mask rcnn

1.总结：

1.mask rcnn 是object dection和semantic segmentation的结合

参考：https://blog.csdn.net/wangdongwei0/article/details/83110305#1%E3%80%81Introduction

2.相比于FCIS，FCIS使用全卷机网络，同时预测物体classes、boxes、masks，速度更快，但是对于重叠物体的分割效果不好。

3.mask rcnn 对小物体检测有优势，因为采用了roi align 减少误差。 

2.基本步骤：

1.Mask R-CNN基本结构：与Faster RCNN采用了相同的two-state步骤：首先是找出RPN，然后对RPN找到的每个RoI进行分类、定位、并找到binary mask。

3.ROI pooling、ROI align 的区别：

 https://www.cnblogs.com/wangyong/p/8523814.html

https://blog.csdn.net/jiongnima/article/details/79094159

4.训练自己的数据：

1.图片标注工具http://www.robots.ox.ac.uk/~vgg/software/via/via.html

2.生成一个json格式的文件，使用notepad++ 的json插件查看

3.下载源码（tensorflow版）

https://github.com/matterport/Mask_RCNN

4.在samples下，有一个气球案例，复制一份，修改部分代码

import os
import sys
import json
import datetime
import numpy as np
import skimage.draw

# Root directory of the project
ROOT_DIR = os.path.abspath("../../")

# Import Mask RCNN
sys.path.append(ROOT_DIR)  # To find local version of the library
from mrcnn.config import Config
from mrcnn import model as modellib, utils

# Path to trained weights file
COCO_WEIGHTS_PATH = os.path.join(ROOT_DIR, "mask_rcnn_coco.h5")

# Directory to save logs and model checkpoints, if not provided
# through the command line argument --logs
DEFAULT_LOGS_DIR = os.path.join(ROOT_DIR, "logs")


############################################################
#  Configurations
############################################################


class CircleConfig(Config):
    """Configuration for training on the toy  dataset.
    Derives from the base Config class and overrides some values.
    """
    # Give the configuration a recognizable name
    NAME = "circle"

    # We use a GPU with 12GB memory, which can fit two images.
    # Adjust down if you use a smaller GPU.
    IMAGES_PER_GPU = 1

    # Number of classes (including background)
    NUM_CLASSES = 1 + 1  # Background + center ,point

    # Number of training steps per epoch
    STEPS_PER_EPOCH = 100

    # Skip detections with < 90% confidence
    DETECTION_MIN_CONFIDENCE = 0.9

    IMAGE_MIN_DIM = 3072
    IMAGE_MAX_DIM = 3072

    IMAGE_CHANNEL_COUNT = 3

    # Image mean (RGB)
    MEAN_PIXEL = np.array([182,182,182])

############################################################
#  Dataset
############################################################

class CircleDataset(utils.Dataset):

    def load_circle(self, dataset_dir, subset):
        """Load a subset of the Balloon dataset.
        dataset_dir: Root directory of the dataset.
        subset: Subset to load: train or val
        """
        # Add classes. We have only one class to add.
        self.add_class("circle", 1, "center")


        # Train or validation dataset?
        assert subset in ["train", "val"]
        dataset_dir = os.path.join(dataset_dir, subset)


        annotations = json.load(open(os.path.join(dataset_dir, "via_export_json.json"), encoding='UTF-8'))
        annotations = list(annotations.values())  # don't need the dict keys

        # The VIA tool saves images in the JSON even if they don't have any
        # annotations. Skip unannotated images.
        annotations = [a for a in annotations if a['regions']]

        # Add images
        for a in annotations:
            # Get the x, y coordinaets of points of the polygons that make up
            # the outline of each object instance. These are stores in the
            # shape_attributes (see json format above)
            # The if condition is needed to support VIA versions 1.x and 2.x.

            rects = [r['shape_attributes'] for r in a['regions']]
            # print(a['regions'])
            name = [r['region_attributes']['name'] for r in a['regions']]

