def __get_image_information(single_line):
line_string = bytes.decode(single_line.numpy(), encoding="utf-8")
line_list = line_string.strip().split(" ")
image_file, image_height, image_width = line_list[:3]
image_height, image_width = str_to_int(image_height), str_to_int(
image_width)
boxes = []
num_of_boxes = (len(line_list) - 3) / 5
if int(num_of_boxes) == num_of_boxes:
num_of_boxes = int(num_of_boxes)
else:
raise ValueError("num_of_boxes must be 'int'.")
for index in range(num_of_boxes):
if index < MAX_BOXES_PER_IMAGE:
xmin = str_to_int(line_list[3 + index * 5])
ymin = str_to_int(line_list[3 + index * 5 + 1])
xmax = str_to_int(line_list[3 + index * 5 + 2])
ymax = str_to_int(line_list[3 + index * 5 + 3])
class_id = int(line_list[3 + index * 5 + 4])
xmin, ymin, xmax, ymax = resize_box(image_height, image_width,
xmin, ymin, xmax, ymax)
boxes.append([xmin, ymin, xmax, ymax, class_id])
num_padding_boxes = MAX_BOXES_PER_IMAGE - num_of_boxes
if num_padding_boxes > 0:
for i in range(num_padding_boxes):
boxes.append([0, 0, 0, 0, -1])
boxes_array = np.array(
boxes, dtype=np.float32) # shape: (MAX_BOXES_PER_IMAGE, 5)
return image_file, boxes_array
def __process_coord(self, x_min, y_min, x_max, y_max):
x_min = str_to_int(x_min)
y_min = str_to_int(y_min)
x_max = str_to_int(x_max)
y_max = str_to_int(y_max)
return int(x_min), int(y_min), int(x_max), int(y_max)