FFmpeg进阶教程之给视频添加文字水印

2022-11-13 19:15

短信预约 -IT技能 免费直播动态提醒

前言

和图片水印一样，很多时候为了声明视频的原创性，我们会给视频添加文字水印进行版权保护。添加文字水印和添加图片水印的流程相似，但又略有不同，这里介绍一下如何通过FFmpeg给视频添加文字水印。添加文字水印的流程图如下图所示:

文字水印配置项

在讲文字水印之前先介绍一下文字水印支持的那些配置，方便大家的使用。

项目	介绍
使用格式	drawtext=fontfile=font_f:text=text1…(通过冒号分割配置项,通过=给配置项赋值)
fontfile	用于绘制文本的字体文件的正确路径，强制参数
text	要绘制的文本字符串，必须是UTF-8编码的字符序列
x,y	绘制的位置的起始坐标值
fontcolor	字体颜色名称或0xRRGGBB[AA]格式的颜色，默认为黑色
fontsize	要绘制的文本字体大小，默认值为16
tabsize	用于呈现选项卡的空间大小，默认值为4
line_h,lh	每个文本行的高度
main_h,h,H	输入的高度
main_w,w,W	输入的宽度

常用的配置项主要有这些，如果需要其他的配置可以参考官方文档介绍。

文字水印关键点

中文的支持

和QT一样，FFmpeg绘制文字水印也存在中文乱码的问题。在windows下解决中文乱码主要需要以下几点:

1.将源码文件修改为utf-8编码

2.将编译编码类型修改为utf-8编码对应的配置如下:

#pragma execution_character_set("utf-8")

同时我们还应该确保使用的字体支持中文。

字体路径问题

指定字体文件路径是强制参数，可以使用绝对路径和相对路径

//使用工程相对路径下的字体
fontfile=.//simsun.ttc
//使用D盘绝对路径下的字体,要对斜杠进行转义
fontfile=D\\\\:simun.ttc

定义滤镜实现

文字水印对应的绘制流程图如下图所示:

文字水印滤镜的实现如下:

int InitFilter(AVCodecContext * codecContext)
{
	char args[512];
	int ret = 0;
	//缓存输入和缓存输出
	const AVFilter *bufferclass="lazy" data-src = avfilter_get_by_name("buffer");
	const AVFilter *buffersink = avfilter_get_by_name("buffersink");

	//创建输入输出参数
	AVFilterInOut *outputs = avfilter_inout_alloc();
	AVFilterInOut *inputs = avfilter_inout_alloc();

	//滤镜的描述
	//使用simhei字体,绘制的字体大小为100,文本内容为"鬼灭之刃",绘制位置为(100,100)
	//绘制的字体颜色为白色
	string  filters_descr = "drawtext=fontfile=.//simsun.ttc:fontsize=100:text=鬼灭之刃:x=100:y=100:fontcolor=0xFFFFFF";
	enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_YUV420P, AV_PIX_FMT_YUV420P };

	//创建滤镜容器
	filter_graph = avfilter_graph_alloc();
	if (!outputs || !inputs || !filter_graph)
	{
		ret = AVERROR(ENOMEM);
		goto end;
	}

	//初始化数据帧的格式
	sprintf_s(args, sizeof(args),
		"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",
		codecContext->width, codecContext->height, codecContext->pix_fmt,
		codecContext->time_base.num, codecContext->time_base.den,
		codecContext->sample_aspect_ratio.num, codecContext->sample_aspect_ratio.den);

	//输入数据缓存
	ret = avfilter_graph_create_filter(&bufferclass="lazy" data-src_ctx, bufferclass="lazy" data-src, "in",
		args, NULL, filter_graph);

	if (ret < 0) {
		goto end;
	}

	//输出数据缓存
	ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out",
		NULL, NULL, filter_graph);

	if (ret < 0)
	{
		av_log(NULL, AV_LOG_ERROR, "Cannot create buffer sink\n");
		goto end;
	}

	//设置元素样式
	ret = av_opt_set_int_list(buffersink_ctx, "pix_fmts", pix_fmts,
		AV_PIX_FMT_YUV420P, 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->name = av_strdup("in");
	outputs->filter_ctx = bufferclass="lazy" data-src_ctx;
	outputs->pad_idx = 0;
	outputs->next = NULL;

	inputs->name = av_strdup("out");
	inputs->filter_ctx = buffersink_ctx;
	inputs->pad_idx = 0;
	inputs->next = NULL;

	//初始化滤镜
	if ((ret = avfilter_graph_parse_ptr(filter_graph, filters_descr.c_str(),
		&inputs, &outputs, NULL)) < 0)
		goto end;

