Dempster-Shafer Multifeature Fusion for Pedestrian Detection
Pedestrian detection is of great importance for ensuring traffic safety. In recent years, many works employing image-based shape features to recognize pedestrians have been reported. However, previous pedestrian detectors were in many cases not sufficient to achieve satisfactory results under comple...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SAGE Publishing
2015-01-01
|
Series: | Advances in Mechanical Engineering |
Online Access: | https://doi.org/10.1155/2014/454058 |
id |
doaj-5dc11273e4ed419cab8f91f201a54ee6 |
---|---|
record_format |
Article |
spelling |
doaj-5dc11273e4ed419cab8f91f201a54ee62020-11-25T03:06:33ZengSAGE PublishingAdvances in Mechanical Engineering1687-81322015-01-01710.1155/2014/45405810.1155_2014/454058Dempster-Shafer Multifeature Fusion for Pedestrian DetectionHua Cui0Lingling Peng1Huansheng Song2Guofeng Wang3Jiancheng Li4Lu Guo5Chao Yuan6 School of Information Engineering, Chang'an University, Xi'an 710064, China School of Information Engineering, Chang'an University, Xi'an 710064, China School of Information Engineering, Chang'an University, Xi'an 710064, China China Highway Engineering Consulting Corporation, Beijing 100097, China China Highway Engineering Consulting Corporation, Beijing 100097, China School of Information Engineering, Chang'an University, Xi'an 710064, China School of Information Engineering, Chang'an University, Xi'an 710064, ChinaPedestrian detection is of great importance for ensuring traffic safety. In recent years, many works employing image-based shape features to recognize pedestrians have been reported. However, previous pedestrian detectors were in many cases not sufficient to achieve satisfactory results under complex weather conditions and complex scenarios. As a solution this paper exploits two video-based motion feature descriptors and applies such motion features to the detection task in addition to four classical shape features with the aim of significantly improving the detection performance. Our motion features are defined as the trajectory smoothness degree and motion vector field, which are derived from our proposed point tracking strategy beyond tough target segmentation. And then the appealing Dempster-Shafer theory of evidence (D-S theory) is applied to fuse these features, due to the fact that D-S theory is better than the classical Bayesian approach in handling the information with lack of prior probabilities. The proposed automatic pedestrian detection algorithm is evaluated on real data and in real traffic scenes under various weather conditions. Theoretical analysis and experiment results consistently show that the proposed method outperforms SVM-based multifeature fusion approach for pedestrian detection in terms of recognition ability and robustness in various real traffic scenes.https://doi.org/10.1155/2014/454058 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Hua Cui Lingling Peng Huansheng Song Guofeng Wang Jiancheng Li Lu Guo Chao Yuan |
spellingShingle |
Hua Cui Lingling Peng Huansheng Song Guofeng Wang Jiancheng Li Lu Guo Chao Yuan Dempster-Shafer Multifeature Fusion for Pedestrian Detection Advances in Mechanical Engineering |
author_facet |
Hua Cui Lingling Peng Huansheng Song Guofeng Wang Jiancheng Li Lu Guo Chao Yuan |
author_sort |
Hua Cui |
title |
Dempster-Shafer Multifeature Fusion for Pedestrian Detection |
title_short |
Dempster-Shafer Multifeature Fusion for Pedestrian Detection |
title_full |
Dempster-Shafer Multifeature Fusion for Pedestrian Detection |
title_fullStr |
Dempster-Shafer Multifeature Fusion for Pedestrian Detection |
title_full_unstemmed |
Dempster-Shafer Multifeature Fusion for Pedestrian Detection |
title_sort |
dempster-shafer multifeature fusion for pedestrian detection |
publisher |
SAGE Publishing |
series |
Advances in Mechanical Engineering |
issn |
1687-8132 |
publishDate |
2015-01-01 |
description |
Pedestrian detection is of great importance for ensuring traffic safety. In recent years, many works employing image-based shape features to recognize pedestrians have been reported. However, previous pedestrian detectors were in many cases not sufficient to achieve satisfactory results under complex weather conditions and complex scenarios. As a solution this paper exploits two video-based motion feature descriptors and applies such motion features to the detection task in addition to four classical shape features with the aim of significantly improving the detection performance. Our motion features are defined as the trajectory smoothness degree and motion vector field, which are derived from our proposed point tracking strategy beyond tough target segmentation. And then the appealing Dempster-Shafer theory of evidence (D-S theory) is applied to fuse these features, due to the fact that D-S theory is better than the classical Bayesian approach in handling the information with lack of prior probabilities. The proposed automatic pedestrian detection algorithm is evaluated on real data and in real traffic scenes under various weather conditions. Theoretical analysis and experiment results consistently show that the proposed method outperforms SVM-based multifeature fusion approach for pedestrian detection in terms of recognition ability and robustness in various real traffic scenes. |
url |
https://doi.org/10.1155/2014/454058 |
work_keys_str_mv |
AT huacui dempstershafermultifeaturefusionforpedestriandetection AT linglingpeng dempstershafermultifeaturefusionforpedestriandetection AT huanshengsong dempstershafermultifeaturefusionforpedestriandetection AT guofengwang dempstershafermultifeaturefusionforpedestriandetection AT jianchengli dempstershafermultifeaturefusionforpedestriandetection AT luguo dempstershafermultifeaturefusionforpedestriandetection AT chaoyuan dempstershafermultifeaturefusionforpedestriandetection |
_version_ |
1724673771286036480 |