def generate():
propcountarr = []
with open(config.prop_json) as f:
jsonprop = json.load(f)
rgb_files = [f for f in listdir(rgb_folder) if isfile(join(rgb_folder, f))]
for i in range(0, len(rgb_files)):
#print(rgb_files[i])
image_name = rgb_files[i] #jsonprop["proposals"][i]["file"]
image_name_png = image_name.replace("jpg", "png")
count = 0
for j in range(0, len(jsonprop)): #[i]["objects"])):
if (jsonprop[j]["file"] == image_name_png):
propcount = 0
for k in range(0, len(jsonprop[j]["objects"])):
propcount = propcount + 1
x = jsonprop[j]["objects"][k]["x"]
y = jsonprop[j]["objects"][k]["y"]
z = jsonprop[j]["objects"][k]["z"]
xmiw, ymiw, xmaw, ymaw = core_functions.get_bounding_box_WASSIMEA(
x, y, z)
prop_object = [xmiw, ymiw, xmaw, ymaw, y, x]
headpoint = [prop_object[4], prop_object[5]]
image, headpoint_new = core_functions.get_neg_roi(
rgb_folder + image_name, prop_object, headpoint)
if config.mod == True:
image = mod.get_channels(image, headpoint_new)
cv2.normalize(image, image, 0, 255, cv2.NORM_MINMAX)
width, height = image.shape[1], image.shape[0]
if (width > 30 and height > 30):
#c1,c2,c3,mod = core_functions.get_channels(image, headpoint_new)
cv2.imwrite(
out_folder + image_name.replace(".jpg", "_id_") +
str(count) + ".jpg", image)
count = count + 1
zabre = np.mean(propcountarr)
pingit = 1
def process_image(filename, jsongt, jsonprop):
#png_image = cv2.imread(filename, -1)
#counter = 0
filename_only_jpg = os.path.basename(filename).replace(".png", ".jpg")
filename_only_png = filename_only_jpg.replace("jpg", "png")
#print(filename_only_jpg)
gtarray = common_functions.get_gt_annotations(filename_only_jpg, jsongt)
proparray = common_functions.get_proposals(filename_only_png, jsonprop)
flag_array = np.zeros(len(proparray))
#if "videos_holland_orbbec_2018-02-21_CAM2_1519231983220" in filename:
# im = cv2.imread(filename)
# for gt in gtarray:
# cv2.rectangle(im,(gt[0], gt[1]), (gt[2], gt[3]),(0,255,0), 1)
# for prop in proparray:
# cv2.rectangle(im,(prop[0], prop[1]), (prop[2], prop[3]),(0,0,255), 1)
#
# cv2.imshow("im",im)
# cv2.waitKey()
for gt_object in gtarray:
id = gt_object[6]
if 1 == 1: #common_functions.check_box_boundaries(gt_object[0],gt_object[1],gt_object[2],gt_object[3]):
list_valid_proposals = []
list_valid_proposals_ious = []
for prop_object in proparray:
#if(flag_array[proparray.index(prop_object)] != 1):
iou = core_functions.bb_intersection_over_union(
gt_object, prop_object)
#if(common_functions.check_if_box_is_contained(gt_object,prop_object)):
if (iou > 0.75):
list_valid_proposals.append(prop_object)
list_valid_proposals_ious.append(iou)
if (len(list_valid_proposals_ious) > 0):
#list_acceptable_proposals = []
#idx = np.argsort(list_valid_proposals_ious)
#
#list_valid_proposals_ious = np.array(list_valid_proposals_ious)[idx]
#list_valid_proposals = np.array(list_valid_proposals)[idx]
#list_valid_proposals_ious = list_valid_proposals_ious.tolist()
#list_valid_proposals = list_valid_proposals.tolist()
for i in range(0, len(list_valid_proposals)):
#highest_iou_index = np.argmax(list_valid_proposals_ious)
#prop_object = list_valid_proposals[highest_iou_index]
#flag_array[proparray.index(list_valid_proposals[i])] = 1
image, head_new, right_shoulder_new, left_shoulder_new, headpoint_new = core_functions.get_new_roi_with_annotations(
