图像增强库Albumentations使用总结

我不是女神ヾ 2023-01-18 14:21 13阅读 0赞

**目录**

摘要

Albumentations 的 pip 安装

基准测试结果

Spatial-level transforms（空间层次转换）

支持的列表

简单的使用案例

详细使用案例

VerticalFlip 围绕X轴垂直翻转输入。

Blur模糊输入图像

HorizontalFlip 围绕y轴水平翻转输入

Flip水平，垂直或水平和垂直翻转输入。

Transpose， 通过交换行和列来转置输入。

RandomCrop 随机裁剪

RandomGamma 随机灰度系数

RandomRotate90 将输入随机旋转90度，N次

ShiftScaleRotate 随机平移，缩放和旋转输入。

CenterCrop 裁剪图像的中心部分

GridDistortion网格失真

ElasticTransform 弹性变换

RandomGridShuffle把图像切成网格单元随机排列。

HueSaturationValue随机更改图像的颜色，饱和度和值。

PadIfNeeded 填充图像

RGBShift，对图像RGB的每个通道随机移动值。

GaussianBlur 使用随机核大小的高斯滤波器对图像进行模糊处理

CLAHE自适应直方图均衡

ChannelShuffle随机重新排列输入RGB图像的通道。

InvertImg反色

Cutout 随机擦除

RandomFog随机雾化

GridDropout网格擦除

--------------------

# 摘要 #

albumentations包是一种针对数据增强专门写的API，里面基本包含大量的数据增强手段，其特点：

1、Albumentations支持所有常见的计算机视觉任务，如分类、语义分割、实例分割、目标检测和姿态估计。

2、该库提供了一个简单统一的API，用于处理所有数据类型:图像(rbg图像、灰度图像、多光谱图像)、分割掩码、边界框和关键点。

3、该库包含70多种不同的增强功能，可以从现有数据中生成新的训练样本。

4、Albumentations快。我们对每个新版本进行基准测试，以确保增强功能提供最大的速度。

5、它与流行的深度学习框架(如PyTorch和TensorFlow)一起工作。顺便说一下，Albumentations是PyTorch生态系统的一部分。

6、由专家写的。作者既有生产计算机视觉系统的工作经验，也有参与竞争性机器学习的经验。许多核心团队成员是Kaggle Masters和Grandmasters。

7、该库广泛应用于工业、深度学习研究、机器学习竞赛和开源项目。

# Albumentations 的 pip 安装 #

pip install albumentations

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70][]

# **基准测试结果** #

测试使用ImageNet验证集的前2000张图像在Intel Xeon Gold 6140 CPU运行基准测试的结果。所有输出都被转换为带有np的连续NumPy数组。uint8数据类型。表格显示了在单个核上每秒可以处理的图像数量;高越好。

