解複用FLV檔案(基於FFMPEG解析FLV(h264+aac))
阿新 • • 發佈：2018-11-11
技術在於交流、溝通，轉載請註明出處並保持作品的完整性。
原文： https://blog.csdn.net/hiwubihe/article/details/82346759
[本系列相關文章]
- H264和音訊流打包成PS流 (MPEG2-PS)
- PS流解複用成H264和音訊流(ES提取)
- H264和音訊流打包成TS流 (MPEG2-TS)
- TS流解複用成H264和音訊流(ES提取)
- FLV檔案格式基礎
- 解複用FLV檔案(基於FFMPEG解析FLV(h264+aac))
- 解複用FLV檔案(不用FFMPEG,C++實現)
本篇介紹基於FFMPEG解析FLV檔案，FLV由H264視訊和AAC音訊組成。FFMPEG 解複用出的音視訊AVPacket，直接寫入檔案播放不了。FLV格式最常用的封裝組合就是H264+AAC,FFMPEG解複用的結果AVPacket只包含實際的壓縮音視訊資料，不包括解碼必須的資訊，H264的SPS+PPS,AAC的視訊頭ADTS等。這就需要從FLV中提取SPS+PPS資訊，和ADTS資訊加到音視訊幀前面。
1.FLV檔案AVCDecoderConfigurationRecord
封裝FLV，一般第一個VideoTag就是AVCDecoderConfigurationRecord結構，該結構就是H264的SPS+PPS資訊
AVCDecoderConfigurationRecord定義如下
aligned(8) class AVCDecoderConfigurationRecord { unsigned int(8) configurationVersion = 1; unsigned int(8) AVCProfileIndication; unsigned int(8) profile_compatibility; unsigned int(8) AVCLevelIndication; bit(6) reserved = ‘111111’b; unsigned int(2) lengthSizeMinusOne; bit(3) reserved = ‘111’b; unsigned int(5) numOfSequenceParameterSets; for (i=0; i< numOfSequenceParameterSets; i++) { unsigned int(16) sequenceParameterSetLength ; bit(8*sequenceParameterSetLength) sequenceParameterSetNALUnit; } unsigned int(8) numOfPictureParameterSets; for (i=0; i< numOfPictureParameterSets; i++) { unsigned int(16) pictureParameterSetLength; bit(8*pictureParameterSetLength) pictureParameterSetNALUnit; } bit(6) reserved = ‘111111’b; unsigned int(2) chroma_format; bit(5) reserved = ‘11111’b; unsigned int(3) bit_depth_luma_minus8; bit(5) reserved = ‘11111’b; unsigned int(3) bit_depth_chroma_minus8; unsigned int(8) numOfSequenceParameterSetExt; for (i=0; i< numOfSequenceParameterSetExt; i++) { unsigned int(16) sequenceParameterSetExtLength; bit(8*sequenceParameterSetExtLength) sequenceParameterSetExtNALUnit; } }
2.FLV檔案AudioSpecificConfig
封裝FLV，一般第一個AudioTag就是AudioSpecificConfig結構，該結構就是AAC封裝頭ADTS需要的資訊，取樣率，通道數，樣本深度等資訊。
AudioSpecificConfig在所有封裝AAC的檔案結構中都有，如FLV/MP4。 AudioSpecificConfig結構比較複雜。一般結構就是2個位元組。在MP4檔案中一般叫AES,一般結構如下表
| 總體結構 | 5 bits: object type
if (object type == 31)
6 bits + 32: object type
4 bits: frequency index
if (frequency index == 15)
24 bits: frequency
4 bits: channel configuration
var bits: AOT Specific Config |
| ObjectType |
|
| 取樣率索引 |
|
| Channels |
|
解析AES程式碼
typedef struct
{
int write_adts;
int objecttype;
int sample_rate_index;
int channel_conf;
}ADTSContext;
int aac_decode_extradata(ADTSContext *adts, unsigned char *pbuf, int bufsize)
{
int aot, aotext, samfreindex;
int i, channelconfig;
unsigned char *p = pbuf;
if (!adts || !pbuf || bufsize<2)
{
return -1;
}
aot = (p[0]>>3)&0x1f;
if (aot == 31)
{
aotext = (p[0]<<3 | (p[1]>>5)) & 0x3f;
aot = 32 + aotext;
samfreindex = (p[1]>>1) & 0x0f;
if (samfreindex == 0x0f)
{
channelconfig = ((p[4]<<3) | (p[5]>>5)) & 0x0f;
}
else
{
channelconfig = ((p[1]<<3)|(p[2]>>5)) & 0x0f;
}
}
else
