Generalized Passerby Detection via Wheel Feature Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vision-based active protection technologies for vulnerable road users, such as bicycle and motorcycle riders, are inefficient due to the time-consuming process of collecting samples and training classifiers, limiting their detection efficiency.
Innovation Solution
A method and apparatus that detect a generalized passerby by identifying a common wheel feature in an input image, selecting an image window, and inputting it into an upper-body classifier, with optional pre-processing using pyramid down-sampling for whole-body detection, reducing the need for extensive sample collection and classifier training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to detect bicycle riders and motorcycle riders by collecting samples and training classifiers separately, then detection accuracy can be maintained, but the time required for sample collection and classifier training increases significantly
Solution Approach 1:
The patent creates a unified generalized passerby classifier that can detect multiple types of vulnerable road users (pedestrians, bicycle riders, motorcycle riders) using a single training process. This universal classifier eliminates the need for separate sample collection and training for each user type, significantly reducing time while maintaining detection accuracy across different categories
Solution Approach 2:
The patent merges the detection of different vulnerable road user types into a single integrated system. By combining multiple detection tasks into one unified classifier trained on aggregated data, the system achieves efficient multi-category detection without the time penalty of separate training processes for each category
2Measurement precision
If separate classifiers are trained for bicycle riders and motorcycle riders, then specific detection accuracy for each type is improved, but the overall detection efficiency decreases due to multiple training processes
Solution Approach 1:
The unified generalized passerby classifier is designed to handle multiple user types (pedestrians, bicycle riders, motorcycle riders) within a single model architecture and training process. This universal approach maintains the ability to accurately detect specific user types while dramatically improving overall detection efficiency by eliminating redundant training operations
3Measurement precision
If extensive sample collection is performed to ensure comprehensive feature extraction, then classifier training accuracy is improved, but the time required for feature extraction and training increases
Solution Approach 1:
The patent combines sample data from multiple vulnerable road user types into a single aggregated training dataset. By merging these samples and training a unified classifier on the combined data, the system achieves comprehensive feature extraction across all user types without the time cost of processing and training separate datasets for each category
Data Source
AI summary
A method for detecting a generalized passerby includes: acquiring an input image; determining whether a preset common feature of a wheel exists in the input image; selecting an image window at left side or right side or upper side of a center of a region where the preset common feature of the wheel is located in a case that the preset common feature of the wheel exists in the input image; inputting the selected image window into a preset upper-body classifier; detecting whether an upper body of a passerby exists in the selected image window and outputting a first detection result.