<table> 
 <thead> 
  <tr> 
   <th>albumentations<br> 0.5.0</th> 
   <th>imgaug<br> 0.4.0</th> 
   <th>torchvision (Pillow-SIMD backend)<br> 0.7.0</th> 
   <th>keras<br> 2.4.3</th> 
   <th>augmentor<br> 0.2.8</th> 
   <th>solt<br> 0.1.9</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td>HorizontalFlip</td> 
   <td><strong>9909</strong></td> 
   <td>2821</td> 
   <td>2267</td> 
   <td>873</td> 
   <td>2301</td> 
   <td>6223</td> 
  </tr> 
  <tr> 
   <td>VerticalFlip</td> 
   <td><strong>4374</strong></td> 
   <td>2218</td> 
   <td>1952</td> 
   <td>4339</td> 
   <td>1968</td> 
   <td>3562</td> 
  </tr> 
  <tr> 
   <td>Rotate</td> 
   <td><strong>371</strong></td> 
   <td>296</td> 
   <td>163</td> 
   <td>27</td> 
   <td>60</td> 
   <td>345</td> 
  </tr> 
  <tr> 
   <td>ShiftScaleRotate</td> 
   <td><strong>635</strong></td> 
   <td>437</td> 
   <td>147</td> 
   <td>28</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>Brightness</td> 
   <td><strong>2751</strong></td> 
   <td>1178</td> 
   <td>419</td> 
   <td>229</td> 
   <td>418</td> 
   <td>2300</td> 
  </tr> 
  <tr> 
   <td>Contrast</td> 
   <td><strong>2756</strong></td> 
   <td>1213</td> 
   <td>352</td> 
   <td>-</td> 
   <td>348</td> 
   <td>2305</td> 
  </tr> 
  <tr> 
   <td>BrightnessContrast</td> 
   <td><strong>2738</strong></td> 
   <td>699</td> 
   <td>195</td> 
   <td>-</td> 
   <td>193</td> 
   <td>1179</td> 
  </tr> 
  <tr> 
   <td>ShiftRGB</td> 
   <td><strong>2757</strong></td> 
   <td>1176</td> 
   <td>-</td> 
   <td>348</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>ShiftHSV</td> 
   <td><strong>597</strong></td> 
   <td>284</td> 
   <td>58</td> 
   <td>-</td> 
   <td>-</td> 
   <td>137</td> 
  </tr> 
  <tr> 
   <td>Gamma</td> 
   <td><strong>2844</strong></td> 
   <td>-</td> 
   <td>382</td> 
   <td>-</td> 
   <td>-</td> 
   <td>946</td> 
  </tr> 
  <tr> 
   <td>Grayscale</td> 
   <td><strong>5159</strong></td> 
   <td>428</td> 
   <td>709</td> 
   <td>-</td> 
   <td>1064</td> 
   <td>1273</td> 
  </tr> 
  <tr> 
   <td>RandomCrop64</td> 
   <td><strong>175886</strong></td> 
   <td>3018</td> 
   <td>52103</td> 
   <td>-</td> 
   <td>41774</td> 
   <td>20732</td> 
  </tr> 
  <tr> 
   <td>PadToSize512</td> 
   <td><strong>3418</strong></td> 
   <td>-</td> 
   <td>574</td> 
   <td>-</td> 
   <td>-</td> 
   <td>2874</td> 
  </tr> 
  <tr> 
   <td>Resize512</td> 
   <td>1003</td> 
   <td>634</td> 
   <td><strong>1036</strong></td> 
   <td>-</td> 
   <td>1016</td> 
   <td>977</td> 
  </tr> 
  <tr> 
   <td>RandomSizedCrop_64_512</td> 
   <td><strong>3191</strong></td> 
   <td>939</td> 
   <td>1594</td> 
   <td>-</td> 
   <td>1529</td> 
   <td>2563</td> 
  </tr> 
  <tr> 
   <td>Posterize</td> 
   <td><strong>2778</strong></td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>Solarize</td> 
   <td><strong>2762</strong></td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>Equalize</td> 
   <td>644</td> 
   <td>413</td> 
   <td>-</td> 
   <td>-</td> 
   <td><strong>735</strong></td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>Multiply</td> 
   <td><strong>2727</strong></td> 
   <td>1248</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>MultiplyElementwise</td> 
   <td>118</td> 
   <td><strong>209</strong></td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
  <tr> 
   <td>ColorJitter</td> 
   <td><strong>368</strong></td> 
   <td>78</td> 
   <td>57</td> 
   <td>-</td> 
   <td>-</td> 
   <td>-</td> 
  </tr> 
 </tbody> 
</table>

Python and library versions: Python 3.8.6 (default, Oct 13 2020, 20:37:26) \[GCC 8.3.0\], numpy 1.19.2, pillow-simd 7.0.0.post3, opencv-python 4.4.0.44, scikit-image 0.17.2, scipy 1.5.2.

