FFmpeg YUV视频序列编码为视频

上一篇已经写了如何配置好开发环境，这次就先小试牛刀，来个视频的编码。搞视频处理的朋友肯定比较熟悉YUV视频序列，很多测试库提供的视频数据都是YUV视频序列，我们这里就用用YUV视频序列来做视频。关于YUV视频序列，我就不多讲了，可以看书学习，通常的视频序列都是YUV420格式的。

步骤也就那几步，添加视频流，打开编码器，开辟相应的内存空间，然后就可以打开YUV序列逐帧写入数据了，so easy！记得最后要做好文件的关闭和内存的释放，因为FFmpeg是c风格的（不知道新版本是否是c++风格的），这些工作都需要自己做好啊。过多的说明是没用的，直接上代码：

这里我补充一下，大多数的视频格式好像只支持YUV格式的视频帧AVFrame，我试图直接把RGB的视频序列直接编码到视频这条路好像走不通，都需要把RGB的视频帧再转成YUV视频帧才行，不知道高手有没有其他高见。

#include <stdio.h>
#include <string.h>

extern "C"
{
#include <libavcodecavcodec.h>
#include <libavformatavformat.h>
#include <libswscaleswscale.h>
};

void main(int argc, char ** argv)
{
    AVFormatContext* oc;
    AVOutputFormat* fmt;
    AVStream* video_st;
    double video_pts;
    uint8_t* video_outbuf;
    uint8_t* picture_buf;
    AVFrame* picture;
//     AVFrame* pictureRGB;
    int size;
    int ret;
    int video_outbuf_size;

    FILE *fin = fopen("akiyo_qcif.yuv", "rb"); //视频源文件 

    const char* filename = "test.mpg";
//     const char* filename;
//     filename = argv[1];

    av_register_all();

//     avcodec_init(); // 初始化codec库
//     avcodec_register_all(); // 注册编码器

    fmt = guess_format(NULL, filename, NULL);
    oc = av_alloc_format_context();
    oc->oformat = fmt;
    snprintf(oc->filename, sizeof(oc->filename), "%s", filename);

    video_st = NULL;
    if (fmt->video_codec != CODEC_ID_NONE)
    {
        AVCodecContext* c;
        video_st = av_new_stream(oc, 0);
        c = video_st->codec;
        c->codec_id = fmt->video_codec;
        c->codec_type = CODEC_TYPE_VIDEO;
        c->bit_rate = 400000;
        c->width = 176;
        c->height = 144;
        c->time_base.num = 1;
        c->time_base.den = 25; 
        c->gop_size = 12;
        c->pix_fmt = PIX_FMT_YUV420P;
        if (c->codec_id == CODEC_ID_MPEG2VIDEO)
        {
            c->max_b_frames = 2;
        }
        if (c->codec_id == CODEC_ID_MPEG1VIDEO)
        {
            c->mb_decision = 2;
        }
        if (!strcmp(oc->oformat->name, "mp4") || !strcmp(oc->oformat->name, "mov") || !strcmp(oc->oformat->name, "3gp"))
        {
            c->flags |= CODEC_FLAG_GLOBAL_HEADER;
        }
    }

    if (av_set_parameters(oc, NULL)<0)
    {
        return;
    }
    
    dump_format(oc, 0, filename, 1);
    if (video_st)
    {
        AVCodecContext* c;
        AVCodec* codec;
        c = video_st->codec;
        codec = avcodec_find_encoder(c->codec_id);
        if (!codec)
        {
            return;
        }
        if (avcodec_open(c, codec) < 0)
        {
            return;
        }
        if (!(oc->oformat->flags & AVFMT_RAWPICTURE))
        {
            video_outbuf_size = 200000;
            video_outbuf = (uint8_t*)av_malloc(video_outbuf_size);
        }
        picture = avcodec_alloc_frame();
        size = avpicture_get_size(c->pix_fmt, c->width, c->height);
        picture_buf = (uint8_t*)av_malloc(size);
        if (!picture_buf)
        {
            av_free(picture);
        }
        avpicture_fill((AVPicture*)picture, picture_buf, c->pix_fmt, c->width, c->height);
    }

    if (!(fmt->flags & AVFMT_NOFILE))
    {
        if (url_fopen(&oc->pb, filename, URL_WRONLY) < 0)
        {
            return;
        }
    }
    av_write_header(oc);

