Recherche avancée

Sur d’autres sites (6345)

• ffmpeg intensity histogram adjustment

  30 septembre 2016, par jlarsch

  I am using ffmpeg for background correction of a video and I would like to improve the intensity scaling of the output.

  My gray scale videos have dark moving objects on a light background. In 8 bit pixel intensities, the light background has pixel values around 240, the dark objects have intensities of around 120.

  outside of ffmpeg, I generate a background image by taking the median frame over some number of frames.
  Then, I use ffmpeg to blend/divide each frame by the background image. (I want to use division, not subtraction of the background).
  [there is also some cropping in my ffmpeg command, but it is irrelevant to my question]

  'ffmpeg.exe', '-i', u'inputVideo.avi', '-i', u'bgMed.tif', '-y', '-r', '160', '-filter_complex', "[1:0] setsar=sar=1 [1sared]; [0:0][1sared] blend=all_mode='divide':repeatlast=1,format=gray,split=1 [int1];[int1]crop=1097:1097:12:11:[out1]", '-map', '[out1]', '-c:v', 'libxvid', '-q:v', '5', '-g', '10', u'outputVideo'

  This procedure is basically working but the resulting video frames look too washed out. This is probably expected ? I am guessing ffmpeg does the division and produces an internal float result which it then maps back to an 8 bit output. I would like to stretch the histogram of the result from the division. It would be preferable to stretch before the mapping to 8 bit for a finer dynamic range.

  In my example, I am assuming that the background division produces a result frame that has a value close to 1 for the background, and values close to 0.5 for the dark objects. Then, ffmpeg seems to be mapping the full range 0-1 into 8 bit 0-255. I would like it to map the range 0.5-1 of the division result into the 8 bit range of the output. is this possible somehow ? Or how else can I achieve a similar result ?

• C# get dominant color in an image

  22 janvier 2017, par CK13

  I’m building a program that makes screenshots from a video.
  It extracts frames from the video (with ffmpeg) and then combines them into one file.

  All works fine, except sometimes I get (almost) black images, mostly in the beginning and ending of the video.

  A possible solution I can think of is to detect if the extracted frame is dark. If it is dark, extract another frame from a slightly different time.

  How can I detect if the extracted frame is dark/black ? Or is there another way I can solve this ?

  private void getScreenshots_Click(object sender, EventArgs e)
  
  {
  
      int index = 0;
  
      foreach (string value in this.filesList.Items)
  
      {
  
          string file = selectedFiles[index] + "\\" + value;
  
  
  
          // ------------------------------------------------
  
          //   MediaInfo
  
          // ------------------------------------------------
  
          // https://github.com/Nicholi/MediaInfoDotNet
  
          //
  
          // get file width, height, and frame count
  
          //
  
          // get aspect ratio of the video
  
          // and calculate height of thumbnail 
  
          // using width and aspect ratio
  
          //
  
          MediaInfo MI = new MediaInfo();
  
          MI.Open(file);
  
          var width = MI.Get(StreamKind.Video, 0, "Width");
  
          var height = MI.Get(StreamKind.Video, 0, "Height");
  
          decimal d = Decimal.Parse(MI.Get(StreamKind.Video, 0, "Duration"));
  
          decimal frameCount = Decimal.Parse(MI.Get(StreamKind.Video, 0, "FrameCount"));
  
          MI.Close();
  
          decimal ratio = Decimal.Divide(Decimal.Parse(width), Decimal.Parse(height));
  
          int newHeight = Decimal.ToInt32(Decimal.Divide(newWidth, ratio));
  
          decimal startTime = Decimal.Divide(d, totalImages);
  
          //totalImages - number of thumbnails the final image will have
  
          for (int x = 0; x < totalImages; x++)
  
          {
  
              // increase the time where the thumbnail is taken on each iteration
  
              decimal newTime = Decimal.Multiply(startTime, x);
  
              string time = TimeSpan.FromMilliseconds(double.Parse(newTime.ToString())).ToString(@"hh\:mm\:ss");
  
