Vehicle Video Computing System for Reducing False Positives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle tracking systems using image-based processing are resource-intensive and prone to false positives due to misidentification of environmental objects as vehicles, especially in aerial vehicle environments.
Innovation Solution
A digital video computing system that identifies salient image features, calculates global and local motion vectors, and analyzes candidate image patches to differentiate between vehicle motion and environmental artifacts, thereby reducing false positives and resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image-based vehicle tracking is performed using conventional processing methods, then vehicle detection capability is provided, but computational resource consumption increases significantly
Solution Approach 1:
The patent segments the image processing task by first identifying salient features (edges, corners, blobs) rather than processing entire images. This feature-based segmentation reduces computational load while maintaining vehicle detection capability. The system divides the tracking task into: feature detection in current frame, feature matching with previous frame, and vehicle determination based on feature displacement patterns.
Solution Approach 2:
The patent extracts only the essential information needed for vehicle tracking by taking out salient image features from the complete image data. Instead of processing all pixels, the system extracts key feature points and uses only their positions and movements for tracking decisions, significantly reducing computational resource requirements while preserving detection reliability.
2Reliability
If conventional image processing is used to detect and track vehicles, then tracking functionality is provided, but false positive detections increase due to misidentification of environmental objects
Solution Approach 1:
The patent implements feedback through frame-to-frame feature matching and motion consistency verification. The system compares salient features in the current frame with features in the previous frame, tracks their displacement patterns, and uses this feedback to distinguish moving vehicles from stationary environmental objects. Objects that don't exhibit consistent motion patterns across multiple frames are filtered out, reducing false positives.
Solution Approach 2:
The patent applies dynamics by focusing on motion analysis - vehicles are identified based on their dynamic movement patterns rather than static appearance. The system tracks feature displacement between frames and determines vehicles based on dynamic characteristics (consistent motion, velocity patterns) rather than relying solely on static image recognition, which reduces misidentification of stationary objects like clouds or terrain features.
Data Source
AI summary
A digital video computing system receives two or more frames depicting an environment from a camera system of a vehicle. For a salient image feature identified in the two or more frames, a global motion vector is calculated that is indicative of movement of the feature at least partially attributable to movement of the vehicle. A local motion vector is calculated that is indicative of movement of the feature independent from the movement of the vehicle. Based on the local motion vector, the salient image feature is determined to have an apparent motion relative to the environment that is independent from the movement of the vehicle. A candidate image patch is identified including the salient image feature. The candidate image patch is analyzed to output a likelihood that the candidate image patch depicts a second vehicle.