A multi feature based on-road vehicle recognition
- Authors
- 신현철
- Issue Date
- Nov-2011
- Publisher
- IEEE
- Keywords
- Canny edge detectors; Multi features; Hypothesis verifications; Recognition rates; Bag-of-Features; Vehicle detection systems; Hypothesis generation; Horizontal edge; Information technology; Weather conditions; Vehicles; Computer science; Vehicle recognit
- Citation
- IEEE International Conference on Computer Sciences and Convergence Information Technology (ICCIT 201, v. , no. , pp.173 - 178
- Journal Title
- IEEE International Conference on Computer Sciences and Convergence Information Technology (ICCIT 201
- Start Page
- 173
- End Page
- 178
- URI
- https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/36458
- Abstract
- Vehicle recognition techniques are used for recognition of vehicles and to alert driver from dangerous situations that may cause accidents. In this paper, we introduced Difference of BiGaussian (DoBG) based edge detection method. This method is proved to be better than other famous edge based methods like canny edge detector. It was observed that horizontal edges are strong heuristic for vehicle recognition. Therefore, in hypothesis generation, we use Horizontal Edge Filtering (HEF) on DoBG edge map to filter long horizontal edges. Moreover, images are segmented to detect vehicles far from camera and to detect vehicles which are overtaking from right and left side of vehicle containing the camera. In Hypothesis verification, we use Bag-of-Features (BoF) with K nearest neighbor's algorithm for verification of generated hypothesis. Main focus of this paper is to improve the performance of vehicle detection systems by combination of DoBG and BoF. Our method is tested on different weather conditions (like Sunny/cloudy) in daytime (at afternoon/evening) and it shows recognition rate of 98.5% on average on roads inside a city and on highways. © 2011 AICIT.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - COLLEGE OF ENGINEERING SCIENCES > SCHOOL OF ELECTRICAL ENGINEERING > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/36458)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.