FFmpeg进阶: 给视频添加图片水印

文章目录

定义水印滤镜
定义输入和输出
项目工程
使用效果

为了声明视频的原创性，防止别人抄袭搬运，很多时候我们会给视频资源添加水印进行版权保护。FFmpeg的滤镜模块支持给视频资源添加水印。这里就介绍一下如何通过FFmpeg库给视频资源添加图片水印。视频添加水印的流程如下图所示:

首先我们对原始视频文件进行解封装，分离出视频流和音频流。音频流不做处理直接输出到文件中。视频流解析之后，通过滤镜将解析出来的视频帧和图片数据进行组合拼接之后进行输出。将输出的音视频流进行封装之后输出的视频文件就携带了对应的图片水印。

定义水印滤镜

水印滤镜主要实现两个功能
1.对原始的图片尺寸进行缩放
2.将缩放后的图片和原始的视频帧进行组合
对应的滤镜实现如下所示:

//滤镜首先将图片缩放到50*50
//然后将缩放之后的图片放到原始帧的(15,15)的位置
const char *filter_descr = "[in]scale=50:50[scl];[in1][scl]overlay=15:15";
static int init_filters(const char *filters_descr)
{char args[512];char logoargs[512];int ret = 0;//定义使用到的滤镜const AVFilter *buffersrc = avfilter_get_by_name("buffer");const AVFilter *buffersrc1 = avfilter_get_by_name("buffer");const AVFilter *buffersink = avfilter_get_by_name("buffersink");AVFilterInOut *outputs = avfilter_inout_alloc();AVFilterInOut *overlayFrameInOut = avfilter_inout_alloc();AVFilterInOut *inputs = avfilter_inout_alloc();AVRational time_base = inputFormatContext->streams[videoStreamIndex]->time_base;enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_YUV420P, AV_PIX_FMT_NONE };filterGraph = avfilter_graph_alloc();if (!outputs || !inputs || !filterGraph) {ret = AVERROR(ENOMEM);goto end;}//原始帧的格式snprintf(args, sizeof(args),"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",outputVideoCodecContext->width, outputVideoCodecContext->height, outputVideoCodecContext->pix_fmt,time_base.num, time_base.den,outputVideoCodecContext->sample_aspect_ratio.num, outputVideoCodecContext->sample_aspect_ratio.den);//图片的格式snprintf(logoargs, sizeof(logoargs),"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",logoDecoderContext->width, logoDecoderContext->height, logoDecoderContext->pix_fmt,time_base.num, time_base.den,outputVideoCodecContext->sample_aspect_ratio.num, outputVideoCodecContext->sample_aspect_ratio.den);//缩放滤镜的输入(图片)ret = avfilter_graph_create_filter(&bufferSourceContext, buffersrc, "in",logoargs, NULL, filterGraph);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot create buffer source\n");goto end;}//水印滤镜的输入(原始帧)ret = avfilter_graph_create_filter(&bufferSource1Context, buffersrc1, "in1",args, NULL, filterGraph);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot create buffer source\n");goto end;}//设置滤镜输出ret = avfilter_graph_create_filter(&bufferSinkContext, buffersink, "out",NULL, NULL, filterGraph);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot create buffer sink\n");goto end;}//设置输出的像素格式ret = av_opt_set_int_list(bufferSinkContext, "pix_fmts", pix_fmts,AV_PIX_FMT_NONE, AV_OPT_SEARCH_CHILDREN);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot set output pixel format\n");goto end;}//设置滤镜的端点//outputs变量意指bufferSourceContext滤镜的输出引脚(outputpad)//src缓冲区(bufferSourceContext滤镜)的输出必须连到filters_descr中第一个//滤镜的输输入filtersdescr中第一个滤镜的输入标号未指定，故默认为//"in"此处将buffersrc_ctx的输出标号也设为"in"outputs->name = av_strdup("in");outputs->filter_ctx = bufferSourceContext;outputs->pad_idx = 0;outputs->next = overlayFrameInOut;overlayFrameInOut->name = av_strdup("in1");overlayFrameInOut->filter_ctx = bufferSource1Context;overlayFrameInOut->pad_idx = 0;overlayFrameInOut->next = NULL;//inputs变量意指bufferSinkContext滤镜的输入引脚(inputpad)//sink缓冲区(bufferSinkContext滤镜)的输入必须连到filters_descr中最后//一个滤镜的输出filtersdescr中最后一个滤镜的输出标号未指定，故//默认为"out"此处将bufferSinkContext的输出标号也设为"out"inputs->name = av_strdup("out");inputs->filter_ctx = bufferSinkContext;inputs->pad_idx = 0;inputs->next = NULL;// 在filters_descr字符串中，如果第一个滤镜未指定输入标号，则假定为"in"// 如果最后一个滤镜未指定输出标号，则假定为"out"。几个实参说明如下：// @filter_graph  滤镜容器// @filters_descr 描述滤镜的字符串// @inputs        指向滤镜图中输入链表// @outputs       指向滤镜图的输出链表// 调用前：filterGraph包含三个滤镜两个buffersrcctx和一个buffersink_ctx// 调用后：filtersdescr描述的滤镜图插入到filter_graph中，buffersrcctx连接到filtersdescr// 的输入，filtersdescr的输出连接到buffersink_ctx，filters_descr只进行了解析而不// 建立内部滤镜间的连接。filters_desc与filter_graph间的连接是利用AVFilterInOut inputs// 和AVFilterInOut inputs连接起来的，AVFilterInOut是一个链表，最终可用的连在一起的// 滤镜链/滤镜图就是通过这个链表串在一起的。if ((ret = avfilter_graph_parse_ptr(filterGraph, filters_descr,&inputs, &outputs, NULL)) < 0)goto end;if ((ret = avfilter_graph_config(filterGraph, NULL)) < 0)goto end;//输出滤镜的调用流程char *graph_str = avfilter_graph_dump(filterGraph, NULL);printf("%s", graph_str);end:avfilter_inout_free(&inputs);avfilter_inout_free(&outputs);return ret;
}