    for (int i=0; i<300; i++)
    {
        if (video_st)
        {
            video_pts = (double)(video_st->pts.val * video_st->time_base.num / video_st->time_base.den);
        }
        else
        {
            video_pts = 0.0;
        }
        if (!video_st/* || video_pts >= 5.0*/)
        {
            break;
        }
        AVCodecContext* c;
        c = video_st->codec;
        size = c->width * c->height;

        if (fread(picture_buf, 1, size*3/2, fin) < 0)
        {
            break;
        }
        
        picture->data[0] = picture_buf;  // 亮度
        picture->data[1] = picture_buf+ size;  // 色度 
        picture->data[2] = picture_buf+ size*5/4; // 色度 

        // 如果是rgb序列，可能需要如下代码
//         SwsContext* img_convert_ctx;
//         img_convert_ctx = sws_getContext(c->width, c->height, PIX_FMT_RGB24, c->width, c->height, c->pix_fmt, SWS_BICUBIC, NULL, NULL, NULL);
//         sws_scale(img_convert_ctx, pictureRGB->data, pictureRGB->linesize, 0, c->height, picture->data, picture->linesize);

        if (oc->oformat->flags & AVFMT_RAWPICTURE)
        {
            AVPacket pkt;
            av_init_packet(&pkt);
            pkt.flags |= PKT_FLAG_KEY;
            pkt.stream_index = video_st->index;
            pkt.data = (uint8_t*)picture;
            pkt.size = sizeof(AVPicture);
            ret = av_write_frame(oc, &pkt);
        }
        else
        {
            int out_size = avcodec_encode_video(c, video_outbuf, video_outbuf_size, picture);
            if (out_size > 0)
            {
                AVPacket pkt;
                av_init_packet(&pkt);
                pkt.pts = av_rescale_q(c->coded_frame->pts, c->time_base, video_st->time_base);
                if (c->coded_frame->key_frame)
                {
                    pkt.flags |= PKT_FLAG_KEY;
                }
                pkt.stream_index = video_st->index;
                pkt.data = video_outbuf;
                pkt.size = out_size;
                ret = av_write_frame(oc, &pkt);
            }
        }
    }

    if (video_st)
    {
        avcodec_close(video_st->codec);
//         av_free(picture->data[0]);
        av_free(picture);
        av_free(video_outbuf);
//         av_free(picture_buf);
    }
    av_write_trailer(oc);
    for (int i=0; i<oc->nb_streams; i++)
    {
        av_freep(&oc->streams[i]->codec);
        av_freep(&oc->streams[i]);
    }
    if (!(fmt->flags & AVFMT_NOFILE))
    {
        url_fclose(oc->pb);
    }
    av_free(oc);
}

上一篇介绍了视频编码的小例子，视频解码跟编码差不多，只是要在视频文件中寻找视频流，找到后对流逐帧解码，就这样简单。闲言少叙，上code：

int main(int argc, char *argv[])
{
    AVFormatContext *pFormatCtx;
    int             i, videoStream;
    AVCodecContext  *pCodecCtx;
    AVCodec         *pCodec;
    AVFrame         *pFrame; 
    AVFrame         *pFrameRGB;
    AVPacket        packet;
    int             frameFinished;
    int             numBytes;
    uint8_t         *buffer;
    struct SwsContext *img_convert_ctx;

    if(argc < 2)
    {
        printf("Please provide a movie file
");
        return -1;
    }
    // Register all formats and codecs
    // 初始化ffmpeg库
    av_register_all();

    // Open video file
    if(av_open_input_file(&pFormatCtx, argv[1], NULL, 0, NULL)!=0)
        return -1; // Couldn't open file

    // Retrieve stream information
    // 查找文件的流信息
    if(av_find_stream_info(pFormatCtx)<0)
        return -1; // Couldn't find stream information

    // Dump information about file onto standard error
    // dump只是一个调试函数，输出文件的音、视频流的基本信息：帧率、分辨率、音频采样等等
    dump_format(pFormatCtx, 0, argv[1], 0);

    // Find the first video stream
    // 遍历文件的流，找到第一个视频流，并记录流的编码信息
    videoStream=-1;
    for(i=0; i<pFormatCtx->nb_streams; i++)
    {
        if(pFormatCtx->streams[i]->codec->codec_type==CODEC_TYPE_VIDEO)
        {
            videoStream=i;
            break;
        }
    }
    if(videoStream==-1)
        return -1; // Didn't find a video stream

