An Improved Faster R-CNN for Small Object Detection
With the increase of training data and the improvement of machine performance, the object detection method based on convolutional neural network (CNN) has become the mainstream algorithm in field of the current object detection. However, due to the complex background, occlusion and low resolution, t...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8786135/ |
id |
doaj-6f030c0adb2f459c970defc47858cf6b |
---|---|
record_format |
Article |
spelling |
doaj-6f030c0adb2f459c970defc47858cf6b2021-04-05T17:07:48ZengIEEEIEEE Access2169-35362019-01-01710683810684610.1109/ACCESS.2019.29327318786135An Improved Faster R-CNN for Small Object DetectionChangqing Cao0Bo Wang1https://orcid.org/0000-0002-7612-702XWenrui Zhang2Xiaodong Zeng3Xu Yan4Zhejun Feng5Yutao Liu6Zengyan Wu7School of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an, ChinaWith the increase of training data and the improvement of machine performance, the object detection method based on convolutional neural network (CNN) has become the mainstream algorithm in field of the current object detection. However, due to the complex background, occlusion and low resolution, there are still problems of small object detection. In this paper, we propose an improved algorithm based on faster region-based CNN (Faster R-CNN) for small object detection. Using the two-stage detection idea, in the positioning stage, we propose an improved loss function based on intersection over Union (IoU) for bounding box regression, and use bilinear interpolation to improve the regions of interest (RoI) pooling operation to solve the problem of positioning deviation, in the recognition stage, we use the multi-scale convolution feature fusion to make the feature map contain more information, and use the improved non-maximum suppression (NMS) algorithm to avoid loss of overlapping objects. The results show that the proposed algorithm has good performance on traffic signs whose resolution is in the range of (0, 32], the algorithm's recall rate reaches 90%, and the accuracy rate reaches 87%. Detection performance is significantly better than Faster R- CNN. Therefore, our algorithm is an effective way to detect small objects.https://ieeexplore.ieee.org/document/8786135/CNNfaster R-CNNsmall object detection |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Changqing Cao Bo Wang Wenrui Zhang Xiaodong Zeng Xu Yan Zhejun Feng Yutao Liu Zengyan Wu |
spellingShingle |
Changqing Cao Bo Wang Wenrui Zhang Xiaodong Zeng Xu Yan Zhejun Feng Yutao Liu Zengyan Wu An Improved Faster R-CNN for Small Object Detection IEEE Access CNN faster R-CNN small object detection |
author_facet |
Changqing Cao Bo Wang Wenrui Zhang Xiaodong Zeng Xu Yan Zhejun Feng Yutao Liu Zengyan Wu |
author_sort |
Changqing Cao |
title |
An Improved Faster R-CNN for Small Object Detection |
title_short |
An Improved Faster R-CNN for Small Object Detection |
title_full |
An Improved Faster R-CNN for Small Object Detection |
title_fullStr |
An Improved Faster R-CNN for Small Object Detection |
title_full_unstemmed |
An Improved Faster R-CNN for Small Object Detection |
title_sort |
improved faster r-cnn for small object detection |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
With the increase of training data and the improvement of machine performance, the object detection method based on convolutional neural network (CNN) has become the mainstream algorithm in field of the current object detection. However, due to the complex background, occlusion and low resolution, there are still problems of small object detection. In this paper, we propose an improved algorithm based on faster region-based CNN (Faster R-CNN) for small object detection. Using the two-stage detection idea, in the positioning stage, we propose an improved loss function based on intersection over Union (IoU) for bounding box regression, and use bilinear interpolation to improve the regions of interest (RoI) pooling operation to solve the problem of positioning deviation, in the recognition stage, we use the multi-scale convolution feature fusion to make the feature map contain more information, and use the improved non-maximum suppression (NMS) algorithm to avoid loss of overlapping objects. The results show that the proposed algorithm has good performance on traffic signs whose resolution is in the range of (0, 32], the algorithm's recall rate reaches 90%, and the accuracy rate reaches 87%. Detection performance is significantly better than Faster R- CNN. Therefore, our algorithm is an effective way to detect small objects. |
topic |
CNN faster R-CNN small object detection |
url |
https://ieeexplore.ieee.org/document/8786135/ |
work_keys_str_mv |
AT changqingcao animprovedfasterrcnnforsmallobjectdetection AT bowang animprovedfasterrcnnforsmallobjectdetection AT wenruizhang animprovedfasterrcnnforsmallobjectdetection AT xiaodongzeng animprovedfasterrcnnforsmallobjectdetection AT xuyan animprovedfasterrcnnforsmallobjectdetection AT zhejunfeng animprovedfasterrcnnforsmallobjectdetection AT yutaoliu animprovedfasterrcnnforsmallobjectdetection AT zengyanwu animprovedfasterrcnnforsmallobjectdetection AT changqingcao improvedfasterrcnnforsmallobjectdetection AT bowang improvedfasterrcnnforsmallobjectdetection AT wenruizhang improvedfasterrcnnforsmallobjectdetection AT xiaodongzeng improvedfasterrcnnforsmallobjectdetection AT xuyan improvedfasterrcnnforsmallobjectdetection AT zhejunfeng improvedfasterrcnnforsmallobjectdetection AT yutaoliu improvedfasterrcnnforsmallobjectdetection AT zengyanwu improvedfasterrcnnforsmallobjectdetection |
_version_ |
1721540253913710592 |