整个滤镜的调用流程如下图所示:

定义输入和输出

1.定义视频输入

static int open_meida_file(std::string file_path)
{int ret;AVCodec *dec;//打开媒体文件if ((ret = avformat_open_input(&inputFormatContext, file_path.c_str(), NULL, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot open input file\n");return ret;}if ((ret = avformat_find_stream_info(inputFormatContext, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find stream information\n");return ret;}//查找音视频流ret = av_find_best_stream(inputFormatContext, AVMEDIA_TYPE_VIDEO, -1, -1, &dec, 0);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find a video stream in the input file\n");return ret;}videoStreamIndex = ret;ret = av_find_best_stream(inputFormatContext, AVMEDIA_TYPE_AUDIO, -1, -1, NULL, 0);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find a audio stream in the input file\n");return ret;}audioStreamIndex = ret;//输出视频上下文outputVideoCodecContext = avcodec_alloc_context3(outputVideoCodec);if (!outputVideoCodecContext)return AVERROR(ENOMEM);//依据输入媒体文件创建输出视频编码器的上下文outputVideoCodecContext->time_base = inputFormatContext->streams[videoStreamIndex]->time_base;outputVideoCodecContext->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;outputVideoCodecContext->sample_fmt = AV_SAMPLE_FMT_S16;outputVideoCodecContext->width = inputFormatContext->streams[videoStreamIndex]->codecpar->width;outputVideoCodecContext->height = inputFormatContext->streams[videoStreamIndex]->codecpar->height;outputVideoCodecContext->bit_rate = inputFormatContext->streams[videoStreamIndex]->codecpar->bit_rate;outputVideoCodecContext->pix_fmt = AV_PIX_FMT_YUV420P;outputVideoCodecContext->profile = FF_PROFILE_H264_MAIN;outputVideoCodecContext->level = 41;outputVideoCodecContext->thread_count = 8;//视频解码器的上下文videoDecoderContext = avcodec_alloc_context3(dec);if (!videoDecoderContext)return AVERROR(ENOMEM);avcodec_parameters_to_context(videoDecoderContext, inputFormatContext->streams[videoStreamIndex]->codecpar);//打开解码器if ((ret = avcodec_open2(videoDecoderContext, dec, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot open video decoder\n");return ret;}return 0;
}

