Estimating Tool and Stream Processing Time

Individual tool processing time is displayed in the Database Overview, as shown below:

Processing Time is the average processing time per image in the last processing task, and it is the sum of the processing time and the post-processing time.

The processing time for a stream containing multiple tools is not available through the VisionPro Deep Learning GUI, and you cannot estimate it by summing the tool execution time of the tools in the stream because the processing time includes the time required to prepare and transmit view information between tools.

When considering stream processing, remember that the processing of tools in a stream is always serialized when you call Stream.Process(). Only one tool is processed at a time unless you explicitly process tools individually using Tool.Process().

Throughput

Within VisionPro Deep Learning, throughput refers to the total number of images that can be processed per unit of time. If your application can process multiple streams concurrently using different threads, it may be possible to improve system throughput, although individual tool processing is slower.