YUV420P、YUV420SP、NV12、NV21和RGB互相转换并存储为JPEG以及PNG图片

音视频实践学习

android全平台编译ffmpeg以及x264与fdk-aac实践
ubuntu下使用nginx和nginx-rtmp-module配置直播推流服务器
android全平台编译ffmpeg合并为单个库实践
android-studio使用cmake编译ffmpeg实践
android全平台编译ffmpeg视频解码器实践
android全平台编译ffmpeg支持命令行实践
android全平台编译ffmpeg视频推流实践
android平台下音频编码之编译LAME库转码PCM为MP3
ubuntu平台下编译vlc-android视频播放器实践
图解YU12、I420、YV12、NV12、NV21、YUV420P、YUV420SP、YUV422P、YUV444P的区别
图解RGB565、RGB555、RGB16、RGB24、RGB32、ARGB32等格式的区别
YUV420P、YUV420SP、NV12、NV21和RGB互相转换并存储为JPEG以及PNG图片
android全平台编译libyuv库实现YUV和RGB的转换

RGB转YUV420P

这里就不新建android工程来测试了，直接使用Clion来运行测试输出

首先将我们目标的JPEG图片转换为RGB文件：

ffmpeg -i main.jpg -s 510x510 -pix_fmt rgb24 rgb24.rgb

开始运行转换RGB24转换为YUV，这里要注意的一点就是，RGB24实际占用的内存为width * height * 3，而YUV420P实际占用的内存为width * height * 3 / 2

unsigned char ClipValue(unsigned char x, unsigned char min_val, unsigned char max_val) {    if (x > max_val) {        return max_val;    } else if (x < min_val) {        return min_val;    } else {        return x;    }}void RGB2YUV420P(unsigned char *rgb24, int width, int height, unsigned char *yuv420p) {    unsigned char *ptrY, *ptrU, *ptrV;    memset(yuv420p, 0, width * height * 3 / 2);    ptrY = yuv420p;    ptrU = yuv420p + width * height;    ptrV = ptrU + (width * height * 1 / 4);    unsigned char y, u, v, r, g, b;    int index = 0;    for (int j = 0; j < height; j++) {        for (int i = 0; i < width; i++) {            index = width * j * 3 + i * 3;            r = rgb24[index];            g = rgb24[index + 1];            b = rgb24[index + 2];            y = (unsigned char) ((66 * r + 129 * g + 25 * b + 128) >> 8) + 16;            u = (unsigned char) ((-38 * r - 74 * g + 112 * b + 128) >> 8) + 128;            v = (unsigned char) ((112 * r - 94 * g - 18 * b + 128) >> 8) + 128;            *(ptrY++) = ClipValue(y, 0, 255);            if (j % 2 == 0 && i % 2 == 0) {                *(ptrU++) = ClipValue(u, 0, 255);            } else if (i % 2 == 0) {                *(ptrV++) = ClipValue(v, 0, 255);            }        }    }}

void rgb24_to_yuv420(char *rgbPath, char *yuvPath, int width, int height) {    FILE *fp_rgb = fopen(rgbPath, "rb+");    FILE *fp_yuv = fopen(yuvPath, "wb+");        unsigned char *rgb24_data = (unsigned char *) malloc(width * height * 3);    unsigned char *yuv420_data = (unsigned char *) malloc(width * height * 3 / 2);    fread(rgb24_data, 1, width * height * 3, fp_rgb);    RGB2YUV420P(rgb24_data, width, height, yuv420_data);    fwrite(yuv420_data, 1, width * height * 3 / 2, fp_yuv);    free(rgb24_data);    free(yuv420_data);    fclose(fp_rgb);    fclose(fp_yuv);}

我们可以使用ffplay来播放我们最终生成的rgb24.yuv文件

YUV420P、YUV420SP、NV12、NV21和RGB互相转换并存储为JPEG以及PNG图片_第1张图片

YUV420P转RGB

你也可以根据上面的例子一样，直接使用Clion转换完之后，使用雷神的yuvplayer来查看，这里笔者直接新建了一个native-yuv2rgb，将最终的yuv文件转换为RGB数据之后基于ANativeWindow渲染出来。

YUV420P、YUV420SP、NV12、NV21和RGB互相转换并存储为JPEG以及PNG图片_第2张图片

定义好java层的接口

package com.onzhou.graphic.yuv2rgb;import android.view.Surface;public class NativeYUV2RGB {    static {        System.loadLibrary("native-yuv2rgb");    }    public native void yuv2rgb(String imagePath, int width, int height, Surface surface);}