{
samfreindex = ((p[0]<<1)|p[1]>>7) & 0x0f;
if (samfreindex == 0x0f)
{
channelconfig = (p[4]>>3) & 0x0f;
}
else
{
channelconfig = (p[1]>>3) & 0x0f;
}
}
#ifdef AOT_PROFILE_CTRL
if (aot < 2) aot = 2;
#endif
adts->objecttype = aot-1;
adts->sample_rate_index = samfreindex;
adts->channel_conf = channelconfig;
adts->write_adts = 1;
return 0;
} 構造AES程式碼
//獲取音訊AES值
bool GetDecoderSpecificInfo(AdtsHeadInfo stAdtsHeadInfo,unsigned char*pBuffer,unsigned long &lSizeOfDecoderSpecificInfo)
{
lSizeOfDecoderSpecificInfo = 2;
memset(pBuffer,0,lSizeOfDecoderSpecificInfo);
bits_buffer_t bw;
bits_initwrite (&bw, lSizeOfDecoderSpecificInfo, pBuffer);
bits_write (&bw, 5, stAdtsHeadInfo.iProfile);
bits_write (&bw, 4, stAdtsHeadInfo.iSampleRateIndex);
bits_write (&bw, 4, stAdtsHeadInfo.iChans);
memcpy(pBuffer, bw.p_data, lSizeOfDecoderSpecificInfo);
return true;
}
3.FFMPEG處理原理
在av_read_frame函式呼叫中，如果讀到AVCDecoderConfigurationRecord或者AudioSpecificConfig結構時，會把資訊儲存在每個stream的streams[index]->codec->extradata擴充套件資料中。處理H264資料時，提供一個叫 “h264_mp4toannexb”位元流過濾器，在av_read_frame讀取到正常資料包時，需要呼叫位元流過濾器處理每個資料包，其實就是判斷是否是I幀，是I幀的話，解析AVCDecoderConfigurationRecord結構成SPS+PPS,然後把SPS+PPS儲存加到I幀前面。可以參考FFMPEG原始碼，版本3.2.4.r86064。下面程式碼就是迴圈處理接受到的資料，新增SPS/PPS。
static int h264_mp4toannexb_filter(AVBSFContext *ctx, AVPacket *out)
{
H264BSFContext *s = ctx->priv_data;
AVPacket *in;
uint8_t unit_type;
int32_t nal_size;
uint32_t cumul_size = 0;
const uint8_t *buf;
const uint8_t *buf_end;
int buf_size;
int ret = 0, i;
//讀取的資料包複製到in
ret = ff_bsf_get_packet(ctx, &in);
if (ret < 0)
return ret;
/* nothing to filter */
//不需要過濾 直接賦值
if (!s->extradata_parsed)
{
av_packet_move_ref(out, in);
av_packet_free(&in);
return 0;
}
buf = in->data;
buf_size = in->size;
buf_end = in->data + in->size;
do
{
ret= AVERROR(EINVAL);
//長度錯誤
if (buf + s->length_size > buf_end)
goto fail;
//獲取NALU長度
for (nal_size = 0, i = 0; i<s->length_size; i++)
nal_size = (nal_size << 8) | buf[i];
//向後移動s->length_size
buf += s->length_size;
//NALU type
unit_type = *buf & 0x1f;
//長度出錯
if (nal_size > buf_end - buf || nal_size < 0)
goto fail;
//SPS資料 新的IDR標誌 SPS直接拷貝加到快取中
if (unit_type == 7)
s->idr_sps_seen = s->new_idr = 1;
//PPS資料 新的IDR標誌
else if (unit_type == 8)
{
s->idr_pps_seen = s->new_idr = 1;
/* if SPS has not been seen yet, prepend the AVCC one to PPS */
//當前是收到pps包 前面卻沒有SPS包，需要從擴充套件資料中把SPS加上
//如果有sps 直接走到 一般資料拷貝else
if (!s->idr_sps_seen)
{
if (s->sps_offset == -1)
av_log(ctx, AV_LOG_WARNING, "SPS not present in the stream, nor in AVCC, stream may be unreadable\n");
else
{
//把sps和pps 都拷貝到資料快取中
if ((ret = alloc_and_copy(out,
ctx->par_out->extradata + s->sps_offset,
s->pps_offset != -1 ? s->pps_offset : ctx->par_out->extradata_size - s->sps_offset,
buf, nal_size)) < 0)
goto fail;
s->idr_sps_seen = 1;
goto next_nal;
}
}
}
/* if this is a new IDR picture following an IDR picture, reset the idr flag.
