Ground truth free retinal vessel segmentation by learning from simple pixels
Abstract Retinal vessel segmentation is fundamental for the automatic retinal image analysis and ocular disease screening. This paper aims to learn a ground truth free feature aggregation strategy for the vessel segmentation. Five vesselness maps modelling the vessels'profile, appearance, and s...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-05-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12098 |
| _version_ | 1811344463563325440 |
|---|---|
| author | Beiji Zou Hongpu Fu Zailiang Chen Qing Liu |
| author_facet | Beiji Zou Hongpu Fu Zailiang Chen Qing Liu |
| author_sort | Beiji Zou |
| collection | DOAJ |
| description | Abstract Retinal vessel segmentation is fundamental for the automatic retinal image analysis and ocular disease screening. This paper aims to learn a ground truth free feature aggregation strategy for the vessel segmentation. Five vesselness maps modelling the vessels'profile, appearance, and shape are first generated. Together, the histogram of the local binary pattern and the green colour are extracted. In each vesselness map, the pixels with large vesselness values are regarded as simple positive samples. The pixels with small vesselness values are regarded as simple negative samples, and the pixels with mediocre values are treated as difficult pixels. The simple positive samples and simple negative samples near the difficult pixels consist of the training dataset while the rest vesselness maps together with the local binary pattern histogram, and green colour channel are used as the features to learn a strong classifier. Then, without leveraging any ground truth, multiple kernel boosting is used to combine four support vector machine kernels to learn a strong vessel model for each image. Applying this learnt model to the pixels with mediocre values in the single vesselness map, their label will be determined. Totally, five strong vessel models are learnt. Finally, pixels with the majority supports from the strong vessel models are labelled as vessel pixels. The proposed method achieves accuracy of 94.83%, sensitivity of 72.59%, and specificity of 98.11% on DRIVE dataset, and accuracy of 95.51%, sensitivity of 78.09%, and specificity of 97.56% on STARE. It outperforms the state‐of‐the‐art unsupervised methods and achieves comparable performances to the supervised methods. |
| first_indexed | 2024-04-13T19:47:43Z |
| format | Article |
| id | doaj.art-fa0f6b17ea1c4cdf8f90ddcf495f4e61 |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-04-13T19:47:43Z |
| publishDate | 2021-05-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-fa0f6b17ea1c4cdf8f90ddcf495f4e612022-12-22T02:32:40ZengWileyIET Image Processing1751-96591751-96672021-05-011561210122010.1049/ipr2.12098Ground truth free retinal vessel segmentation by learning from simple pixelsBeiji Zou0Hongpu Fu1Zailiang Chen2Qing Liu3School of Computer Science Central South University Changsha ChinaSchool of Computer Science Central South University Changsha ChinaSchool of Computer Science Central South University Changsha ChinaSchool of Computer Science Central South University Changsha ChinaAbstract Retinal vessel segmentation is fundamental for the automatic retinal image analysis and ocular disease screening. This paper aims to learn a ground truth free feature aggregation strategy for the vessel segmentation. Five vesselness maps modelling the vessels'profile, appearance, and shape are first generated. Together, the histogram of the local binary pattern and the green colour are extracted. In each vesselness map, the pixels with large vesselness values are regarded as simple positive samples. The pixels with small vesselness values are regarded as simple negative samples, and the pixels with mediocre values are treated as difficult pixels. The simple positive samples and simple negative samples near the difficult pixels consist of the training dataset while the rest vesselness maps together with the local binary pattern histogram, and green colour channel are used as the features to learn a strong classifier. Then, without leveraging any ground truth, multiple kernel boosting is used to combine four support vector machine kernels to learn a strong vessel model for each image. Applying this learnt model to the pixels with mediocre values in the single vesselness map, their label will be determined. Totally, five strong vessel models are learnt. Finally, pixels with the majority supports from the strong vessel models are labelled as vessel pixels. The proposed method achieves accuracy of 94.83%, sensitivity of 72.59%, and specificity of 98.11% on DRIVE dataset, and accuracy of 95.51%, sensitivity of 78.09%, and specificity of 97.56% on STARE. It outperforms the state‐of‐the‐art unsupervised methods and achieves comparable performances to the supervised methods.https://doi.org/10.1049/ipr2.12098Optical, image and video signal processingComputer vision and image processing techniquesOther topics in statisticsOptical and laser radiation (medical uses)Patient diagnostic methods and instrumentationBiology and medical computing |
| spellingShingle | Beiji Zou Hongpu Fu Zailiang Chen Qing Liu Ground truth free retinal vessel segmentation by learning from simple pixels IET Image Processing Optical, image and video signal processing Computer vision and image processing techniques Other topics in statistics Optical and laser radiation (medical uses) Patient diagnostic methods and instrumentation Biology and medical computing |
| title | Ground truth free retinal vessel segmentation by learning from simple pixels |
| title_full | Ground truth free retinal vessel segmentation by learning from simple pixels |
| title_fullStr | Ground truth free retinal vessel segmentation by learning from simple pixels |
| title_full_unstemmed | Ground truth free retinal vessel segmentation by learning from simple pixels |
| title_short | Ground truth free retinal vessel segmentation by learning from simple pixels |
| title_sort | ground truth free retinal vessel segmentation by learning from simple pixels |
| topic | Optical, image and video signal processing Computer vision and image processing techniques Other topics in statistics Optical and laser radiation (medical uses) Patient diagnostic methods and instrumentation Biology and medical computing |
| url | https://doi.org/10.1049/ipr2.12098 |
| work_keys_str_mv | AT beijizou groundtruthfreeretinalvesselsegmentationbylearningfromsimplepixels AT hongpufu groundtruthfreeretinalvesselsegmentationbylearningfromsimplepixels AT zailiangchen groundtruthfreeretinalvesselsegmentationbylearningfromsimplepixels AT qingliu groundtruthfreeretinalvesselsegmentationbylearningfromsimplepixels |