            # print(name)
            name_dict = {"center": 1,  }
            name_id = [name_dict[a] for a in name]

            # load_mask() needs the image size to convert polygons to masks.
            # Unfortunately, VIA doesn't include it in JSON, so we must read
            # the image. This is only managable since the dataset is tiny.
            image_path = os.path.join(dataset_dir, a['filename'])
            image = skimage.io.imread(image_path,0)
            # print(image.shape)
            height, width = image.shape[:2]
            # print(height,width)
            self.add_image(
                "circle",
                image_id=a['filename'],  # use file name as a unique image id
                path=image_path,
                class_id=name_id,
                width=width, height=height,
                polygons=rects)

    def load_mask(self, image_id):
        """Generate instance masks for an image.
       Returns:
        masks: A bool array of shape [height, width, instance count] with
            one mask per instance.
        class_ids: a 1D array of class IDs of the instance masks.
        """
        # If not a balloon dataset image, delegate to parent class.
        image_info = self.image_info[image_id]
        if image_info["source"] != "circle":
            return super(self.__class__, self).load_mask(image_id)

        # Convert polygons to a bitmap mask of shape
        # [height, width, instance_count]
        name_id = image_info["class_id"]
        # print(name_id)
        info = self.image_info[image_id]
        # print(info)
        mask = np.zeros([info["width"], info["height"], len(info["polygons"])],
                        dtype=np.uint8)
        class_ids = np.array(name_id, dtype=np.int32)
        for i, p in enumerate(info["polygons"]):
            # Get indexes of pixels inside the polygon and set them to 1
            # rr, cc = skimage.draw.rectangle((p['y'], p['x']), extent=(p['height'], p['width']))
            rr, cc = skimage.draw.rectangle((p['y'], p['x']), extent=(p['height'], p['width']))
            mask[rr, cc, i] = 1

        # Return mask, and array of class IDs of each instance. Since we have
        # one class ID only, we return an array of 1s
        return (mask.astype(np.bool), class_ids)

    def image_reference(self, image_id):
        """Return the path of the image."""
        info = self.image_info[image_id]
        if info["source"] == "circle":
            return info["path"]
        else:
            super(self.__class__, self).image_reference(image_id)


def train(model):
    """Train the model."""
    # Training dataset.
    dataset_train = CircleDataset()
    dataset_train.load_circle(args.dataset, "train")
    # print(args.dataset)
    dataset_train.prepare()

    # Validation dataset
    dataset_val = CircleDataset()
    dataset_val.load_circle(args.dataset, "val")
    dataset_val.prepare()

    # *** This training schedule is an example. Update to your needs ***
    # Since we're using a very small dataset, and starting from
    # COCO trained weights, we don't need to train too long. Also,
    # no need to train all layers, just the heads should do it.
    print("Training network heads")
    model.train(dataset_train, dataset_val,
                learning_rate=config.LEARNING_RATE,
                epochs=30,
                layers='all')


def color_splash(image, mask):
    """Apply color splash effect.
    image: RGB image [height, width, 3]
    mask: instance segmentation mask [height, width, instance count]

    Returns result image.
    """
    # Make a grayscale copy of the image. The grayscale copy still
    # has 3 RGB channels, though.
    gray = skimage.color.gray2rgb(skimage.color.rgb2gray(image)) * 255
    # Copy color pixels from the original color image where mask is set
    if mask.shape[-1] > 0:
        # We're treating all instances as one, so collapse the mask into one layer
        mask = (np.sum(mask, -1, keepdims=True) >= 1)
        splash = np.where(mask, image, gray).astype(np.uint8)
    else:
        splash = gray.astype(np.uint8)
    return splash


def detect_and_color_splash(model, image_path=None, video_path=None):
    assert image_path or video_path

    # Image or video?
    if image_path:
        # Run model detection and generate the color splash effect
        print("Running on {}".format(args.image))
        # Read image
        image = skimage.io.imread(args.image)
        # Detect objects
        r = model.detect([image], verbose=1)[0]
        # Color splash
        splash = color_splash(image, r['masks'])
        # Save output
        file_name = "splash_{:%Y%m%dT%H%M%S}.png".format(datetime.datetime.now())
        skimage.io.imsave(file_name, splash)
    elif video_path:
        import cv2
        # Video capture
        vcapture = cv2.VideoCapture(video_path)
        width = int(vcapture.get(cv2.CAP_PROP_FRAME_WIDTH))
        height = int(vcapture.get(cv2.CAP_PROP_FRAME_HEIGHT))
        fps = vcapture.get(cv2.CAP_PROP_FPS)

