Object Detection Region Expansion for Accuracy and Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques fail to detect individuals when a part of the person lies outside the user-set region, leading to incomplete detection results and increased processing load when applying detection processing to the entire image.
Innovation Solution
An image processing apparatus that sets a second region larger than the user-designated first region, including it, to perform detection processing, using size information to ensure detection accuracy and reduce processing load by expanding the detection area based on maximum and minimum size settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If detection processing is performed on the user-set region only, then processing load is reduced, but detection accuracy deteriorates when part of the person lies outside the region
Solution Approach 1:
The system performs preliminary actions by setting a detection area that is expanded from the user-set region based on the maximum size of the object. This preliminary expansion ensures that even if part of the person lies outside the original user-set region, the detection processing can still capture the complete object, thereby resolving the contradiction between reduced processing load and maintained detection accuracy.
2Reliability
If detection processing is performed on the entire image, then detection accuracy is improved, but processing load increases
Solution Approach 1:
The system applies local quality by setting the detection area selectively based on the user-set region and object size information. Instead of uniformly processing the entire image, the detection processing is concentrated in the locally optimized detection area that is just large enough to capture the complete object. This resolves the contradiction by improving detection accuracy only where needed while keeping processing load manageable.
3Reliability
If the detection area is expanded to ensure complete detection, then detection accuracy is improved, but processing load increases
Solution Approach 1:
The system changes parameters by dynamically adjusting the detection area size based on the maximum size of the object. The detection area is set to be larger than the user-set region by a specific margin determined by the object size parameters. This parameter-based adjustment ensures complete detection while optimizing processing load, resolving the contradiction between detection accuracy and processing efficiency.
Data Source
AI summary
An image processing apparatus sets a maximum size of an object that is to be included in a detection result of detection processing to detect the object from an image captured by an imaging unit, and sets a second region based on a position of a first region, designated by a user, in the image and the set maximum size. The second region, larger than the first region, includes the first region, and is subjected to the detection processing.