U-Net based Multi-level Texture Suppression for Vessel Segmentation in Low Contrast Regions. Upadhyay, K., Agrawal, M., & Vashist, P. In 2020 28th European Signal Processing Conference (EUSIPCO), pages 1304-1308, Aug, 2020.
![U-Net based Multi-level Texture Suppression for Vessel Segmentation in Low Contrast Regions [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper doi abstract bibtex
Paper doi abstract bibtex Segmentation of retinal blood vessels is important for diagnosis of many retinal diseases. Precise segmentation of complete vessel-map is still a challenge in low contrast regions of fundus images. Vessel pixels belonging to these regions, such as, fine vessel-endings and boundaries of vessels, get merged in the neighboring vessel-like texture. This paper proposes a novel retinal vessel segmentation algorithm which handles the background vessel-like texture in a sophisticated manner without harming the vessel pixels. In this work, first we enhance all possible vessel-like features of fundus at different `levels' using 2-D Gabor wavelet and Gaussian matched filtering. At each `level', texture is suppressed using Local Laplacian filter while preserving the vessel edges. The resulting images are combined to produce a maximum response image with enhanced vessels of different thicknesses and suppressed texture. This handcrafted image is used to train the deep U-net model for further suppression of non-vessel pixels. Proposed segmentation method is tested on publicly available DRIVE and STARE databases. The algorithm has produced state-of-the-art results. It has performed outstandingly well in terms of sensitivity measure which is most affected with the correct segmentation of fine vessels and vessel-boundary pixels present in low-contrast regions.
@InProceedings{9287387,
author = {K. Upadhyay and M. Agrawal and P. Vashist},
booktitle = {2020 28th European Signal Processing Conference (EUSIPCO)},
title = {U-Net based Multi-level Texture Suppression for Vessel Segmentation in Low Contrast Regions},
year = {2020},
pages = {1304-1308},
abstract = {Segmentation of retinal blood vessels is important for diagnosis of many retinal diseases. Precise segmentation of complete vessel-map is still a challenge in low contrast regions of fundus images. Vessel pixels belonging to these regions, such as, fine vessel-endings and boundaries of vessels, get merged in the neighboring vessel-like texture. This paper proposes a novel retinal vessel segmentation algorithm which handles the background vessel-like texture in a sophisticated manner without harming the vessel pixels. In this work, first we enhance all possible vessel-like features of fundus at different `levels' using 2-D Gabor wavelet and Gaussian matched filtering. At each `level', texture is suppressed using Local Laplacian filter while preserving the vessel edges. The resulting images are combined to produce a maximum response image with enhanced vessels of different thicknesses and suppressed texture. This handcrafted image is used to train the deep U-net model for further suppression of non-vessel pixels. Proposed segmentation method is tested on publicly available DRIVE and STARE databases. The algorithm has produced state-of-the-art results. It has performed outstandingly well in terms of sensitivity measure which is most affected with the correct segmentation of fine vessels and vessel-boundary pixels present in low-contrast regions.},
keywords = {biomedical optical imaging;blood vessels;diseases;edge detection;eye;feature extraction;filtering theory;image colour analysis;image enhancement;image filtering;image segmentation;image texture;matched filters;medical image processing;U-net based multilevel texture suppression;low contrast regions;retinal blood vessels;retinal diseases;precise segmentation;complete vessel-map;fundus images;vessel pixels;vessel-endings;neighboring vessel-like texture;retinal vessel segmentation algorithm;background vessel-like texture;vessel edges;maximum response image;enhanced vessels;suppressed texture;nonvessel pixels;segmentation method;correct segmentation;fine vessels;vessel-boundary pixels;low-contrast regions;Image segmentation;Matched filters;Sensitivity;Simulation;Signal processing algorithms;Signal processing;Retinal vessels;segmentation;texture suppression;multiscale;U-net},