	//滤镜生效
	if ((ret = avfilter_graph_config(filter_graph, NULL)) < 0)
		goto end;
end:
	//释放对应的输入输出
	avfilter_inout_free(&inputs);
	avfilter_inout_free(&outputs);
	return ret;
}

项目工程源码

给视频文件添加文字水印的工程源码如下，欢迎参考，如有问题欢迎反馈。

#pragma execution_character_set("utf-8")
#include <string>
#include <iostream>
#include <thread>
#include <memory>
#include <iostream>
#include <fstream>

extern "C"
{
#include <libavcodec/avcodec.h>
#include <libavdevice/avdevice.h>
#include <libavfilter/avfilter.h>
#include <libavformat/avformat.h>
#include <libavformat/avio.h>
#include <libavutil/avutil.h>
#include <libswresample/swresample.h>
#include <libswscale/swscale.h>
#include <libavutil/frame.h>
#include <libavutil/imgutils.h>
#include <libavformat/avformat.h>
#include <libavutil/time.h>
#include <libavfilter/avfilter.h>
#include <libavfilter/buffersink.h>
#include <libavfilter/bufferclass="lazy" data-src.h>
#include <libswscale/swscale.h>
#include <libswresample/swresample.h>
#include <libavdevice/avdevice.h>
}

using namespace std;

//输入媒体文件的上下文
AVFormatContext * input_format_ctx = nullptr;

//输出媒体文件的上下文
AVFormatContext* output_format_ctx;

//输出视频编码器
AVCodecContext*	ouput_video_encode_ctx = NULL;

//音视频解码器
AVCodecContext *video_decode_ctx = NULL;
AVCodecContext *audio_decode_ctx = NULL;

//视频索引和音频索引
int video_stream_index = -1;
int audio_stream_index = -1;


//输出编码器
static AVCodec *  output_video_codec;

//滤镜容器和缓存
AVFilterGraph * filter_graph = nullptr;
AVFilterContext *buffersink_ctx = nullptr;;
AVFilterContext *bufferclass="lazy" data-src_ctx = nullptr;;
AVPacket packet;

//起始时间
static int64_t startTime;

int OpenOutput(char *fileName)
{
	//创建输出流,输出flv格式视频
	int ret = 0;
	ret = avformat_alloc_output_context2(&output_format_ctx, NULL, "flv", fileName);
	if (ret < 0)
	{
		return -1;
	}
	
	//打开输出流
	ret = avio_open(&output_format_ctx->pb, fileName, AVIO_FLAG_READ_WRITE);
	if (ret < 0)
	{
		return -2;
	}

	//创建输出流
	for (int index = 0; index < input_format_ctx->nb_streams; index++)
	{
		if (index == video_stream_index)
		{
			AVStream * stream = avformat_new_stream(output_format_ctx, output_video_codec);
			avcodec_parameters_from_context(stream->codecpar, ouput_video_encode_ctx);
			stream->codecpar->codec_tag = 0;
		}
		else if (index == audio_stream_index)
		{
			AVStream * stream = avformat_new_stream(output_format_ctx, NULL);
			stream->codecpar = input_format_ctx->streams[audio_stream_index]->codecpar;
			stream->codecpar->codec_tag = 0;
		}
	}
	//写文件头
	ret = avformat_write_header(output_format_ctx, nullptr);
	if (ret < 0)
	{
		return -3;
	}
	if (ret >= 0)
		cout << "open output stream successfully" << endl;
	return ret;
}

//初始化输出视频的编码器
int InitEncoderCodec(int iWidth, int iHeight)
{
	output_video_codec = avcodec_find_encoder(AV_CODEC_ID_H264);
	if (NULL == output_video_codec)
	{
		return  -1;
	}
	//指定编码器的参数
	ouput_video_encode_ctx = avcodec_alloc_context3(output_video_codec);
	ouput_video_encode_ctx->time_base = input_format_ctx->streams[video_stream_index]->time_base;
	ouput_video_encode_ctx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;
	ouput_video_encode_ctx->sample_fmt = AV_SAMPLE_FMT_S16;
	ouput_video_encode_ctx->width = iWidth;
	ouput_video_encode_ctx->height = iHeight;
	ouput_video_encode_ctx->bit_rate = input_format_ctx->streams[video_stream_index]->codecpar->bit_rate;
	ouput_video_encode_ctx->pix_fmt = (AVPixelFormat)*output_video_codec->pix_fmts;
	ouput_video_encode_ctx->profile = FF_PROFILE_H264_MAIN;
	ouput_video_encode_ctx->level = 41;
	ouput_video_encode_ctx->thread_count = 8;
	return 0;
}


int InitFilter(AVCodecContext * codecContext)
{
	char args[512];
	int ret = 0;
	//缓存输入和缓存输出
	const AVFilter *bufferclass="lazy" data-src = avfilter_get_by_name("buffer");
	const AVFilter *buffersink = avfilter_get_by_name("buffersink");