接下来就是核心的native层实现

/** * YUV420P转RGB24 * @param data * @param rgb * @param width * @param height */void YUV420P_TO_RGB24(unsigned char *data, unsigned char *rgb, int width, int height) {    int index = 0;    unsigned char *ybase = data;    unsigned char *ubase = &data[width * height];    unsigned char *vbase = &data[width * height * 5 / 4];    for (int y = 0; y < height; y++) {        for (int x = 0; x < width; x++) {            //YYYYYYYYUUVV            u_char Y = ybase[x + y * width];            u_char U = ubase[y / 2 * width / 2 + (x / 2)];            u_char V = vbase[y / 2 * width / 2 + (x / 2)];            rgb[index++] = Y + 1.402 * (V - 128); //R            rgb[index++] = Y - 0.34413 * (U - 128) - 0.71414 * (V - 128); //G            rgb[index++] = Y + 1.772 * (U - 128); //B        }    }}

上面这个方法的主要作用就是将输入的YUV数据转换为RGB，转换公式可以参考之前的博客，或者百度一下。

接下来我们需要通过ANativeWindow_fromSurface获取应用的窗口，然后读取YUV文件转换为最终的RGB数据，接着将最终的RGB数据写入到窗口的buffer中去，完成渲染。

void drawYUV(const char *path, int width, int height, ANativeWindow_Buffer buffer) {    FILE *file = fopen(path, "rb");    unsigned char *yuvData = new unsigned char[width * height * 3 / 2];    fread(yuvData, 1, width * height * 3 / 2, file);    unsigned char *rgb24 = new unsigned char[width * height * 3];    //YUV420P转RGB24    YUV420P_TO_RGB24(yuvData, rgb24, width, height);    uint32_t *line = (uint32_t *) buffer.bits;    for (int y = 0; y < height; y++) {        for (int x = 0; x < width; x++) {            int index = y * width + x;            line[x] = rgb24[index * 3 + 2] << 16                      | rgb24[index * 3 + 1] << 8                      | rgb24[index * 3];        }        line = line + buffer.stride;    }    //释放内存    delete[] yuvData;    delete[] rgb24;    //关闭文件句柄    fclose(file);}void yuv2rgb(JNIEnv *env, jobject obj, jstring jpegPath, jint width, jint height, jobject surface) {    const char *path = env->GetStringUTFChars(jpegPath, 0);    //获取目标surface    ANativeWindow *window = ANativeWindow_fromSurface(env, surface);    if (NULL == window) {        ThrowException(env, "java/lang/RuntimeException", "unable to get native window");        return;    }    //默认的是RGB_565    int32_t result = ANativeWindow_setBuffersGeometry(window, 0, 0, WINDOW_FORMAT_RGBA_8888);    if (result < 0) {        ThrowException(env, "java/lang/RuntimeException", "unable to set buffers geometry");        //释放窗口        ANativeWindow_release(window);        window = NULL;        return;    }    ANativeWindow_acquire(window);    ANativeWindow_Buffer buffer;    //锁定窗口的绘图表面    if (ANativeWindow_lock(window, &buffer, NULL) < 0) {        ThrowException(env, "java/lang/RuntimeException", "unable to lock native window");        //释放窗口        ANativeWindow_release(window);        window = NULL;        return;    }    //绘制YUV    drawYUV(path, width, height, buffer);    //解锁窗口的绘图表面    if (ANativeWindow_unlockAndPost(window) < 0) {        ThrowException(env, "java/lang/RuntimeException",                       "unable to unlock and post to native window");    }    env->ReleaseStringUTFChars(jpegPath, path);    //释放    ANativeWindow_release(window);}

在启动的Activity中，读取目标的YUV文件，将路径以及surface传递给native层完成转换以及渲染。

    public void onYUV420PToRGB24(View view) {        File file = new File(getExternalFilesDir(null), "yuv420p.yuv");        mNativeYUV2RGB.yuv2rgb(file.getAbsolutePath(), NativeYUV2RGB.Type.YUV420P_TO_RGB24, 510, 510, mSurfaceView.getHolder().getSurface());    }

