图像处理之特殊灰度算法技巧

图像处理之特殊灰度算法技巧

介绍几种特殊的灰度算法滤镜，将彩色图像转换为灰度图像。其中涉及到的有基于阈值的图

像二值化，弗洛伊德.斯坦德伯格抖动算法，基于阈值的部分灰度化

基础知识- 怎么把RGB转换为单色的[0 ~256]之间的灰度，最常用的转换公式如下：

Gray = 0.299 * red + 0.587 * green + 0.114 * blue;

1. 基于像素平均值的图像阈值二值化算法：

处理流程：

a. 首先将彩色图像转换为灰度图像

b. 计算灰度图像的算术平均值– M

c. 以M为阈值，完成对灰度图二值化( 大于阈值M,像素点赋值为白色，否则赋值为黑

色)

图像效果：



关键代码：

public BufferedImage filter(BufferedImage src, BufferedImage dest) {
int width = src.getWidth();
int height = src.getHeight();

if ( dest == null )
dest = createCompatibleDestImage( src, null );
src = super.filter(src, dest);

int[] inPixels = new int[width*height];
int[] outPixels = new int[width*height];
getRGB(src, 0, 0, width, height, inPixels );

// calculate means of pixel
int index = 0;
double redSum = 0, greenSum = 0, blueSum = 0;
double total = height * width;
for(int row=0; row<height; row++) {
int ta = 0, tr = 0, tg = 0, tb = 0;
for(int col=0; col<width; col++) {
index = row * width + col;
ta = (inPixels[index] >> 24) & 0xff;
tr = (inPixels[index] >> 16) & 0xff;
tg = (inPixels[index] >> 8) & 0xff;
tb = inPixels[index] & 0xff;
redSum += tr;
greenSum += tg;
blueSum +=tb;
}
}
int means = (int)(redSum / total);
System.out.println(" threshold average value = " + means);

// dithering
for(int row=0; row<height; row++) {
int ta = 0, tr = 0, tg = 0, tb = 0;
for(int col=0; col<width; col++) {
index = row * width + col;
ta = (inPixels[index] >> 24) & 0xff;
tr = (inPixels[index] >> 16) & 0xff;
tg = (inPixels[index] >> 8) & 0xff;
tb = inPixels[index] & 0xff;
if(tr >=means) {
tr = tg = tb = 255;
} else {
tr = tg = tb = 0;
}
outPixels[index] = (ta << 24) | (tr << 16) | (tg << 8) | tb;

}
}
setRGB( dest, 0, 0, width, height, outPixels );
return dest;
}

2. 基于错误扩散的Floyd-Steinberg抖动算法

关于什么是Floyd-Steinberg抖动，参见这里

http://en.wikipedia.org/wiki/Floyd–Steinberg_dithering

图像效果：



关键代码：

@Override
public BufferedImage filter(BufferedImage src, BufferedImage dest) {
int width = src.getWidth();
int height = src.getHeight();

if ( dest == null )
dest = createCompatibleDestImage( src, null );
src = super.filter(src, dest);

int[] inPixels = new int[width*height];
int[] outPixels = new int[width*height];
getRGB( src, 0, 0, width, height, inPixels );
int index = 0;
for(int row=0; row<height; row++) {
for(int col=0; col<width; col++) {
index = row * width + col;
int r1 = (inPixels[index] >> 16) & 0xff;
int g1 = (inPixels[index] >> 8) & 0xff;
int b1 = inPixels[index] & 0xff;
int cIndex = getCloseColor(r1, g1, b1);
outPixels[index] = (255 << 24) | (COLOR_PALETTE[cIndex][0] << 16) | (COLOR_PALETTE[cIndex][1] << 8) | COLOR_PALETTE[cIndex][2];
int er = r1 - COLOR_PALETTE[cIndex][0];
int eg = g1 - COLOR_PALETTE[cIndex][1];
int eb = b1 -  COLOR_PALETTE[cIndex][2];
int k = 0;

