((x + w < resized.shape[1]) and (y + h < resized.shape[0])):
cv2.rectangle(resized, (x, y), (x + w, y + h), (255, 255, 255),
2)
cv2.rectangle(thresh, (x, y), (x + w, y + h), (255, 255, 255),
2)
# cvf.save_image(path=PACK_PATH+"/images/"+str(tmp_file)+"/", filename="opened.png", image=dilated)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="contour.png",
image=thresh)
cvf.save_image(path=PACK_PATH + "/images/" + str(tmp_file) + "/",
filename="resized.png",
image=resized)
cvf.save_image(path=PACK_PATH + "/images/",
filename="resized" + str(tmp_file) + ".png",
image=resized)