A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images
The tube contours in two-dimensional images are important cues for optical three-dimensional reconstruction. Aiming at the practical problems encountered in the application of tube contour detection under complex background, a fully convolutional network (FCN)-based tube contour detection method is...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/12/4095 |
id |
doaj-951df1f46f2340fa9e39794312ee59ba |
---|---|
record_format |
Article |
spelling |
doaj-951df1f46f2340fa9e39794312ee59ba2021-07-01T00:09:27ZengMDPI AGSensors1424-82202021-06-01214095409510.3390/s21124095A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure ImagesXiaoqi Cheng0Junhua Sun1Fuqiang Zhou2School of Instrumentation Science and Opto-electronics Engineering, Beihang University, Beijing 100191, ChinaSchool of Instrumentation Science and Opto-electronics Engineering, Beihang University, Beijing 100191, ChinaSchool of Instrumentation Science and Opto-electronics Engineering, Beihang University, Beijing 100191, ChinaThe tube contours in two-dimensional images are important cues for optical three-dimensional reconstruction. Aiming at the practical problems encountered in the application of tube contour detection under complex background, a fully convolutional network (FCN)-based tube contour detection method is proposed. Multi-exposure (ME) images are captured as the input of FCN in order to get information of tube contours in different dynamic ranges, and the U-Net type architecture is adopted by the FCN to achieve pixel-level dense classification. In addition, we propose a new loss function that can help eliminate the adverse effects caused by the positional deviation and jagged morphology of tube contour labels. Finally, we introduce a new dataset called multi-exposure tube contour dataset (METCD) and a new evaluation metric called dilate inaccuracy at optimal dataset scale (DIA-ODS) to reach an overall evaluation of our proposed method. The experimental results show that the proposed method can effectively improve the integrity and accuracy of tube contour detection in complex scenes.https://www.mdpi.com/1424-8220/21/12/4095fully convolutional networktube contour detectionmulti-exposure imagesU-Netdilation operation |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Xiaoqi Cheng Junhua Sun Fuqiang Zhou |
spellingShingle |
Xiaoqi Cheng Junhua Sun Fuqiang Zhou A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images Sensors fully convolutional network tube contour detection multi-exposure images U-Net dilation operation |
author_facet |
Xiaoqi Cheng Junhua Sun Fuqiang Zhou |
author_sort |
Xiaoqi Cheng |
title |
A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images |
title_short |
A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images |
title_full |
A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images |
title_fullStr |
A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images |
title_full_unstemmed |
A Fully Convolutional Network-Based Tube Contour Detection Method Using Multi-Exposure Images |
title_sort |
fully convolutional network-based tube contour detection method using multi-exposure images |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2021-06-01 |
description |
The tube contours in two-dimensional images are important cues for optical three-dimensional reconstruction. Aiming at the practical problems encountered in the application of tube contour detection under complex background, a fully convolutional network (FCN)-based tube contour detection method is proposed. Multi-exposure (ME) images are captured as the input of FCN in order to get information of tube contours in different dynamic ranges, and the U-Net type architecture is adopted by the FCN to achieve pixel-level dense classification. In addition, we propose a new loss function that can help eliminate the adverse effects caused by the positional deviation and jagged morphology of tube contour labels. Finally, we introduce a new dataset called multi-exposure tube contour dataset (METCD) and a new evaluation metric called dilate inaccuracy at optimal dataset scale (DIA-ODS) to reach an overall evaluation of our proposed method. The experimental results show that the proposed method can effectively improve the integrity and accuracy of tube contour detection in complex scenes. |
topic |
fully convolutional network tube contour detection multi-exposure images U-Net dilation operation |
url |
https://www.mdpi.com/1424-8220/21/12/4095 |
work_keys_str_mv |
AT xiaoqicheng afullyconvolutionalnetworkbasedtubecontourdetectionmethodusingmultiexposureimages AT junhuasun afullyconvolutionalnetworkbasedtubecontourdetectionmethodusingmultiexposureimages AT fuqiangzhou afullyconvolutionalnetworkbasedtubecontourdetectionmethodusingmultiexposureimages AT xiaoqicheng fullyconvolutionalnetworkbasedtubecontourdetectionmethodusingmultiexposureimages AT junhuasun fullyconvolutionalnetworkbasedtubecontourdetectionmethodusingmultiexposureimages AT fuqiangzhou fullyconvolutionalnetworkbasedtubecontourdetectionmethodusingmultiexposureimages |
_version_ |
1721349354267082752 |