        # Define codec and create video writer
        file_name = "splash_{:%Y%m%dT%H%M%S}.avi".format(datetime.datetime.now())
        vwriter = cv2.VideoWriter(file_name,
                                  cv2.VideoWriter_fourcc(*'MJPG'),
                                  fps, (width, height))

        count = 0
        success = True
        while success:
            print("frame: ", count)
            # Read next image
            success, image = vcapture.read()
            if success:
                # OpenCV returns images as BGR, convert to RGB
                image = image[..., ::-1]
                # Detect objects
                r = model.detect([image], verbose=0)[0]
                # Color splash
                splash = color_splash(image, r['masks'])
                # RGB -> BGR to save image to video
                splash = splash[..., ::-1]
                # Add image to video writer
                vwriter.write(splash)
                count += 1
        vwriter.release()
    print("Saved to ", file_name)


############################################################
#  Training
############################################################

if __name__ == '__main__':
    import argparse

    # Parse command line arguments
    parser = argparse.ArgumentParser(
        description='Train Mask R-CNN to detect balloons.')
    parser.add_argument("command",
                        metavar="<command>",
                        default="train",
                        help="'train' or 'splash'")
    parser.add_argument('--dataset', required=False,
                        metavar="/path/to/balloon/dataset/",
                        default="C:/Users/Administrator/Downloads/Mask_RCNN-master/datasets/circle",
                        help='Directory of the Balloon dataset')
    parser.add_argument('--weights', required=False,
                        metavar="/path/to/weights.h5",
                        default="coco",
                        help="Path to weights .h5 file or 'coco'")
    parser.add_argument('--logs', required=False,
                        default=DEFAULT_LOGS_DIR,
                        metavar="/path/to/logs/",
                        help='Logs and checkpoints directory (default=logs/)')
    parser.add_argument('--image', required=False,
                        metavar="path or URL to image",
                        help='Image to apply the color splash effect on')
    parser.add_argument('--video', required=False,
                        metavar="path or URL to video",
                        help='Video to apply the color splash effect on')
    args = parser.parse_args()

    # Validate arguments
    if args.command == "train":
        assert args.dataset, "Argument --dataset is required for training"
    elif args.command == "splash":
        assert args.image or args.video, \
            "Provide --image or --video to apply color splash"

    print("Weights: ", args.weights)
    print("Dataset: ", args.dataset)
    print("Logs: ", args.logs)

    # Configurations
    if args.command == "train":
        config = CircleConfig()
    else:
        class InferenceConfig(CircleConfig):
            # Set batch size to 1 since we'll be running inference on
            # one image at a time. Batch size = GPU_COUNT * IMAGES_PER_GPU
            GPU_COUNT = 1
            IMAGES_PER_GPU = 1


        config = InferenceConfig()
    config.display()

    # Create model
    if args.command == "train":
        model = modellib.MaskRCNN(mode="training", config=config,
                                  model_dir=args.logs)
    else:
        model = modellib.MaskRCNN(mode="inference", config=config,
                                  model_dir=args.logs)

    # Select weights file to load
    if args.weights.lower() == "coco":
        weights_path = COCO_WEIGHTS_PATH
        # Download weights file
        if not os.path.exists(weights_path):
            utils.download_trained_weights(weights_path)
    elif args.weights.lower() == "last":
        # Find last trained weights
        weights_path = model.find_last()
    elif args.weights.lower() == "imagenet":
        # Start from ImageNet trained weights
        weights_path = model.get_imagenet_weights()
    else:
        weights_path = args.weights

    # Load weights
    print("Loading weights ", weights_path)


    # Train or evaluate
    if args.command == "train":
        train(model)
    elif args.command == "splash":
        detect_and_color_splash(model, image_path=args.image,
                                video_path=args.video)
    else:
        print("'{}' is not recognized. "
              "Use 'train' or 'splash'".format(args.command))