filename, list_valid_proposals[i], gt_object)
if config.mod == True:
if ("video_fifth_orbbec_2018-01-19_lunch_CAM1_1516381409554.png"
in filename):
x = 1
#image = mod.get_channels(image,headpoint_new)#.astype('uint8')
image = mod.get_channels(image, headpoint_new)
#roi1copy = roi1.copy()
#b,g,r = cv2.split(image)
#b = cv2.normalize(b, None, alpha=0, beta=255, norm_type=cv2.NORM_MINMAX).astype('uint8')
#g = g.astype('uint8')
#r = cv2.normalize(r, None, alpha=0, beta=255, norm_type=cv2.NORM_MINMAX).astype('uint8')
#b = (b/256).astype('uint8')
#g = (g/256).astype('uint8')
#r = (r/256).astype('uint8')
#image = cv2.merge((b,g,r)).astype('uint8')
#cv2.imshow("im",image)
#cv2.waitKey()
#cv2.normalize(image, image, 0, 255, cv2.NORM_MINMAX)
#image = image.astype('uint8')
if image is not None:
encode_param = [int(cv2.IMWRITE_JPEG_QUALITY), 95]
result, encimg = cv2.imencode('.jpg', image,
encode_param)
image = cv2.imdecode(encimg, 1)
flag_array[proparray.index(
list_valid_proposals[i])] = 1
width, height = image.shape[1], image.shape[0]
if (width > 2 and height >
2): # and num_zeros < (width*height)/ 3):
data['frames'].append({
'file':
filename_only_png.replace(".png", "_id_") +
str(id) + "_index_" + str(i) + ".png",
'width':
width,
'height':
height,
'annotations': [{
'label':
'Head',
'x':
head_new[0],
'y':
head_new[1],
'width':
head_new[2] - head_new[0],
'height':
head_new[3] - head_new[1]
}, {
'label': 'Headpoint',
'x': headpoint_new[0],
'y': headpoint_new[1]
}, {
'label': 'Right Shoulder',
'x': right_shoulder_new[1],
'y': right_shoulder_new[0]
}, {
'label': 'Left Shoulder',
'x': left_shoulder_new[1],
'y': left_shoulder_new[0]
}]
})
cv2.imwrite(
pos_out_folder +
filename_only_png.replace(".png", "_id_") +
str(id) + "_index_" + str(i) + ".png", image)
for i in range(0, len(flag_array)):
if flag_array[i] == 0:
headpoint = [proparray[i][4], proparray[i][5]]
image, headpoint_new = core_functions.get_neg_roi(
filename, proparray[i], headpoint)
if config.mod == True:
image = mod.get_channels(image, headpoint_new)
if image is not None:
encode_param = [int(cv2.IMWRITE_JPEG_QUALITY), 95]
result, encimg = cv2.imencode('.jpg', image, encode_param)
image = cv2.imdecode(encimg, 1)
cv2.imwrite(
neg_out_folder +
filename_only_png.replace(".png", "_id_") + str(i) +
".png", image)
def process_image(filename, jsongt, jsonprop):
#png_image = cv2.imread(filename, -1)
#counter = 0
filename_only_png = os.path.basename(filename)
filename_only_jpg = filename_only_png.replace("png", "jpg")
print(filename_only_png)
gtarray = common_functions.get_gt_annotations(filename_only_jpg, jsongt)
proparray = common_functions.get_proposals(filename_only_png, jsonprop)
#for gt_object in gtarray:
if 1 == 1:
gt_object = gtarray[0]
id = gt_object[6]
if common_functions.check_box_boundaries(gt_object[0], gt_object[1],
gt_object[2], gt_object[3]):
for prop_object in proparray:
iou = core_functions.bb_intersection_over_union(
gt_object, prop_object)
if (iou > 0.5):
contained = True
image, head_new, right_shoulder_new, left_shoulder_new, headpoint_new = core_functions.get_new_roi_with_annotations(
filename, prop_object, gt_object)
num_zeros = (image == 0).sum()
width, height = image.shape[1], image.shape[0]
if (width > 2 and height > 2 and num_zeros <
(width * height) / 3):