最终显示的效果图：

YUV420P、YUV420SP、NV12、NV21和RGB互相转换并存储为JPEG以及PNG图片_第3张图片

NV12转RGB24

之前的博客也说了，NV12(NV12: YYYYYYYYUVUV =>YUV420SP)本质上也属于YUV420SP，在进行这个例子之前，我们先拿到一张最原始的NV12的文件。

ffmpeg -i main.jpg -s 510x510 -pix_fmt nv12 nv12.yuv

转换过程跟上述类似，下面给出具体的转换公式：

/** * NV12属于YUV420SP格式 * @param data * @param rgb * @param width * @param height */void NV12_TO_RGB24(unsigned char *data, unsigned char *rgb, int width, int height) {    int index = 0;    unsigned char *ybase = data;    unsigned char *ubase = &data[width * height];    for (int y = 0; y < height; y++) {        for (int x = 0; x < width; x++) {            //YYYYYYYYUVUV            u_char Y = ybase[x + y * width];            u_char U = ubase[y / 2 * width + (x / 2) * 2];            u_char V = ubase[y / 2 * width + (x / 2) * 2 + 1];            rgb[index++] = Y + 1.402 * (V - 128); //R            rgb[index++] = Y - 0.34413 * (U - 128) - 0.71414 * (V - 128); //G            rgb[index++] = Y + 1.772 * (U - 128); //B        }    }}

NV21转RGB24

同上面所述，NV21(NV21: YYYYYYYYVUVU =>YUV420SP)同样上也属于YUV420SP，这个格式通常也默认是android手机相机的格式，在进行这个例子之前，我们先拿到一张最原始的NV21的文件。

ffmpeg -i main.jpg -s 510x510 -pix_fmt nv21 nv21.yuv

转换公式：

/** * NV12属于YUV420SP格式,android相机默认格式 * @param data * @param rgb * @param width * @param height */void NV21_TO_RGB24(unsigned char *data, unsigned char *rgb, int width, int height) {    int index = 0;    unsigned char *ybase = data;    unsigned char *ubase = &data[width * height];    for (int y = 0; y < height; y++) {        for (int x = 0; x < width; x++) {            //YYYYYYYYVUVU            u_char Y = ybase[x + y * width];            u_char U = ubase[y / 2 * width + (x / 2) * 2 + 1];            u_char V = ubase[y / 2 * width + (x / 2) * 2];            rgb[index++] = Y + 1.402 * (V - 128); //R            rgb[index++] = Y - 0.34413 * (U - 128) - 0.71414 * (V - 128); //G            rgb[index++] = Y + 1.772 * (U - 128); //B        }    }}

项目地址：native-yuv2rgb
https://github.com/byhook/graphic4android

存储为JPEG图片

YUV数据是无法直接存储为JPEG图片的，这里需要先转换为RGB数据，然后利用libjpeg-turbo库，完成JPEG转换。

YUV420P转RGB24：

void YUV420P_TO_RGB24(unsigned char *yuv420p, unsigned char *rgb24, int width, int height) {    int index = 0;    for (int y = 0; y < height; y++) {        for (int x = 0; x < width; x++) {            int indexY = y * width + x;            int indexU = width * height + y / 2 * width / 2 + x / 2;            int indexV = width * height + width * height / 4 + y / 2 * width / 2 + x / 2;            u_char Y = yuv420p[indexY];            u_char U = yuv420p[indexU];            u_char V = yuv420p[indexV];            rgb24[index++] = Y + 1.402 * (V - 128); //R            rgb24[index++] = Y - 0.34413 * (U - 128) - 0.71414 * (V - 128); //G            rgb24[index++] = Y + 1.772 * (U - 128); //B        }    }}