    // Get a pointer to the codec context for the video stream
    // 得到视频流编码的上下文指针
    pCodecCtx=pFormatCtx->streams[videoStream]->codec;

    // construct the scale context, conversing to PIX_FMT_RGB24
    // 根据编码信息设置渲染格式
    img_convert_ctx = sws_getContext(pCodecCtx->width, pCodecCtx->height, pCodecCtx->pix_fmt, 
            pCodecCtx->width, pCodecCtx->height, PIX_FMT_RGB24, SWS_BICUBIC, NULL, NULL, NULL);
    if(img_convert_ctx == NULL)
    {
        fprintf(stderr, "Cannot initialize the conversion context!
");
//         exit(1);
        return -1;
    }

    // Find the decoder for the video stream
    // 在库里面查找支持该格式的解码器
    pCodec=avcodec_find_decoder(pCodecCtx->codec_id);
    if(pCodec==NULL)
    {
        fprintf(stderr, "Unsupported codec!
");
        return -1; // Codec not found
    }
    // Open codec
    // 打开解码器
    if(avcodec_open(pCodecCtx, pCodec)<0)
        return -1; // Could not open codec

    // Allocate video frame
    // 分配一个帧指针，指向解码后的原始帧
    pFrame=avcodec_alloc_frame();

    // Allocate an AVFrame structure
    // 分配一个帧指针，指向存放转换成rgb后的帧
    pFrameRGB=avcodec_alloc_frame();
    if(pFrameRGB==NULL)
        return -1;

    // Determine required buffer size and allocate buffer
    numBytes=avpicture_get_size(PIX_FMT_RGB24, pCodecCtx->width, pCodecCtx->height);
    buffer=(uint8_t *)av_malloc(numBytes*sizeof(uint8_t));    // buffer = new uint8_t[numBytes];

    // Assign appropriate parts of buffer to image planes in pFrameRGB
    // Note that pFrameRGB is an AVFrame, but AVFrame is a superset
    // of AVPicture
    // 给pFrameRGB帧附加上分配的内存
    avpicture_fill((AVPicture *)pFrameRGB, buffer, PIX_FMT_RGB24, pCodecCtx->width, pCodecCtx->height);

    // Read frames and save first five frames to disk
    i=0;
    while(av_read_frame(pFormatCtx, &packet)>=0)    // 读取一个帧
    {
        // Is this a packet from the video stream?
        if(packet.stream_index==videoStream)
        {
            // Decode video frame
            // 解码该帧
            avcodec_decode_video(pCodecCtx, pFrame, &frameFinished, packet.data, packet.size);

            // Did we get a video frame?
            if(frameFinished)
            {
                // Convert the image from its native format to RGB
                // img_convert((AVPicture *)pFrameRGB, PIX_FMT_RGB24, 
                //                 (AVPicture*)pFrame, pCodecCtx->pix_fmt, pCodecCtx->width, 
                //              pCodecCtx->height);
                
                // 把该帧转换成rgb

                // 如果只提取关键帧，加上这句
                // if (pFrame->key_frame == 1)
                sws_scale(img_convert_ctx, pFrame->data, pFrame->linesize, 0, pCodecCtx->height, pFrameRGB->data, pFrameRGB->linesize);    
                // Save the frame to disk
                // 保存前5帧
                if(++i<=5)
                {
//                     char pic[200];
//                     sprintf(pic,"pic%d.bmp",i);
//                     av_create_bmp(pic, pFrameRGB->data[0], pCodecCtx->width, pCodecCtx->height, 24);
                    SaveFrame(pFrameRGB, pCodecCtx->width, pCodecCtx->height, i);
                }
            }
        }

        // Free the packet that was allocated by av_read_frame
        // 释放读取的帧内存
        av_free_packet(&packet);
    }

    // Free the RGB image
    av_free(buffer);
    av_free(pFrameRGB);

    // Free the YUV frame
    av_free(pFrame);

    // Close the codec
    avcodec_close(pCodecCtx);

    // Close the video file
    av_close_input_file(pFormatCtx);

    return 0;
}

FFmpeg浅尝辄止（四）——音频的解码和编码

音频和视频其实是一样的，在文件中寻找音频流，然后解压出来，得到音频帧的数据，同样也可以按照设定的编码格式进行压缩，我这里把音频的解码和编码做成了两个工程，也是直接上代码：