2.定义图片输入

//打开图片资源
static int open_logo_file(std::string logo_path)
{int ret = -1;AVCodec *logoDec;//打开文件解析上下文if ((ret = avformat_open_input(&logoFormatContext, logo_path.c_str(), NULL, NULL)) < 0){av_log(NULL, AV_LOG_ERROR, "Cannot open logo input file\n");return ret;}if ((ret = avformat_find_stream_info(logoFormatContext, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find logo stream information\n");return ret;}ret = av_find_best_stream(logoFormatContext, AVMEDIA_TYPE_VIDEO, -1, -1, &logoDec, 0);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find a video stream in the logo input file\n");return ret;}//图片解码器上下文logoDecoderContext = avcodec_alloc_context3(logoDec);if (!logoDecoderContext)return AVERROR(ENOMEM);avcodec_parameters_to_context(logoDecoderContext, logoFormatContext->streams[0]->codecpar);//打开解码器if ((ret = avcodec_open2(logoDecoderContext, logoDec, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot open video decoder\n");return ret;}return 0;
}

3.定义视频输出

//初始化输出流
static int initializeOutputFile(std::string output_path)
{int error;//视频流avformat_alloc_output_context2(&outputFormatContext, NULL, "flv", output_path.c_str());if (!outputFormatContext){av_log(NULL, AV_LOG_ERROR, "Could not create output context!");return -1;}videoStream = avformat_new_stream(outputFormatContext, outputVideoCodec);if (!videoStream){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not create Video Stream");return -1;}avcodec_parameters_from_context(videoStream->codecpar,outputVideoCodecContext);videoStream->codecpar->codec_tag = 0;//音频流audioStream = avformat_new_stream(outputFormatContext, NULL);if (!audioStream){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not create Live Audio Stream");return -1;}audioStream->codecpar = inputFormatContext->streams[audioStreamIndex]->codecpar;audioStream->codecpar->codec_tag = 0;//打开输出IOerror = avio_open(&outputFormatContext->pb, output_path.c_str(), AVIO_FLAG_WRITE);if (error < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not open output url");return -2;}//输出文件头error = avformat_write_header(outputFormatContext, NULL);if (error < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not write header to output format context");return -1;}return 0;
}//将加水印之后的图像帧输出到文件中
static int output_frame(AVFrame *frame, AVRational time_base)
{int x, y;uint8_t *p0, *p;int64_t delay;int code;int gotOutput;AVPacket packet = { 0 };av_init_packet(&packet);int ret = avcodec_send_frame(outputVideoCodecContext, frame);if (ret < 0) {printf("Error sending a frame for encoding\n");return -1;}while (ret >= 0) {ret = avcodec_receive_packet(outputVideoCodecContext, &packet);if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {return (ret == AVERROR(EAGAIN)) ? 0 : 1;}else if (ret < 0) {printf("Error during encoding\n");exit(1);}gotOutput = 1;AVRational avTimeBaseQ = { 1, AV_TIME_BASE };int64_t ptsTime = av_rescale_q(frame->pts, inputFormatContext->streams[videoStreamIndex]->time_base, avTimeBaseQ);int64_t nowTime = av_gettime() - startTime;if ((ptsTime > nowTime)) {int64_t sleepTime = ptsTime - nowTime;av_usleep((sleepTime));}else{printf("not sleeping\n");}if (gotOutput){packet.pts = av_rescale_q_rnd(packet.pts, time_base, videoStream->time_base,(AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));packet.dts = av_rescale_q_rnd(packet.dts, time_base, videoStream->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));packet.stream_index = videoStream->index;code = av_interleaved_write_frame(outputFormatContext, &packet);av_packet_unref(&packet);if (code < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] Writing Live Stream Interleaved Frame");}if (ret < 0) {exit(1);}av_packet_unref(&packet);}}
}

项目工程

完整的调用流程如下图所示:

