CIMaskToAlpha无法保留真实硬度（模糊度），求线条透明化不增粗方案

问题原因 #

CIMaskToAlpha滤镜会将图像灰度值转换为alpha通道，同时强制把RGB通道设为白色。原线条边缘的抗锯齿灰色区域会变成半透明白色，视觉上就显得线条变粗，不符合你“保留线条硬度、不增粗”的需求。

方案一：使用CIColorMatrix自定义alpha通道（Core Image实现） #

通过CIColorMatrix单独调整alpha通道，保留原图像的RGB颜色，避免引入额外白色边缘：

extension UIImage {
    func cleared() -> UIImage {
        let entryImage = self
        let context = CIContext()
        guard let cgImage = entryImage.cgImage else { return self }
        let ciImage = CIImage(cgImage: cgImage)
        
        guard let colorMatrixFilter = CIFilter(name: "CIColorMatrix") else { return self }
        
        // 基于RGB灰度值生成alpha：黑色（灰度0）对应alpha=0，白色（灰度255）对应alpha=1
        let grayToAlphaVector = CIVector(x: 1/3.0, y: 1/3.0, z: 1/3.0, w: 0)
        
        colorMatrixFilter.setValue(ciImage, forKey: kCIInputImageKey)
        // 保持RGB通道不变
        colorMatrixFilter.setValue(CIVector(x: 1, y: 0, z: 0, w: 0), forKey: "inputRVector")
        colorMatrixFilter.setValue(CIVector(x: 0, y: 1, z: 0, w: 0), forKey: "inputGVector")
        colorMatrixFilter.setValue(CIVector(x: 0, y: 0, z: 1, w: 0), forKey: "inputBVector")
        // 用灰度值控制alpha通道
        colorMatrixFilter.setValue(grayToAlphaVector, forKey: "inputAVector")
        
        guard let result = colorMatrixFilter.outputImage else { return self }
        guard let cgResult = context.createCGImage(result, from: result.extent) else { return self }
        
        return UIImage(cgImage: cgResult, scale: entryImage.scale, orientation: entryImage.imageOrientation)
    }
}

方案二：手动操作像素（精确控制） #

直接通过CGContext遍历像素，仅调整alpha通道，完全保留原图像颜色信息，适合需要精细调整阈值的场景：

extension UIImage {
    func cleared() -> UIImage {
        guard let cgImage = self.cgImage else { return self }
        let width = cgImage.width
        let height = cgImage.height
        let bytesPerPixel = 4
        let bytesPerRow = bytesPerPixel * width
        let bitsPerComponent = 8
        let colorSpace = CGColorSpaceCreateDeviceRGB()
        let bitmapInfo = CGImageAlphaInfo.premultipliedLast.rawValue
        
        guard let context = CGContext(data: nil, width: width, height: height, bitsPerComponent: bitsPerComponent, bytesPerRow: bytesPerRow, space: colorSpace, bitmapInfo: bitmapInfo) else { return self }
        
        context.draw(cgImage, in: CGRect(x: 0, y: 0, width: width, height: height))
        
        guard let data = context.data else { return self }
        let pixels = data.bindMemory(to: UInt8.self, capacity: width * height * bytesPerPixel)
        
        for i in 0..<width*height {
            let offset = i * bytesPerPixel
            let r = pixels[offset]
            let g = pixels[offset+1]
            let b = pixels[offset+2]
            let originalAlpha = pixels[offset+3]
            
            // 计算像素灰度值
            let grayValue = (Int(r) + Int(g) + Int(b)) / 3
            // 灰度越低（越黑），alpha越低，实现透明化
            let alphaFactor = CGFloat(grayValue) / 255.0
            let newAlpha = UInt8(CGFloat(originalAlpha) * alphaFactor)
            
            pixels[offset+3] = newAlpha
        }
        
        guard let newCGImage = context.makeImage() else { return self }
        return UIImage(cgImage: newCGImage, scale: self.scale, orientation: self.imageOrientation)
    }
}

方案优势说明 #

两种方案均仅调整alpha通道，保留原图像的RGB颜色：

• 黑色线条区域随灰度降低逐渐透明，完全符合“黑色像素转透明”的需求
• 线条边缘的模糊抗锯齿部分对应转为半透明，完美保留线条的硬度和原始粗细，不会出现视觉上的增粗效果

内容的提问来源于stack exchange，提问作者Jabed Dhali