random_erasing.py

from __future__ import absolute_import

from torchvision.transforms import *

from PIL import Image
import random
import math
import numpy as np
from torch import nn


class ColorJitter(object):
    def __init__(self):
        self.color_jitter = transforms.RandomChoice([
            transforms.ColorJitter(brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5),
        ])

    def __call__(self, image):
        h, w = 288, 144
        mask = np.ones((h, w, 3), dtype=np.uint8)
        for i in range(h//36):
            for j in range(h//36):
                if (i + j)%2==1:
                    mask[i*36:i*36+36,j*36:j*36+36, :] = 0
                else:
                    mask[i*36:i*36+36,j*36:j*36+36, :] = 1
    
        img = self.color_jitter(image) * mask + image * (1 - mask)
        #img = image * mask + image * (1 - mask)
        return img


class RandomErasing(object):
    """ Randomly selects a rectangle region in an image and erases its pixels.
        'Random Erasing Data Augmentation' by Zhong et al.
        See https://arxiv.org/pdf/1708.04896.pdf
    Args:
         probability: The probability that the Random Erasing operation will be performed.
         sl: Minimum proportion of erased area against input image.
         sh: Maximum proportion of erased area against input image.
         r1: Minimum aspect ratio of erased area.
         mean: Erasing value. 
    """
    
    def __init__(self, probability = 0.5, sl = 0.02, sh = 0.4, r1 = 0.3, mean=[0.4914, 0.4822, 0.4465]):
        self.probability = probability
        self.mean = mean
        self.sl = sl
        self.sh = sh
        self.r1 = r1
       
    def __call__(self, img):

        if random.uniform(0, 1) > self.probability:
            return img

        for attempt in range(100):
            area = img.size()[1] * img.size()[2]
       
            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)))

            if w < img.size()[2] and h < img.size()[1]:
                x1 = random.randint(0, img.size()[1] - h)
                y1 = random.randint(0, img.size()[2] - w)
                if img.size()[0] == 3:
                    img[0, x1:x1+h, y1:y1+w] = self.mean[0]
                    img[1, x1:x1+h, y1:y1+w] = self.mean[1]
                    img[2, x1:x1+h, y1:y1+w] = self.mean[2]
                else:
                    img[0, x1:x1+h, y1:y1+w] = self.mean[0]
                return img

        return img