* Just check first_mb_in_slice to be 0 as this is the simplest solution.
* This could be checking idr_pic_id instead, but would complexify the parsing. */
if (!s->new_idr && unit_type == 5 && (buf[1] & 0x80))
s->new_idr = 1;
/* prepend only to the first type 5 NAL unit of an IDR picture, if no sps/pps are already present */
//新的IDR
if (s->new_idr && unit_type == 5 && !s->idr_sps_seen && !s->idr_pps_seen)
{
if ((ret=alloc_and_copy(out,
ctx->par_out->extradata, ctx->par_out->extradata_size,
buf, nal_size)) < 0)
goto fail;
s->new_idr = 0;
/* if only SPS has been seen, also insert PPS */
}
else if (s->new_idr && unit_type == 5 && s->idr_sps_seen && !s->idr_pps_seen)
{
if (s->pps_offset == -1)
{
av_log(ctx, AV_LOG_WARNING, "PPS not present in the stream, nor in AVCC, stream may be unreadable\n");
if ((ret = alloc_and_copy(out, NULL, 0, buf, nal_size)) < 0)
goto fail;
}
else if ((ret = alloc_and_copy(out,
ctx->par_out->extradata + s->pps_offset, ctx->par_out->extradata_size - s->pps_offset,
buf, nal_size)) < 0)
goto fail;
}
else
{
if ((ret=alloc_and_copy(out, NULL, 0, buf, nal_size)) < 0)
goto fail;
if (!s->new_idr && unit_type == 1)
{
s->new_idr = 1;
s->idr_sps_seen = 0;
s->idr_pps_seen = 0;
}
}
next_nal:
buf += nal_size;
cumul_size += nal_size + s->length_size;
}
while (cumul_size < buf_size);
ret = av_packet_copy_props(out, in);
if (ret < 0)
goto fail;
fail:
if (ret < 0)
av_packet_unref(out);
av_packet_free(&in);
return ret;
}4.解析AudioSpecificConfig
AAC也會把AudioSpecificConfig儲存到streams[index]->codec->extradata擴充套件資料中，但是FFMPEG 沒有提供加ADTS的位元流過濾器，需要自己解析AudioSpecificConfig，然後新增到AAC幀前面。
//解析AudioSpecificConfig
int aac_decode_extradata(ADTSContext *adts, unsigned char *pbuf, int bufsize)
{
int aot, aotext, samfreindex;
int i, channelconfig;
unsigned char *p = pbuf;
if (!adts || !pbuf || bufsize<2)
{
return -1;
}
aot = (p[0]>>3)&0x1f;
if (aot == 31)
{
aotext = (p[0]<<3 | (p[1]>>5)) & 0x3f;
aot = 32 + aotext;
samfreindex = (p[1]>>1) & 0x0f;
if (samfreindex == 0x0f)
{
channelconfig = ((p[4]<<3) | (p[5]>>5)) & 0x0f;
}
else
{
channelconfig = ((p[1]<<3)|(p[2]>>5)) & 0x0f;
}
}
else
{
samfreindex = ((p[0]<<1)|p[1]>>7) & 0x0f;
if (samfreindex == 0x0f)
{
channelconfig = (p[4]>>3) & 0x0f;
}
else
{
channelconfig = (p[1]>>3) & 0x0f;
}
}
#ifdef AOT_PROFILE_CTRL
if (aot < 2) aot = 2;
#endif
adts->objecttype = aot-1;
adts->sample_rate_index = samfreindex;
adts->channel_conf = channelconfig;
adts->write_adts = 1;
return 0;
} 5.DEMO程式
程式運用ffmpeg把h264+aac封裝的FLV，解複用成兩個檔案，h264檔案和AAC檔案。
/*******************************************************************************
Copyright (c) wubihe Tech. Co., Ltd. All rights reserved.
