Image Processing System for Moving Body Detection with Mask Area Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems for monitoring cameras often misdetect moving bodies due to environmental changes such as sunlight reflections and shadows, leading to false alerts, especially when a mask area is set to prevent misdetection, resulting in missed detections of moving bodies within these areas.
Innovation Solution
An image processing system that includes an object detecting unit, an object-occurrence-position detecting unit, and a valid-object determining unit to accurately identify moving bodies by determining if the object's occurrence position is outside a set mask area, thereby reducing misreporting of alerts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a mask area is set to prevent misdetection of environmental changes, then false alerts are reduced, but moving bodies within the mask area are missed
Solution Approach 1:
The detection area is segmented into a mask area (where environmental changes occur) and a non-mask area. The system segments the detection process into two paths: one for detecting objects that move from non-mask to mask area, and another for detecting objects entirely within the non-mask area. This segmentation allows the system to maintain high reliability by excluding the problematic mask area from standard detection while still capturing objects that traverse into it from the non-mask area.
Solution Approach 2:
The system introduces an intermediary detection mechanism that monitors the boundary between mask and non-mask areas. Objects are detected by their movement across this boundary rather than by direct detection within the mask area itself. This intermediary approach allows the system to capture objects that would otherwise be missed while maintaining the protective effect of the mask area.
2Measurement precision
If standard moving body detection is applied to the entire image area, then all moving objects are detected, but environmental changes cause false detections
Solution Approach 1:
The problematic mask area is extracted and removed from the standard detection process. By taking out the region where environmental changes cause false detections, the system eliminates the source of false alerts. The detection algorithm processes only the non-mask area, ensuring that all detected objects are genuine moving bodies rather than environmental artifacts.
Solution Approach 2:
Instead of detecting objects throughout the entire image and then filtering out false detections, the system inverts the approach by detecting only in the non-mask area and then identifying objects that move into the mask area. This inversion transforms the problem from one of elimination (removing false detections) to one of selection (capturing genuine detections).
Data Source
AI summary
An image processing system includes: an object detecting unit that detects a moving body object from image data of an image of a predetermined area; an object-occurrence-position detecting unit that detects an occurrence position of the object detected by the object detecting unit; and a valid-object determining unit that determines that the object detected by the object detecting unit is a valid object when the object is present in a mask area set as a non-detection target in the image of the predetermined area and the occurrence position of the object in the mask area detected by the object-occurrence-position detecting unit is outside the mask area.