Python----图像数据增强翻转变换规则修剪高斯模糊随机旋转直方图均值化

╰+攻爆jí腚メ 2022-11-02 06:02 8阅读 0赞

import math
    import os
    import numpy as np
    import cv2
    import random
    from tqdm import tqdm
    import logging
    
    logger = logging.getLogger(__name__)
    class DataAugmentation:
        def rule_crop(img, box_ratio=(3. / 4, 3. / 4), location_type='LT', resize_w=256, resize_h=256):
            """
            按照一定规则进行裁剪, 直接在原图尺寸上操作，不对原图进行
            :param img:
            :param box_ratio: 剪切的 比例：  （宽度上的比例， 高度上的比例）
            :param location_type: 具体在=哪个位置： 以下其中一个：
                    LR : 左上角
                    RT : 右上角
                    LB : 左下角
                    RB : 右下角
                    CC : 中心
            :param resize_w: 输出图的width
            :param resize_h: 输出图的height
            :return:
            """
            assert location_type in ('LT', 'RT', 'LB', 'RB', 'CC'), 'must have a location .'
            is_gray = False
            if len(img.shape) == 3:
                h, w, c = img.shape
            elif len(img.shape) == 2:
                h, w = img.shape
                is_gray = True
    
            crop_w, crop_h = int(w * box_ratio[0]), int(h * box_ratio[1])
            crop_img = np.zeros([10, 10])
            if location_type == 'LT':
                crop_img = img[:crop_h, :crop_w, :] if not is_gray else img[:crop_h, :crop_w]
            elif location_type == 'RT':
                crop_img = img[:crop_h:, w - crop_w:, :] if not is_gray else img[:crop_h:, w - crop_w:]
            elif location_type == 'LB':
                crop_img = img[h - crop_h:, :crop_w, :] if not is_gray else img[h - crop_h:, :crop_w]
            elif location_type == 'RB':
                crop_img = img[h - crop_h:, w - crop_w:, :] if not is_gray else img[h - crop_h:, w - crop_w:]
            elif location_type == 'CC':
                start_h = (h - crop_h) // 2
                start_w = (w - crop_w) // 2
                crop_img = img[start_h:start_h + crop_h, start_w:start_w + crop_w, :] if not is_gray else img[
                                                                                                          start_h:start_h + crop_h,
                                                                                                          start_w:start_w + crop_w]
            resize = cv2.resize(crop_img, (resize_w, resize_h))
            return resize
    
        def flip_Horizontal (img):
            """
            水平翻转
            :param img:
            :return:
            """
            flip = np.random.choice(2) * 2 - 1  # -1 / 1
            img = img[:, ::flip, :]
            return img
        def flip_Vertical (img):
            """
            垂直翻转
            :param img:
            :return:
            """
            flip = np.random.choice(2) * 2 - 1  # -1 / 1
            img = img[::flip, :, :]
            return img
        def flip_Horizontal_Vertical(img):
            """
            水平垂直
            :param img:
            :return:
            """
            flip = np.random.choice(2) * 2 - 1  # -1 / 1
            img = img[::flip, ::flip, :]
            return img
    
        # 随机锐化增强
        def random_USM(img, gamma=0.):
            """
            USM锐化增强算法可以去除一些细小的干扰细节和图像噪声，比一般直接使用卷积锐化算子得到的图像更可靠。
                output = 原图像−w∗高斯滤波(原图像)/(1−w)
                其中w为上面所述的系数，取值范围为0.1~0.9，一般取0.6。
            :param img:
            :param gamma:
            :return:
            """
            blur = cv2.GaussianBlur(img, (0, 0), 25)
            img_sharp = cv2.addWeighted(img, 1.5, blur, -0.3, gamma)
            return img_sharp
    
        def random_noise(img, rand_range=(3, 20)):
            """
            随机噪声
            :param img:
            :param rand_range: (min, max)
            :return:
            """
            img = np.asarray(img, np.float)
            sigma = random.randint(*rand_range)
            nosie = np.random.normal(0, sigma, size=img.shape)
            img += nosie
            img = np.uint8(np.clip(img, 0, 255))
            return img
    