data['frames'].append({
'file':
filename_only_png.replace(".png", "_id_") +
str(id) + ".jpg",
'width':
width,
'height':
height,
'annotations': [{
'label': 'Head',
'x': head_new[0],
'y': head_new[1],
'width': head_new[2] - head_new[0],
'height': head_new[3] - head_new[1]
}, {
'label': 'Headpoint',
'x': headpoint_new[0],
'y': headpoint_new[1]
}, {
'label': 'Right Shoulder',
'x': right_shoulder_new[1],
'y': right_shoulder_new[0]
}, {
'label': 'Left Shoulder',
'x': left_shoulder_new[1],
'y': left_shoulder_new[0]
}]
})
#startpy = time.clock()
#mod = core_functions.get_channels(image, headpoint_new)
#timepy = time.clock() - startpy
startcy = time.clock()
#image = np.ascontiguousarray(image)
mod1 = mod.get_channels(image, headpoint_new)
#cv2.imshow("x",mod1)
#cv2.waitKey()
timecy = time.clock() - startcy
#cv2.rectangle(mod,(head_new[0],head_new[1]),(head_new[2] ,head_new[3]),(255,0,0),3)
#cv2.circle(mod,(right_shoulder_new[0], right_shoulder_new[1]), 5, (0,255,0), -1)
#cv2.circle(mod,(left_shoulder_new[0], left_shoulder_new[1]), 5, (0,255,0), -1)
#cv2.imwrite(out_folder + filename.replace(".png","") + "/" + str(id) + "/c1.png",c1)
#cv2.imwrite(out_folder + filename.replace(".png","") + "/" + str(id) + "/c2.jpg",c2)
#cv2.imwrite(out_folder + filename.replace(".png","") + "/" + str(id) + "/c3.jpg",c3)
#cv2.imwrite(out_folder + filename.replace(".png","") + "/" + str(id) + "/mod.jpg",mod)
#cv2.imwrite("mod.jpg", mod)
cv2.imwrite(
"/home/wassimea/Desktop/chl_work/mod_scripts/modtest.jpg",
mod1)
cv2.imshow("ayre", mod1)
cv2.waitKey()
print(timecy)
break
def evaluate_3c(index):
config = singleton.configuration._instance.config
data = {}
data['results'] = []
tf.logging.set_verbosity(tf.logging.FATAL)
logging.getLogger("tensorflow").setLevel(logging.ERROR)
pingit = common_functions.Detector(index, config)
images_folder = config.rgb_folder
display_folder = config.display_folder
images = [
f for f in listdir(images_folder) if isfile(join(images_folder, f))
]
jsonprop = config.jsonprop
jsongt = config.jsongt
iou_props_total = 0
iou_detections_total = 0
total_detections = 0
total_gt = 0
total_tp = 0
total_fp = 0
total_pckh_precision = 0
master_count = 0
#val_folder = "/Data2TB/chl_data/mod/val/pos_neg_png_48/"
images = [
f for f in listdir(images_folder) if isfile(join(images_folder, f))
]
pos = 0
neg = 0
for image in images:
img = cv2.imread(images_folder + image, -1)
#img = cv2.imread(val_folder + "video_fifth_2018-04-23_CAM1_1524502510848_id_791_index_0.png")
positive_props = []
detections_regressed = []
img = cv2.imread(images_folder + image, -1)
#img = cv2.imread("/Data2TB/chl_data/mod/val/pos_png_48/video_fifth_2018-04-23_CAM1_1524502858816_id_831_index_0.png",-1)
##roi_detect = img.reshape(1, 48,48,3)
##roi_detect = (roi_detect.astype('float32') - 127.5)/128
##roi_detect = roi_detect[...,::-1]
proposals, gtobjects = common_functions.get_proposal_and_gt_boxes_mod(
image, jsonprop, jsongt)
display_image = cv2.imread(display_folder + image)
#
#
count = 0
propcount = 0
for prop in proposals:
total_detections += len(proposals)
imgtemp = img.copy()
#cv2.imshow("tmp",imgtemp)
#cv2.waitKey()
xmin = prop[0][0]
ymin = prop[0][1]
xmax = prop[0][2]
ymax = prop[0][3]
headpoint = prop[1]
xmin, ymin, xmax, ymax = common_functions.refine_bounding_box(
