Pedestrian Detection Using Lane-Based Region of Interest
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pedestrian detection methods in vehicles face slow processing speeds and low accuracy due to high data processing requirements and motion-based detection analysis, especially when pedestrians are still or at low resolution.
Innovation Solution
A pedestrian detecting apparatus and method that uses lane information to set a Region Of Interest (ROI) and divide it into upper and lower body regions, employing a search window to detect pedestrians efficiently, thereby reducing processing time and enhancing detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full search method based on SVM classifier is used for pedestrian detection, then detection accuracy is improved, but processing speed deteriorates
Solution Approach 1:
The patent divides the image processing into multiple stages: first stage uses a simple classifier to quickly eliminate negative regions, second stage applies a more accurate classifier only to candidate regions. This segmentation of detection stages allows achieving high accuracy while maintaining processing speed by avoiding exhaustive search throughout the entire image.
Solution Approach 2:
The patent performs preliminary filtering of candidate regions using a simple and fast classifier before applying the more accurate but computationally intensive SVM classifier. This preliminary action reduces the search space significantly, allowing the high-accuracy classifier to operate only on relevant regions, thus resolving the speed-accuracy tradeoff.
2Adaptability or versatility
If motion-based detection analysis is performed on the whole region of input image, then detection coverage is improved, but processing speed deteriorates
Solution Approach 1:
The patent extracts and analyzes motion information only from candidate pedestrian regions identified through lane-based ROI setting, rather than analyzing motion across the entire input image. This extraction of motion features from specific regions of interest maintains detection coverage while significantly reducing computational burden and processing speed requirements.
3Measurement precision
If high resolution input image is used for pedestrian detection, then detection accuracy is improved, but data processing volume increases
Solution Approach 1:
The patent applies different processing qualities to different regions of the image: high-resolution detailed analysis is applied only to candidate pedestrian regions within the ROI, while other regions receive minimal or no processing. This local quality approach maintains detection accuracy in critical areas while reducing overall data processing volume.
Data Source
AI summary
Provided is an image processing apparatus for detecting pedestrians. The image processing apparatus includes a lane detecting module configured to extract a lane coordinate value from an input image and a pedestrian detecting module configured to set, as a pedestrian region of interest (ROI), a region between a first line passing through ends of first left and right lanes and a second line passing through ends of second left and right lanes which are respectively disposed above the left and right lanes, and search for the pedestrian ROI by using a predetermined window to detect a pedestrian region having a pedestrian feature.