语义分割 图像增强（数据扩充）教程

参考博客1：https://blog.csdn.net/qianbin3200896/article/details/88656524?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-3.channel_param&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-3.channel_param
参考博客2：https://blog.csdn.net/xuhe93/article/details/105826868

博客1的方法很简单：pip install Augmentor 在本地cmd终端下输入，第二点呢 这是默认安装在你pycharm中的默认下环境中，我的默认是3.7py，所以我得切换到3.7下的环境执行。

 #导入数据增强工具
import Augmentor
 
#确定原始图像存储路径以及掩码文件存储路径
p = Augmentor.Pipeline("test1")
p.ground_truth("test2")
 
#图像旋转： 按照概率0.8执行，最大左旋角度10，最大右旋角度10
p.rotate(probability=0.8, max_left_rotation=10, max_right_rotation=10)
 
#图像左右互换： 按照概率0.5执行
p.flip_left_right(probability=0.5)
 
#图像放大缩小： 按照概率0.8执行，面积为原始图0.85倍
p.zoom_random(probability=0.3, percentage_area=0.85)
 
#最终扩充的数据样本数
p.sample(20)

执行后，最终在test1文件目录下会生成一个output文件夹，该文件夹中即为增强过后的图片，如下图所示：

标签和转换后的图有编号，可以对应上。主要是旋转，对称上下左右翻转，缩放。

在看看第二种方法：代码源博客不全，参考我下面的代码把：

'''''


#2   数据增强代码测试
import glob
from keras.preprocessing.image import ImageDataGenerator,load_img,img_to_array,array_to_img
import os
import Augmentor
import cv2

class myAugmentation(object):
    """
    一个用于图像增强的类：
    首先：分别读取训练的图片和标签，然后将图片和标签合并用于下一个阶段使用
    然后：使用Keras的预处理来增强图像
    最后：将增强后的图片分解开，分为训练图片和训练标签
    """

    def __init__(self, train_path="D:/shujuzengqiang/1", label_path="D:/shujuzengqiang/2", merge_path="D:/shujuzengqiang/merge", aug_merge_path="D:/shujuzengqiang/aug_merge",
                 aug_train_path="D:/shujuzengqiang/aug_train", aug_label_path="D:/shujuzengqiang/aug_label"):

        """
        使用glob从路径中得到所有的“.img_type”文件，初始化类：__init__()
        """
        self.train_imgs = glob.glob(train_path + "/*" )
        self.label_imgs = glob.glob(label_path + "/*" )
        self.train_path = train_path
        self.label_path = label_path
        self.merge_path = merge_path
        self.aug_merge_path = aug_merge_path
        self.aug_train_path = aug_train_path
        self.aug_label_path = aug_label_path
        self.slices = len(self.train_imgs)
        self.datagen = ImageDataGenerator(
            rotation_range=180,
            width_shift_range=0.2,
            height_shift_range=0.2,
            shear_range=0.05,
            zoom_range=0.1,
            horizontal_flip=True,
            vertical_flip=True,
            fill_mode='nearest')
        if not os.path.exists(self.merge_path):
            os.mkdir(self.merge_path)
        if not os.path.exists(self.aug_merge_path):
            os.mkdir(self.aug_merge_path)
        if not os.path.exists(self.aug_label_path):
            os.mkdir(self.aug_label_path)
        if not os.path.exists(self.aug_train_path):
            os.mkdir(self.aug_train_path)

    def Augmentation(self):

        """
        Start augmentation.....
        """
        trains = self.train_imgs
        labels = self.label_imgs
        path_train = self.train_path
        path_label = self.label_path
        path_merge = self.merge_path
        path_aug_merge = self.aug_merge_path
        if len(trains) != len(labels) or len(trains) == 0 or len(trains) == 0:
            print("trains can't match labels")
            return 0
        for i in range(len(trains)):
            img_t = load_img(trains[i])
            img_l = load_img(labels[i])
            x_t = img_to_array(img_t)
            x_l = img_to_array(img_l)

            x_t[:, :, 2] = x_l[:, :, 0]
            img_tmp = array_to_img(x_t)
            img_tmp.save(path_merge + "/" + str(i) + ".png")
            img = x_t
            img = img.reshape((1,) + img.shape)
            savedir = path_aug_merge + "/" + str(i)
            if not os.path.lexists(savedir):
                os.mkdir(savedir)
            self.doAugmentate(img, savedir, str(i))

    def doAugmentate(self, img, save_to_dir, save_prefix, batch_size=1, save_format='png', imgnum=10):
        # 增强一张图片的方法
        """
        augmentate one image
        """
        datagen = self.datagen
        i = 0
        for batch in datagen.flow(img,
                                  batch_size=batch_size,
                                  save_to_dir=save_to_dir,
                                  save_prefix=save_prefix,
                                  save_format=save_format):
            i += 1
            if i >= imgnum:
                break

    def splitMerge(self):
        # 将合在一起的图片分开
        """
        split merged image apart
        """
        path_merge = self.aug_merge_path
        path_train = self.aug_train_path
        path_label = self.aug_label_path

        for i in range(self.slices):
            path = path_merge + "/" + str(i)
            train_imgs = glob.glob(path + "/*.png")
            savedir = path_train + "/" + str(i)
            if not os.path.lexists(savedir):
                os.mkdir(savedir)
            savedir = path_label + "/" + str(i)

            if not os.path.lexists(savedir):
                os.mkdir(savedir)
            for imgname in train_imgs:

                midname = imgname[imgname.rindex("\\") + 1:]
                img = cv2.imread(imgname)
                img_train = img[:, :, 2]  # cv2 read image rgb->bgr
                img_label = img[:, :, 0]
                cv2.imwrite(path_train + "/" + str(i) + "/" + midname, img_train)
                cv2.imwrite(path_label + "/" + str(i) + "/" + midname, img_label)


if __name__ == "__main__":
    aug = myAugmentation()
    aug.Augmentation()
    aug.splitMerge()


'''''

里面只需要改路径就行，看看每个文件夹代表什么。每个图片都生成自己单独的文件夹，里面是这个图片的一系列变化，


主要的效果就是各种角度的旋转，平移，
def doAugmentate(self, img, save_to_dir, save_prefix, batch_size=1, save_format=‘tif’, imgnum=10):
修改10 可以变换生成的图像数目。
当然数据增强是针对数据集较小的时候，很多标注的数据集一般都几百张，数据增强可以有效提高数据量，可以扩充训练数据集。但也并非万能的，有时过度信任数据增强会带来负面效果，还会增加网络训练时间。需酌情使用。

阅读参考3：https://blog.csdn.net/JNingWei/article/details/79219838?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.edu_weight&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.edu_weight