--------------------------------------------------------------------------------
Date Created: 2014-10-25
Author: wubihe QQ:1269122125 Email:[email protected]
Description: 解複用flv儲存成h264檔案和aac檔案
--------------------------------------------------------------------------------
Modification History
DATE AUTHOR DESCRIPTION
--------------------------------------------------------------------------------
********************************************************************************/
#include <stdio.h>
#define __STDC_CONSTANT_MACROS
extern "C"
{
#include "libavformat/avformat.h"
};
//封裝格式MKV/MP4/FLV中如果有AAC的情況，首先這些封裝格式中包含AudioSpecificConfig，
//儲存在音訊流的AVCodecContext->extradata 裡面，需要解析然後封裝ADTS即可
#define DEMUXER_AAC 1
#define DEMUXER_MP3 0
#define ADTS_HEADER_SIZE (7)
//FLV封裝音視訊 AAC封裝在第一個AAC TAG會封裝一個AudioSpecificConfig結構
//AudioSpecificConfig解析結果儲存在該結構體中
typedef struct
{
int write_adts;
int objecttype;
int sample_rate_index;
int channel_conf;
}ADTSContext;
//解析AudioSpecificConfig
int aac_decode_extradata(ADTSContext *adts, unsigned char *pbuf, int bufsize)
{
int aot, aotext, samfreindex;
int i, channelconfig;
unsigned char *p = pbuf;
if (!adts || !pbuf || bufsize<2)
{
return -1;
}
aot = (p[0]>>3)&0x1f;
if (aot == 31)
{
aotext = (p[0]<<3 | (p[1]>>5)) & 0x3f;
aot = 32 + aotext;
samfreindex = (p[1]>>1) & 0x0f;
if (samfreindex == 0x0f)
{
channelconfig = ((p[4]<<3) | (p[5]>>5)) & 0x0f;
}
else
{
channelconfig = ((p[1]<<3)|(p[2]>>5)) & 0x0f;
}
}
else
{
samfreindex = ((p[0]<<1)|p[1]>>7) & 0x0f;
if (samfreindex == 0x0f)
{
channelconfig = (p[4]>>3) & 0x0f;
}
else
{
channelconfig = (p[1]>>3) & 0x0f;
}
}
#ifdef AOT_PROFILE_CTRL
if (aot < 2) aot = 2;
#endif
adts->objecttype = aot-1;
adts->sample_rate_index = samfreindex;
adts->channel_conf = channelconfig;
adts->write_adts = 1;
return 0;
}
//新增ADTS頭
int aac_set_adts_head(ADTSContext *acfg, unsigned char *buf, int size)
{
unsigned char byte;
if (size < ADTS_HEADER_SIZE)
{
return -1;
}
buf[0] = 0xff;
buf[1] = 0xf1;
byte = 0;
byte |= (acfg->objecttype & 0x03) << 6;
byte |= (acfg->sample_rate_index & 0x0f) << 2;
byte |= (acfg->channel_conf & 0x07) >> 2;
buf[2] = byte;
byte = 0;
byte |= (acfg->channel_conf & 0x07) << 6;
byte |= (ADTS_HEADER_SIZE + size) >> 11;
buf[3] = byte;
byte = 0;
byte |= (ADTS_HEADER_SIZE + size) >> 3;
buf[4] = byte;
byte = 0;
byte |= ((ADTS_HEADER_SIZE + size) & 0x7) << 5;
byte |= (0x7ff >> 6) & 0x1f;
buf[5] = byte;
byte = 0;
byte |= (0x7ff & 0x3f) << 2;
buf[6] = byte;
return 0;
}
int main(int argc, char* argv[])
{