#include <stdio.h>
#include <stdlib.h>

extern "C"
{
#include <libavcodecavcodec.h>
#include <libavformatavformat.h>
#include <libswscaleswscale.h>
}

int main(char arg,char *argv[])
{
    char *filename = argv[1];

    av_register_all();    //注册所有可解码类型
    AVFormatContext *pInFmtCtx=NULL;    //文件格式
    AVCodecContext *pInCodecCtx=NULL;    //编码格式 
    if (av_open_input_file(&pInFmtCtx, filename, NULL, 0, NULL)!=0)    //获取文件格式
        printf("av_open_input_file error
");
    if (av_find_stream_info(pInFmtCtx) < 0)    //获取文件内音视频流的信息
        printf("av_find_stream_info error
");

    unsigned int j;
    // Find the first audio stream

    int audioStream = -1;
    for (j=0; j<pInFmtCtx->nb_streams; j++)    //找到音频对应的stream
    {
        if (pInFmtCtx->streams[j]->codec->codec_type == CODEC_TYPE_AUDIO)
        {
            audioStream = j;
            break;
        }
    }
    if (audioStream == -1)
    {
        printf("input file has no audio stream
");
        return 0; // Didn't find a audio stream
    }
    printf("audio stream num: %d
",audioStream);
    pInCodecCtx = pInFmtCtx->streams[audioStream]->codec; //音频的编码上下文
    AVCodec *pInCodec = NULL;

    pInCodec = avcodec_find_decoder(pInCodecCtx->codec_id); //根据编码ID找到用于解码的结构体
    if (pInCodec == NULL)
    {
        printf("error no Codec found
");
        return -1 ; // Codec not found
    }

    if(avcodec_open(pInCodecCtx, pInCodec)<0)//将两者结合以便在下面的解码函数中调用pInCodec中的对应解码函数
    {
        printf("error avcodec_open failed.
");
        return -1; // Could not open codec

    }

    static AVPacket packet;

    printf(" bit_rate = %d 
", pInCodecCtx->bit_rate);
    printf(" sample_rate = %d 
", pInCodecCtx->sample_rate);
    printf(" channels = %d 
", pInCodecCtx->channels);
    printf(" code_name = %s 
", pInCodecCtx->codec->name);
    printf(" block_align = %d
",pInCodecCtx->block_align);

    uint8_t *pktdata;
    int pktsize;
    int out_size = AVCODEC_MAX_AUDIO_FRAME_SIZE*100;
    uint8_t * inbuf = (uint8_t *)malloc(out_size);
    FILE* pcm;
    pcm = fopen("result.pcm","wb");
    long start = clock();
    while (av_read_frame(pInFmtCtx, &packet) >= 0)//pInFmtCtx中调用对应格式的packet获取函数
    {
        if(packet.stream_index==audioStream)//如果是音频
        {
            pktdata = packet.data;
            pktsize = packet.size;
            while(pktsize>0)
            {
                out_size = AVCODEC_MAX_AUDIO_FRAME_SIZE*100;
                //解码
                int len = avcodec_decode_audio2(pInCodecCtx, (short*)inbuf, &out_size, pktdata, pktsize);
                if (len < 0)
                {
                    printf("Error while decoding.
");
                    break;
                }
                if(out_size > 0)
                {
                    fwrite(inbuf,1,out_size,pcm);//pcm记录
                    fflush(pcm);
                }
                pktsize -= len;
                pktdata += len;
            }
        } 
        av_free_packet(&packet);
    }
    long end = clock();
    printf("cost time :%f
",double(end-start)/(double)CLOCKS_PER_SEC);
    free(inbuf);
    fclose(pcm);
    if (pInCodecCtx!=NULL)
    {
        avcodec_close(pInCodecCtx);
    }
    av_close_input_file(pInFmtCtx);

    return 0;
}

解码后的文件保存为result.pcm中，现在用这个文件压缩出新的音频文件，代码如下：

void main()
{
    int16_t *samples;
    uint8_t *audio_outbuf;
    int audio_outbuf_size;
    int audio_input_frame_size;
    double audio_pts;
    
