Adaptive Threshold Object Detection for HOV Lanes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated methods for detecting objects in HOV/HOT lanes suffer from low accuracy due to varying conditions and image quality, relying on static threshold values that are inefficient and expensive for law enforcement deployment.
Innovation Solution
A method using a dynamic distance ratio threshold, calculated from training images, to identify outlier scores in images, allowing for accurate detection of objects by comparing scores to a predefined distance ratio, rather than fixed scalar thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If static threshold values are used for object detection, then the detection method is simple to implement, but the detection accuracy is low due to varying conditions and image quality
Solution Approach 1:
The patent transforms the static threshold into a dynamic threshold by using the box plot method to calculate adaptive thresholds based on the distribution of detection scores in each image. The threshold is no longer a fixed value but adapts to the specific characteristics of each image, thereby improving detection accuracy under varying conditions while maintaining computational efficiency through statistical methods.
Solution Approach 2:
The patent changes the parameter of the threshold from a fixed scalar value to a dynamically calculated value based on the statistical distribution of scores. By using the box plot method to determine thresholds that adapt to each image's characteristics, the system improves detection accuracy without requiring complex machine learning models, thus balancing accuracy and complexity.
2Measurement precision
If law enforcement officers are deployed to manually count people in HOV/HOT lanes, then detection accuracy is high, but the cost and inefficiency increase significantly
Solution Approach 1:
The patent implements an automated detection system that performs people counting and verification independently without requiring manual intervention from law enforcement officers. The system uses adaptive thresholding to automatically determine whether vehicles meet HOV/HOT lane requirements, thereby maintaining high detection accuracy while dramatically improving enforcement efficiency and reducing operational costs.
3Productivity
If current facial detection methods are used to automate detection, then the need for manual enforcement is reduced, but the detection consistency and accuracy deteriorate due to varying conditions
Solution Approach 1:
The patent introduces dynamic threshold adjustment based on the box plot method, allowing the detection system to adapt to varying image conditions and quality. This dynamic approach maintains automated detection capability while improving consistency by adjusting thresholds according to the specific characteristics of each image, thereby resolving the reliability issue.
Solution Approach 2:
The patent incorporates feedback mechanisms by using the distribution of detection scores to inform threshold selection. The box plot method analyzes the score distribution and uses this feedback to determine appropriate thresholds, creating a closed-loop system that improves detection consistency under varying conditions while maintaining automated operation.
Data Source
AI summary
A method, non-transitory computer readable medium, and apparatus for detecting an object in an image are disclosed. For example, the method receives the image, calculates a score for each one of a plurality of locations in the image, performs a box plot of the score of the each one of the plurality of locations of the image, identifies an outlier score that falls outside of the box plot, determines that a distance ratio of the outlier score is less than a predefined distance ratio and detects the object in a location of the plurality of locations of the image corresponding to the outlier score.