Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining
Object detection is a significant issue in visual surveillance. Faster region-based convolutional neural network (R-CNN) is a typical object detection algorithm of deep learning; however, neither its generalization ability nor its detection accuracy of small object is high. In this paper, an effecti...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2018-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8314823/ |
id |
doaj-73f3deb4c811449899afa651b24c373d |
---|---|
record_format |
Article |
spelling |
doaj-73f3deb4c811449899afa651b24c373d2021-03-29T20:52:58ZengIEEEIEEE Access2169-35362018-01-016199591996710.1109/ACCESS.2018.28151498314823Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example MiningJun Chu0Zhixian Guo1Lu Leng2https://orcid.org/0000-0002-5667-224XSchool of Software, Nanchang Hangkong University, Nanchang, ChinaSchool of Information Engineering, Nanchang Hangkong University, Nanchang, ChinaSchool of Software, Nanchang Hangkong University, Nanchang, ChinaObject detection is a significant issue in visual surveillance. Faster region-based convolutional neural network (R-CNN) is a typical object detection algorithm of deep learning; however, neither its generalization ability nor its detection accuracy of small object is high. In this paper, an effective object detection algorithm is proposed for the small and occluded objects, which is based on multi-layer convolution feature fusion (MCFF) and online hard example mining (OHEM). First, the candidate regions are generated with region proposal network optimized by MCFF. Then, an effective OHEM algorithm is employed to train the region-based ConvNet detector. The hard examples are automatically selected to improve training efficiency. The avoidance of invalid examples accelerates the convergence speed of the model training. The experiments are performed on KITTI data set in intelligent traffic scenario. The proposed method outperforms the popular methods, such as Faster R-CNN, Regionlets, in terms of the overall detection accuracy. Furthermore, our method is good at the detection of small and occluded objects.https://ieeexplore.ieee.org/document/8314823/Deep leaningmulti-layer convolution feature fusionobject detectiononline hard example miningregion proposal network |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jun Chu Zhixian Guo Lu Leng |
spellingShingle |
Jun Chu Zhixian Guo Lu Leng Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining IEEE Access Deep leaning multi-layer convolution feature fusion object detection online hard example mining region proposal network |
author_facet |
Jun Chu Zhixian Guo Lu Leng |
author_sort |
Jun Chu |
title |
Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining |
title_short |
Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining |
title_full |
Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining |
title_fullStr |
Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining |
title_full_unstemmed |
Object Detection Based on Multi-Layer Convolution Feature Fusion and Online Hard Example Mining |
title_sort |
object detection based on multi-layer convolution feature fusion and online hard example mining |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2018-01-01 |
description |
Object detection is a significant issue in visual surveillance. Faster region-based convolutional neural network (R-CNN) is a typical object detection algorithm of deep learning; however, neither its generalization ability nor its detection accuracy of small object is high. In this paper, an effective object detection algorithm is proposed for the small and occluded objects, which is based on multi-layer convolution feature fusion (MCFF) and online hard example mining (OHEM). First, the candidate regions are generated with region proposal network optimized by MCFF. Then, an effective OHEM algorithm is employed to train the region-based ConvNet detector. The hard examples are automatically selected to improve training efficiency. The avoidance of invalid examples accelerates the convergence speed of the model training. The experiments are performed on KITTI data set in intelligent traffic scenario. The proposed method outperforms the popular methods, such as Faster R-CNN, Regionlets, in terms of the overall detection accuracy. Furthermore, our method is good at the detection of small and occluded objects. |
topic |
Deep leaning multi-layer convolution feature fusion object detection online hard example mining region proposal network |
url |
https://ieeexplore.ieee.org/document/8314823/ |
work_keys_str_mv |
AT junchu objectdetectionbasedonmultilayerconvolutionfeaturefusionandonlinehardexamplemining AT zhixianguo objectdetectionbasedonmultilayerconvolutionfeaturefusionandonlinehardexamplemining AT luleng objectdetectionbasedonmultilayerconvolutionfeaturefusionandonlinehardexamplemining |
_version_ |
1724193950334451712 |