Image Processing Apparatus Subject Tracking Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems face challenges in accurately tracking subjects when multiple types of subjects overlap, leading to degraded tracking performance and accuracy, particularly in cases where a person riding a vehicle is detected as both a human and a vehicle subject.
Innovation Solution
An image processing apparatus and method that detects subjects of different types and selects the appropriate detection result for tracking, using a control circuit to prioritize either the first or second type of subject based on detection reliability and overlap criteria, ensuring continuous tracking even when one subject type is no longer detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a subject is detected using multiple detection types (e.g., both vehicle and human detection), then the detection coverage is improved, but the tracking reliability deteriorates due to overlapping detection results causing confusion in subject identification
Solution Approach 1:
The patent segments the detection results by type (vehicle detection results and human detection results) and processes them through separate detection circuits. This segmentation allows the system to maintain multiple detection capabilities while managing them independently, reducing confusion from overlapping results.
Solution Approach 2:
The patent dynamically switches between different detection results based on detection reliability. The control circuit selects which detection result to use for tracking (vehicle or human) depending on which one reliably detects the subject in the current frame, allowing adaptive response to changing detection conditions.
2Stability of the object's composition
If the system prioritizes one subject type over another in overlapping areas, then the tracking stability is improved, but the detection accuracy deteriorates for the non-prioritized subject type
Solution Approach 1:
The system dynamically adjusts which subject type is prioritized based on real-time detection reliability rather than using a fixed priority scheme. When vehicle detection is reliable, it prioritizes vehicle tracking; when human detection is more reliable, it switches to human tracking. This dynamic approach maintains both tracking stability and detection accuracy.
Solution Approach 2:
The control circuit uses feedback from detection reliability assessment to determine which detection result to use for tracking. By continuously evaluating the reliability of each detection type and adjusting the selection accordingly, the system maintains accurate tracking without permanently sacrificing detection accuracy for either subject type.
3Duration of action of stationary object
If the system switches between different subject types based on detection results, then the tracking continuity is improved, but the device complexity increases due to the need for selection logic
Solution Approach 1:
The control circuit implements dynamic selection between detection results based on reliability assessment. This dynamic switching mechanism enables tracking continuity by adapting to which detection type is currently most reliable, while the complexity is managed through a unified control logic that handles both vehicle and human detection results.
Solution Approach 2:
The control circuit is designed with multi-functionality to handle both vehicle detection results and human detection results through a single unified selection mechanism. This universal approach allows the same control logic to manage tracking for either subject type, reducing overall system complexity compared to having separate tracking systems for each subject type.
Data Source
AI summary
An image processing apparatus detects a subject of a first type and a subject of a second type to an image. When executing tracking processing of a subject based on a detection result of the detection circuit, if a same subject is detected as a subject of the first type and a subject of the second type, the image processing apparatus selects either the detection result regarding the subject of the first type or the detection result regarding the subject of the second type is to be used to perform the tracking processing of the subject.