    const char* filename = "test.wav";
    FILE *fin = fopen("result.pcm", "rb"); //音频源文件 
    AVOutputFormat *fmt;
    AVFormatContext *oc;
    AVStream * audio_st;
    av_register_all();
    fmt = guess_format(NULL, filename, NULL);
    oc = av_alloc_format_context();
    oc->oformat = fmt;
    snprintf(oc->filename, sizeof(oc->filename), "%s", filename);
    audio_st = NULL;

    if (fmt->audio_codec != CODEC_ID_NONE)
    {
        AVCodecContext *c;
        audio_st = av_new_stream(oc, 1);
        c = audio_st->codec;
        c->codec_id = fmt->audio_codec;
        c->codec_type = CODEC_TYPE_AUDIO;
        c->bit_rate = 128000;
        c->sample_rate = 44100;
        c->channels = 2;
    }
    if (av_set_parameters(oc, NULL) < 0)
    {
        return;
    }
    dump_format(oc, 0, filename, 1);
    if (audio_st)
    {
        AVCodecContext* c;
        AVCodec* codec;
        c = audio_st->codec;
        codec = avcodec_find_encoder(c->codec_id);
        avcodec_open(c, codec);
        audio_outbuf_size = 10000;
        audio_outbuf = (uint8_t*)av_malloc(audio_outbuf_size);
        if (c->frame_size <= 1)
        {
            audio_input_frame_size = audio_outbuf_size / c->channels;
            switch (audio_st->codec->codec_id)
            {
            case CODEC_ID_PCM_S16LE:
            case CODEC_ID_PCM_S16BE:
            case CODEC_ID_PCM_U16LE:
            case CODEC_ID_PCM_U16BE:
                audio_input_frame_size >>= 1;
                break;
            default:
                break;
            }
        }
        else
        {
            audio_input_frame_size = c->frame_size;
        }
        samples = (int16_t*)av_malloc(audio_input_frame_size*2*c->channels);
    }
    if (!fmt->flags & AVFMT_NOFILE)
    {
        if (url_fopen(&oc->pb, filename, URL_WRONLY) < 0)
        {
            return;
        }
    }
    av_write_header(oc);
    for (;;)
    {
        if (audio_st)
        {
            audio_pts = (double)audio_st->pts.val * audio_st->time_base.num / audio_st->time_base.den;
        }
        else
        {
            audio_pts = 0.0;
        }
        if (!audio_st || audio_pts >= 360.0)
        {
            break;
        }
        if (fread(samples, 1, audio_input_frame_size*2*audio_st->codec->channels, fin) <= 0)
        {
            break;
        }
        AVCodecContext* c;
        AVPacket pkt;
        av_init_packet(&pkt);
        c = audio_st->codec;
        pkt.size = avcodec_encode_audio(c, audio_outbuf, audio_outbuf_size, samples);
        pkt.pts = av_rescale_q(c->coded_frame->pts, c->time_base, audio_st->time_base);
        pkt.flags |= PKT_FLAG_KEY;
        pkt.stream_index = audio_st->index;
        pkt.data = audio_outbuf;
        if (av_write_frame(oc, &pkt) != 0)
        {
            return;
        }
    }
    if (audio_st)
    {
        avcodec_close(audio_st->codec);
        av_free(samples);
        av_free(audio_outbuf);
    }
    av_write_trailer(oc);
    for (int i=0; i<oc->nb_streams; i++)
    {
        av_freep(&oc->streams[i]->codec);
        av_freep(&oc->streams[i]);
    }
    if (!(fmt->flags & AVFMT_NOFILE))
    {
        url_fclose(oc->pb);
    }
    av_free(oc);
    fclose(fin);
}

对应的下载链接：

http://download.csdn.net/detail/yang_xian521/4398576

音频解码：http://download.csdn.net/detail/yang_xian521/4399219

音频编码：http://download.csdn.net/detail/yang_xian521/4399293

至此，我已经实现了视频的解码编码，音频的解码编码，相信有心人肯定可以在此基础上实现视频音频的同步压缩，不过要再看一些网上的资料，也是很轻松 的。至于视频的显示播放，大多数是要结合SDL实现，由于我只是浅尝辄止，只实现了图片的显示和简单的视频播放，比起我之前推荐的几个链接是弱爆了的，就 不写下去了，这个系列就到这里吧，希望能给像我一样的一些入门级选手点帮助。也欢迎大神来多多指教。

(附： 例子文件见360云盘的  所有文件-> work -> 音视频处理)