虽然还是没有搞出来,但总感觉快了哈哈(哪来的自信)
1、RTP协议接受数据
#region 1-RTP协议变量声明 RTPSession session; RTPReceiver receiver; RTPParticipant participant; private Dictionary<uint, List<RTPPacket>> Clients; #endregion #region 对RTP进行初始化,并接收数据,调用之后就可以接收数据了 session = new RTPSession(); receiver = new RTPReceiver(); IPEndPoint rtpEp = new IPEndPoint(IPAddress.Parse("192.168.1.109"), 5000); participant = new RTPParticipant(rtpEp); receiver.AddParticipant(participant); session.NewRTPPacket = new RTPSession.NewRTPPacket_Callback(NewRTPPacket); session.AddReceiver(receiver); Clients = new Dictionary<uint, List<RTPPacket>>(); #endregion
其中NewRTPPackt是
public delegate bool NewRTPPacket_Callback( RTPPacket packet )
类型的委托。packet为接收到的RTP包,我们就对这些包进行处理得到想要的帧,然后再把帧进行解码,得到想要的图像(我是这样理解的)
2、H.264进行解码
我从网络上搜索到了一个海思的DLL,可以对H.264进行解码
#region 解码器相关变量声明 /// <summary> /// 数据的句柄 /// </summary> IntPtr pData; /// <summary> /// 这是解码器属性信息 /// </summary> public H264Dec.hiH264_DEC_ATTR_S decAttr; /// <summary> /// 这是解码器输出图像信息 /// </summary> public H264Dec.hiH264_DEC_FRAME_S _decodeFrame = new H264Dec.hiH264_DEC_FRAME_S(); /// <summary> /// 解码器句柄 /// </summary> public IntPtr _decHandle; #endregion #region 解码器相关初始化,一般在窗口load中进行初始化 decAttr = new H264Dec.hiH264_DEC_ATTR_S(); decAttr.uPictureFormat = 0; decAttr.uStreamInType = 0; decAttr.uPicWidthInMB = 480 >> 4; decAttr.uPicHeightInMB = 640 >> 4; decAttr.uBufNum = 8; decAttr.uWorkMode = 16; //创建、初始化解码器句柄 _decHandle = H264Dec.Hi264DecCreate(ref decAttr); //_decodeFrame = new H264Dec.hiH264_DEC_FRAME_S(); #endregion //这一写代码就是h264解码的代码,其中未声明的函数和变量会在下面进行声明给出,主要是讲YUV转为RGB,在保存为Bitmap文件 if (H264Dec.Hi264DecAU(_decHandle, pData, (uint)newData.Length, 0, ref _decodeFrame, 0) == 0) { if (_decodeFrame.bError == 0) { //策画 y u v 的长度 var yLength = _decodeFrame.uHeight * _decodeFrame.uYStride; var uLength = _decodeFrame.uHeight * _decodeFrame.uUVStride / 2; var vLength = uLength; var yBytes = new byte[yLength]; var uBytes = new byte[uLength]; var vBytes = new byte[vLength]; var decodedBytes = new byte[yLength + uLength + vLength]; //_decodeFrame 是解码后的数据对象,里面包含 YUV 数据、宽度、高度等信息 Marshal.Copy(_decodeFrame.pY, yBytes, 0, (int)yLength); Marshal.Copy(_decodeFrame.pU, uBytes, 0, (int)uLength); Marshal.Copy(_decodeFrame.pV, vBytes, 0, (int)vLength); //将从 _decodeFrame 中取出的 YUV 数据放入 decodedBytes 中 Array.Copy(yBytes, decodedBytes, yLength); Array.Copy(uBytes, 0, decodedBytes, yLength, uLength); Array.Copy(vBytes, 0, decodedBytes, yLength + uLength, vLength); ConvertYUV2RGB(yuv, rgb, width, height); ConvertYUV2RGB(decodedBytes, rgb, width, height); // 写 BMP 文件。 WriteBMP(rgb, width, height, string.Format("E:\\test\\yuv2bmp_{0}.bmp", index++)); } }
其中pData为需要的一帧数据,因为pData为Intptr类型,而一帧数据是byte[]类型,所以我从网上查了查怎么转换,下面是代码,newData是byte【】,pData是intptr类型。
GCHandle hObject = GCHandle.Alloc(newData, GCHandleType.Pinned);
pData = hObject.AddrOfPinnedObject();
H264解码类
