xmax = xmax + col * stride
ymax = ymax + row * stride
xmin, ymin, xmax, ymax = transform_coord(
xmin, ymin, xmax, ymax, height, width,
each_trans)
updated_metric_box.append(
[xmin, ymin, xmax, ymax])
boxes += updated_metric_box
scores += metric_scores
np_boxes = np.array(boxes)
np_scores = np.array(scores)
boxes, scores = non_max_suppression_fast(
np_boxes, 0.75, np_scores)
metric_json[file_name]['boxes'] = boxes
metric_json[file_name]['scores'] = scores
[
cv2.rectangle(gt_box_img, (xmin, ymin), (xmax, ymax),
(0, 0, 0), 5)
for xmin, ymin, xmax, ymax in boxes
]
# cv2.namedWindow('win',cv2.WINDOW_NORMAL)
# cv2.imshow('win',gt_box_img)
# cv2.waitKey(0)
# cv2.destroyAllWindows()
cv2.imwrite(os.path.join(result_path, file_name), gt_box_img)
with open(predicted_json_path, 'w') as f:
# print (metric_json)
# print (output_dict['detection_boxes'])
# print (output_dict['detection_classes'])
# print (output_dict['detection_scores'])
# print ("Plot")
# Plot Image
# img_cv2_bgr = cv2.cvtColor(image_np,cv2.COLOR_RGB2BGR)
# cv2.imwrite(detection_out_path + str(count) + '.jpg',img_cv2_bgr)
# count += 1
# whole_img
# cv2.imshow('win',img_cv2_bgr)
# cv2.waitKey(0)
# cv2.destroyAllWindows()
boxes, scores = non_max_suppression_fast(
np.array(boxes), 0.75, np.array(scores))
[
cv2.rectangle(whole_img, (xmin, ymin), (xmax, ymax),
(0, 255, 0), 3)
for xmin, ymin, xmax, ymax in boxes
]
print("File_name:{},No.of boxes:{}".format(
file_name, len(boxes)))
metric_json[file_name]['boxes'] = boxes
metric_json[file_name]['scores'] = scores
yellow_cell_img_path = os.path.join(image_yellow_path,
file_name).replace(
'.bmp', '.jpg')