The following is an example configuration that demonstrates how to detect text in ingested images. The configuration includes a transformation task to draw the region(s) and an encoding task to output the resulting image(s).
[Ingest] IngestEngine=Image [Image] Type=Image [Analysis] AnalysisEngine0=DetectText [DetectText] Type=TextDetection DetectionThreshold=50 [Output] OutputEngine0=Response [Response] Type=Response Input=DetectText.Result [Transform] TransformEngine0=Draw [Draw] Type=Draw Input=DetectText.ResultWithSource Color=Red Thickness=3 [Encoding] EncodingEngine0=EncodeImage [EncodeImage] Type=ImageEncoder ImageInput=Draw.Output OutputPath=output/%session.token%/%segment.sequence%-%source.filename%
|