# Spatial-level transforms（空间层次转换） #

空间级转换将同时改变输入图像和附加目标，如掩模、边界框和关键点。下表显示了每个转换支持哪些附加目标。

<table> 
 <thead> 
  <tr> 
   <th>Transform</th> 
   <th>Image</th> 
   <th>Masks</th> 
   <th>BBoxes</th> 
   <th>Keypoints</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/transforms/#albumentations.augmentations.geometric.transforms.Affine" rel="nofollow">Affine</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.CenterCrop" rel="nofollow">CenterCrop</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.CoarseDropout" rel="nofollow">CoarseDropout</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.Crop" rel="nofollow">Crop</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.CropAndPad" rel="nofollow">CropAndPad</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.CropNonEmptyMaskIfExists" rel="nofollow">CropNonEmptyMaskIfExists</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/transforms/#albumentations.augmentations.geometric.transforms.ElasticTransform" rel="nofollow">ElasticTransform</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Flip" rel="nofollow">Flip</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.GridDistortion" rel="nofollow">GridDistortion</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.GridDropout" rel="nofollow">GridDropout</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.HorizontalFlip" rel="nofollow">HorizontalFlip</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/imgaug/transforms/#albumentations.imgaug.transforms.IAAAffine" rel="nofollow">IAAAffine</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/imgaug/transforms/#albumentations.imgaug.transforms.IAAPiecewiseAffine" rel="nofollow">IAAPiecewiseAffine</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Lambda" rel="nofollow">Lambda</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/resize/#albumentations.augmentations.geometric.resize.LongestMaxSize" rel="nofollow">LongestMaxSize</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.MaskDropout" rel="nofollow">MaskDropout</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/core/transforms_interface/#albumentations.core.transforms_interface.NoOp" rel="nofollow">NoOp</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.OpticalDistortion" rel="nofollow">OpticalDistortion</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.PadIfNeeded" rel="nofollow">PadIfNeeded</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/transforms/#albumentations.augmentations.geometric.transforms.Perspective" rel="nofollow">Perspective</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.RandomCrop" rel="nofollow">RandomCrop</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.RandomCropNearBBox" rel="nofollow">RandomCropNearBBox</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomGridShuffle" rel="nofollow">RandomGridShuffle</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.RandomResizedCrop" rel="nofollow">RandomResizedCrop</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/rotate/#albumentations.augmentations.geometric.rotate.RandomRotate90" rel="nofollow">RandomRotate90</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/resize/#albumentations.augmentations.geometric.resize.RandomScale" rel="nofollow">RandomScale</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.RandomSizedBBoxSafeCrop" rel="nofollow">RandomSizedBBoxSafeCrop</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>&nbsp;</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/crops/transforms/#albumentations.augmentations.crops.transforms.RandomSizedCrop" rel="nofollow">RandomSizedCrop</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/resize/#albumentations.augmentations.geometric.resize.Resize" rel="nofollow">Resize</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/rotate/#albumentations.augmentations.geometric.rotate.Rotate" rel="nofollow">Rotate</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/rotate/#albumentations.augmentations.geometric.rotate.SafeRotate" rel="nofollow">SafeRotate</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/transforms/#albumentations.augmentations.geometric.transforms.ShiftScaleRotate" rel="nofollow">ShiftScaleRotate</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/geometric/resize/#albumentations.augmentations.geometric.resize.SmallestMaxSize" rel="nofollow">SmallestMaxSize</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Transpose" rel="nofollow">Transpose</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
  <tr> 
   <td><a href="https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.VerticalFlip" rel="nofollow">VerticalFlip</a></td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
   <td>✓</td> 
  </tr> 
 </tbody> 
</table>