doi = {10.23919/Eusipco47968.2020.9287387},
issn = {2076-1465},
month = {Aug},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2020/pdfs/0001304.pdf},
}Downloads: 0
{"_id":"8bk8bwEFtiWgYqRNz","bibbaseid":"upadhyay-agrawal-vashist-unetbasedmultileveltexturesuppressionforvesselsegmentationinlowcontrastregions-2020","authorIDs":[],"author_short":["Upadhyay, K.","Agrawal, M.","Vashist, P."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["K."],"propositions":[],"lastnames":["Upadhyay"],"suffixes":[]},{"firstnames":["M."],"propositions":[],"lastnames":["Agrawal"],"suffixes":[]},{"firstnames":["P."],"propositions":[],"lastnames":["Vashist"],"suffixes":[]}],"booktitle":"2020 28th European Signal Processing Conference (EUSIPCO)","title":"U-Net based Multi-level Texture Suppression for Vessel Segmentation in Low Contrast Regions","year":"2020","pages":"1304-1308","abstract":"Segmentation of retinal blood vessels is important for diagnosis of many retinal diseases. Precise segmentation of complete vessel-map is still a challenge in low contrast regions of fundus images. Vessel pixels belonging to these regions, such as, fine vessel-endings and boundaries of vessels, get merged in the neighboring vessel-like texture. This paper proposes a novel retinal vessel segmentation algorithm which handles the background vessel-like texture in a sophisticated manner without harming the vessel pixels. In this work, first we enhance all possible vessel-like features of fundus at different `levels' using 2-D Gabor wavelet and Gaussian matched filtering. At each `level', texture is suppressed using Local Laplacian filter while preserving the vessel edges. The resulting images are combined to produce a maximum response image with enhanced vessels of different thicknesses and suppressed texture. This handcrafted image is used to train the deep U-net model for further suppression of non-vessel pixels. Proposed segmentation method is tested on publicly available DRIVE and STARE databases. The algorithm has produced state-of-the-art results. It has performed outstandingly well in terms of sensitivity measure which is most affected with the correct segmentation of fine vessels and vessel-boundary pixels present in low-contrast regions.","keywords":"biomedical optical imaging;blood vessels;diseases;edge detection;eye;feature extraction;filtering theory;image colour analysis;image enhancement;image filtering;image segmentation;image texture;matched filters;medical image processing;U-net based multilevel texture suppression;low contrast regions;retinal blood vessels;retinal diseases;precise segmentation;complete vessel-map;fundus images;vessel pixels;vessel-endings;neighboring vessel-like texture;retinal vessel segmentation algorithm;background vessel-like texture;vessel edges;maximum response image;enhanced vessels;suppressed texture;nonvessel pixels;segmentation method;correct segmentation;fine vessels;vessel-boundary pixels;low-contrast regions;Image segmentation;Matched filters;Sensitivity;Simulation;Signal processing algorithms;Signal processing;Retinal vessels;segmentation;texture suppression;multiscale;U-net","doi":"10.23919/Eusipco47968.2020.9287387","issn":"2076-1465","month":"Aug","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2020/pdfs/0001304.pdf","bibtex":"@InProceedings{9287387,\n author = {K. Upadhyay and M. Agrawal and P. Vashist},\n booktitle = {2020 28th European Signal Processing Conference (EUSIPCO)},\n title = {U-Net based Multi-level Texture Suppression for Vessel Segmentation in Low Contrast Regions},\n year = {2020},\n pages = {1304-1308},\n abstract = {Segmentation of retinal blood vessels is important for diagnosis of many retinal diseases. Precise segmentation of complete vessel-map is still a challenge in low contrast regions of fundus images. Vessel pixels belonging to these regions, such as, fine vessel-endings and boundaries of vessels, get merged in the neighboring vessel-like texture. This paper proposes a novel retinal vessel segmentation algorithm which handles the background vessel-like texture in a sophisticated manner without harming the vessel pixels. In this work, first we enhance all possible vessel-like features of fundus at different `levels' using 2-D Gabor wavelet and