A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes
The detection capability of deep learning-based stereo matching in industrial applications is inherently limited due to challenges posed by weak texture and inconsistent reflectance, making it difficult to accurately recover complex surface details. To achieve accurate measurements, this paper prese...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10387681/ |
| _version_ | 1827308341982920704 |
|---|---|
| author | Yunxuan Liu Kai Yang Xinyu Li Zijian Bai Yingying Wan Liming Xie |
| author_facet | Yunxuan Liu Kai Yang Xinyu Li Zijian Bai Yingying Wan Liming Xie |
| author_sort | Yunxuan Liu |
| collection | DOAJ |
| description | The detection capability of deep learning-based stereo matching in industrial applications is inherently limited due to challenges posed by weak texture and inconsistent reflectance, making it difficult to accurately recover complex surface details. To achieve accurate measurements, this paper presents an end-to-end speckle stereo matching network that incorporates fringe, Gray code, and speckle projection patterns. The model is trained using a high-precision dataset consisting of thousands of pairs generated through binocular Gray code-assisted phase shifting. After establishing local correspondences between the left and right images using speckle patterns, the images are used as inputs to the network. The proposed network consists of two siamese 2D feature extraction networks. One network is dedicated to cost volume computation, while the other focuses on weight refinement feature extraction. The former network incorporates a lightweight module for extracting high-dimensional fusion features. These features are obtained from different dilation scales and randomly concatenated along the channel dimension. Patch convolution is utilized to effectively adapt to pixel features at various levels, reducing redundancy within the cost volume and improving the network’s capacity to learn from ill-posed regions. Experimental results demonstrate that the proposed network achieves a significant improvement of approximately 10.7% in matching accuracy compared to state-of-the-art networks on public datasets. Furthermore, this method exhibits outstanding matching results when applied to diverse industrial scenarios. The reconstruction error for the radius of optical standard spheres is below 0.06-mm, which meets the demands of the majority of industrial applications. |
| first_indexed | 2024-04-24T18:56:40Z |
| format | Article |
| id | doaj.art-cfbbd9b379024a7da3590f1e0a5f6d05 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-24T18:56:40Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-cfbbd9b379024a7da3590f1e0a5f6d052024-03-26T17:34:37ZengIEEEIEEE Access2169-35362024-01-01126777678910.1109/ACCESS.2024.335213610387681A Robust End-to-End Speckle Stereo Matching Network for Industrial ScenesYunxuan Liu0https://orcid.org/0009-0009-7698-621XKai Yang1https://orcid.org/0000-0002-1211-4262Xinyu Li2Zijian Bai3Yingying Wan4Liming Xie5School of Physical Science and Technology, Southwest Jiaotong University, Chengdu, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu, ChinaThe detection capability of deep learning-based stereo matching in industrial applications is inherently limited due to challenges posed by weak texture and inconsistent reflectance, making it difficult to accurately recover complex surface details. To achieve accurate measurements, this paper presents an end-to-end speckle stereo matching network that incorporates fringe, Gray code, and speckle projection patterns. The model is trained using a high-precision dataset consisting of thousands of pairs generated through binocular Gray code-assisted phase shifting. After establishing local correspondences between the left and right images using speckle patterns, the images are used as inputs to the network. The proposed network consists of two siamese 2D feature extraction networks. One network is dedicated to cost volume computation, while the other focuses on weight refinement feature extraction. The former network incorporates a lightweight module for extracting high-dimensional fusion features. These features are obtained from different dilation scales and randomly concatenated along the channel dimension. Patch convolution is utilized to effectively adapt to pixel features at various levels, reducing redundancy within the cost volume and improving the network’s capacity to learn from ill-posed regions. Experimental results demonstrate that the proposed network achieves a significant improvement of approximately 10.7% in matching accuracy compared to state-of-the-art networks on public datasets. Furthermore, this method exhibits outstanding matching results when applied to diverse industrial scenarios. The reconstruction error for the radius of optical standard spheres is below 0.06-mm, which meets the demands of the majority of industrial applications.https://ieeexplore.ieee.org/document/10387681/Stereo matchingGray code-assisted phase shiftinghigh robustness industrial imaginggroup-wise volumecorrection map |
| spellingShingle | Yunxuan Liu Kai Yang Xinyu Li Zijian Bai Yingying Wan Liming Xie A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes IEEE Access Stereo matching Gray code-assisted phase shifting high robustness industrial imaging group-wise volume correction map |
| title | A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes |
| title_full | A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes |
| title_fullStr | A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes |
| title_full_unstemmed | A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes |
| title_short | A Robust End-to-End Speckle Stereo Matching Network for Industrial Scenes |
| title_sort | robust end to end speckle stereo matching network for industrial scenes |
| topic | Stereo matching Gray code-assisted phase shifting high robustness industrial imaging group-wise volume correction map |
| url | https://ieeexplore.ieee.org/document/10387681/ |
| work_keys_str_mv | AT yunxuanliu arobustendtoendspecklestereomatchingnetworkforindustrialscenes AT kaiyang arobustendtoendspecklestereomatchingnetworkforindustrialscenes AT xinyuli arobustendtoendspecklestereomatchingnetworkforindustrialscenes AT zijianbai arobustendtoendspecklestereomatchingnetworkforindustrialscenes AT yingyingwan arobustendtoendspecklestereomatchingnetworkforindustrialscenes AT limingxie arobustendtoendspecklestereomatchingnetworkforindustrialscenes AT yunxuanliu robustendtoendspecklestereomatchingnetworkforindustrialscenes AT kaiyang robustendtoendspecklestereomatchingnetworkforindustrialscenes AT xinyuli robustendtoendspecklestereomatchingnetworkforindustrialscenes AT zijianbai robustendtoendspecklestereomatchingnetworkforindustrialscenes AT yingyingwan robustendtoendspecklestereomatchingnetworkforindustrialscenes AT limingxie robustendtoendspecklestereomatchingnetworkforindustrialscenes |