所谓黑白图片其实就是对图像进行去色，也就是灰度化，只要调用之前写过的灰度化函数即可。

底片的原理就是对RGB取补色即可。也就是将当前像素点的RGB值分别与255之差后的值作为当前点的RGB值，即

R = 255 – R；G = 255 – G；B = 255 – B；

浮雕其实就是一种边缘检测，在图像处理基础那个系列中，介绍过的sobel算子，Prewitt算子 或者是拉普拉斯算子我觉得都有着类似的效果

本文采用了新的梯度检测算法，相比之前介绍的边缘检测算子，效果稍弱，但是视觉效果好。

底片的代码：

public Image filter() {

if(this.img.gray)

return this.img; // Grayscale images can not be processed

for (int y = 0; y < this.img.h; y++) {

for (int x = 0; x < this.img.w; x++) {

int c = this.img.data[x + y * this.img.w];

int A = (c >> 24) & 0xff;

int R = (c >> 16) & 0xFF;

int G = (c >> 8) & 0xFF;

int B = (c >> 0) & 0xFF;

R = 255 - R;

G = 255 - G;

B = 255 - B;

this.img.data[x + y * this.img.w] = (A << 24) | (R << 16) | (G << 8) | B;

}

}

return this.img;

}

浮雕的代码：

public Image filter() {

if(this.img.gray)

return this.img; // Grayscale images can not be processed

int[] d = new int[this.img.w*this.img.h]; //what's the fuck. it must be writed as this. not int[] d = this.img.data;

for (int y = 0; y < this.img.h -1; y++) {

for (int x = 0; x < this.img.w -1; x++) {

int pc = this.img.data[x + y * this.img.w];

int PR = (pc >> 16) & 0xFF;

int PG = (pc >> 8) & 0xFF;

int PB = (pc >> 0) & 0xFF;

int c = this.img.data[x + 1 + (y+1) * this.img.w];

int R = (c >> 16) & 0xFF;

int G = (c >> 8) & 0xFF;

int B = (c >> 0) & 0xFF;

R = PR - R + 128;

G = PG - G + 128;

B = PB - B + 128;

d[x + y * this.img.w] = (255 << 24) | (R << 16) | (G << 8) | B;

}

}

this.img.data = d;

this.img.toGray();

return this.img;

}

处理后的效果

其实除了浮雕之外，都可以用颜色矩阵来实现。