darknet训练图像分类图像预处理random_distort_image研究

更多darknet训练图像分类图像预处理

random_distort_image()

void random_distort_image(image im, float hue, float saturation, float exposure)
{
    float dhue = rand_uniform_strong(-hue, hue);
    float dsat = rand_scale(saturation);
    float dexp = rand_scale(exposure);
    distort_image(im, dhue, dsat, dexp);
}

distort_image 

扭曲图像

void distort_image(image im, float hue, float sat, float val)
{
    if (im.c >= 3)
    {
        rgb_to_hsv(im);
        scale_image_channel(im, 1, sat);
        scale_image_channel(im, 2, val);
        int i;
        for(i = 0; i < im.w*im.h; ++i){
            im.data[i] = im.data[i] + hue;
            if (im.data[i] > 1) im.data[i] -= 1;
            if (im.data[i] < 0) im.data[i] += 1;
        }
        hsv_to_rgb(im);
    }
    else
    {
        scale_image_channel(im, 0, val);
    }
    constrain_image(im);
}

rgb_to_hsv()

rgb转换为hsv，rgb易于显示但不易于图像处理，所以需要转换为hsb

RGB、HSV和HSL颜色空间

参考网址：

// http://www.cs.rit.edu/~ncs/color/t_convert.html

// http://www.cs.rit.edu/~ncs/color/t_convert.html
void rgb_to_hsv(image im)
{
    assert(im.c == 3);
    int i, j;
    float r, g, b;
    float h, s, v;
    for(j = 0; j < im.h; ++j){
        for(i = 0; i < im.w; ++i){
            r = get_pixel(im, i , j, 0);
            g = get_pixel(im, i , j, 1);
            b = get_pixel(im, i , j, 2);
            float max = three_way_max(r,g,b);
            float min = three_way_min(r,g,b);
            float delta = max - min;
            v = max;
            if(max == 0){
                s = 0;
                h = 0;
            }else{
                s = delta/max;
                if(r == max){
                    h = (g - b) / delta;
                } else if (g == max) {
                    h = 2 + (b - r) / delta;
                } else {
                    h = 4 + (r - g) / delta;
                }
                if (h < 0) h += 6;
                h = h/6.;
            }
            set_pixel(im, i, j, 0, h);
            set_pixel(im, i, j, 1, s);
            set_pixel(im, i, j, 2, v);
        }
    }
}

scale_image_channel

缩放图像通道

void scale_image_channel(image im, int c, float v)
{
    int i, j;
    for(j = 0; j < im.h; ++j){
        for(i = 0; i < im.w; ++i){
            float pix = get_pixel(im, i, j, c);
            pix = pix*v;
            set_pixel(im, i, j, c, pix);
        }
    }
}

constrain_image（）

void constrain_image(image im)
{
    int i;
    for(i = 0; i < im.w*im.h*im.c; ++i){
        if(im.data[i] < 0) im.data[i] = 0;
        if(im.data[i] > 1) im.data[i] = 1;
    }
}