Imgproc.cvtColor(rgb, gray, Imgproc.COLOR_RGB2GRAY)可以快速的实现图片脱色，其通过简单的计算将三通道图像转换成单通道图像，经典的灰度转换公式如下：
但是转换过的图像存在对比度丢失，无法区分颜色的问题。OpenCV给我们提供另外一种脱色方法Photo.decolor，旨在最大限度的保留图片对比度。
论文详情：www.cse.cuhk.edu.hk/~leojia/pro…
论文PDF：www.computer.org/csdl/pds/ap…
相比于简单的使用R/G/B线性组合，作者构造了更为复杂的多项式拟合，向量空间的基底为：[r,g,b,rg,rb,gb,r2,g2,b2]，核心就是求取灰度化函数：
public static void decolor(Mat src, Mat grayscale, Mat color_boost)
/**
* 脱色
* author: yidong
* 2020/12/5
*/
class DecolorActivity : AppCompatActivity() {
private val mBinding: ActivityDecolorBinding by lazy {
ActivityDecolorBinding.inflate(layoutInflater)
}
private val mList = mutableListOf<ImageTextObject>()
private lateinit var mAdapter: ImageTextAdapter
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(mBinding.root)
mAdapter = ImageTextAdapter(this, mList)
mBinding.container.adapter = mAdapter
val bgr = Utils.loadResource(this, R.drawable.ceil)
val rgb = Mat()
Imgproc.cvtColor(bgr, rgb, Imgproc.COLOR_BGR2RGB)
val gray = Mat()
val dst = Mat()
val boost = Mat()
mBinding.isLoading = true
GlobalScope.launch(Dispatchers.IO) {
Imgproc.cvtColor(rgb, gray, Imgproc.COLOR_RGB2GRAY)
Photo.decolor(rgb, dst, boost)
launch(Dispatchers.Main) {
mList.add(ImageTextObject(rgb, "原图"))
mList.add(ImageTextObject(gray, "RGB2GRAY"))
mList.add(ImageTextObject(dst, "DeColor"))
mList.add(ImageTextObject(boost, "ColorBoosting"))
mAdapter.notifyItemRangeInserted(0, 4)
mBinding.isLoading = false
}
}
}
}
分别采用decolor和RGB2GRAY实现的脱色效果图，在对比度和颜色保留上效果相差很大。
