图像均匀裁剪并修改对应的labelme标注文件

因为在大图上进行使用labelme进行了标注，但在实际训练过程中发现大图（大概4000x4000）训练的结果并不好，小目标尺寸相比之下非常小（50x50），所以考虑将大图进行裁剪后再进行训练，同时根据已经标注好的json格式文件生成对应的小图的标注。

#将一整张大图切分成多张小图并修改对应的json标注

import cv2
import json
import os

# 读取大图的JSON文件
def splitImgandJson(IPath_json, IPath_img, OPath_json, OPath_img):

    events = os.listdir(IPath_json)

    #循环处理每一张图像和标注
    for onevent in events:
        name = onevent.split('.json')[0]
        img_name = name + '.jpg'

        with open(os.path.join(IPath_json, onevent), 'r') as json_file:
            large_json_data = json.load(json_file)

        # 读取大图的JPEG图像
        large_image = cv2.imread(os.path.join(IPath_img, img_name))

        # 定义每个小图的大小
   
        crop_width, crop_height = 500, 500  # 可以根据需要调整大小

        # 循环裁剪图像并生成新的JSON文件
        L_H = large_image.shape[0]
        L_W = large_image.shape[1]
        for i in range(0, L_H, crop_height):
            for j in range(0, L_W, crop_width):
                # 裁剪图像
                small_image = large_image[i:i + crop_height, j:j + crop_width]

                # 创建新的JSON数据，基于大图的JSON数据进行调整
                small_json_data = large_json_data.copy()

                # 遍历大图标注看是否存在在小图中
                new_target = []  # 替换'shapes'
                for items in large_json_data[('shapes')]:
                    # 提取边界框坐标
                    x1, y1 = items["points"][0]
                    x2, y2 = items["points"][1]

                    # 判断目标框在小图中的位置
                    # 1、目标框全在小图中
                    new_item = items.copy()
                    if (j <= x1) and (x2 <= j + crop_width) and (i <= y1) and (y2 <= i + crop_height):
                        new_item['points'] = [[x1 - j, y1 - i], [x2 - j, y2 - i]]
                        new_target.append(new_item)
                        continue
                    # 2、某一点在小图中并且所占面积超过目标面积的40%（比例可以根据需要调整）
                    elif ((j < x1 < j + crop_width) and (i < y1 < i + crop_height)) or (
                            (j < x2 < j + crop_width) and (i < y1 < i + crop_height)) or (
                            (j < x1 < j + crop_width) and (i < y2 < i + crop_height)) or ((
                            j < x2 < j + crop_width) and (i < y2 < i + crop_height)):
                        x_min = max(j, x1)-j
                        y_min = max(i, y1)-i
                        x_max = min(x2, j + crop_width)-j
                        y_max = min(y2, i + crop_height)-i
                        if (x_max - x_min) * (y_max - y_min) > 0.4 * (x2 - x1) * (y2 - y1):
                            new_item['points'] = [[x_min, y_min], [x_max, y_max]]
                            new_target.append(new_item)
                            continue
                    else:
                        continue
                if new_target != []:
                    small_json_data['shapes'] = new_target
                    small_json_data["imageWidth"] = crop_width
                    small_json_data["imageHeight"] = crop_height

                    # 保存小图的JSON文件
                    small_json_filename = f'{name}_{i}_{j}.json'
                    with open(os.path.join(OPath_json, small_json_filename), 'w') as small_json_file:
                        json.dump(small_json_data, small_json_file, indent=4)

                    # 保存小图的JPEG图像
                    small_image_filename = f'{name}_{i}_{j}.jpg'
                    cv2.imwrite(os.path.join(OPath_img, small_image_filename), small_image)

                # 完成后，你会得到多个小图像和对应的JSON文件



if __name__ == '__main__':
    IPath_json = 'D:\Projects\Python\dataB\Json'
    IPath_img = 'D:\Projects\Python\dataB\JPEGImgs'
    OPath_json = 'D:\Projects\Python\dataB\json_s500'
    OPath_img = 'D:\Projects\Python\dataB\img_s500'

    if not os.path.exists(OPath_json):
        os.mkdir(OPath_json)
    if not os.path.exists(OPath_img):
        os.mkdir(OPath_img)

    splitImgandJson(IPath_json,IPath_img,OPath_json,OPath_img)