Dynamic Patch Region Positioning for Human Flow Counting Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing human flow analysis techniques inaccurately count the number of persons passing over a detection line due to the fixed method of image division, which does not account for the relationship between the detection line and divided regions, leading to incomplete detection of object passages.
Innovation Solution
An image processing apparatus and method that sets patch regions on a display screen based on a detection line, with a detection region for object detection and a counting region within it, to accurately count intersection points between movement vectors and the detection line, ensuring accurate counting by dynamically determining the position of patch regions to encompass the detection line.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the image is divided into fixed regions regardless of detection line position, then the processing can be simplified and parallelized, but the counting accuracy deteriorates because passages crossing region boundaries are missed
Solution Approach 1:
The patent applies dynamics by making the image division dynamic rather than fixed. The divided regions are adjusted based on the detection line position, allowing the system to adapt to different scenarios while maintaining parallel processing capabilities. This resolves the contradiction by enabling both simplified processing and accurate counting.
Solution Approach 2:
The patent applies local quality by creating different divided regions tailored to specific areas relative to the detection line. Instead of uniform division, the system creates detection regions and counting regions with specific characteristics suited for their functional requirements, improving counting accuracy while maintaining processing efficiency.
2Reliability
If the detection region is expanded to cover the entire screen, then all passages can be detected, but the processing time increases significantly
Solution Approach 1:
The patent applies segmentation by dividing the screen into multiple smaller detection regions and counting regions. This allows the system to process only relevant areas rather than the entire screen, reducing processing time while maintaining detection completeness through strategic positioning of regions around the detection line.
Solution Approach 2:
The patent extracts and focuses computational resources on specific regions that are most relevant to counting passages across the detection line. By taking out only the necessary detection and counting regions rather than processing the entire screen, the system achieves both reliability and efficiency.
3Measurement precision
If patch regions are dynamically positioned based on detection line, then counting accuracy improves, but the device complexity increases
Solution Approach 1:
The patent uses dynamics to adjust patch region positions based on detection line location, improving counting accuracy. The dynamic positioning ensures that counting regions are always appropriately placed relative to the detection line, resolving the contradiction between accuracy and complexity.
Solution Approach 2:
The patent creates a universal framework for region division that can adapt to any detection line position. The same basic structure of detection regions and counting regions serves multiple purposes and works across different scenarios, reducing the effective complexity despite the dynamic positioning capability.
Data Source
AI summary
An image processing apparatus comprises a setting unit configured to set, based on a detection line that is set in a display screen, a patch region in the display screen, the patch region including a detection region serving as a region for detecting a position of an object, and a counting region encompassed by the detection region, and a counting unit configured to count, from among intersection points between a movement vector in the detection region of an object moving inside the display screen or a locus based on the movement vector and the detection line in the detection region, intersection points located in the counting region.