import os

def check_dataset_match(images_dir, labels_dir):
    """检查图片和标注文件数量是否匹配"""
    # 获取图片文件列表
    image_files = [f for f in os.listdir(images_dir) if f.lower().endswith('.jpg')]
    # 获取标注文件列表（排除classes.txt）
    label_files = [f for f in os.listdir(labels_dir) 
                  if f.lower().endswith('.txt') and f != 'classes.txt']
    
    # 提取文件名（不带扩展名）
    image_names = {os.path.splitext(f)[0] for f in image_files}
    label_names = {os.path.splitext(f)[0] for f in label_files}
    
    # 找出不匹配的文件
    missing_images = label_names - image_names
    missing_labels = image_names - label_names
    
    print(f"图片数量: {len(image_files)}")
    print(f"标注数量: {len(label_files)}")
    print(f"匹配状态: {'匹配' if len(image_files) == len(label_files) else '不匹配'}")
    
    if missing_images:
        print("\n缺少对应的图片文件:")
        for name in sorted(missing_images):
            print(f"- {name}.jpg")
    
    if missing_labels:
        print("\n缺少对应的标注文件:")
        for name in sorted(missing_labels):
            print(f"- {name}.txt")

if __name__ == "__main__":
    images_dir = r'F:\myprojects\啾啾救援识别系统\JJCarDetection\datasets\猫狗\train\images'
    labels_dir = r'F:\myprojects\啾啾救援识别系统\JJCarDetection\datasets\猫狗\train\labels'
    
    print("正在检查数据集匹配情况...")
    check_dataset_match(images_dir, labels_dir)
    print("\n检查完成")