if(row + 1 < height && col - 1 > 0) {
k = (row + 1) * width + col - 1;
r1 = (inPixels[k] >> 16) & 0xff;
g1 = (inPixels[k] >> 8) & 0xff;
b1 = inPixels[k] & 0xff;
r1 += (int)(er * kernelData[0]);
g1 += (int)(eg * kernelData[0]);
b1 += (int)(eb * kernelData[0]);
inPixels[k] = (255 << 24) | (clamp(r1) << 16) | (clamp(g1) << 8) | clamp(b1);
}

if(col + 1 < width) {
k = row * width + col + 1;
r1 = (inPixels[k] >> 16) & 0xff;
g1 = (inPixels[k] >> 8) & 0xff;
b1 = inPixels[k] & 0xff;
r1 += (int)(er * kernelData[3]);
g1 += (int)(eg * kernelData[3]);
b1 += (int)(eb * kernelData[3]);
inPixels[k] = (255 << 24) | (clamp(r1) << 16) | (clamp(g1) << 8) | clamp(b1);
}

if(row + 1 < height) {
k = (row + 1) * width + col;
r1 = (inPixels[k] >> 16) & 0xff;
g1 = (inPixels[k] >> 8) & 0xff;
b1 = inPixels[k] & 0xff;
r1 += (int)(er * kernelData[1]);
g1 += (int)(eg * kernelData[1]);
b1 += (int)(eb * kernelData[1]);
inPixels[k] = (255 << 24) | (clamp(r1) << 16) | (clamp(g1) << 8) | clamp(b1);
}

if(row + 1 < height && col + 1 < width) {
k = (row + 1) * width + col + 1;
r1 = (inPixels[k] >> 16) & 0xff;
g1 = (inPixels[k] >> 8) & 0xff;
b1 = inPixels[k] & 0xff;
r1 += (int)(er * kernelData[2]);
g1 += (int)(eg * kernelData[2]);
b1 += (int)(eb * kernelData[2]);
inPixels[k] = (255 << 24) | (clamp(r1) << 16) | (clamp(g1) << 8) | clamp(b1);
}
}
}
setRGB( dest, 0, 0, width, height, outPixels );
return dest;
}

3. 选择性灰度算法

计算选择的颜色与像素灰度颜色之间的几何距离值，跟阈值比较决定是否像素点为灰度

值，可以得到一些让你意想不到的图像处理效果！

图像效果 (Main Color = GREEN, 阈值 = 200)

原图：



处理以后



关键代码：

public BufferedImage filter(BufferedImage src, BufferedImage dest) {
int width = src.getWidth();
int height = src.getHeight();

if ( dest == null )
dest = createCompatibleDestImage( src, null );

int[] inPixels = new int[width*height];
int[] outPixels = new int[width*height];
getRGB( src, 0, 0, width, height, inPixels );
int index = 0;
for(int row=0; row<height; row++) {
int ta = 0, tr = 0, tg = 0, tb = 0;
for(int col=0; col<width; col++) {
index = row * width + col;
ta = (inPixels[index] >> 24) & 0xff;
tr = (inPixels[index] >> 16) & 0xff;
tg = (inPixels[index] >> 8) & 0xff;
tb = inPixels[index] & 0xff;
int gray = (int)(0.299 * (double)tr + 0.587 * (double)tg + 0.114 * (double)tb);
double distance = getDistance(tr, tg, tb);
if(distance < threshold) {
double k = distance / threshold;
int[] rgb = getAdjustableRGB(tr, tg, tb, gray, (float)k);
tr = rgb[0];
tg = rgb[1];
tb = rgb[2];
outPixels[index] = (ta << 24) | (tr << 16) | (tg << 8) | tb;
} else {
outPixels[index] = (ta << 24) | (gray << 16) | (gray << 8) | gray;
}

}
}
setRGB( dest, 0, 0, width, height, outPixels );
return dest;
}

创建新的目标Image

public BufferedImage createCompatibleDestImage(BufferedImage src, ColorModel dstCM) {
if ( dstCM == null )
dstCM = src.getColorModel();
return new BufferedImage(dstCM, dstCM.createCompatibleWritableRaster(src.getWidth(), src.getHeight()), dstCM.isAlphaPremultiplied(), null);
}