Pedestrian Detection Using HOG and Granule Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pedestrian detection systems, such as those using histograms of oriented gradients (HOG), struggle with accurately identifying pedestrians in disorderly environments due to limitations in providing texture information, leading to potential misjudgments.
Innovation Solution
A pedestrian detection system comprising an image-taking mechanism and an image analyzer that combines HOG characteristics with granule-based relative characteristics, using a composite characteristic-extracting module and a support vector machine (SVM) to analyze images for pedestrian presence, employing a granule brightness-averaging unit and relative value-encoding unit to produce a 10-dimensional vector for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If HOG characteristics are used to detect pedestrians, then translation and orientation tolerance is improved, but texture information capability deteriorates
Solution Approach 1:
The patent combines HOG characteristics with granule-based relative characteristics into a composite feature representation. The HOG module extracts silhouette and edge information that provides translation and orientation tolerance, while the granule-based module extracts brightness variation patterns that capture texture information. By merging these two complementary feature sets, the system achieves both adaptability to pose variations and sensitivity to texture details, resolving the contradiction between the two requirements.
2Difficulty of detecting and measuring
If HOG is used to describe edges, then edge detection capability is improved, but distinction capability in disorderly environments deteriorates
Solution Approach 1:
The system merges HOG edge detection results with granule-based brightness variation analysis. While HOG effectively identifies edges and contours, the granule-based module divides the image into granules and analyzes brightness variations within and between granules, providing additional contextual information that helps distinguish pedestrians from background clutter in disorderly environments. This combination improves measurement precision without sacrificing edge detection capability.
3Device complexity
If single characteristic extraction is used, then system complexity is reduced, but detection accuracy deteriorates
Solution Approach 1:
The characteristic extraction system is segmented into two independent modules: a HOG characteristic extraction module and a granule-based relative characteristic extraction module. Each module independently extracts specific types of features from the image. This segmentation allows the system to maintain relatively simple individual modules while achieving high detection accuracy through the complementary combination of multiple feature types, effectively managing complexity while improving precision.
Data Source
AI summary
A pedestrian detector includes an image-taking mechanism and an image analyzer. The image-taking mechanism is provided on a vehicle and used to take an image in front of the vehicle. The image analyzer is connected to the image-taking mechanism and used to receive the image from the image-taking mechanism and analyze the image to determine whether the image conforms to a pedestrian to determine whether there is a pedestrian in front of the vehicle.