OpenCV Python实现矩形图像转曲线并保留透明度的技术求助

要实现带透明度的曲线变形，核心是将RGB通道与Alpha通道同步处理，避免变形过程中丢失透明信息。以下是针对你的场景的完整实现：

关键思路 #

1. 将带Alpha通道的叠加图拆分为RGB通道和单独的Alpha通道
2. 对RGB通道和Alpha通道应用相同的曲线变形变换
3. 合并变形后的RGB与Alpha通道，再与背景图叠加

完整代码 #

import cv2
import numpy as np

def rotate_image_with_alpha(image, angle):
    """带Alpha通道的图像旋转"""
    h, w = image.shape[:2]
    center = (w // 2, h // 2)
    rotation_matrix = cv2.getRotationMatrix2D(center, angle, 1.0)
    cos = np.abs(rotation_matrix[0, 0])
    sin = np.abs(rotation_matrix[0, 1])
    new_w = int((h * sin) + (w * cos))
    new_h = int((h * cos) + (w * sin))
    rotation_matrix[0, 2] += (new_w / 2) - center[0]
    rotation_matrix[1, 2] += (new_h / 2) - center[1]
    # 对RGB和Alpha通道分别旋转
    rgb = cv2.warpAffine(image[:, :, :3], rotation_matrix, (new_w, new_h), flags=cv2.INTER_LINEAR, borderMode=cv2.BORDER_CONSTANT, borderValue=(0,0,0))
    alpha = cv2.warpAffine(image[:, :, 3], rotation_matrix, (new_w, new_h), flags=cv2.INTER_LINEAR, borderMode=cv2.BORDER_CONSTANT, borderValue=0)
    return cv2.merge((rgb, alpha))

def curve_distort_with_alpha(image, curve_strength=0.05):
    """带Alpha通道的曲线变形（适配背景海报的弧形区域）"""
    h, w = image.shape[:2]
    # 创建目标坐标映射
    map_x = np.zeros((h, w), dtype=np.float32)
    map_y = np.zeros((h, w), dtype=np.float32)
    
    for y in range(h):
        # 计算横向偏移，模拟上凸弧形
        offset = int(curve_strength * w * np.sin((y / h) * np.pi))
        for x in range(w):
            map_x[y, x] = x + offset
            map_y[y, x] = y
    
    # 对RGB和Alpha通道分别应用变形
    rgb = cv2.remap(image[:, :, :3], map_x, map_y, interpolation=cv2.INTER_LINEAR, borderMode=cv2.BORDER_CONSTANT, borderValue=(0,0,0))
    alpha = cv2.remap(image[:, :, 3], map_x, map_y, interpolation=cv2.INTER_LINEAR, borderMode=cv2.BORDER_CONSTANT, borderValue=0)
    
    return cv2.merge((rgb, alpha))

def merge_image(bg, overlay, x, y):
    """带Alpha通道的图像叠加"""
    h, w = overlay.shape[:2]
    # 确保叠加区域在背景范围内
    if x + w > bg.shape[1] or y + h > bg.shape[0]:
        raise ValueError("Overlay exceeds background bounds")
    
    # 提取Alpha通道并归一化
    alpha = overlay[:, :, 3] / 255.0
    alpha_inv = 1.0 - alpha
    
    # 逐通道混合
    for c in range(3):
        bg[y:y+h, x:x+w, c] = (alpha * overlay[:, :, c] + alpha_inv * bg[y:y+h, x:x+w, c]).astype(np.uint8)
    
    return bg

# 主流程
bg = cv2.imread("bg_layer.webp", cv2.IMREAD_UNCHANGED)
poster = cv2.imread("7.jpeg", cv2.IMREAD_UNCHANGED)

# 1. 调整尺寸
poster = cv2.resize(poster, (500, 700))
# 2. 曲线变形（保持透明度）
poster = curve_distort_with_alpha(poster, curve_strength=0.08)
# 3. 旋转
poster = rotate_image_with_alpha(poster, -30)
# 4. 叠加并显示
result = merge_image(bg, poster, 450, 450)

cv2.imshow("Result", result)
cv2.waitKey(0)
cv2.destroyAllWindows()
# 保存结果
cv2.imwrite("final_result.png", result)

代码说明 #

• curve_distort_with_alpha函数：通过创建坐标映射实现弧形变形，分别处理RGB和Alpha通道，确保透明度信息不丢失
• rotate_image_with_alpha函数：优化了带Alpha通道的旋转逻辑，避免透明区域出现黑边
• merge_image函数：基于Alpha通道的逐通道混合，保证叠加效果自然

你可以根据背景海报的实际弧度调整curve_strength参数，数值越大弧度越明显。

内容的提问来源于stack exchange，提问作者Rishabh