extern "C"
{#include <libavcodec/avcodec.h>
#include <libavformat/avformat.h>
#include <libavfilter/buffersink.h>
#include <libavfilter/buffersrc.h>
#include <libavutil/opt.h>
#include <libavutil/mathematics.h>
#include <libavutil/time.h>
#include <libavutil/imgutils.h>
#include <libswscale/swscale.h>
#include <stdio.h>
#include <stdlib.h>
}#include <string>
#include <thread>//水印的大小为50pix*50pix
//放的位置为15*15
const char *filter_descr = "[in]scale=50:50[scl];[in1][scl]overlay=15:15";static AVFormatContext *outputFormatContext;
static AVFormatContext *inputFormatContext;   //输入媒体文件上下文
static AVFormatContext *logoFormatContext;    //图标文件的上下文static AVCodecContext *videoDecoderContext;   //视频解码器上下文
static AVCodecContext *logoDecoderContext;
static AVCodecContext *outputVideoCodecContext;static AVCodec *outputVideoCodec;
static AVCodec *outputAudioCodec;//输出音频流和视频流
static AVStream *audioStream;
static AVStream *videoStream;AVFilterContext *bufferSinkContext;
AVFilterContext *bufferSourceContext;
AVFilterContext *bufferSource1Context;
AVFilterGraph *filterGraph;//视频流索引和音频流索引
static int videoStreamIndex = -1;
static int audioStreamIndex = -1;
static int64_t lastPts = AV_NOPTS_VALUE;
static int64_t startTime; //起始时间static std::string input_meida_path;  //媒体文件的地址
static std::string logo_path;         //图标文件的地址
static std::string output_meida_path; //输出媒体文件的地址//解析图片帧的线程
static std::thread decodeTask;//水印帧
static AVFrame* logoFrame;
static AVFrame *filt_frame;//打开图片的解码器
static int open_logo_file(std::string logo_path)
{int ret = -1;AVCodec *logoDec;if ((ret = avformat_open_input(&logoFormatContext, logo_path.c_str(), NULL, NULL)) < 0){av_log(NULL, AV_LOG_ERROR, "Cannot open logo input file\n");return ret;}if ((ret = avformat_find_stream_info(logoFormatContext, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find logo stream information\n");return ret;}ret = av_find_best_stream(logoFormatContext, AVMEDIA_TYPE_VIDEO, -1, -1, &logoDec, 0);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find a video stream in the logo input file\n");return ret;}//图片解码器上下文logoDecoderContext = avcodec_alloc_context3(logoDec);if (!logoDecoderContext)return AVERROR(ENOMEM);avcodec_parameters_to_context(logoDecoderContext, logoFormatContext->streams[0]->codecpar);//打开解码器if ((ret = avcodec_open2(logoDecoderContext, logoDec, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot open video decoder\n");return ret;}return 0;
}static int open_meida_file(std::string file_path)
{int ret;AVCodec *dec;if ((ret = avformat_open_input(&inputFormatContext, file_path.c_str(), NULL, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot open input file\n");return ret;}if ((ret = avformat_find_stream_info(inputFormatContext, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find stream information\n");return ret;}//查找音视频流ret = av_find_best_stream(inputFormatContext, AVMEDIA_TYPE_VIDEO, -1, -1, &dec, 0);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find a video stream in the input file\n");return ret;}videoStreamIndex = ret;ret = av_find_best_stream(inputFormatContext, AVMEDIA_TYPE_AUDIO, -1, -1, NULL, 0);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot find a audio stream in the input file\n");return ret;}audioStreamIndex = ret;//输出的视频上下文outputVideoCodecContext = avcodec_alloc_context3(outputVideoCodec);if (!outputVideoCodecContext)return AVERROR(ENOMEM);//依据输入媒体文件创建输出视频编码器的上下文outputVideoCodecContext->time_base = inputFormatContext->streams[videoStreamIndex]->time_base;outputVideoCodecContext->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;outputVideoCodecContext->sample_fmt = AV_SAMPLE_FMT_S16;outputVideoCodecContext->width = inputFormatContext->streams[videoStreamIndex]->codecpar->width;outputVideoCodecContext->height = inputFormatContext->streams[videoStreamIndex]->codecpar->height;outputVideoCodecContext->bit_rate = inputFormatContext->streams[videoStreamIndex]->codecpar->bit_rate;outputVideoCodecContext->pix_fmt = AV_PIX_FMT_YUV420P;outputVideoCodecContext->profile = FF_PROFILE_H264_MAIN;outputVideoCodecContext->level = 41;outputVideoCodecContext->thread_count = 8;//视频解码器的上下文videoDecoderContext = avcodec_alloc_context3(dec);if (!videoDecoderContext)return AVERROR(ENOMEM);avcodec_parameters_to_context(videoDecoderContext, inputFormatContext->streams[videoStreamIndex]->codecpar);//打开解码器if ((ret = avcodec_open2(videoDecoderContext, dec, NULL)) < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot open video decoder\n");return ret;}return 0;
