Face Region Detection Using Segmented Template Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face region detection systems experience decreased tracking performance and increased processing load as the size of the reference template image increases, leading to inefficient face region detection.
Innovation Solution
A face region detection device and method that sets a face region detected from a first captured image as a reference image, searches for a similar region in a second captured image, and sets the similar region as the face region with the same size and position as the reference image, using the positional relationship between the two images to reduce processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the size of the reference template image is increased to improve detection accuracy, then the tracking performance deteriorates due to increased processing time
Solution Approach 1:
The patent divides the reference template image into multiple sub-regions and performs template matching independently on each sub-region. This segmentation allows parallel processing of smaller image portions, reducing the overall processing time while maintaining detection accuracy by combining results from all sub-regions.
Solution Approach 2:
The patent performs template matching on only a portion of the reference template image (specific sub-regions) rather than the entire image. This partial action approach reduces processing time by focusing computational resources on critical areas while still achieving sufficient detection accuracy for tracking purposes.
2Measurement precision
If the size of the reference template image is increased to improve detection accuracy, then the device complexity increases due to higher processing load
Solution Approach 1:
The patent segments the reference template image into multiple sub-regions, allowing the processing system to handle smaller, more manageable image portions independently. This reduces the computational complexity and processing load on the device while maintaining detection accuracy through aggregation of sub-region results.
Solution Approach 2:
The patent applies template matching to only specific sub-regions of the reference image rather than processing the entire image. This partial processing approach reduces the processing load and device complexity while achieving sufficient tracking accuracy for practical applications.
Data Source
AI summary
A system controller of a digital camera has a function of detecting a face region from a captured image which is output from an imaging element, and sets a face region detected from a first captured image, as a reference image, searches for a similar region similar to a specific image which is a part of the reference image from a second captured image output after the first captured image, and detects a first region including the similar region in the second captured image and having the same size as the reference image, as the face region. The position of the specific image in the reference image and the position of the similar region in the first region coincide with each other.