public class H264Dec { public const int HI_SUCCESS = 0; public const int HI_FAILURE = -1; public const int HI_LITTLE_ENDIAN = 1234; public const int HI_BIG_ENDIAN = 4321; public const int HI_DECODER_SLEEP_TIME = 60000; public const int HI_H264DEC_OK = 0; public const int HI_H264DEC_NEED_MORE_BITS = -1; public const int HI_H264DEC_NO_PICTURE = -2; public const int HI_H264DEC_ERR_HANDLE = -3; [DllImport("hi_h264dec_w.dll", EntryPoint = "Hi264DecImageEnhance", CallingConvention = CallingConvention.Cdecl)] public static extern int Hi264DecImageEnhance(IntPtr hDec, ref hiH264_DEC_FRAME_S pDecFrame, uint uEnhanceCoeff); [DllImport("hi_h264dec_w.dll", EntryPoint = "Hi264DecCreate", CallingConvention = CallingConvention.Cdecl)] public static extern IntPtr Hi264DecCreate(ref hiH264_DEC_ATTR_S pDecAttr); [DllImport("hi_h264dec_w.dll", EntryPoint = "Hi264DecDestroy", CallingConvention = CallingConvention.Cdecl)] public static extern void Hi264DecDestroy(IntPtr hDec); [DllImport("hi_h264dec_w.dll", EntryPoint = "Hi264DecGetInfo", CallingConvention = CallingConvention.Cdecl)] public static extern int Hi264DecGetInfo(ref hiH264_LIBINFO_S pLibInfo); /// <summary> /// 对输入的一段码流进行解码并按帧输出图像 /// </summary> /// <param name="hDec">解码器句柄</param> /// <param name="pStream">码流起始地址</param> /// <param name="iStreamLen">码流长度</param> /// <param name="ullPTS">时间戳信息</param> /// <param name="pDecFrame">图像信息</param> /// <param name="uFlags">解码模式 0:正常解码;1、解码完毕并要求解码器输出残留图像</param> /// <returns></returns> [DllImport("hi_h264dec_w.dll", EntryPoint = "Hi264DecFrame", CallingConvention = CallingConvention.Cdecl)] public static extern int Hi264DecFrame(IntPtr hDec, IntPtr pStream, uint iStreamLen, ulong ullPTS, ref hiH264_DEC_FRAME_S pDecFrame, uint uFlags); [DllImport("hi_h264dec_w.dll", EntryPoint = "Hi264DecAU", CallingConvention = CallingConvention.Cdecl)] public static extern int Hi264DecAU(IntPtr hDec, IntPtr pStream, uint iStreamLen, ulong ullPTS, ref hiH264_DEC_FRAME_S pDecFrame, uint uFlags); /// <summary> /// 解码器属性信息。 /// </summary> [StructLayout(LayoutKind.Sequential)] public struct hiH264_DEC_ATTR_S { /// <summary> /// 解码器输出图像格式,目前解码库只支持YUV420图像格式 /// </summary> public uint uPictureFormat; /// <summary> /// 输入码流格式 0x00: 目前解码库只支持以“00 00 01”为nalu分割符的流式H.264码流 /// </summary> public uint uStreamInType; /// <summary> /// 图像宽度 /// </summary> public uint uPicWidthInMB; /// <summary> /// 图像高度 /// </summary> public uint uPicHeightInMB; /// <summary> /// 参考帧数目 /// </summary> public uint uBufNum; /// <summary> /// 解码器工作模式 /// </summary> public uint uWorkMode; /// <summary> /// 用户私有数据 /// </summary> public IntPtr pUserData; /// <summary> /// 保留字 /// </summary> public uint uReserved; } /// <summary> /// 解码器输出图像信息数据结构 /// </summary> [StructLayout(LayoutKind.Sequential)] public struct hiH264_DEC_FRAME_S { /// <summary> /// Y分量地址 /// </summary> public IntPtr pY; /// <summary> /// U分量地址 /// </summary> public IntPtr pU; /// <summary> /// V分量地址 /// </summary> public IntPtr pV; /// <summary> /// 图像宽度(以像素为单位) /// </summary> public uint uWidth; /// <summary> /// 图像高度(以像素为单位) /// </summary> public uint uHeight; /// <summary> /// 输出Y分量的stride (以像素为单位) /// </summary> public uint uYStride; /// <summary> /// 输出UV分量的stride (以像素为单位) /// </summary> public uint uUVStride; /// <summary> /// 图像裁减信息:左边界裁减像素数 /// </summary> public uint uCroppingLeftOffset; /// <summary> /// 图像裁减信息:右边界裁减像素数 /// </summary> public uint