}static int init_filters(const char *filters_descr)
{char args[512];char logoargs[512];int ret = 0;//定义使用到的滤镜const AVFilter *buffersrc = avfilter_get_by_name("buffer");const AVFilter *buffersrc1 = avfilter_get_by_name("buffer");const AVFilter *buffersink = avfilter_get_by_name("buffersink");AVFilterInOut *outputs = avfilter_inout_alloc();AVFilterInOut *overlayFrameInOut = avfilter_inout_alloc();AVFilterInOut *inputs = avfilter_inout_alloc();AVRational time_base = inputFormatContext->streams[videoStreamIndex]->time_base;enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_YUV420P, AV_PIX_FMT_NONE };filterGraph = avfilter_graph_alloc();if (!outputs || !inputs || !filterGraph) {ret = AVERROR(ENOMEM);goto end;}//原始帧的格式snprintf(args, sizeof(args),"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",outputVideoCodecContext->width, outputVideoCodecContext->height, outputVideoCodecContext->pix_fmt,time_base.num, time_base.den,outputVideoCodecContext->sample_aspect_ratio.num, outputVideoCodecContext->sample_aspect_ratio.den);//图片帧的格式snprintf(logoargs, sizeof(logoargs),"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",logoDecoderContext->width, logoDecoderContext->height, logoDecoderContext->pix_fmt,time_base.num, time_base.den,outputVideoCodecContext->sample_aspect_ratio.num, outputVideoCodecContext->sample_aspect_ratio.den);//缩放滤镜的输入(图片)ret = avfilter_graph_create_filter(&bufferSourceContext, buffersrc, "in",logoargs, NULL, filterGraph);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot create buffer source\n");goto end;}//水印滤镜的输入ret = avfilter_graph_create_filter(&bufferSource1Context, buffersrc1, "in1",args, NULL, filterGraph);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot create buffer source\n");goto end;}//设置滤镜输出ret = avfilter_graph_create_filter(&bufferSinkContext, buffersink, "out",NULL, NULL, filterGraph);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot create buffer sink\n");goto end;}//设置输出的像素格式ret = av_opt_set_int_list(bufferSinkContext, "pix_fmts", pix_fmts,AV_PIX_FMT_NONE, AV_OPT_SEARCH_CHILDREN);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Cannot set output pixel format\n");goto end;}//设置滤镜的端点//outputs变量意指bufferSourceContext滤镜的输出引脚(outputpad)//src缓冲区(bufferSourceContext滤镜)的输出必须连到filters_descr中第一个//滤镜的输输入filtersdescr中第一个滤镜的输入标号未指定，故默认为//"in"此处将buffersrc_ctx的输出标号也设为"in"outputs->name = av_strdup("in");outputs->filter_ctx = bufferSourceContext;outputs->pad_idx = 0;outputs->next = overlayFrameInOut;overlayFrameInOut->name = av_strdup("in1");overlayFrameInOut->filter_ctx = bufferSource1Context;overlayFrameInOut->pad_idx = 0;overlayFrameInOut->next = NULL;//inputs变量意指bufferSinkContext滤镜的输入引脚(inputpad)//sink缓冲区(bufferSinkContext滤镜)的输入必须连到filters_descr中最后//一个滤镜的输出filtersdescr中最后一个滤镜的输出标号未指定，故//默认为"out"此处将bufferSinkContext的输出标号也设为"out"inputs->name = av_strdup("out");inputs->filter_ctx = bufferSinkContext;inputs->pad_idx = 0;inputs->next = NULL;// 在filters_descr字符串中，如果第一个滤镜未指定输入标号，则假定为"in"；// 如果最后一个滤镜未指定输出标号，则假定为"out"。几个实参说明如下：// @filter_graph  滤镜容器// @filters_descr 描述滤镜的字符串// @inputs        指向滤镜图中输入链表// @outputs       指向滤镜图的输出链表// 调用前：filterGraph包含三个滤镜两个buffersrcctx和一个buffersink_ctx// 调用后：filtersdescr描述的滤镜图插入到filter_graph中，buffersrcctx连接到filtersdescr// 的输入，filtersdescr的输出连接到buffersink_ctx，filters_descr只进行了解析而不// 建立内部滤镜间的连接。filters_desc与filter_graph间的连接是利用AVFilterInOut inputs// 和AVFilterInOut inputs连接起来的，AVFilterInOut是一个链表，最终可用的连在一起的// 滤镜链/滤镜图就是通过这个链表串在一起的。if ((ret = avfilter_graph_parse_ptr(filterGraph, filters_descr,&inputs, &outputs, NULL)) < 0)goto end;if ((ret = avfilter_graph_config(filterGraph, NULL)) < 0)goto end;//输出滤镜的调用流程char *graph_str = avfilter_graph_dump(filterGraph, NULL);printf("%s", graph_str);end:avfilter_inout_free(&inputs);avfilter_inout_free(&outputs);return ret;
