Learning parallel and hierarchical mechanisms for edge detection
Edge detection is one of the fundamental components of advanced computer vision tasks, and it is essential to preserve computational resources while ensuring a certain level of performance. In this paper, we propose a lightweight edge detection network called the Parallel and Hierarchical Network (P...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-07-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2023.1194713/full |
| _version_ | 1797773084646703104 |
|---|---|
| author | Ling Zhou Chuan Lin Chuan Lin Chuan Lin Xintao Pang Xintao Pang Xintao Pang Hao Yang Hao Yang Yongcai Pan Yongcai Pan Yuwei Zhang Yuwei Zhang |
| author_facet | Ling Zhou Chuan Lin Chuan Lin Chuan Lin Xintao Pang Xintao Pang Xintao Pang Hao Yang Hao Yang Yongcai Pan Yongcai Pan Yuwei Zhang Yuwei Zhang |
| author_sort | Ling Zhou |
| collection | DOAJ |
| description | Edge detection is one of the fundamental components of advanced computer vision tasks, and it is essential to preserve computational resources while ensuring a certain level of performance. In this paper, we propose a lightweight edge detection network called the Parallel and Hierarchical Network (PHNet), which draws inspiration from the parallel processing and hierarchical processing mechanisms of visual information in the visual cortex neurons and is implemented via a convolutional neural network (CNN). Specifically, we designed an encoding network with parallel and hierarchical processing based on the visual information transmission pathway of the “retina-LGN-V1” and meticulously modeled the receptive fields of the cells involved in the pathway. Empirical evaluation demonstrates that, despite a minimal parameter count of only 0.2 M, the proposed model achieves a remarkable ODS score of 0.781 on the BSDS500 dataset and ODS score of 0.863 on the MBDD dataset. These results underscore the efficacy of the proposed network in attaining superior edge detection performance at a low computational cost. Moreover, we believe that this study, which combines computational vision and biological vision, can provide new insights into edge detection model research. |
| first_indexed | 2024-03-12T22:00:09Z |
| format | Article |
| id | doaj.art-89c31d1ea2254a2cb5f59d105e663a1c |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-03-12T22:00:09Z |
| publishDate | 2023-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-89c31d1ea2254a2cb5f59d105e663a1c2023-07-25T07:57:42ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2023-07-011710.3389/fnins.2023.11947131194713Learning parallel and hierarchical mechanisms for edge detectionLing Zhou0Chuan Lin1Chuan Lin2Chuan Lin3Xintao Pang4Xintao Pang5Xintao Pang6Hao Yang7Hao Yang8Yongcai Pan9Yongcai Pan10Yuwei Zhang11Yuwei Zhang12Key Laboratory of AI and Information Processing (Hechi University), Education Department of Guangxi Zhuang Autonomous Region, Hechi University, Yizhou, ChinaKey Laboratory of AI and Information Processing (Hechi University), Education Department of Guangxi Zhuang Autonomous Region, Hechi University, Yizhou, ChinaSchool of Automation, Guangxi University of Science and Technology, Liuzhou, ChinaGuangxi Key Laboratory of Automobile Components and Vehicle Technology, Guangxi University of Science and Technology, Liuzhou, ChinaKey Laboratory of AI and Information Processing (Hechi University), Education Department of Guangxi Zhuang Autonomous Region, Hechi University, Yizhou, ChinaSchool of Automation, Guangxi University of Science and Technology, Liuzhou, ChinaGuangxi Key Laboratory of Automobile Components and Vehicle Technology, Guangxi University of Science and Technology, Liuzhou, ChinaSchool of Automation, Guangxi University of Science and Technology, Liuzhou, ChinaGuangxi Key Laboratory of Automobile Components and Vehicle Technology, Guangxi University of Science and Technology, Liuzhou, ChinaSchool of Automation, Guangxi University of Science and Technology, Liuzhou, ChinaGuangxi Key Laboratory of Automobile Components and Vehicle Technology, Guangxi University of Science and Technology, Liuzhou, ChinaSchool of Automation, Guangxi University of Science and Technology, Liuzhou, ChinaGuangxi Key Laboratory of Automobile Components and Vehicle Technology, Guangxi University of Science and Technology, Liuzhou, ChinaEdge detection is one of the fundamental components of advanced computer vision tasks, and it is essential to preserve computational resources while ensuring a certain level of performance. In this paper, we propose a lightweight edge detection network called the Parallel and Hierarchical Network (PHNet), which draws inspiration from the parallel processing and hierarchical processing mechanisms of visual information in the visual cortex neurons and is implemented via a convolutional neural network (CNN). Specifically, we designed an encoding network with parallel and hierarchical processing based on the visual information transmission pathway of the “retina-LGN-V1” and meticulously modeled the receptive fields of the cells involved in the pathway. Empirical evaluation demonstrates that, despite a minimal parameter count of only 0.2 M, the proposed model achieves a remarkable ODS score of 0.781 on the BSDS500 dataset and ODS score of 0.863 on the MBDD dataset. These results underscore the efficacy of the proposed network in attaining superior edge detection performance at a low computational cost. Moreover, we believe that this study, which combines computational vision and biological vision, can provide new insights into edge detection model research.https://www.frontiersin.org/articles/10.3389/fnins.2023.1194713/fulledge detectionconvolutional neural networkparallel processing mechanismhierarchical processing mechanismlightweight methods |
| spellingShingle | Ling Zhou Chuan Lin Chuan Lin Chuan Lin Xintao Pang Xintao Pang Xintao Pang Hao Yang Hao Yang Yongcai Pan Yongcai Pan Yuwei Zhang Yuwei Zhang Learning parallel and hierarchical mechanisms for edge detection Frontiers in Neuroscience edge detection convolutional neural network parallel processing mechanism hierarchical processing mechanism lightweight methods |
| title | Learning parallel and hierarchical mechanisms for edge detection |
| title_full | Learning parallel and hierarchical mechanisms for edge detection |
| title_fullStr | Learning parallel and hierarchical mechanisms for edge detection |
| title_full_unstemmed | Learning parallel and hierarchical mechanisms for edge detection |
| title_short | Learning parallel and hierarchical mechanisms for edge detection |
| title_sort | learning parallel and hierarchical mechanisms for edge detection |
| topic | edge detection convolutional neural network parallel processing mechanism hierarchical processing mechanism lightweight methods |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2023.1194713/full |
| work_keys_str_mv | AT lingzhou learningparallelandhierarchicalmechanismsforedgedetection AT chuanlin learningparallelandhierarchicalmechanismsforedgedetection AT chuanlin learningparallelandhierarchicalmechanismsforedgedetection AT chuanlin learningparallelandhierarchicalmechanismsforedgedetection AT xintaopang learningparallelandhierarchicalmechanismsforedgedetection AT xintaopang learningparallelandhierarchicalmechanismsforedgedetection AT xintaopang learningparallelandhierarchicalmechanismsforedgedetection AT haoyang learningparallelandhierarchicalmechanismsforedgedetection AT haoyang learningparallelandhierarchicalmechanismsforedgedetection AT yongcaipan learningparallelandhierarchicalmechanismsforedgedetection AT yongcaipan learningparallelandhierarchicalmechanismsforedgedetection AT yuweizhang learningparallelandhierarchicalmechanismsforedgedetection AT yuweizhang learningparallelandhierarchicalmechanismsforedgedetection |