        def gaussianBlue(img, ks=(7, 7), stdev=1.5):
            """
            高斯模糊, 可以对图像进行平滑处理，去除尖锐噪声
            :param img:
            :param ks:  卷积核
            :param stdev: 标准差
            :return:
            """
            return cv2.GaussianBlur(img, (7, 7), 1.5)
    
        def equalize_hist(img):
            """
            直方图均衡化
            :param img:
            :return:
            """
            gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
            hist = cv2.equalizeHist(gray)
            rgb = cv2.cvtColor(hist, cv2.COLOR_GRAY2RGB)
            return rgb
    
        def random_rotate(img, angle_range=(-10, 10)):
            """
            随机旋转
            :param img:
            :param angle_range:  旋转角度范围 (min,max)   >0 表示逆时针，
            :return:
            """
            height, width = img.shape[:2]  # 获取图像的高和宽
            center = (width / 2, height / 2)  # 取图像的中点
            angle = random.randrange(*angle_range, 1)
            M = cv2.getRotationMatrix2D(center, angle, 1.0)  # 获得图像绕着某一点的旋转矩阵
            # cv2.warpAffine()的第二个参数是变换矩阵,第三个参数是输出图像的大小
            rotated = cv2.warpAffine(img, M, (height, width))
            return rotated
    
        # 颜色抖动(亮度\色度\饱和度\对比度)  color
        def adjust_contrast_bright(img, contrast=1.2, brightness=100):
            """
            调整亮度与对比度
            dst = img * contrast + brightness
            :param img:
            :param contrast: 对比度   越大越亮
            :param brightness: 亮度  0~100
            :return:
            """
            # 像素值会超过0-255， 因此需要截断
            return np.uint8(np.clip((contrast * img + brightness), 0, 255))
    def imageOps(func_name,dir_path,save_path):
        funcMap = {"crop": DataAugmentation.rule_crop,
                    "flip_Horizontal": DataAugmentation.flip_Horizontal,
                   "flip_Vertical": DataAugmentation.flip_Vertical,
                   "flip_Horizontal_Vertical": DataAugmentation.flip_Horizontal_Vertical,
                   "random_USM":DataAugmentation.random_USM,
                   "random_noise":DataAugmentation.random_noise,
                   "gaussianBlue":DataAugmentation.gaussianBlue,
                    "equalize_hist":DataAugmentation.equalize_hist,
                    "random_rotate":DataAugmentation.random_rotate,
                   "adjust_contrast_bright":DataAugmentation.adjust_contrast_bright}
        if funcMap.get(func_name) is None:
            logger.error("%s is not exist", func_name)
            return -1
        i=0
        for root, dirs, files in os.walk(dir_path):
            # print(root,dirs,files)
            for Ufile in tqdm(files):
                # Ufile是文件名
                img_path = os.path.join(root, Ufile)  # 文件的所在路径
                # print(img_path)
                # File = root.split('/')[-1]  # 文件所在文件夹的名字, 也就是label
                # print(File)
                # 读取image和label数据
                img_data = cv2.imread(img_path)
                # print(save_img_path)
                img_data = funcMap[func_name](img_data)
                i=i+1
                save_img_path = os.path.join(save_path, str(Ufile.split(".")[0]) + str(func_name) + ".jpg")
                print(save_img_path)
                cv2.imwrite(save_img_path,img_data)
    opsList = {"crop","flip_Horizontal","flip_Vertical","flip_Horizontal_Vertical","random_USM","random_noise",
               "gaussianBlue","equalize_hist","random_rotate","adjust_contrast_bright"}
    
    dir_path = r"C:\Users\admin\Desktop\Machine Learning\flowers\npy_data\Gray_Flowers\train\Anemone"
    save_path = r"C:\Users\admin\Desktop\Machine Learning\flowers\npy_data\Gray_Flowers\train\A"
    for ops_name in opsList:
        func_name = ops_name
        imageOps(func_name,dir_path,save_path)