# 支持的列表 #

*  [Blur][]
 *  [CLAHE][]
 *  [ChannelDropout][]
 *  [ChannelShuffle][]
 *  [ColorJitter][]
 *  [Downscale][]
 *  [Emboss][]
 *  [Equalize][]
 *  [FDA][]
 *  [FancyPCA][]
 *  [FromFloat][]
 *  [GaussNoise][]
 *  [GaussianBlur][]
 *  [GlassBlur][]
 *  [HistogramMatching][]
 *  [HueSaturationValue][]
 *  [ISONoise][]
 *  [ImageCompression][]
 *  [InvertImg][]
 *  [MedianBlur][]
 *  [MotionBlur][]
 *  [MultiplicativeNoise][]
 *  [Normalize][]
 *  [Posterize][]
 *  [RGBShift][]
 *  [RandomBrightnessContrast][]
 *  [RandomFog][]
 *  [RandomGamma][]
 *  [RandomRain][]
 *  [RandomShadow][]
 *  [RandomSnow][]
 *  [RandomSunFlare][]
 *  [RandomToneCurve][]
 *  [Sharpen][]
 *  [Solarize][]
 *  [Superpixels][]
 *  [ToFloat][]
 *  [ToGray][]
 *  [ToSepia][]

# 简单的使用案例 #

import albumentations as A
    import cv2
    
    import matplotlib.pyplot as plt
    
    # Declare an augmentation pipeline
    transform = A.Compose([
        A.RandomCrop(width=512, height=512),
        A.HorizontalFlip(p=0.8),
        A.RandomBrightnessContrast(p=0.5),
    ])
    
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    
    # Augment an image
    transformed = transform(image=image)
    transformed_image = transformed["image"]
    plt.imshow(transformed_image)
    plt.show()

原始图像：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 1][]

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 2][]

# 详细使用案例 #

*  ## VerticalFlip 围绕X轴垂直翻转输入。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.VerticalFlip(always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('Blur后的图像')
    plt.imshow(transformed_image)
    plt.show()

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 3][]

*  ## Blur模糊输入图像 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.Blur(blur_limit=15,always_apply=False, p=1)(image=image) 
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('Blur后的图像')
    plt.imshow(transformed_image)
    plt.show()

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 4][]

*  ## HorizontalFlip 围绕y轴水平翻转输入 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.HorizontalFlip(always_apply=False, p=1)(image=image) 
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('HorizontalFlip后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 5][]

*  ## Flip水平，垂直或水平和垂直翻转输入。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.Flip(always_apply=False, p=1)(image=image) 
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('Flip后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果有一定的随机性，如下图：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 6][]

*  ## Transpose， 通过交换行和列来转置输入。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.Transpose(always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('Transpose后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 7][]

*  ## RandomCrop 随机裁剪 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.RandomCrop(512, 512,always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('RandomCrop后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 8][]

*  ##  RandomGamma 随机灰度系数 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.RandomGamma(gamma_limit=(20, 20), eps=None, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('RandomGamma后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 9][]

*  ## RandomRotate90 将输入随机旋转90度，N次 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.RandomRotate90(always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('RandomRotate90后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 10][]

*  ## ShiftScaleRotate 随机平移，缩放和旋转输入。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    #解决中文显示问题
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.ShiftScaleRotate(shift_limit=0.0625, scale_limit=0.1, rotate_limit=45, interpolation=1, border_mode=4, value=None, mask_value=None, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')   #第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title('ShiftScaleRotate后的图像')
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 11][]

*  ## CenterCrop 裁剪图像的中心部分 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.CenterCrop(256, 256, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("CenterCrop后的图像")
    plt.imshow(transformed_image)
    plt.show()

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 12][]

*  ## GridDistortion网格失真 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.GridDistortion(num_steps=10, distort_limit=0.3,border_mode=4, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("GridDistortion后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 13][]

*  ## ElasticTransform 弹性变换 ##
 *      import albumentations as A
        import cv2
        import numpy as np
        import matplotlib.pyplot as plt
        
        # 解决中文显示问题
        plt.rcParams['font.sans-serif'] = ['SimHei']
        plt.rcParams['axes.unicode_minus'] = False
        # Read an image with OpenCV and convert it to the RGB colorspace
        image = cv2.imread("aa.jpg")
        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
        # Augment an image
        transformed = A.ElasticTransform(alpha=5, sigma=50, alpha_affine=50, interpolation=1, border_mode=4,always_apply=False, p=1)(image=image)
        transformed_image = transformed["image"]
        plt.subplot(1, 2, 1)
        plt.title('原图')  # 第一幅图片标题
        plt.imshow(image)
        plt.subplot(1, 2, 2)
        plt.title("ElasticTransform后的图像")
        plt.imshow(transformed_image)
        plt.show()
    
