Tensorflow调用实例分割模型并显示mask

本教程主要分享调用在tensorflow框架下训练好的实例分割模型进行目标检测，并用opencv显示检测的mask和bbox。
该教程代码是在tensorflow的object detection API框架下运行的。

import numpy as np
import os
import tensorflow as tf
import cv2
import time
from object_detection.utils import ops as utils_ops
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util

MODEL_NAME = 'mask_rcnn_inception_v2_coco'
PATH_TO_FROZEN_GRAPH = "exported_model/frozen_inference_graph.pb"
PATH_TO_LABELS = "mouth_dataset/label_map.pbtxt"


def load_image_into_numpy_array(image):
    (im_width, im_height) = image.size
    return np.array(image.getdata()).reshape(
        (im_height, im_width, 3)).astype(np.uint8)


def net_init(graph):
    with graph.as_default():
        # with tf.Session() as sess:
        # Get handles to input and output tensors
        ops = tf.get_default_graph().get_operations()
        all_tensor_names = {output.name for op in ops for output in op.outputs}
        tensor_dict = {}
        for key in [
            'num_detections', 'detection_boxes', 'detection_scores',
            'detection_classes', 'detection_masks'
        ]:
            tensor_name = key + ':0'
            if tensor_name in all_tensor_names:
                tensor_dict[key] = tf.get_default_graph().get_tensor_by_name(
                    tensor_name)
        if 'detection_masks' in tensor_dict:
            # The following processing is only for single image
            detection_boxes = tf.squeeze(tensor_dict['detection_boxes'], [0])
            detection_masks = tf.squeeze(tensor_dict['detection_masks'], [0])
            # Reframe is required to translate mask from box coordinates to image coordinates and fit the image size.
            real_num_detection = tf.cast(tensor_dict['num_detections'][0], tf.int32)
            detection_boxes = tf.slice(detection_boxes, [0, 0], [real_num_detection, -1])
            detection_masks = tf.slice(detection_masks, [0, 0, 0], [real_num_detection, -1, -1])
            detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(detection_masks, detection_boxes,
                                                                                  960, 544)
            # detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(detection_masks, detection_boxes, image.shape[0], image.shape[1])
            detection_masks_reframed = tf.cast(tf.greater(detection_masks_reframed, 0.5), tf.uint8)
            # Follow the convention by adding back the batch dimension
            tensor_dict['detection_masks'] = tf.expand_dims(detection_masks_reframed, 0)
        image_tensor = tf.get_default_graph().get_tensor_by_name('image_tensor:0')
        return image_tensor, tensor_dict


def run_inference_for_single_image(image_tensor, tensor_dict, graph, image, sess):
    with graph.as_default():
            output_dict = sess.run(tensor_dict, feed_dict={image_tensor: np.expand_dims(image, 0)})

            # all outputs are float32 numpy arrays, so convert types as appropriate
            output_dict['num_detections'] = int(output_dict['num_detections'][0])
            output_dict['detection_classes'] = output_dict[
                'detection_classes'][0].astype(np.uint8)
            output_dict['detection_boxes'] = output_dict['detection_boxes'][0]
            output_dict['detection_scores'] = output_dict['detection_scores'][0]
            if 'detection_masks' in output_dict:
                output_dict['detection_masks'] = output_dict['detection_masks'][0]
    return output_dict


detection_graph = tf.Graph()
with detection_graph.as_default():
    od_graph_def = tf.GraphDef()
    with tf.gfile.GFile(PATH_TO_FROZEN_GRAPH, 'rb') as fid:
        serialized_graph = fid.read()
        od_graph_def.ParseFromString(serialized_graph)
        tf.import_graph_def(od_graph_def, name='')
sess = tf.Session(graph=detection_graph)
# category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)
category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS)

# Size, in inches, of the output images.
#IMAGE_SIZE = (12, 8)

image_tensor, tensor_dict = net_init(detection_graph)
while 1:
    t0 = time.time()
    rgb = cv2.imread('mouth_dataset/test/1_107.jpg')
    output_dict = run_inference_for_single_image(image_tensor, tensor_dict, detection_graph, rgb, sess)
    # Visualization of the results of a detection.
    vis_util.visualize_boxes_and_labels_on_image_array(
        rgb,
        output_dict['detection_boxes'],
        output_dict['detection_classes'],
        output_dict['detection_scores'],
        category_index,
        instance_masks=output_dict.get('detection_masks'),
        use_normalized_coordinates=True,
        line_thickness=8)
    t1 = time.time()
    print("take cost %f s" % (t1 - t0))
    cv2.imshow('test', rgb)
    k = cv2.waitKey(1) & 0xff
    if k == ord('q') or k == 27:
        break

• 如果要运行自己的模型和图像，需要修改代码;
  （1）第10行：MODEL_NAME = ‘mask_rcnn_inception_v2_coco’，换成自己模型的config文件；
  （2）第11行：PATH_TO_FROZEN_GRAPH = “exported_model/frozen_inference_graph.pb”，换成自己训练好的pb文件；
  （3）第12行：PATH_TO_LABELS = “mouth_dataset/label_map.pbtxt”，换成自己数据的label文件；
  （4）第86行：rgb = cv2.imread(‘mouth_dataset/test/1_107.jpg’)，换成自己需要检测的图片文件；
  （5）第44行：detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(detection_masks, detection_boxes, 960, 544)，960，544换成自己检测图片的尺寸。