Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network
To accurately extract high precision edges in complex background images,this paper proposes an improved single-pixel edge extraction algorithm.In the improved fully convolutional neural network,this method adds an auxiliary output layer and adopts a multi-scale input method to coarsely extract multi...
| Published in: | Jisuanji gongcheng |
|---|---|
| Main Author: | |
| Format: | Article |
| Language: | English |
| Published: |
Editorial Office of Computer Engineering
2020-01-01
|
| Subjects: | |
| Online Access: | https://www.ecice06.com/fileup/1000-3428/PDF/20200137.pdf |
| _version_ | 1848649843296174080 |
|---|---|
| author | LIU Chang, ZHANG Jian, LIN Jianping |
| author_facet | LIU Chang, ZHANG Jian, LIN Jianping |
| author_sort | LIU Chang, ZHANG Jian, LIN Jianping |
| collection | DOAJ |
| container_title | Jisuanji gongcheng |
| description | To accurately extract high precision edges in complex background images,this paper proposes an improved single-pixel edge extraction algorithm.In the improved fully convolutional neural network,this method adds an auxiliary output layer and adopts a multi-scale input method to coarsely extract multi-pixel edges of an image.Then the watershed algorithm is used to refine and relocate the multi-pixel edges to obtain a high precision single-pixel edge of an image.Application results on magnetic tile images show that the algorithm has strong robustness and can extract complete continuous high precision single-pixel edges. |
| format | Article |
| id | doaj-758d008eae15415a97f497c01c2782db |
| institution | Directory of Open Access Journals |
| issn | 1000-3428 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Editorial Office of Computer Engineering |
| record_format | Article |
| spelling | doaj-758d008eae15415a97f497c01c2782db2025-11-03T05:51:46ZengEditorial Office of Computer EngineeringJisuanji gongcheng1000-34282020-01-0146126227010.19678/j.issn.1000-3428.0053574Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural NetworkLIU Chang, ZHANG Jian, LIN Jianping0School of Mechanical Engineering, Tongji University, Shanghai 201804, ChinaTo accurately extract high precision edges in complex background images,this paper proposes an improved single-pixel edge extraction algorithm.In the improved fully convolutional neural network,this method adds an auxiliary output layer and adopts a multi-scale input method to coarsely extract multi-pixel edges of an image.Then the watershed algorithm is used to refine and relocate the multi-pixel edges to obtain a high precision single-pixel edge of an image.Application results on magnetic tile images show that the algorithm has strong robustness and can extract complete continuous high precision single-pixel edges.https://www.ecice06.com/fileup/1000-3428/PDF/20200137.pdfsingle-pixel edge detection|fully convolutional neural network|watershed algorithm|distance error|magnetic tile image |
| spellingShingle | LIU Chang, ZHANG Jian, LIN Jianping Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network single-pixel edge detection|fully convolutional neural network|watershed algorithm|distance error|magnetic tile image |
| title | Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network |
| title_full | Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network |
| title_fullStr | Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network |
| title_full_unstemmed | Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network |
| title_short | Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network |
| title_sort | single pixel edge extraction of image based on improved fully convolutional neural network |
| topic | single-pixel edge detection|fully convolutional neural network|watershed algorithm|distance error|magnetic tile image |
| url | https://www.ecice06.com/fileup/1000-3428/PDF/20200137.pdf |
| work_keys_str_mv | AT liuchangzhangjianlinjianping singlepixeledgeextractionofimagebasedonimprovedfullyconvolutionalneuralnetwork |