    运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 14][]

*  ## RandomGridShuffle把图像切成网格单元随机排列。 ##
 *      import albumentations as A
        import cv2
        import numpy as np
        import matplotlib.pyplot as plt
        
        # 解决中文显示问题
        plt.rcParams['font.sans-serif'] = ['SimHei']
        plt.rcParams['axes.unicode_minus'] = False
        # Read an image with OpenCV and convert it to the RGB colorspace
        image = cv2.imread("aa.jpg")
        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
        # Augment an image
        transformed = A.RandomGridShuffle(grid=(3, 3), always_apply=False, p=1) (image=image)
        transformed_image = transformed["image"]
        plt.subplot(1, 2, 1)
        plt.title('原图')  # 第一幅图片标题
        plt.imshow(image)
        plt.subplot(1, 2, 2)
        plt.title("RandomGridShuffle后的图像")
        plt.imshow(transformed_image)
        plt.show()
    
    运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 15][]

*  ## HueSaturationValue随机更改图像的颜色，饱和度和值。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.HueSaturationValue(hue_shift_limit=20, sat_shift_limit=30, val_shift_limit=20, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("HueSaturationValue后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 16][]

*  ## PadIfNeeded 填充图像 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.PadIfNeeded(min_height=2048, min_width=2048, border_mode=4, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("PadIfNeeded后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：  
![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 17][]

*  ## RGBShift，对图像RGB的每个通道随机移动值。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.RGBShift(r_shift_limit=10, g_shift_limit=20, b_shift_limit=20, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("RGBShift后的图像")
    plt.imshow(transformed_image)
    plt.show()

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 18][]

*  ## GaussianBlur 使用随机核大小的高斯滤波器对图像进行模糊处理 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.GaussianBlur(blur_limit=11, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("GaussianBlur后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 19][]

*  ## CLAHE自适应直方图均衡 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.CLAHE(clip_limit=4.0, tile_grid_size=(8, 8), always_apply=False, p=0.5)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("CLAHE后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果:

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 20][]

*  ## ChannelShuffle随机重新排列输入RGB图像的通道。 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.ChannelShuffle(always_apply=False, p=0.5)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("ChannelShuffle后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 21][]

*  ## InvertImg反色 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.InvertImg(always_apply=False, p=0.5)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("InvertImg后的图像")
    plt.imshow(transformed_image)
    plt.show()

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 22][]

*  ## Cutout 随机擦除 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.Cutout(num_holes=20, max_h_size=20, max_w_size=20, fill_value=0, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("Cutout后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 23][]

*  ## RandomFog随机雾化 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.RandomFog(fog_coef_lower=0.3, fog_coef_upper=1, alpha_coef=0.08, always_apply=False, p=1)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("RandomFog后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 24][]

*  ## GridDropout网格擦除 ##

import albumentations as A
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    
    # 解决中文显示问题
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # Read an image with OpenCV and convert it to the RGB colorspace
    image = cv2.imread("aa.jpg")
    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    # Augment an image
    transformed = A.GridDropout(ratio=0.5, unit_size_min=None, unit_size_max=None, holes_number_x=None, holes_number_y=None,
                                shift_x=0, shift_y=0, always_apply=False, p=0.5)(image=image)
    transformed_image = transformed["image"]
    plt.subplot(1, 2, 1)
    plt.title('原图')  # 第一幅图片标题
    plt.imshow(image)
    plt.subplot(1, 2, 2)
    plt.title("GridDropout后的图像")
    plt.imshow(transformed_image)
    plt.show()

