Pedestrian Behavior Prediction Using Shape Feature Accumulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pedestrian behavior prediction technologies fail to accurately and rapidly predict pedestrian movements, especially discontinuous movements such as sudden direction changes and rushes onto the road, due to insufficient environmental recognition and precision.
Innovation Solution
A pedestrian behavior predicting device that detects movement changes by analyzing shape information from partial images, normalizing and accumulating feature data, and using optical flow to estimate both continuous and discontinuous movements, allowing for precise prediction of pedestrian behavior based on historical position data and environmental context.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional pedestrian detection methods are used, then pedestrian detection is achieved, but discontinuous movement (sudden direction changes and road rushes) cannot be predicted
Solution Approach 1:
The prediction system is segmented into two specialized modules: continuous movement estimation means for linear predictable movements and discontinuous movement estimation means for sudden direction changes. Each module specializes in specific movement patterns, allowing the system to handle diverse pedestrian behaviors effectively
Solution Approach 2:
The system dynamically switches between continuous and discontinuous movement estimation based on detected movement changes. When sudden direction changes are detected, the system activates the discontinuous movement estimation module to provide accurate predictions for unpredictable behaviors
2Measurement precision
If shape information analysis is performed, then movement change detection precision is improved, but processing time increases
Solution Approach 1:
The system extracts only essential shape information (feature amount distribution) from pedestrian images and accumulates this extracted data over time. By focusing on key shape characteristics rather than processing entire images, the system achieves high detection precision while maintaining efficient processing speed
Solution Approach 2:
Shape information is accumulated in advance over time periods before prediction is needed. This preliminary accumulation of shape data allows the system to rapidly detect movement changes when they occur, without requiring intensive real-time processing
Data Source
Figure 1
Figure 2
Figure 3
AI summary
According to the present invention, a pedestrian is detected from an imaged image and a partial image including the pedestrian is extracted, shape information of the pedestrian acquired from the extracted partial image is accumulated and the shape information of a predetermined time before and the current shape information are compared using the accumulated shape information to detect change in the movement of the pedestrian, discontinuous movement estimating information indicating a discontinuous movement of the pedestrian that occurs following the change in the movement of the pedestrian is acquired from a storage means at the time the change in the movement of the pedestrian is detected, and a behavior of the pedestrian is predicted using the acquired discontinuous movement estimating information.