Fisheye Human Body Detection Using Area-Specific Color Reference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In top-down images captured by fisheye cameras, the variation in color information of detected persons due to their position leads to difficulties in accurately distinguishing human bodies from other objects, resulting in erroneous detections.
Innovation Solution
An information processing device that stores reference color information for each divided area as the similarity between the detected area and the similarity between the detected area and the similarity between the detected area and the similarity between the acquired reference color information of the detected human body candidate and the detected human body determination unit the similarity between the similarity of the detected human body candidate and the similarity between the detected human body candidate and the reference color information corresponding to the detection area, thereby accurately determining whether the detected human body candidate is a human body.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If color information from captured images is used for human body detection, then detection capability is provided, but erroneous detections occur due to color variation depending on position in the image
Solution Approach 1:
The patent applies local quality by creating position-specific reference color information for different areas of the image. Instead of using a single uniform color reference for the entire image, the system divides the image into multiple areas and stores distinct reference color information for each area, accounting for the fact that color appearance varies by position in fisheye images. This allows the detection to adapt to local color characteristics at each position.
2Measurement precision
If reference color information is stored for each divided area, then detection accuracy is improved, but device complexity increases due to larger storage requirements
Solution Approach 1:
The patent applies segmentation by dividing the image into multiple areas and creating separate reference color information for each area. This segmentation allows the system to manage color variation by position while keeping each individual reference value relatively simple. The storage structure is organized by area division, making the complexity manageable and systematic rather than requiring a completely complex solution.
Data Source
AI summary
An information processing device includes a storage unit that stores color information of a human body as reference color information, a detection unit that detects a human body candidate from a captured image captured by a fisheye camera, and a human body determination unit that acquires, from the storage unit, the reference color information corresponding to a detection area where the human body candidate is detected and then determines whether the human body candidate is a human body on the basis of a similarity between the acquired reference color information of the detection area and color information of the human body candidate.