Example Configuration

The following diagram shows a simple configuration for performing OCR on a video stream and sending the results to IDOL Server. The setup includes an ingest engine, an analysis engine, and an output engine. These are usually the minimum required for any configuration.

  1. Media Server receives a Process action, which starts a new session. The LibAV ingest engine receives video from the source specified in the action. The LibAV engine produces an image track and audio track.
  2. The image track is used as the input for the OCR analysis engine.
  3. The audio track cannot be used by the OCR analysis engine so is discarded.
  4. The OCR engine produces several output tracks. The Result track contains the OCR results and is used as the input for the IDOL output task.
  5. The other tracks produced by the OCR engine are not required and so are discarded.
  6. The IDOL output engine indexes the data produced by Media Server into IDOL.

A Media Server configuration that matches this process is shown below.

[Ingest]
IngestEngine=LibAvIngest

[LibAvIngest]
Type=libav

[Analysis]
AnalysisEngine0=OCRtext

[OCRtext]
Type=ocr
Input=Default_Image

[Output]
OutputEngine0=IDOLOutput

[IDOLOutput]
Input=OCRtext.Result
Type=idol
Mode=time
OutputInterval=60s
XslTemplate=./xsl/toIDX.xsl

_HP_HTML5_bannerTitle.htm