如何将mp4文件解复用并且解码为单独的.yuv图像序列以及.pcm音频采样数据?

2023-10-22 13:53:52 浏览数 (1)

一.初始化解复用器

  在音视频的解复用的过程中,有一个非常重要的结构体AVFormatContext,即输入文件的上下文句柄结构,代表当前打开的输入文件或流。我们可以将输入文件的路径以及AVFormatContext **format_ctx 传入函数avformat_open_input(),就可以打开对应的音视频文件或流。接下来再调用avformat_find_stream_info()函数去解析输入文件中的音视频流信息,打开对应的解码器,读取文件头的信息进行解码, 然后在解码过程中将一些参数的信息保存到AVStream结构对应的成员中。之后,我们便可以通过AVStream去初始化编解码器的上下文结构,下面给出代码:

代码语言:javascript复制
static AVFormatContext *format_ctx= nullptr;
static AVCodecContext *video_dec_ctx= nullptr,*audio_dec_ctx= nullptr;
static int32_t video_stream_index=-1;
static int32_t audio_stream_index=-1;
static AVStream *video_stream= nullptr,*audio_stream= nullptr;
static FILE *output_video_file= nullptr,*output_audio_file= nullptr;
static AVPacket *pkt= nullptr;
static AVFrame *frame= nullptr;

static int open_codec_context(int32_t *stream_idx,AVCodecContext **dec_ctx,AVFormatContext *fmt_ctx,enum AVMediaType type){
    int ret,stream_index;
    AVStream *st= nullptr;
    const AVCodec *dec= nullptr;
    ret= av_find_best_stream(fmt_ctx,type,-1,-1, nullptr,0);
    if(ret<0){
        cerr<<"Error:Could not find "<<string(av_get_media_type_string(type))<<" stream in input file."<<endl;
        return ret;
    }
    else{
        stream_index=ret;
        st=fmt_ctx->streams[stream_index];
        //find decoder for the stream
        dec= avcodec_find_decoder(st->codecpar->codec_id);
        if(!dec){
            cerr<<"Error:Failed to find codec:"<<string(av_get_media_type_string(type))<<endl;
            return -1;
        }
        *dec_ctx= avcodec_alloc_context3(dec);
        if(!*dec_ctx){
            cerr<<"Error:Failed to alloc codec context:"<<string(av_get_media_type_string(type))<<endl;
            return -1;
        }
        if((ret= avcodec_parameters_to_context(*dec_ctx,st->codecpar))<0){
            cerr<<"Error:Failed to copy codec parameters to decoder context."<<endl;
            return ret;
        }
        if((ret=avcodec_open2(*dec_ctx,dec, nullptr))<0){
            cerr<<"Error:Could not open "<<string(av_get_media_type_string(type))<<" codec."<<endl;
            return ret;
        }
        *stream_idx=stream_index;
    }
    return 0;
}
int32_t init_demuxer(const char *input_name,const char *video_output_name,const char *audio_output_name){
    if(strlen(input_name)==0){
        cerr<<"Error:empty input file name."<<endl;
        exit(-1);
    }
    int32_t result= avformat_open_input(&format_ctx,input_name, nullptr, nullptr);
    if(result<0){
        cerr<<"Error:avformat_open_input failed."<<endl;
        exit(-1);
    }
    result= avformat_find_stream_info(format_ctx, nullptr);
    if(result<0){
        cerr<<"Error:avformat_find_stream_info failed."<<endl;
        exit(-1);
    }
    result= open_codec_context(&video_stream_index,&video_dec_ctx,format_ctx,AVMEDIA_TYPE_VIDEO);
    if(result>=0){
        video_stream=format_ctx->streams[video_stream_index];
        output_video_file=fopen(video_output_name,"wb");
        if(!output_video_file){
            cerr<<"Error:failed to open video output file."<<endl;
            return -1;
        }
    }
    result= open_codec_context(&audio_stream_index,&audio_dec_ctx,format_ctx,AVMEDIA_TYPE_AUDIO);
    if(result>=0){
        audio_stream=format_ctx->streams[audio_stream_index];
        output_audio_file=fopen(audio_output_name,"wb");
        if(!output_audio_file){
            cerr<<"Error:failed to open audio output file."<<endl;
            return -1;
        }
    }
    av_dump_format(format_ctx,0,input_name,0);
    if(!audio_stream&&!video_stream){
        cerr<<"Error:Could not find audio or video stream in the input,aborting"<<endl;
        return -1;
    }
    pkt=av_packet_alloc();
    if(!pkt){
        cerr<<"Error:could not alloc packet."<<endl;
        return -1;
    }
    frame=av_frame_alloc();
    if(!frame){
        cerr<<"Error:could not alloc frame."<<endl;
        return -1;
    }
    if(video_stream){
        cout<<"Demuxing video from file "<<string(input_name)<<" into "<<string(video_output_name)<<endl;
    }
    if(audio_stream){
        cout<<"Demuxing audio from file "<<string(input_name)<<" into "<<string(audio_output_name)<<endl;
    }
    return 0;
}

