图像分类——OpenCV常用图像增强方法

图像增广

常用图像增广方法主要有：左右翻转(上下翻转对于许多目标并不常用)，随机裁剪，变换颜色(亮度，对比度，饱和度和色调)等等，拟用opencv-python实现部分数据增强方法。

结构如下：

class FunctionClass:
    def __init__(self, parameter):
        self.parameter=parameter

    def __call__(self, img):       

import cv2
import numpy as np
from matplotlib import pyplot as plt
%matplotlib inline

filename = 'work/1.jpg'
## [Load an image from a file]
img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
plt.imshow(img)

print(img.shape)

(350, 350, 3)

1.图片缩放

class Resize:
    def __init__(self, size):
        self.size=size

    def __call__(self, img):

        # 此处插入代码
        return cv2.resize(img, self.size)


resize=Resize( (600, 600))
img2=resize(img)
plt.imshow(img2)

2.图片翻转

class Flip:
    def __init__(self, mode):
        self.mode=mode

    def __call__(self, img):

        # 此处插入代码
        return cv2.flip(img, self.mode)

img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
flip=Flip(mode=0)
img2=flip(img)
plt.imshow(img2)

3图片旋转

class Rotate:
    def __init__(self, degree,size):
        self.degree=degree
        self.size=size

    def __call__(self, img):

        # 此处插入代码
        rows, cols = img.shape[:2]
        M = cv2.getRotationMatrix2D((cols / 2, rows / 2), 45, self.size)
        return cv2.warpAffine(img, M, (cols, rows))


img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
rotate=Rotate( 45, 0.7)
img2=rotate(img)
plt.imshow(img2)

4.图片亮度调节

class Brightness:
    def __init__(self,brightness_factor):
        self.brightness_factor=brightness_factor

    def __call__(self, img):

        # 此处插入代码
        return np.uint8(np.clip((self.brightness_factor * img + 125*(1-self.brightness_factor)), 0, 255))



img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
brightness=Brightness(0.6)
img2=brightness(img)
plt.imshow(img2)

5.图片随机擦除

import random
import math

class RandomErasing(object):
    def __init__(self, EPSILON=0.5, sl=0.02, sh=0.4, r1=0.3,
                 mean=[0., 0., 0.]):
        self.EPSILON = EPSILON
        self.mean = mean
        self.sl = sl
        self.sh = sh
        self.r1 = r1

    def __call__(self, img):
        if random.uniform(0, 1) > self.EPSILON:
            return img

        for attempt in range(100):
            area = img.shape[0] * img.shape[1]

            target_area = random.uniform(self.sl, self.sh) * area
            aspect_ratio = random.uniform(self.r1, 1 / self.r1)

            h = int(round(math.sqrt(target_area * aspect_ratio)))
            w = int(round(math.sqrt(target_area / aspect_ratio)))

            
            # 此处插入代码
            if w < img.shape[0] and h < img.shape[1]:
                x1 = random.randint(0, img.shape[1] - h)
                y1 = random.randint(0, img.shape[0] - w)
                if img.shape[2] == 3:
                    img[ x1:x1 + h, y1:y1 + w, 0] = self.mean[0]
                    img[ x1:x1 + h, y1:y1 + w, 1] = self.mean[1]
                    img[ x1:x1 + h, y1:y1 + w, 2] = self.mean[2]
                else:
                    img[x1:x1 + h, y1:y1 + w,0] = self.mean[0]
                return img

            return img

img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
erase = RandomErasing()
img2=erase(img)
plt.imshow(img2)    

6.图片随机裁剪

class RandCropImage(object):
    def __init__(self, size, scale=None, ratio=None, interpolation=-1):

        self.interpolation = interpolation if interpolation >= 0 else None
        if type(size) is int:
            self.size = (size, size)  # (h, w)
        else:
            self.size = size

        self.scale = [0.08, 1.0] if scale is None else scale
        self.ratio = [3. / 4., 4. / 3.] if ratio is None else ratio

    def __call__(self, img):
        size = self.size
        scale = self.scale
        ratio = self.ratio

        aspect_ratio = math.sqrt(random.uniform(*ratio))
        w = 1. * aspect_ratio
        h = 1. / aspect_ratio

        img_h, img_w = img.shape[:2]

        bound = min((float(img_w) / img_h) / (w**2),
                    (float(img_h) / img_w) / (h**2))
        scale_max = min(scale[1], bound)
        scale_min = min(scale[0], bound)

        target_area = img_w * img_h * random.uniform(scale_min, scale_max)
        target_size = math.sqrt(target_area)
        w = int(target_size * w)
        h = int(target_size * h)

        i = random.randint(0, img_w - w)
        j = random.randint(0, img_h - h)

        img = img[j:j + h, i:i + w, :]
        if self.interpolation is None:
            return cv2.resize(img, size)
        else:
            return cv2.resize(img, size, interpolation=self.interpolation)

img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
crop = RandCropImage(350)
size, interpolation=self.interpolation)

img = cv2.imread(filename)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
crop = RandCropImage(350)
plt.imshow(crop(img))

相关链接: 飞桨领航团图像分类零基础训练营