}//将加水印之后的图像帧输出到文件中
static int output_frame(AVFrame *frame, AVRational time_base)
{int x, y;uint8_t *p0, *p;int64_t delay;int code;int gotOutput;AVPacket packet = { 0 };av_init_packet(&packet);int ret = avcodec_send_frame(outputVideoCodecContext, frame);if (ret < 0) {printf("Error sending a frame for encoding\n");return -1;}while (ret >= 0) {ret = avcodec_receive_packet(outputVideoCodecContext, &packet);if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {return (ret == AVERROR(EAGAIN)) ? 0 : 1;}else if (ret < 0) {printf("Error during encoding\n");exit(1);}gotOutput = 1;AVRational avTimeBaseQ = { 1, AV_TIME_BASE };int64_t ptsTime = av_rescale_q(frame->pts, inputFormatContext->streams[videoStreamIndex]->time_base, avTimeBaseQ);int64_t nowTime = av_gettime() - startTime;if ((ptsTime > nowTime)) {int64_t sleepTime = ptsTime - nowTime;av_usleep((sleepTime));}else{printf("not sleeping\n");}if (gotOutput){packet.pts = av_rescale_q_rnd(packet.pts, time_base, videoStream->time_base,(AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));packet.dts = av_rescale_q_rnd(packet.dts, time_base, videoStream->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));packet.stream_index = videoStream->index;code = av_interleaved_write_frame(outputFormatContext, &packet);av_packet_unref(&packet);if (code < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] Writing Live Stream Interleaved Frame");}if (ret < 0) {exit(1);}av_packet_unref(&packet);}}
}static int initializeOutputFile(std::string output_path)
{int error;//视频流avformat_alloc_output_context2(&outputFormatContext, NULL, "flv", output_path.c_str());if (!outputFormatContext){av_log(NULL, AV_LOG_ERROR, "Could not create output context!");return -1;}videoStream = avformat_new_stream(outputFormatContext, outputVideoCodec);if (!videoStream){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not create Video Stream");return -1;}avcodec_parameters_from_context(videoStream->codecpar,outputVideoCodecContext);videoStream->codecpar->codec_tag = 0;//音频流audioStream = avformat_new_stream(outputFormatContext, NULL);if (!audioStream){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not create Live Audio Stream");return -1;}audioStream->codecpar = inputFormatContext->streams[audioStreamIndex]->codecpar;audioStream->codecpar->codec_tag = 0;//打开输出IOerror = avio_open(&outputFormatContext->pb, output_path.c_str(), AVIO_FLAG_WRITE);if (error < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not open output url");return -2;}//输出文件头error = avformat_write_header(outputFormatContext, NULL);if (error < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] Could not write header to output format context");return -1;}return 0;
}int main(int argc, char **argv)
{int ret;AVPacket packet;AVFrame *frame;int error;avformat_network_init();if (argc != 4) {printf("usage argv[1] input file path argv[2] logo path argv[3] outputpath");exit(1);}//媒体文件地址和图标地址input_meida_path = std::string(argv[1]);logo_path = std::string(argv[2]);output_meida_path = std::string(argv[3]);//输出视频上下文outputVideoCodec = avcodec_find_encoder(AV_CODEC_ID_H264);if (!outputVideoCodec){printf("Can't find h264 encoder");exit(1);}//输出音频上下文outputAudioCodec = avcodec_find_encoder(AV_CODEC_ID_AAC);if (!outputVideoCodec){printf("Can't find aac encoder");exit(1);}//数据帧和滤镜帧frame = av_frame_alloc();filt_frame = av_frame_alloc();logoFrame = av_frame_alloc();if (!frame || !filt_frame) {printf("Could not allocate frame or filter frame");exit(1);}//打开媒体文件if ((ret = open_meida_file(input_meida_path)) < 0)goto end;//打开图标文件if ((ret = open_logo_file(logo_path)) < 0)goto end;//初始化滤镜if ((ret = init_filters(filter_descr)) < 0)goto end;//打开输出的编码器error = avcodec_open2(outputVideoCodecContext, outputVideoCodec, NULL);if (error < 0){printf("[ERROR] avcodec_open2");exit(1);}error = av_opt_set(outputVideoCodecContext->priv_data, "preset", "ultrafast", 0);if (error < 0){printf("[ERROR] av_opt_set preset");exit(1);}//初始化输出媒体文件ret = initializeOutputFile(output_meida_path);if (ret < 0){av_log(NULL, AV_LOG_ERROR, "[ERROR] initializeOutputFile()\n");exit(ret);}startTime = av_gettime();decodeTask.swap(std::thread([&] {bool ret = true;while (ret){AVPacket avPacket;av_init_packet(&avPacket);av_read_frame(logoFormatContext, &avPacket);ret = avcodec_send_packet(logoDecoderContext, &avPacket);ret = avcodec_receive_frame(logoDecoderContext, logoFrame);if (ret) break;}}));decodeTask.join();AVFrame* target_logo_frame = av_frame_alloc();av_frame_make_writable(target_logo_frame);target_logo_frame->format = videoDecoderContext->pix_fmt;target_logo_frame->width = logoFrame->width;target_logo_frame->height = logoFrame->height;target_logo_frame->pts = 0;ret = av_image_alloc(target_logo_frame->data, target_logo_frame->linesize, target_logo_frame->width, target_logo_frame->height,videoDecoderContext->pix_fmt, 32);//获取图片处理对象的上下文struct SwsContext *sws_ctx = sws_getContext(logoFrame->width, logoFrame->height, logoDecoderContext->pix_fmt, logoFrame->width, logoFrame->height,videoDecoderContext->pix_fmt, SWS_BILINEAR, NULL, NULL, NULL);//将图片数据添加到帧中uint8_t *src_planes[4];int src_linesize[4];av_image_fill_arrays(src_planes, src_linesize, (const uint8_t *)target_logo_frame->data, logoDecoderContext->pix_fmt,logoFrame->width, logoFrame->height, 1);sws_scale(sws_ctx, (const uint8_t *const *)src_planes, src_linesize, 0,logoFrame->height, target_logo_frame->data, target_logo_frame->linesize);//读取数据包while (1) {if ((ret = av_read_frame(inputFormatContext, &packet)) < 0)break;//处理视频流if (packet.stream_index == videoStreamIndex) { ret = avcodec_send_packet(videoDecoderContext, &packet);if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Error while sending a packet to the decoder\n");break;}while (ret >= 0) {ret = avcodec_receive_frame(videoDecoderContext, frame);if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {break;}else if (ret < 0) {av_log(NULL, AV_LOG_ERROR, "Error while receiving a frame from the decoder\n");goto end;}frame->pts = frame->best_effort_timestamp;logoFrame->pts = frame->best_effort_timestamp;//将原始帧添加到滤镜图像中//if (av_buffersrc_add_frame_flags(bufferSourceContext, frame, AV_BUFFERSRC_FLAG_KEEP_REF) < 0)if (av_buffersrc_add_frame_flags(bufferSourceContext, logoFrame, AV_BUFFERSRC_FLAG_KEEP_REF) < 0){av_log(NULL, AV_LOG_ERROR, "Error while feeding the filtergraph\n");break;}//添加图像信息if (av_buffersrc_add_frame_flags(bufferSource1Context, frame, AV_BUFFERSRC_FLAG_KEEP_REF) < 0) {av_log(NULL, AV_LOG_ERROR, "Error while feeding the filtergraph1\n");break;}/* pull filtered frames from the filtergraph */while (1) {ret = av_buffersink_get_frame(bufferSinkContext, filt_frame);if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)break;if (ret < 0)goto end;output_frame(filt_frame, bufferSinkContext->inputs[0]->time_base);av_frame_unref(filt_frame);}av_frame_unref(frame);}}//音频包直接输出else {packet.pts = av_rescale_q_rnd(packet.pts, inputFormatContext->streams[audioStreamIndex]->time_base, audioStream->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));packet.dts = av_rescale_q_rnd(packet.dts, inputFormatContext->streams[audioStreamIndex]->time_base, audioStream->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));packet.stream_index = audioStream->index;av_interleaved_write_frame(outputFormatContext, &packet);}av_packet_unref(&packet);}
end://添加数据结束字符,清理缓存error = av_write_trailer(outputFormatContext);avfilter_graph_free(&filterGraph);avcodec_free_context(&videoDecoderContext);avcodec_free_context(&logoDecoderContext);avformat_close_input(&inputFormatContext);av_frame_free(&frame);av_frame_free(&filt_frame);av_frame_free(&logoFrame);if (ret < 0 && ret != AVERROR_EOF) {exit(1);}exit(0);
}