然后将转换后的RGB数据通过libjpeg-turbo写入文件：

int YUV2JPEG::RGB24_TO_JPEG(const char *filename, u_char *rgbData, int image_width,                            int image_height,                            int quality) {    jpeg_compress_struct jpegCompress;    jpeg_error_mgr jpegError;    jpegCompress.err = jpeg_std_error(&jpegError);    //初始化压缩对象    jpeg_create_compress(&jpegCompress);    FILE *pFile = fopen(filename, "wb");    if (pFile == NULL) {        return -1;    }    jpeg_stdio_dest(&jpegCompress, pFile);    //设置压缩参数    jpegCompress.client_data = &pFile;    jpegCompress.image_width = image_width;    jpegCompress.image_height = image_height;    jpegCompress.input_components = 3;    jpegCompress.in_color_space = JCS_RGB;    jpeg_set_defaults(&jpegCompress);    //指定亮度及色度质量    jpegCompress.q_scale_factor[0] = jpeg_quality_scaling(100);    jpegCompress.q_scale_factor[1] = jpeg_quality_scaling(100);    //图像采样率,默认为2 * 2    jpegCompress.comp_info[0].v_samp_factor = 2;    jpegCompress.comp_info[0].h_samp_factor = 2;    //设置压缩质量    jpeg_set_quality(&jpegCompress, quality, TRUE);  //100 is the highest    //开始压缩图片    jpeg_start_compress(&jpegCompress, TRUE);    JSAMPROW row_pointer[1];    //从上到下,设置每一个像素    for (unsigned int i = 0; i < jpegCompress.image_height; i++) {        row_pointer[0] = rgbData + i * jpegCompress.image_width * 3;        jpeg_write_scanlines(&jpegCompress, row_pointer, 1);    }    //完成压缩    jpeg_finish_compress(&jpegCompress);    fclose(pFile);    pFile = NULL;    jpeg_destroy_compress(&jpegCompress);    return 0;}void YUV2JPEG::YUV420P_TO_JPEG(const char *yuvPath, const char *jpegPath, int width, int height) {    FILE *file = fopen(yuvPath, "rb");    unsigned char *yuv420p = new unsigned char[width * height * 3 / 2];    fread(yuv420p, 1, width * height * 3 / 2, file);    unsigned char *rgb24 = new unsigned char[width * height * 3];    //YUV420P转RGB24    yuvutils::YUV420P_TO_RGB24(yuv420p, rgb24, width, height);//RGB24转JPEG    RGB24_TO_JPEG(jpegPath, rgb24, width, height, 100);    //释放内存    delete[] yuv420p;    delete[] rgb24;    //关闭文件句柄    fclose(file);}

存储为PNG图片

同样的，跟上面的例子类似，YUV数据也是无法直接存储为PNG图片的，这里需要先转换为RGB数据，然后利用libpng库，完成PNG转换。

bool YUV2PNG::RGB24_TO_PNG(const char *filename, u_char *rgbData, int width, int height,                           int quality) {    FILE *fp;    png_structp png_ptr = NULL;    png_infop info_ptr = NULL;    fp = fopen(filename, "wb");    if (fp) {        png_ptr = png_create_write_struct(PNG_LIBPNG_VER_STRING, NULL, NULL, NULL);        info_ptr = png_create_info_struct(png_ptr);        if (png_ptr == NULL || info_ptr == NULL) {            fclose(fp);            return false;        }        png_init_io(png_ptr, fp);        png_set_IHDR(png_ptr, info_ptr, width, height, 8, PNG_COLOR_TYPE_RGB, PNG_INTERLACE_NONE,                     PNG_COMPRESSION_TYPE_BASE, PNG_FILTER_TYPE_BASE);        png_write_info(png_ptr, info_ptr);        png_set_packing(png_ptr);        uint8_t *data = new uint8_t[width * 3];        memset(data, 255, width * 3);        int nv_start = width * height;        uint32_t rgb_index = 0, pos;        int r, g, b, nv_index, y, ru, rv;        for (uint32_t i = 0; i < height; ++i) {            for (uint32_t j = 0; j < width; ++j) {                data[j * 3 + 0] = rgbData[rgb_index++];                data[j * 3 + 1] = rgbData[rgb_index++];                data[j * 3 + 2] = rgbData[rgb_index++];            }            png_write_row(png_ptr, data);        }        delete data;        png_write_end(png_ptr, info_ptr);        png_destroy_write_struct(&png_ptr, &info_ptr);        fclose(fp);        return true;    }    return false;}void YUV2PNG::YUV420P_TO_PNG(const char *yuvPath, const char *pngPath, int width, int height) {    FILE *file = fopen(yuvPath, "rb");    unsigned char *yuv420p = new unsigned char[width * height * 3 / 2];    fread(yuv420p, 1, width * height * 3 / 2, file);    unsigned char *rgb24 = new unsigned char[width * height * 3];    //YUV420P转RGB24    yuvutils::YUV420P_TO_RGB24(yuv420p, rgb24, width, height);    //RGB24转PNG文件    RGB24_TO_PNG(pngPath, rgb24, width, height, 100);    //释放内存    delete[] yuv420p;    delete[] rgb24;    //关闭文件句柄    fclose(file);}

YUV420P、YUV420SP、NV12、NV21和RGB互相转换并存储为JPEG以及PNG图片_第4张图片

项目地址：native-yuv2image
https://github.com/byhook/graphic4android

参考：
https://blog.csdn.net/leixiaohua1020/article/details/50534150
https://blog.csdn.net/yixianfeng41/article/details/52181578
https://blog.csdn.net/yangyang031213/article/details/84191652