uCroppingRightOffset; /// <summary> /// 图像裁减信息:上边界裁减像素数 /// </summary> public uint uCroppingTopOffset; /// <summary> /// 图像裁减信息:下边界裁减像素数 /// </summary> public uint uCroppingBottomOffset; /// <summary> /// 输出图像在dpb中的序号 /// </summary> public uint uDpbIdx; /// <summary> /// 图像类型:0:帧; 1:顶场; 2:底场 */ /// </summary> public uint uPicFlag; /// <summary> /// 图像类型:0:帧; 1:顶场; 2:底场 */ /// </summary> public uint bError; /// <summary> /// 图像是否为IDR帧:0:非IDR帧;1:IDR帧 /// </summary> public uint bIntra; /// <summary> /// 时间戳 /// </summary> public ulong ullPTS; /// <summary> /// 图像信号 /// </summary> public uint uPictureID; /// <summary> /// 保留字 /// </summary> public uint uReserved; /// <summary> /// 指向用户私有数据 /// </summary> public IntPtr pUserData; } /// <summary> /// 解码库版本、版权和能力集信息。 /// </summary> [StructLayout(LayoutKind.Sequential)] public struct hiH264_LIBINFO_S { /// <summary> /// 主编号 /// </summary> public uint uMajor; /// <summary> /// 次编号 /// </summary> public uint uMinor; /// <summary> /// 发布编号 /// </summary> public uint uRelease; /// <summary> /// 建构编号 /// </summary> public uint uBuild; /// <summary> /// 版本信息 /// </summary> [MarshalAs(UnmanagedType.LPStr)] public string sVersion; /// <summary> /// 版权信息 /// </summary> [MarshalAs(UnmanagedType.LPStr)] public string sCopyRight; /// <summary> /// 解码库能力集 /// </summary> public uint uFunctionSet; /// <summary> /// 支持的输出图像格式 /// </summary> public uint uPictureFormat; /// <summary> /// 输入码流格式 /// </summary> public uint uStreamInType; /// <summary> /// 最大图像宽度(以像素为单位) /// </summary> public uint uPicWidth; /// <summary> /// 最大图像高度(以像素为单位) /// </summary> public uint uPicHeight; /// <summary> /// 最大参考帧数目 /// </summary> public uint uBufNum; /// <summary> /// 保留字 /// </summary> public uint uReserved; } /// <summary> /// 用户私有数据信息。 /// </summary> [StructLayout(LayoutKind.Sequential)] public struct hiH264_USERDATA_S { /// <summary> /// 用户数据类型 /// </summary> public uint uUserDataType; /// <summary> /// 用户数据长度 /// </summary> public uint uUserDataSize; /// <summary> /// 用户数据缓冲区 /// </summary> public IntPtr pData; /// <summary> /// 指向下一段用户数据 /// </summary> public IntPtr pNext; } }
这是YUV转RGB图像。
/// <summary> /// 将转换后的 RGB 图像数据按照 BMP 格式写入文件。 /// </summary> /// <param name="rgbFrame">RGB 格式图像数据。</param> /// <param name="width">图像宽(单位:像素)。</param> /// <param name="height">图像高(单位:像素)。</param> /// <param name="bmpFile"> BMP 文件名。</param> static void WriteBMP(byte[] rgbFrame, int width, int height, string bmpFile) { // 写 BMP 图像文件。 int yu = width * 3 % 4; int bytePerLine = 0; yu = yu != 0 ? 4 - yu : yu; bytePerLine = width * 3 + yu; using (FileStream fs = File.Open(bmpFile, FileMode.Create)) { using (BinaryWriter bw = new BinaryWriter(fs)) { bw.Write('B'); bw.Write('M'); bw.Write(bytePerLine * height + 54); bw.Write(0); bw.Write(54); bw.Write(40); bw.Write(width); bw.Write(height); bw.Write((ushort)1); bw.Write((ushort)24); bw.Write(0); bw.Write(bytePerLine * height); bw.Write(0); bw.Write(0); bw.Write(0); bw.Write(0); byte[] data = new byte[bytePerLine * height]; int gIndex = width * height; int bIndex = gIndex * 2; for (int y = height - 1, j = 0; y >= 0; y--, j++) { for (int x = 0, i = 0; x < width; x++) { data[y * bytePerLine + i++] = rgbFrame[bIndex + j * width + x]; // B data[y * bytePerLine + i++] = rgbFrame[gIndex + j * width + x]; // G data[y * bytePerLine + i++] = rgbFrame[j * width + x]; // R } } bw.Write(data, 0, data.Length); bw.Flush(); } } } /// <summary> /// 将一桢 YUV 格式的图像转换为一桢 RGB 格式图像。 /// </summary> /// <param name="yuvFrame">YUV 格式图像数据。</param> /// <param name="rgbFrame">RGB 格式图像数据。</param> /// <param name="width">图像宽(单位:像素)。</param> /// <param name="height">图像高(单位:像素)。</param> static void ConvertYUV2RGB(byte[] yuvFrame, byte[] rgbFrame, int width, int height) { int uIndex = width * height; int vIndex = uIndex + ((width * height) >> 2); int gIndex = width * height; int bIndex = gIndex * 2; int temp = 0; for (int y = 0; y < height; y++) { for (int x = 0; x < width; x++) { // R分量 temp = (int)(yuvFrame[y * width + x] + (yuvFrame[vIndex + (y / 2) * (width / 2) + x / 2] - 128) * YUV2RGB_CONVERT_MATRIX[0, 2]); rgbFrame[y * width + x] = (byte)(temp < 0 ? 0 : (temp > 255 ? 255 : temp)); // G分量 temp = (int)(yuvFrame[y * width + x] + (yuvFrame[uIndex + (y / 2) * (width / 2) + x / 2] - 128) * YUV2RGB_CONVERT_MATRIX[1, 1] + (yuvFrame[vIndex + (y / 2) * (width / 2) + x / 2] - 128) * YUV2RGB_CONVERT_MATRIX[1, 2]); rgbFrame[gIndex + y * width + x] = (byte)(temp < 0 ? 0 : (temp > 255 ? 255 : temp)); // B分量 temp = (int)(yuvFrame[y * width + x] + (yuvFrame[uIndex + (y / 2) * (width / 2) + x / 2] - 128) * YUV2RGB_CONVERT_MATRIX[2, 1]); rgbFrame[bIndex + y * width + x] = (byte)(temp < 0 ? 0 : (temp > 255 ? 255 : temp)); } } }
3、这可能就是我遇到问题的地方了,怎么把RTPPack中的包数据转换为一帧图像信息,我找到的资料是;
#region 对收到的数据进行处理 if (!Clients.ContainsKey(packet.SSRC))//如果接受端第一次接受到某源的数据,则加入到 { if (Clients.Count < 4)//如果发送端为4,则丢弃包 { Clients.Add(packet.SSRC, new List<RTPPacket> { packet }); //ImagesBoxMapping[ImagesBoxMapping.First(pair => pair.Value == null).Key] = packet.SSRC; } } else { Clients[packet.SSRC].Add(packet); } if (packet.Marker)//如果已经发送完毕 { //丢包检测 var orderPackets = Clients[packet.SSRC].OrderBy(rtpPacket => rtpPacket.SequenceNumber); if (Clients[packet.SSRC].Count != (orderPackets.Last().SequenceNumber - orderPackets.First().SequenceNumber + 1)) { Clients[packet.SSRC].Clear();//清空缓存区 return true; } //1.包重组 var count = Clients[packet.SSRC].Sum(rtpPacket => rtpPacket.DataSize);//数据总数 var newData = new byte[count]; long offSet = 0; foreach (var rtpPacket in Clients[packet.SSRC]) { Array.Copy(rtpPacket.DataPointer, 0, newData, offSet, rtpPacket.DataSize); offSet += rtpPacket.DataSize; } Clients[packet.SSRC].Clear();//清空缓存区
这里我理解的是newData里面就是一帧数据,但我测试了一下不对(晕)。
4、总结
这几天一直想要尽快做出来,却总没有办法深入去研究视频方面的东西。比如得到的包怎么变为一帧,怎么从一帧里面提取需要的数据,什么PPS、SPS、IDR都是什么,虽然知道名词,但总没法很明确的说出来。
我的解码思路是:RTP协议收到包后(这一步没有问题),将包的数据转为帧(这个地方可能出问题了,也可能是传过来的帧数据不符合解码的要求),再把一帧的数据传给H264解码类解码,解码后输出的是YUV,YUV->RGB->图片进行显示就可以了。这是我的思路,但没有成功。如果读者您懂这一方面,还希望给我指导。谢谢
每天写一点点,就能进步一点点.
晚上更新:
H264起始码有时是0x00000001,有时是0x000001,这两种的区别是:一共有两种起始码:3字节的0x000001和4字节的0x00000001,3字节的0x000001只有一种场合下使用,就是一个完整的帧被编为多个slice的时候,包含这些slice的nalu使用3字节起始码。其余场合都是4字节的。而海思的解码库中说的很清楚,只能解0x000001起始码的nalu,而我测试的都是0x00000001四个字节的,所以这方面可能出了点问题。哎,基础只是不好就是容易出现错误。使用VLC.NET开源可以解决RTP发送的H264码流,明天进行总结.