使用效果

原始的视频输入如下图所示:

添加水印之后的效果图如下图所示:

FFmpeg进阶: 给视频添加图片水印相关推荐

用Java为视频添加图片水印（类似直播）
文章目录 1.首先[下载nginx](http://nginx-win.ecsds.eu/download/) 2.[下载ffmpeg](https://pan.baidu.com/s/1LUWeVn ...
PHP实现给视频加图片水印,怎么在视频画面上加图片？如何给视频添加自己的图片作为水印？视频添加图片水印的方法...
今天就是周一啦,昨天周末大家有没有跟好友去玩呢~反正小编是跟同学聚会去了,聚会主题:胖.哈哈哈,不过小编可不跟他们一样只会长胖,小编可是瘦瘦瘦的呢,偷偷的告诉你们,小编减肥瘦了三十斤哦,嘻嘻.好啦,废 ...
Moviepy模块之视频添加图片水印
文章目录前言视频添加图片水印 1.引入库 2.加载视频文件 3.加载水印图片 4.缩放水印图片大小 5.设置水印的位置 5.1 相对于视频的左上角 5.2 相对于视频的左下角 5.3 相对于视频的 ...
iOS GPUImage研究六：为视频添加图片水印
前言: 我们在前面已经做了关于.GPUImage导入方式.滤镜介绍.性能对比.图片处理.捕获照片处理.实时滤镜以及一个比较完整地Demo. 今天: 我们来给摄像头添加一个实时视频水印,很简单,大概有6 ...
视频添加图片水印？教你一招，给视频添加自己的水印
如今短视频行业发展得如火如荼,很多小伙伴都投入短视频的制作中,学会给视频添加水印是很有必要的,如果有小伙伴还不知道视频如何添加水印的话,就来看看小编分享的方法吧!小编演示操作步骤. 第一步,运行视频剪 ...
【Qt+FFmpeg】给视频添加时间水印
ffmpeg编解码中,给本地视频加上时间水印,并保存到本地,使用到的技术是ffmpeg中的avfilter库: 具体效果如下 yuv: mp4 本方法不适合摄像头解码,解码出来糊得不行,本地视频的话会 ...
FFmpeg进阶:给视频添加模糊滤镜
文章目录全部画面模糊处理部分画面模糊处理在视频的录制过程中很多时候为了保护视频中某些敏感信息不泄露,我们会给对应的视频部位添加模糊处理.这时候就需要用到FFmpeg的模糊滤镜了,模糊滤镜有很多种 ...
批量剪辑视频，添加图片水印
小伙伴在平时都会制作一些视频,需要在视频上面添加水印,今天为大家讲解一下如何用电脑批量给视频添加图片水印. 打开软件视频剪辑高手,主界面有三个功能,切换到批量剪辑视频页面. 第一步,添加视频,将需要剪 ...
使用ffmpeg给视频添加图片及文字水印
1. 准备工具 MacBook Pro homebrew ffmpeg 2.工具安装 homebrew安装说起homebrew确实是Mac上非常好用的一个工具,像什么CocoPods,node,ff ...
ffmpeg给视频添加时间水印
ffmpeg给视频添加时间水印通过 drawtext 滤镜模块给视频添加时间水印给视频添加时间水印用来做片源调试,非常方便的查找和定位处理的哪一帧视频片源: 1. 添加本地时间水印 ffmpeg ...