AVFormatContext *ifmt_ctx = NULL;
AVPacket pkt;
int ret, i;
int videoindex=-1,audioindex=-1;
#if DEMUXER_AAC
const char *in_filename = "titanic.flv"; //Input file URL
const char *out_filename_v = "titanic.h264"; //Output file URL
const char *out_filename_a = "titanic.aac";
#endif
#if DEMUXER_MP3
const char *in_filename = "titanic.flv"; //Input file URL
const char *out_filename_v = "titanic.h264"; //Output file URL
const char *out_filename_a = "titanic.mp3";
#endif
av_register_all();
//Input
if ((ret = avformat_open_input(&ifmt_ctx, in_filename, 0, 0)) < 0)
{
printf( "Could not open input file.");
return -1;
}
if ((ret = avformat_find_stream_info(ifmt_ctx, 0)) < 0)
{
printf( "Failed to retrieve input stream information");
return -1;
}
videoindex=-1;
for(i=0; i<ifmt_ctx->nb_streams; i++)
{
if(ifmt_ctx->streams[i]->codec->codec_type==AVMEDIA_TYPE_VIDEO)
{
videoindex=i;
}else if(ifmt_ctx->streams[i]->codec->codec_type==AVMEDIA_TYPE_AUDIO)
{
audioindex=i;
}
}
FILE *fp_audio=fopen(out_filename_a,"wb+");
FILE *fp_video=fopen(out_filename_v,"wb+");
//FLV/MP4/MKV等結構中，h264需要h264_mp4toannexb處理。新增SPS/PPS等資訊。FLV封裝時，可以把
//多個NALU放在一個VIDEO TAG中,結構為4B NALU長度+NALU1+4B NALU長度+NALU2+...,需要做的處理把4B
//長度換成00000001或者000001
AVBitStreamFilterContext* h264bsfc = av_bitstream_filter_init("h264_mp4toannexb");
#if DEMUXER_AAC
ADTSContext stADTSContext;
unsigned char pAdtsHead[7];
#endif
while(av_read_frame(ifmt_ctx, &pkt)>=0)
{
if(pkt.stream_index==videoindex)
{
av_bitstream_filter_filter(h264bsfc, ifmt_ctx->streams[videoindex]->codec, NULL, &pkt.data, &pkt.size, pkt.data, pkt.size, 0);
printf("Write Video Packet. size:%d\tpts:%lld\n",pkt.size,pkt.pts);
fwrite(pkt.data,1,pkt.size,fp_video);
}
else if(pkt.stream_index==audioindex)
{
//AAC在封裝結構MKV/FLV/MP4結構中，需要手動新增ADTS
#if DEMUXER_AAC
aac_decode_extradata(&stADTSContext, ifmt_ctx->streams[audioindex]->codec->extradata, ifmt_ctx->streams[audioindex]->codec->extradata_size);
aac_set_adts_head(&stADTSContext, pAdtsHead, pkt.size);
fwrite(pAdtsHead, 1, 7, fp_audio);
#endif
//一般結構如MP3直接寫檔案即可
printf("Write Audio Packet. size:%d\tpts:%lld\n",pkt.size,pkt.pts);
fwrite(pkt.data,1,pkt.size,fp_audio);
}
av_free_packet(&pkt);
}
av_bitstream_filter_close(h264bsfc);
fclose(fp_video);
fclose(fp_audio);
avformat_close_input(&ifmt_ctx);
if (ret < 0 && ret != AVERROR_EOF)
{
printf( "Error occurred.\n");
return -1;
}
return 0;
}
6.執行結果
生成titanic.h264和titanic.aac兩個檔案，用ffplay.exe可以驗證播放。
編譯環境： Win7_64bit+VS2008
DEMO下載地址:https://download.csdn.net/download/hiwubihe/10643142