Appearance-Based False Positive Removal in Video Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video analytics systems face challenges in accurately distinguishing between true and false positive objects during object tracking, particularly due to issues like moving foliage, umbrellas, and shadows, which can lead to high false positive rates and system inefficiencies.
Innovation Solution
The implementation of an appearance-based false positive detection mechanism that utilizes color characteristics and pixel-level analysis, including the determination of an appearance model and mass center, to differentiate between true and false positive objects by comparing color characteristics across frames and applying foreground pixel fullness analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional blob detection and tracking is used, then object detection coverage is improved, but false positive rate increases
Solution Approach 1:
The patent applies color characteristic analysis to distinguish true objects from false positives. By extracting and comparing color histograms of tracked objects across multiple frames, the system identifies objects with inconsistent color properties (like moving foliage or shadows) and removes them from tracking, thereby reducing false positives while maintaining detection coverage
Solution Approach 2:
The patent changes the parameter set used for object validation by introducing color characteristics alongside spatial and temporal parameters. This multi-parameter approach allows the system to differentiate between genuine objects and false positives based on combined evidence from color consistency, motion patterns, and tracking duration
2Measurement precision
If pixel level analysis is added to distinguish true and false positives, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent applies pixel-level analysis selectively rather than universally. It performs detailed color and texture analysis only on objects that pass initial tracking criteria and exhibit ambiguous characteristics, while accepting or rejecting objects based on simpler criteria when possible. This partial application of complex analysis reduces overall computational burden while maintaining high differentiation accuracy
Data Source
AI summary
Techniques and systems are provided for maintaining blob trackers for one or more video frames. For example, a blob tracker is identified for a current video frame. The blob tracker is associated with a blob detected for the current video frame. The blob includes pixels of at least a portion of one or more objects in the current video frame. A current characteristic of pixels in a region of the current video frame associated with the blob tracker is determined. A previous characteristic of pixels in a region of a previous video frame associated with the blob tracker is also determined. A difference is determined between the current characteristic and the previous characteristic, and a status of the blob tracker is determined based on the determined difference. The status of the blob tracker indicating whether to maintain the blob tracker for the one or more video frames.