xmin, ymin, xmax, ymax)
roi = imgtemp[ymin:ymax, xmin:xmax]
x_roi = headpoint[0] - xmin
y_roi = headpoint[1] - ymin
headpoint_roi = [y_roi, x_roi]
roi = mod.get_channels(roi, headpoint_roi)
# #roi1copy = roi1.copy()
encode_param = [int(cv2.IMWRITE_JPEG_QUALITY), 95]
result, encimg = cv2.imencode('.jpg', roi, encode_param)
roi = cv2.imdecode(encimg, 1)
#cv2.imshow("WAS IT LOS", roi)
#cv2.waitKey()
count += 1
# #continue
# #videos_holland_orbbec_2018-02-21_CAM2_1519232343120_id_66320_index_0.png
width_original = roi.shape[1]
width_diff = width_original - 48
#
roi_detect = cv2.resize(roi, (48, 48))
roi_detect = roi_detect.reshape(1, 48, 48, 3)
roi_detect = (roi_detect.astype('float32') - 127.5) / 128
roi_detect = roi_detect[..., ::-1]
#cls_scores = common_functions.testpredict(roi_detect)
cls_scores = pingit.predict(roi_detect)[0]
#print(cls_scores)
ind = np.argmax(cls_scores)
if (ind == 1):
zabri = 1
pos += 1
#cv2.imshow("pos", img)
#cv2.waitKey()
positive_props.append(prop)
#bb,lm = common_functions.rescale_detections_to_full_image(bb,lm, prop)
#detection = [bb,lm]
#detections_regressed.append(detection)
ious = eval_functions.get_iou_totals_per_image(gtobjects,
positive_props,
images_folder + image)
##shoulder_precisions = eval_functions.get_shoulder_precisions(gtobjects, detections_regressed)
#
iou_props_total += ious[0]
iou_detections_total += ious[1]
total_gt += ious[0]
total_tp += ious[1]
total_fp += ious[2]
if (ious[0] != 0 and ious[1] != 0):
master_count += 1
recall = total_tp / total_gt
miss_rate = total_fp / (total_tp + total_fp)
#av_pckh_precision = total_pckh_precision/(master_count)
data['results'].append({
'at': str(index),
'recall': recall,
'miss_rate': miss_rate
})
with open(
config.results_out + config.mtcnn_model + config.bbox_loss +
".json", 'a') as outfile:
json.dump(data, outfile, indent=4)
return recall, miss_rate
def evaluate_3c(pingit, images_folder):
images_folder = "/Data2TB/correctly_registered/S12/test/depth/"
images = [
f for f in listdir(images_folder) if isfile(join(images_folder, f))
]
jsonprop = "/Data2TB/correctly_registered/S12/test/output.json"
jsongt = "/Data2TB/sample/annotations.json"
list_thresholds = [
0.05, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.6, 0.7, 0.8,
0.9
]
percentage = 0.5
list_precisions = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
list_recalls = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
list_gt = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_gt = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_tn = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_tp = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_fp = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_outl = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_fn = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_pckh_precision = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
total_true_detections_p5 = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
iou_props_total = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
iou_detections_total = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
master_count = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
av_iou_prop = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
