Skip to content
TheChatArchiveThe Chat Archive
Claudeclaude-3-5-sonnet-20241022· AnthropicHugging Face LMArena Human Preference 140K

python代码,对图像进行仿射变换,改变形态

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Geminigemini-2.5-pro’s answer.
  1. User

    python代码,对图像进行仿射变换,改变形态

  2. Claude Assistant

    我来帮你写一个使用OpenCV进行图像仿射变换的Python代码示例。这个代码会展示几种常见的仿射变换效果:

    import cv2
    import numpy as np
    
    def affine_transform_demo(image_path):
        # 读取图像
        img = cv2.imread(image_path)
        rows, cols = img.shape[:2]
    
        # 1. 平移变换
        M_translation = np.float32([[1, 0, 50], [0, 1, 30]])
        translated = cv2.warpAffine(img, M_translation, (cols, rows))
    
        # 2. 旋转变换
        angle = 45
        M_rotation = cv2.getRotationMatrix2D((cols/2, rows/2), angle, 1)
        rotated = cv2.warpAffine(img, M_rotation, (cols, rows))
    
        # 3. 缩放变换
        src_points = np.float32([[0, 0], [cols-1, 0], [0, rows-1]])
        dst_points = np.float32([[cols*0.2, rows*0.1], [cols*0.8, rows*0.2], [cols*0.1, rows*0.9]])
        M_affine = cv2.getAffineTransform(src_points, dst_points)
        affine = cv2.warpAffine(img, M_affine, (cols, rows))
    
        # 显示结果
        cv2.imshow('Original', img)
        cv2.imshow('Translated', translated)
        cv2.imshow('Rotated', rotated)
        cv2.imshow('Affine', affine)
        cv2.waitKey(0)
        cv2.destroyAllWindows()
    
    # 使用示例
    image_path = 'your_image.jpg'  # 替换为你的图片路径
    affine_transform_demo(image_path)
    

    这段代码实现了三种仿射变换:

    1. 平移变换:将图像向右移动50像素,向下移动30像素
    2. 旋转变换:将图像绕中心点旋转45度
    3. 仿射变换:通过指定三个点的原始位置和目标位置进行变换

    要使用这段代码,你需要:

    1. 安装必要的库:
    pip install opencv-python numpy
    
    1. 将image_path替换为你实际的图片路径

    如果你想要调整变换效果,可以修改以下参数:

    • 平移:修改M_translation中的[50, 30]值
    • 旋转:修改angle的值
    • 仿射:修改dst_points中的坐标值

    你还可以添加其他类型的仿射变换,比如:

    # 添加缩放变换
    scale = 1.5
    M_scale = np.float32([[scale, 0, 0], [0, scale, 0]])
    scaled = cv2.warpAffine(img, M_scale, (int(cols*scale), int(rows*scale)))
    

    这个代码会帮助你理解和实现基本的图像仿射变换。需要注意的是,仿射变换会保持平行线之间的平行关系,但可能会改变角度和距离。

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.