流程走通。（效果不行:数据集和数据标注都比较少）

src/validate_dataset.py

@@ -0,0 +1,135 @@
+
+import os
+import sys
+from PIL import Image
+
+# 添加 src 目录到路径
+src_dir = os.path.dirname(os.path.abspath(__file__))
+if src_dir not in sys.path:
+    sys.path.insert(0, src_dir)
+
+from dataset_util import load_dataset_config
+
+
+def validate_images_labels(images_dir, labels_dir):
+    """验证 images 和 labels 目录
+    
+    Returns:
+        list: 错误列表
+    """
+    errors = []
+
+    if not os.path.exists(images_dir):
+        errors.append(f"images 目录不存在: {images_dir}")
+        return errors
+
+    image_files = [f for f in os.listdir(images_dir) if f.lower().endswith(('.jpg', '.jpeg', '.png'))]
+
+    if not image_files:
+        errors.append(f"images 目录中没有图片: {images_dir}")
+        return errors
+
+    for img_file in image_files:
+        img_path = os.path.join(images_dir, img_file)
+        label_file = os.path.splitext(img_file)[0] + '.txt'
+        label_path = os.path.join(labels_dir, label_file)
+
+        if not os.path.exists(label_path):
+            errors.append(f"缺失标注文件: {label_file}")
+            continue
+
+        try:
+            img = Image.open(img_path)
+            img_width, img_height = img.size
+        except Exception as e:
+            errors.append(f"无法打开图片: {img_file}")
+            continue
+
+        with open(label_path, 'r') as f:
+            for line_num, line in enumerate(f, 1):
+                line = line.strip()
+                if not line:
+                    continue
+
+                parts = line.split()
+                if len(parts) != 5:
+                    errors.append(f"{label_file}: 格式错误，行 {line_num}")
+                    continue
+
+                try:
+                    class_id = int(parts[0])
+                    center_x = float(parts[1])
+                    center_y = float(parts[2])
+                    width = float(parts[3])
+                    height = float(parts[4])
+
+                    if not (0 <= center_x <= 1 and 0 <= center_y <= 1):
+                        errors.append(f"{label_file}: 坐标超出范围，行 {line_num}")
+                except ValueError:
+                    errors.append(f"{label_file}: 数字解析错误，行 {line_num}")
+
+    return errors
+
+
+def validate_dataset_from_config(config_path=None):
+    """从配置文件验证数据集
+    
+    Args:
+        config_path: 配置文件路径
+    
+    Returns:
+        list: 错误列表
+    """
+    config = load_dataset_config(config_path)
+    
+    data_dir = config.get('path', './dataset')
+    if not os.path.isabs(data_dir) and config_path:
+        data_dir = os.path.join(os.path.dirname(config_path), data_dir)
+    
+    errors = []
+    
+    # 验证 train/val/test
+    for split in ['train', 'val', 'test']:
+        split_images_dir = os.path.join(data_dir, config.get(split, f'{split}/images'))
+        split_labels_dir = os.path.join(data_dir, config.get(split, f'{split}/images').replace('images', 'labels'))
+        
+        if os.path.exists(split_images_dir):
+            print(f"检查 {split} 集...")
+            split_errors = validate_images_labels(split_images_dir, split_labels_dir)
+            errors.extend(split_errors)
+    
+    return errors
+
+
+def validate_yolo_dataset(config_path=None):
+    """验证 YOLO 数据集
+    
+    Args:
+        config_path: 配置文件路径
+    """
+    print("验证数据集...")
+
+    errors = validate_dataset_from_config(config_path)
+
+    if errors:
+        print(f"\n发现 {len(errors)} 个错误：")
+        for error in errors[:20]:
+            print(f"  - {error}")
+        if len(errors) > 20:
+            print(f"  ... 还有 {len(errors) - 20} 个错误")
+    else:
+        print("\n验证通过！")
+
+    return errors
+
+
+if __name__ == '__main__':
+    import argparse
+
+    parser = argparse.ArgumentParser(description='验证 YOLO 数据集')
+    parser.add_argument('config_path', nargs='?', default=None,
+                        help='dataset.yaml 配置文件路径（默认使用项目根目录的 dataset.yaml）')
+
+    args = parser.parse_args()
+
+    validate_yolo_dataset(args.config_path)