	//创建输入输出参数
	AVFilterInOut *outputs = avfilter_inout_alloc();
	AVFilterInOut *inputs = avfilter_inout_alloc();

	//滤镜的描述
	//使用simhei字体,绘制的字体大小为100,文本内容为"鬼灭之刃",绘制位置为(100,100)
	//绘制的字体颜色为白色
	string  filters_descr = "drawtext=fontfile=.//simsun.ttc:fontsize=100:text=鬼灭之刃:x=100:y=100:fontcolor=0xFFFFFF";
	enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_YUV420P, AV_PIX_FMT_YUV420P };

	//创建滤镜容器
	filter_graph = avfilter_graph_alloc();
	if (!outputs || !inputs || !filter_graph)
	{
		ret = AVERROR(ENOMEM);
		goto end;
	}

	//初始化数据帧的格式
	sprintf_s(args, sizeof(args),
		"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",
		codecContext->width, codecContext->height, codecContext->pix_fmt,
		codecContext->time_base.num, codecContext->time_base.den,
		codecContext->sample_aspect_ratio.num, codecContext->sample_aspect_ratio.den);

	//输入数据缓存
	ret = avfilter_graph_create_filter(&bufferclass="lazy" data-src_ctx, bufferclass="lazy" data-src, "in",
		args, NULL, filter_graph);

	if (ret < 0) {
		goto end;
	}

	//输出数据缓存
	ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out",
		NULL, NULL, filter_graph);

	if (ret < 0)
	{
		av_log(NULL, AV_LOG_ERROR, "Cannot create buffer sink\n");
		goto end;
	}

	//设置元素样式
	ret = av_opt_set_int_list(buffersink_ctx, "pix_fmts", pix_fmts,
		AV_PIX_FMT_YUV420P, 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->name = av_strdup("in");
	outputs->filter_ctx = bufferclass="lazy" data-src_ctx;
	outputs->pad_idx = 0;
	outputs->next = NULL;

	inputs->name = av_strdup("out");
	inputs->filter_ctx = buffersink_ctx;
	inputs->pad_idx = 0;
	inputs->next = NULL;

	//初始化滤镜
	if ((ret = avfilter_graph_parse_ptr(filter_graph, filters_descr.c_str(),
		&inputs, &outputs, NULL)) < 0)
		goto end;

	//滤镜生效
	if ((ret = avfilter_graph_config(filter_graph, NULL)) < 0)
		goto end;
end:
	//释放对应的输入输出
	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 code;
	AVPacket packet = { 0 };
	av_init_packet(&packet);

	int ret = avcodec_send_frame(ouput_video_encode_ctx, frame);
	if (ret < 0) 
	{
		printf("Error sending a frame for encoding\n");
		return -1;
	}
	while (ret >= 0)
	{
		ret = avcodec_receive_packet(ouput_video_encode_ctx, &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);
		}

		AVRational avTimeBaseQ = { 1, AV_TIME_BASE };
		int64_t ptsTime = av_rescale_q(frame->pts, input_format_ctx->streams[video_stream_index]->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");
		}

		packet.pts = av_rescale_q_rnd(packet.pts, time_base, output_format_ctx->streams[video_stream_index]->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));
		packet.dts = av_rescale_q_rnd(packet.dts, time_base, output_format_ctx->streams[video_stream_index]->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));
		packet.stream_index = video_stream_index;
		code = av_interleaved_write_frame(output_format_ctx, &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);
	}
}


int main(int argc, char* argv[])
{
	if (argc != 3)
	{
		printf("usage:%1 input filepath %2 outputfilepath");
		return -1;
	}

	//输入文件地址、输出文件地址
	string fileInput = std::string(argv[1]);
	string fileOutput = std::string(argv[2]);

	//初始化各种配置
	avformat_network_init();
	av_log_set_level(AV_LOG_ERROR);

	//打开输入文件
	int ret = avformat_open_input(&input_format_ctx, fileInput.c_str(), NULL, NULL);
	if (ret < 0)
	{
		return  ret;
	}
	ret = avformat_find_stream_info(input_format_ctx, NULL);
	if (ret < 0)
	{
		return ret;
	}

