Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection
In this paper, a new edge detection method is proposed where multi-scale anisotropic Gaussian kernels (AGKs) are used to obtain an edge map from an input image. The main advantage of the proposed method is that high edge detection accuracy and edge resolution are attained while maintaining good nois...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8943409/ |
| _version_ | 1818857136365502464 |
|---|---|
| author | Yunhong Li Yuandong Bi Weichuan Zhang Changming Sun |
| author_facet | Yunhong Li Yuandong Bi Weichuan Zhang Changming Sun |
| author_sort | Yunhong Li |
| collection | DOAJ |
| description | In this paper, a new edge detection method is proposed where multi-scale anisotropic Gaussian kernels (AGKs) are used to obtain an edge map from an input image. The main advantage of the proposed method is that high edge detection accuracy and edge resolution are attained while maintaining good noise robustness. The proposed method consists of three aspects: First, anisotropic Gaussian directional derivatives (AGDDs) are derived from the AGKs which are used to acquire local intensity variation from an input image with multiple scales. Second, multi-scale AGDD based edge strength maps (ESMs) are fused into a new ESM with high edge resolution and little edge stretch effect which has the ability to solve the contradiction issue between noise robustness and accurate edge extraction. Third, the fused ESM is embedded into the framework of Canny detection for obtaining edge contours. Finally, the criteria on precision-recall curve, detection accuracy, and noise robustness are used to evaluate the proposed detector against four state-of-the-art methods. The experimental results show that our proposed detector outperforms all the other tested edge detection methods. |
| first_indexed | 2024-12-19T08:35:36Z |
| format | Article |
| id | doaj.art-5c03a13f8a874f798817eb64bf9a7efd |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T08:35:36Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-5c03a13f8a874f798817eb64bf9a7efd2022-12-21T20:29:03ZengIEEEIEEE Access2169-35362020-01-0181803181210.1109/ACCESS.2019.29625208943409Multi-Scale Anisotropic Gaussian Kernels for Image Edge DetectionYunhong Li0https://orcid.org/0000-0001-8080-1040Yuandong Bi1https://orcid.org/0000-0001-7087-131XWeichuan Zhang2https://orcid.org/0000-0003-4904-1826Changming Sun3https://orcid.org/0000-0001-5943-1989School of Electronics and Information, Xi’an Polytechnic University, Xi’an, ChinaSchool of Electronics and Information, Xi’an Polytechnic University, Xi’an, ChinaSchool of Electronics and Information, Xi’an Polytechnic University, Xi’an, ChinaCSIRO Data61, Epping, NSW, AustraliaIn this paper, a new edge detection method is proposed where multi-scale anisotropic Gaussian kernels (AGKs) are used to obtain an edge map from an input image. The main advantage of the proposed method is that high edge detection accuracy and edge resolution are attained while maintaining good noise robustness. The proposed method consists of three aspects: First, anisotropic Gaussian directional derivatives (AGDDs) are derived from the AGKs which are used to acquire local intensity variation from an input image with multiple scales. Second, multi-scale AGDD based edge strength maps (ESMs) are fused into a new ESM with high edge resolution and little edge stretch effect which has the ability to solve the contradiction issue between noise robustness and accurate edge extraction. Third, the fused ESM is embedded into the framework of Canny detection for obtaining edge contours. Finally, the criteria on precision-recall curve, detection accuracy, and noise robustness are used to evaluate the proposed detector against four state-of-the-art methods. The experimental results show that our proposed detector outperforms all the other tested edge detection methods.https://ieeexplore.ieee.org/document/8943409/Edge detectionanisotropic Gaussian kernelanisotropic Gaussian directional derivativemulti-scale |
| spellingShingle | Yunhong Li Yuandong Bi Weichuan Zhang Changming Sun Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection IEEE Access Edge detection anisotropic Gaussian kernel anisotropic Gaussian directional derivative multi-scale |
| title | Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection |
| title_full | Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection |
| title_fullStr | Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection |
| title_full_unstemmed | Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection |
| title_short | Multi-Scale Anisotropic Gaussian Kernels for Image Edge Detection |
| title_sort | multi scale anisotropic gaussian kernels for image edge detection |
| topic | Edge detection anisotropic Gaussian kernel anisotropic Gaussian directional derivative multi-scale |
| url | https://ieeexplore.ieee.org/document/8943409/ |
| work_keys_str_mv | AT yunhongli multiscaleanisotropicgaussiankernelsforimageedgedetection AT yuandongbi multiscaleanisotropicgaussiankernelsforimageedgedetection AT weichuanzhang multiscaleanisotropicgaussiankernelsforimageedgedetection AT changmingsun multiscaleanisotropicgaussiankernelsforimageedgedetection |