二.循环读取码流包数据进行解码

  在这里,我们需要调用一个非常重要的函数av_read_frame(),它可以从打开的音视频文件或流中依次读取下一个码流包结构,然后我们将码流包传入解码器进行解码即可,代码如下:

代码语言:javascript复制
static int32_t decode_packet(AVCodecContext *dec,const AVPacket *pkt,bool flushing){
    int32_t result=0;
    result= avcodec_send_packet(dec,pkt);
    if(result<0){
        cerr<<"Error:avcodec_send_packet failed."<<endl;
        return -1;
    }
    while(result>=0){
        result=avcodec_receive_frame(dec,frame);
        if(result<0){
            if(result==AVERROR_EOF||result==AVERROR(EAGAIN)){
                return 0;
            }
            cerr<<"Error:Error during decoding,result="<<result<<endl;
            return result;
        }
        if(dec->codec->type==AVMEDIA_TYPE_VIDEO){
            write_frame_to_yuv(frame);
        }
        else{
            write_samples_to_pcm(frame,audio_dec_ctx);
        }
        if(flushing){
            cout<<"flushing"<<endl;
        }
        av_frame_unref(frame);
    }
    return result;
}
int32_t demuxing(){
    int32_t result=0;
    while(av_read_frame(format_ctx,pkt)>=0){
        cout<<"Read packet,pts:"<<pkt->pts<<",stream:"<<pkt->stream_index<<",size:"<<pkt->size<<endl;
        if(pkt->stream_index==audio_stream_index){
            result= decode_packet(audio_dec_ctx,pkt,false);
        }
        else if(pkt->stream_index==video_stream_index){
            result= decode_packet(video_dec_ctx,pkt,false);
        }
        av_packet_unref(pkt);
        if(result<0){
            break;
        }
    }
    if(video_dec_ctx){
        decode_packet(video_dec_ctx, nullptr,true);
    }
    if(audio_dec_ctx){
        decode_packet(audio_dec_ctx, nullptr,true);
    }
    cout<<"Demuxing succeeded."<<endl;
    return 0;
}

三.将解码后的图像序列以及音频采样数据写入相应的文件

  这个步骤比较简单,不解释,直接上代码:

代码语言:javascript复制
int32_t write_frame_to_yuv(AVFrame* frame){
    uint8_t** pBuf=frame->data;
    int* pStride=frame->linesize;
    for(size_t i=0;i<3;i  ){
        int32_t width=(i==0?frame->width:frame->width/2);
        int32_t height=(i==0?frame->height:frame->height/2);
        for(size_t j=0;j<height;j  ){
            fwrite(pBuf[i],1,width,output_video_file);
            pBuf[i] = pStride[i];
        }
    }
    return 0;
}
int32_t write_samples_to_pcm(AVFrame* frame,AVCodecContext* codec_ctx){
    int data_size= av_get_bytes_per_sample(codec_ctx->sample_fmt);
    if(data_size<0){
        cerr<<"Error:failed to calculate data size."<<endl;
        return -1;
    }
    for(int i=0;i<frame->nb_samples;i  ){
        for(int ch=0;ch<codec_ctx->channels;ch  ){
            fwrite(frame->data[ch] i*data_size,1,data_size,output_audio_file);
        }
    }
    return 0;
}

四.销毁资源,释放内存

代码语言:javascript复制
void destroy_demuxer(){
    avcodec_free_context(&video_dec_ctx);
    avcodec_free_context(&audio_dec_ctx);
    avformat_close_input(&format_ctx);
    if(output_audio_file!= nullptr){
        fclose(output_audio_file);
        output_audio_file= nullptr;
    }
    if(output_video_file!= nullptr){
        fclose(output_video_file);
        output_video_file= nullptr;
    }
}

五.main函数

代码语言:javascript复制
int main(){
    int32_t result=init_demuxer("../input.mp4","../output.yuv","../output.pcm");
    if(result<0){
        return -1;
    }
    result=demuxing();
    if(result<0){
        return -1;
    }
    destroy_demuxer();
    return 0;
}

  到这里,就大功告成了,可以使用以下的命令去播放输出的音视频文件:

  ffplay -ac 2 -ar 44100 -f f32le -i output.pcm

  ffplay -f rawvideo -video_size 1920x1080 -i output.yuv

0 人点赞