  
  
              string outputFile = this.tmpPath + "img-" + index + x + ".jpg";
  
  
  
              // create individual thumbnails with ffmpeg
  
              proc = new Process();
  
              proc.StartInfo.FileName = "ffmpeg.exe";
  
              proc.StartInfo.Arguments = "-y -seek_timestamp 1 -ss " + time + " -i \"" + file + "\" -frames:v 1 -qscale:v 3 \"" + outputFile + "\"";
  
              proc.Start();
  
              proc.WaitForExit();
  
          }
  
  
  
          // set width and height of final image
  
          int w = (this.cols * newWidth) + (this.spacing * this.cols + this.spacing);
  
          int h = (this.rows * newHeight) + (this.spacing * this.rows + this.spacing);
  
  
  
          int left, top, i = 0;
  
          // combine individual thumbnails into one image
  
          using (Bitmap bmp = new Bitmap(w, h))
  
          {
  
              using (Graphics g = Graphics.FromImage(bmp))
  
              {
  
                  g.Clear(this.backgroundColor);
  
                  // this.rows - number of rows
  
                  for (int y = 0; y < this.rows; y++)
  
                  {
  
                      // put images on a column
  
                      // this.cols - number of columns
  
                      // when x = number of columns go to next row
  
                      for (int x = 0; x < this.cols; x++)
  
                      {
  
                          Image imgFromFile = Image.FromFile(this.tmpPath + "img-" + index + i + ".jpg");
  
                          MemoryStream imgFromStream = new MemoryStream();
  
                          imgFromFile.Save(imgFromStream, imgFromFile.RawFormat);
  
                          imgFromFile.Dispose();
  
  
  
                          left = (x * newWidth) + ((x + 1) * this.spacing);
  
                          top = (this.spacing * (y + 1)) + (newHeight * y);
  
                          g.DrawImage(Image.FromStream(imgFromStream), left, top, newWidth, newHeight);
  
                          i++;
  
                      }
  
                  }
  
              }
  
  
  
              // save the final image
  
              bmp.Save(selectedFiles[index] + "\\" + value + ".jpg");
  
          }
  
          index++;
  
      }
  
  }

• FFmpeg programming : assign color transfer characteristics

  8 juillet 2017, par Francis Tesla

  When using ffmpeg library to load an EXR file it appears to dark.

  This seems to be the same problem as this one :

  ffmpeg - getting a dark output when converting .exr sequence to .mp4

  The EXR decoder has a parameter to assign color transfer
  characteristics.

  For sRGB, it is

  ffmpeg -apply_trc iec61966_2_1 -start_frame 1100 -i input.$04d.exr
  output.mp4

  But how to achieve the same result programmatically ?

  Here is my code :

  AVFrame *frame = av_frame_alloc();
  
  avcodec_decode_video2(codecCtx, frame, &frameFinished, &packet);
  
  
  
  ... 
  
  
  
  AVPicture       picture;
  
  avpicture_alloc(&picture, dest_format, dest_width, dest_height);
  
  
  
  SwsContext *swsCtx = NULL;
  
  
  
  swsCtx = sws_getContext(source_width, source_height, source_format, dest_width, dest_height, dest_format, SWS_POINT, NULL, NULL, NULL);
  
  
  
  if (swsCtx == NULL) {
  
      qDebug() << "error calling sws_getContext";
  
  }
  
  
  
  sws_scale(swsCtx, frame->data, frame->linesize, 0, frame->height, picture.data, picture.linesize);
  
  
  
  QImage image(dest_width, dest_height, QImage::Format_ARGB32);
  
  
  
  for (int y = 0; y < dest_height; ++y) {
  
      memcpy(image.scanLine(y), picture.data[0] + y * picture.linesize[0], picture.linesize[0]);
  
  }
  
  
  
  av_frame_free(&frame);
  
  sws_freeContext(swsCtx);

  Thank you,
  Francis