运行结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 25][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 26][]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70]: /images/20221021/10d117c148be4a03847d637f1d8c89be.png
[Blur]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Blur
[CLAHE]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.CLAHE
[ChannelDropout]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ChannelDropout
[ChannelShuffle]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ChannelShuffle
[ColorJitter]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ColorJitter
[Downscale]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Downscale
[Emboss]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Emboss
[Equalize]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Equalize
[FDA]: https://albumentations.ai/docs/api_reference/augmentations/domain_adaptation/#albumentations.augmentations.domain_adaptation.FDA
[FancyPCA]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.FancyPCA
[FromFloat]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.FromFloat
[GaussNoise]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.GaussNoise
[GaussianBlur]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.GaussianBlur
[GlassBlur]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.GlassBlur
[HistogramMatching]: https://albumentations.ai/docs/api_reference/augmentations/domain_adaptation/#albumentations.augmentations.domain_adaptation.HistogramMatching
[HueSaturationValue]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.HueSaturationValue
[ISONoise]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ISONoise
[ImageCompression]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ImageCompression
[InvertImg]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.InvertImg
[MedianBlur]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.MedianBlur
[MotionBlur]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.MotionBlur
[MultiplicativeNoise]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.MultiplicativeNoise
[Normalize]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Normalize
[Posterize]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Posterize
[RGBShift]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RGBShift
[RandomBrightnessContrast]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomBrightnessContrast
[RandomFog]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomFog
[RandomGamma]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomGamma
[RandomRain]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomRain
[RandomShadow]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomShadow
[RandomSnow]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomSnow
[RandomSunFlare]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomSunFlare
[RandomToneCurve]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomToneCurve
[Sharpen]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Sharpen
[Solarize]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Solarize
[Superpixels]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Superpixels
[ToFloat]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ToFloat
[ToGray]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ToGray
[ToSepia]: https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.ToSepia
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 1]: /images/20221021/35f81715090646f48ff23f5305b7dd6f.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 2]: /images/20221021/19beba69d1df4572b8a0e6e9591d2c80.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 3]: /images/20221021/5adca07691144728824691e71d8c06b7.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 4]: /images/20221021/09c18e5a0fdc4760b94d82da76b01139.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 5]: /images/20221021/719b4edad3194078beb28680ecdd4071.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 6]: /images/20221021/0327237ed31f442a963fd3a4e8b65b18.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 7]: /images/20221021/231ac50742b14455b7c5847b179ea8f0.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 8]: /images/20221021/cf253a1043c74073a7f5f38dc984370b.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 9]: /images/20221021/bec31fc60b834aeba6cf9a5e47a2c6f1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 10]: /images/20221021/4fdba5b93b854746aa62a8bcdbd94a26.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 11]: /images/20221021/b9205558218a4be8a2762c52a13b741e.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 12]: /images/20221021/99ede0d80eb447f0abe8feb4743bbb50.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 13]: /images/20221021/5c7cc5f54e1a46a6a40be917f759a03d.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 14]: /images/20221021/aa41759e5870461d911afe652d299b6e.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 15]: /images/20221021/ae8656dce61c4ba6abd06edd1148be92.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 16]: /images/20221021/4781c83ddef241a98a53e6a7d6442096.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 17]: /images/20221021/7ea73fd86efc47a19301f89de922037b.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 18]: /images/20221021/5f8f976062dc4647af0c130083643e29.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 19]: /images/20221021/1c3aa2d60f4c4959a626d200ded971b1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 20]: /images/20221021/6dd6c58d49f145e0a9056d591d97383e.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 21]: /images/20221021/e85efae019e049b99694ac15309d5b2a.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 22]: /images/20221021/ef0b98d3d06e4f28a65453a9b079bd3d.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 23]: /images/20221021/c2c5d52607774378ae8b3e5876b0e51b.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 24]: /images/20221021/299920c34d8e465992e0c0f2ce71445b.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 25]: /images/20221021/b2f11d658f4440d188ba4e27c484ae6c.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2hoaGhoaGhoaGh3d3d3d3d3d3d3_size_16_color_FFFFFF_t_70 26]: /images/20221021/d3431674edca423e9b0c4372211c72eb.png