Back to Aurora Vision Library website

You are here: Start » Function Reference » Computer Vision » Template Matching » MergeMultipleLocationResults

MergeMultipleLocationResults


Header: AVL.h
Namespace: avl
Module: MatchingBasic

Combines results from multiple LocateMultipleObject instances.

Syntax

C++
C#
 
void avl::MergeMultipleLocationResults
(
	const atl::Array<avl::Object2D>& inObjects1,
	const atl::Array<avl::Object2D>& inObjects2,
	const atl::Array<avl::Object2D>& inObjects3,
	const atl::Array<avl::Object2D>& inObjects4,
	float inMinDistance,
	float inMinScore,
	atl::Array<avl::Object2D>& outObjects,
	atl::Array<int>& outIndices
)

Parameters

Name Type Range Default Description
Input value inObjects1 const Array<Object2D>&
Input value inObjects2 const Array<Object2D>&
Input value inObjects3 const Array<Object2D>&
Input value inObjects4 const Array<Object2D>&
Input value inMinDistance float 0.0 - 10.0f
Input value inMinScore float -1.0 - 1.0 0.0f
Output value outObjects Array<Object2D>&
Output value outIndices Array<int>&

Description

Filter selects the best matching object from the given matching objects.

This filter is especially useful for finding big objects by using smaller template models instead of the large model.

This filter will return a valid object location even if object is partially covered.

Examples

Object location found based on two matching objects(red and green).

Remarks

Each template matching object which is used in filter must have set this same reference frame.

See Also