A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement
At present, the problem of pedestrian detection has attracted increasing attention in the field of computer vision. The faster regions with convolutional neural network features (Faster R-CNN) are regarded as one of the most important techniques for studying this problem. However, the detection capa...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-11-01
|
Series: | Algorithms |
Subjects: | |
Online Access: | https://www.mdpi.com/1999-4893/11/12/192 |
id |
doaj-2e8440df0cb142bc8cd1a1fb7dd3ab09 |
---|---|
record_format |
Article |
spelling |
doaj-2e8440df0cb142bc8cd1a1fb7dd3ab092020-11-24T23:30:10ZengMDPI AGAlgorithms1999-48932018-11-01111219210.3390/a11120192a11120192A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE EnhancementHongquan Qu0Meihan Wang1Changnian Zhang2Yun Wei3College of Electronic and Information Engineering, North China University of Technology, Shijingshan District, Beijing 100144, ChinaCollege of Electronic and Information Engineering, North China University of Technology, Shijingshan District, Beijing 100144, ChinaCollege of Electronic and Information Engineering, North China University of Technology, Shijingshan District, Beijing 100144, ChinaBeijing Urban Construction Design & Development Group Co., Ltd., Beijing 100037, ChinaAt present, the problem of pedestrian detection has attracted increasing attention in the field of computer vision. The faster regions with convolutional neural network features (Faster R-CNN) are regarded as one of the most important techniques for studying this problem. However, the detection capability of the model trained by faster R-CNN is susceptible to the diversity of pedestrians’ appearance and the light intensity in specific scenarios, such as in a subway, which can lead to the decline in recognition rate and the offset of target selection for pedestrians. In this paper, we propose the modified faster R-CNN method with automatic color enhancement (ACE), which can improve sample contrast by calculating the relative light and dark relationship to correct the final pixel value. In addition, a calibration method based on sample categories reduction is presented to accurately locate the target for detection. Then, we choose the faster R-CNN target detection framework on the experimental dataset. Finally, the effectiveness of this method is verified with the actual data sample collected from the subway passenger monitoring video.https://www.mdpi.com/1999-4893/11/12/192subway pedestrian detectionsample calibrationfaster R-CNNautomatic color enhancement (ACE)false and miss detection |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Hongquan Qu Meihan Wang Changnian Zhang Yun Wei |
spellingShingle |
Hongquan Qu Meihan Wang Changnian Zhang Yun Wei A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement Algorithms subway pedestrian detection sample calibration faster R-CNN automatic color enhancement (ACE) false and miss detection |
author_facet |
Hongquan Qu Meihan Wang Changnian Zhang Yun Wei |
author_sort |
Hongquan Qu |
title |
A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement |
title_short |
A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement |
title_full |
A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement |
title_fullStr |
A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement |
title_full_unstemmed |
A Study on Faster R-CNN-Based Subway Pedestrian Detection with ACE Enhancement |
title_sort |
study on faster r-cnn-based subway pedestrian detection with ace enhancement |
publisher |
MDPI AG |
series |
Algorithms |
issn |
1999-4893 |
publishDate |
2018-11-01 |
description |
At present, the problem of pedestrian detection has attracted increasing attention in the field of computer vision. The faster regions with convolutional neural network features (Faster R-CNN) are regarded as one of the most important techniques for studying this problem. However, the detection capability of the model trained by faster R-CNN is susceptible to the diversity of pedestrians’ appearance and the light intensity in specific scenarios, such as in a subway, which can lead to the decline in recognition rate and the offset of target selection for pedestrians. In this paper, we propose the modified faster R-CNN method with automatic color enhancement (ACE), which can improve sample contrast by calculating the relative light and dark relationship to correct the final pixel value. In addition, a calibration method based on sample categories reduction is presented to accurately locate the target for detection. Then, we choose the faster R-CNN target detection framework on the experimental dataset. Finally, the effectiveness of this method is verified with the actual data sample collected from the subway passenger monitoring video. |
topic |
subway pedestrian detection sample calibration faster R-CNN automatic color enhancement (ACE) false and miss detection |
url |
https://www.mdpi.com/1999-4893/11/12/192 |
work_keys_str_mv |
AT hongquanqu astudyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT meihanwang astudyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT changnianzhang astudyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT yunwei astudyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT hongquanqu studyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT meihanwang studyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT changnianzhang studyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement AT yunwei studyonfasterrcnnbasedsubwaypedestriandetectionwithaceenhancement |
_version_ |
1725542436371955712 |