	//查找音视频流的索引
	for (int index = 0; index < input_format_ctx->nb_streams; ++index)
	{
		if (index == AVMEDIA_TYPE_AUDIO)
		{
			audio_stream_index = index;
		}
		else if (index == AVMEDIA_TYPE_VIDEO)
		{
			video_stream_index = index;
		}
	}

	//打开视频解码器
	const AVCodec* codec = avcodec_find_decoder(input_format_ctx->streams[video_stream_index]->codecpar->codec_id);
	if (!codec)
	{
		return -1;
	}
	video_decode_ctx = avcodec_alloc_context3(codec);
	if (!video_decode_ctx)
	{
		fprintf(stderr, "Could not allocate video codec context\n");
		return -2;
	}
	avcodec_parameters_to_context(video_decode_ctx, input_format_ctx->streams[video_stream_index]->codecpar);
	if (codec->capabilities & AV_CODEC_CAP_TRUNCATED)
		video_decode_ctx->flags |= AV_CODEC_FLAG_TRUNCATED;

	ret = avcodec_open2(video_decode_ctx, codec, NULL);
	if (ret < 0)
	{
		av_free(video_decode_ctx);
		return -3;
	}

	//初始化视频编码器
	ret = InitEncoderCodec(video_decode_ctx->width, video_decode_ctx->height);
	if (ret < 0)
	{
		return 0;
	}

	//初始化滤镜
	ret = InitFilter(ouput_video_encode_ctx);

	//打开编码器
	ret = avcodec_open2(ouput_video_encode_ctx, output_video_codec, NULL);
	if (ret < 0)
	{
		return  ret;
	}

	//初始化输出
	if (OpenOutput((char *)fileOutput.c_str()) < 0)
	{
		cout << "Open file Output failed!" << endl;
		this_thread::sleep_for(chrono::seconds(10));
		return 0;
	}

	AVFrame* pclass="lazy" data-srcFrame = av_frame_alloc();
	AVFrame*  filterFrame = av_frame_alloc();

	av_init_packet(&packet);
	startTime = av_gettime();

	while (true)
	{
		int ret = av_read_frame(input_format_ctx, &packet);
		if (ret < 0)
		{
			break;
		}
		//视频帧通过滤镜处理之后编码输出
		if (packet.stream_index == video_stream_index)
		{
			int ret = avcodec_send_packet(video_decode_ctx, &packet);
			if (ret < 0)
			{
				break;
			}
			while (ret >= 0) 
			{
				ret = avcodec_receive_frame(video_decode_ctx, pclass="lazy" data-srcFrame);
				if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
				{
					break;
				}
				else if (ret < 0)
				{
					goto End;
				}
				pclass="lazy" data-srcFrame->pts = pclass="lazy" data-srcFrame->best_effort_timestamp;

				//添加到滤镜中
				if (av_bufferclass="lazy" data-src_add_frame_flags(bufferclass="lazy" data-src_ctx, pclass="lazy" data-srcFrame, AV_BUFFERclass="lazy" data-src_FLAG_KEEP_REF) < 0)
				{
					break;
				}
				while (1)
				{
					//获取滤镜输出
					int ret = av_buffersink_get_frame(buffersink_ctx, filterFrame);
					if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
					{
						break;
					}
					else if (ret < 0)
					{
						goto End;
					}
					//编码之后输出
					output_frame(filterFrame, buffersink_ctx->inputs[0]->time_base);
					av_frame_unref(filterFrame);
				}
				av_frame_unref(pclass="lazy" data-srcFrame);
			}
				
		}
		else if (packet.stream_index == audio_stream_index)
		{
			packet.pts = av_rescale_q_rnd(packet.pts, input_format_ctx->streams[audio_stream_index]->time_base, output_format_ctx->streams[audio_stream_index]->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));
			packet.dts = av_rescale_q_rnd(packet.dts, input_format_ctx->streams[audio_stream_index]->time_base, output_format_ctx->streams[audio_stream_index]->time_base, (AVRounding)(AV_ROUND_INF | AV_ROUND_PASS_MINMAX));
			packet.stream_index = audio_stream_index;
			av_interleaved_write_frame(output_format_ctx, &packet);
		}
		av_packet_unref(&packet);
	}
	av_write_trailer(output_format_ctx);

End:
    //结束的时候清理资源
	avfilter_graph_free(&filter_graph);
	if (input_format_ctx != NULL)
	{
		avformat_close_input(&input_format_ctx);
	}
	avcodec_free_context(&video_decode_ctx);
	avcodec_free_context(&ouput_video_encode_ctx);
	return 0;
}