av_iou_detection = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
recall = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
miss_rate = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
precision = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
av_pckh_precision = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
for image in images:
total_proposals = 0
#image = images[i]
positive_props = []
detections_regressed = []
img = cv2.imread(images_folder + image, -1)
proposals, gtobjects = common_functions.get_proposal_and_gt_boxes_mod(
image, jsonprop, jsongt)
count = 0
propcount = 0
list_positive_props = [[], [], [], [], [], [], [], [], [], [], [], [],
[], []]
list_detections_regressed = [[], [], [], [], [], [], [], [], [], [],
[], [], [], []]
for prop in proposals:
total_proposals += 1
imgtemp = img.copy()
#cv2.imshow("tmp",imgtemp)
#cv2.waitKey()
xmin = prop[0][0]
ymin = prop[0][1]
xmax = prop[0][2]
ymax = prop[0][3]
headpoint = prop[1]
xmin, ymin, xmax, ymax = common_functions.refine_bounding_box(
xmin, ymin, xmax, ymax)
roi = imgtemp[ymin:ymax, xmin:xmax]
x_roi = headpoint[0] - xmin
y_roi = headpoint[1] - ymin
headpoint_roi = [y_roi, x_roi]
roi = mod.get_channels(roi, headpoint_roi)
# #roi1copy = roi1.copy()
encode_param = [int(cv2.IMWRITE_JPEG_QUALITY), 95]
result, encimg = cv2.imencode('.jpg', roi, encode_param)
roi = cv2.imdecode(encimg, 1)
width_original = roi.shape[1]
roi_detect = cv2.resize(roi, (48, 48))
roi_detect = roi_detect.reshape(1, 48, 48, 3)
roi_detect = (roi_detect.astype('float32') - 127.5) / 128
roi_detect = roi_detect[..., ::-1]
cls_scores = pingit.predict(roi_detect)[0]
for i in range(len(list_thresholds)):
if (cls_scores[0][1] >= list_thresholds[i]):
list_positive_props[i].append(prop)
#bbtmp,lmtmp = common_functions.rescale_detections_to_full_image(bb,lm, prop)
#list_detections_regressed[i].append([bbtmp,lmtmp])
for i in range(len(list_thresholds)):
results = get_iou_totals_per_image(gtobjects,
list_positive_props[i],
images_folder + image,
percentage)
iou_props_total[i] += results[0]
iou_detections_total[i] += results[1]
total_gt[i] += results[2]
total_tp[i] += results[3]
total_fp[i] += results[4]
total_outl[i] += results[6]
total_fn[i] += results[7]
total_tn[
i] += total_proposals - results[3] - results[4] - results[6]
total_pckh_precision[i] += results[5]
total_true_detections_p5[i] += results[8]
#total_precisions += ious[5]
if (results[0] != 0):
master_count[i] += 1
for i in range(len(list_thresholds)):
av_iou_prop[i] = iou_props_total[i] / master_count[i]
av_iou_detection[i] = iou_detections_total[i] / master_count[i]
recall[i] = total_tp[i] / total_gt[i]
miss_rate[i] = total_fp[i] / (total_fp[i] + total_tn[i])
precision[i] = total_tp[i] / (total_tp[i] + total_fp[i])
av_pckh_precision[i] = total_pckh_precision[i] / (master_count[i])
#recall[i] = total_tp[i] / total_gt[i]
#precision[i] = total_tp[i] / (total_tp[i] + total_fp[i])
print("Threshold: ", round(list_thresholds[i], 2))
#print("Average prop IOU: " , round(av_iou_prop[i],2))
#print("Average detection IOU: " , round(av_iou_detection[i],2))
#print("PCKH: " , round(av_pckh_precision[i],2))
print("Recall: ", round(recall[i], 2))
print("Precision: ", round(precision[i], 2))
print("")
print("")
return av_iou_detection, recall, miss_rate, av_pckh_precision, av_iou_prop