Gaussian matched filtering. At each `level', texture is suppressed using Local Laplacian filter while preserving the vessel edges. The resulting images are combined to produce a maximum response image with enhanced vessels of different thicknesses and suppressed texture. This handcrafted image is used to train the deep U-net model for further suppression of non-vessel pixels. Proposed segmentation method is tested on publicly available DRIVE and STARE databases. The algorithm has produced state-of-the-art results. It has performed outstandingly well in terms of sensitivity measure which is most affected with the correct segmentation of fine vessels and vessel-boundary pixels present in low-contrast regions.},\n keywords = {biomedical optical imaging;blood vessels;diseases;edge detection;eye;feature extraction;filtering theory;image colour analysis;image enhancement;image filtering;image segmentation;image texture;matched filters;medical image processing;U-net based multilevel texture suppression;low contrast regions;retinal blood vessels;retinal diseases;precise segmentation;complete vessel-map;fundus images;vessel pixels;vessel-endings;neighboring vessel-like texture;retinal vessel segmentation algorithm;background vessel-like texture;vessel edges;maximum response image;enhanced vessels;suppressed texture;nonvessel pixels;segmentation method;correct segmentation;fine vessels;vessel-boundary pixels;low-contrast regions;Image segmentation;Matched filters;Sensitivity;Simulation;Signal processing algorithms;Signal processing;Retinal vessels;segmentation;texture suppression;multiscale;U-net},\n doi = {10.23919/Eusipco47968.2020.9287387},\n issn = {2076-1465},\n month = {Aug},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2020/pdfs/0001304.pdf},\n}\n\n","author_short":["Upadhyay, K.","Agrawal, M.","Vashist, P."],"key":"9287387","id":"9287387","bibbaseid":"upadhyay-agrawal-vashist-unetbasedmultileveltexturesuppressionforvesselsegmentationinlowcontrastregions-2020","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2020/pdfs/0001304.pdf"},"keyword":["biomedical optical imaging;blood vessels;diseases;edge detection;eye;feature extraction;filtering theory;image colour analysis;image enhancement;image filtering;image segmentation;image texture;matched filters;medical image processing;U-net based multilevel texture suppression;low contrast regions;retinal blood vessels;retinal diseases;precise segmentation;complete vessel-map;fundus images;vessel pixels;vessel-endings;neighboring vessel-like texture;retinal vessel segmentation algorithm;background vessel-like texture;vessel edges;maximum response image;enhanced vessels;suppressed texture;nonvessel pixels;segmentation method;correct segmentation;fine vessels;vessel-boundary pixels;low-contrast regions;Image segmentation;Matched filters;Sensitivity;Simulation;Signal processing algorithms;Signal processing;Retinal vessels;segmentation;texture suppression;multiscale;U-net"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2020url.bib","creationDate":"2021-02-13T19:41:51.295Z","downloads":0,"keywords":["biomedical optical imaging;blood vessels;diseases;edge detection;eye;feature extraction;filtering theory;image colour analysis;image enhancement;image filtering;image segmentation;image texture;matched filters;medical image processing;u-net based multilevel texture suppression;low contrast regions;retinal blood vessels;retinal diseases;precise segmentation;complete vessel-map;fundus images;vessel pixels;vessel-endings;neighboring vessel-like texture;retinal vessel segmentation algorithm;background vessel-like texture;vessel edges;maximum response image;enhanced vessels;suppressed texture;nonvessel pixels;segmentation method;correct segmentation;fine vessels;vessel-boundary pixels;low-contrast regions;image segmentation;matched filters;sensitivity;simulation;signal processing algorithms;signal processing;retinal vessels;segmentation;texture suppression;multiscale;u-net"],"search_terms":["net","based","multi","level","texture","suppression","vessel","segmentation","low","contrast","regions","upadhyay","agrawal","vashist"],"title":"U-Net based Multi-level Texture Suppression for Vessel Segmentation in Low Contrast Regions","year":2020,"dataSources":["wXzutN6o5hxayPKdC","NBHz6C7PWuqwYyaqa"]}