GoWebScan API
GoWebScanVisionMergeTask Class Reference

Description

Represents a task for merging vision tiles along X-axis from the cameras in a top or bottom group. Specifically, the task accepts a list of vision tiles that correspond to the same Y-position, and merges them into a single group-wide tile according to their respective source locations along the X-axis. Gaps and overlaps between adjacent tiles are processed according to configuration options. The task accepts GoWebScanTileListMsg objects and produces GoWebScanVisionTileMsg objects.

Processing steps:

  1. Clear the background of the output system tile.
  2. Add GoWebScanTileListMsg input tiles to the processing queue
  3. Blend the overlap between tiles. Overlaps are processed by searching for the last valid left element and first valid right element within the overlapping region between two tiles. If overlap exists, the overlapping elements are blended by weighted average.
  4. Fill gaps between tiles. Gaps are processed by searching for the last valid left element and first valid right element within the gap search window. If these elements are closer than the gap threshold, any null elements in between are filled in (nearest neighbour).
Inheritance diagram for GoWebScanVisionMergeTask:
Inheritance graph

Public Member Functions

kStatus GoWebScanVisionMergeTask_Construct (GoWebScanVisionMergeTask *task, GoWebScanProcess process, GoWebScanConfigGroup configGroup, kAlloc allocator)
 Constructs a GoWebScanVisionMergeTask object. More...
 
kStatus GoWebScanVisionMergeTask_Send (GoWebScanVisionMergeTask task, GoWebScanPipeMsg msg)
 Adds a message to the processing queue. More...
 
kStatus GoWebScanVisionMergeTask_SetReceiver (GoWebScanVisionMergeTask task, GoWebScanPipeSendFx send, kPointer context)
 Sets the handler for receiving merged system vision tiles. More...
 

The documentation for this class was generated from the following file: