Unsupervised Method for Retinal Vessel Segmentation Based on Gabor Wavelet and Multiscale Line Detector

Shah, Syed Ayaz Ali; Shahzad, Aamir; Khan, Muhammad Amir; Lu, Cheng-Kai; Tang, Tong Boon

doi:10.1109/access.2019.2954314

Cited by 38 publications

(19 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the STARE dataset, it can be found from the visualization results in Fig 17 that Shah et al [16] proposed a pre-processing method using Gabor wavelets to enhance the green channel of the image, but its metrics for vessel segmentation results on the STARE dataset were not the best when compared with other unsupervised methods. However, observing its visualization results reveals that the method has better segmentation results for vessel ends than the method of Miao et al [12] Secondly, a comparison of the segmentation results of Ground truth and Miao et al shows that Shah et al's method has some unsegmented trunk vessels, which may be the reason for its poor segmentation accuracy.…”

Section: Retinal Segmentation Results Of Different Methods On the Stare Datasetmentioning

confidence: 99%

“…These methods are also collectively referred to as unsupervised learning methods. For example, by using B-COSFIRE filters [15], Gabor filters [16], and Gaussian filters [17] for retinal vessel segmentation, these methods aim to eliminate undesired intensity variations in images and suppress background structure and noise. However, because of the simplicity of vascular feature encoding methods based on filters and other methods, and their lack of effective supervision information will lead to the extraction of coarse vascular information and poor final image segmentation, which cannot meet the needs of clinical applications.…”

Section: Plos Onementioning

confidence: 99%

“…The hardware environment required for the experiments of this method is less demanding, but its use of two B-COSFIRE converter response summation makes the choice of experimental parameters extremely demanding[15]. Shah et al completed using an unoptimized Matlab script with an average time of less than 20 seconds per image[16]. Tian et al used a 1080ti graphics card for training and testing with the software environment PyTorch.…”

mentioning

confidence: 99%

See 2 more Smart Citations

RFARN: Retinal vessel segmentation based on reverse fusion attention residual network

et al. 2021

View full text Add to dashboard Cite

Accurate segmentation of retinal vessels is critical to the mechanism, diagnosis, and treatment of many ocular pathologies. Due to the poor contrast and inhomogeneous background of fundus imaging and the complex structure of retinal fundus images, this makes accurate segmentation of blood vessels from retinal images still challenging. In this paper, we propose an effective framework for retinal vascular segmentation, which is innovative mainly in the retinal image pre-processing stage and segmentation stage. First, we perform image enhancement on three publicly available fundus datasets based on the multiscale retinex with color restoration (MSRCR) method, which effectively suppresses noise and highlights the vessel structure creating a good basis for the segmentation phase. The processed fundus images are then fed into an effective Reverse Fusion Attention Residual Network (RFARN) for training to achieve more accurate retinal vessel segmentation. In the RFARN, we use Reverse Channel Attention Module (RCAM) and Reverse Spatial Attention Module (RSAM) to highlight the shallow details of the channel and spatial dimensions. And RCAM and RSAM are used to achieve effective fusion of deep local features with shallow global features to ensure the continuity and integrity of the segmented vessels. In the experimental results for the DRIVE, STARE and CHASE datasets, the evaluation metrics were 0.9712, 0.9822 and 0.9780 for accuracy (Acc), 0.8788, 0.8874 and 0.8352 for sensitivity (Se), 0.9803, 0.9891 and 0.9890 for specificity (Sp), area under the ROC curve(AUC) was 0.9910, 0.9952 and 0.9904, and the F1-Score was 0.8453, 0.8707 and 0.8185. In comparison with existing retinal image segmentation methods, e.g. UNet, R2UNet, DUNet, HAnet, Sine-Net, FANet, etc., our method in three fundus datasets achieved better vessel segmentation performance and results.

show abstract

Section: Retinal Segmentation Results Of Different Methods On the Stare Datasetmentioning

confidence: 99%

Section: Plos Onementioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

RFARN: Retinal vessel segmentation based on reverse fusion attention residual network

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Talking about MCC, our method outperforms [46] and [23] who achieved best specificity and third-best accuracy respectively. The MCC of our method is in second-place close behind [25].…”

Section: B Comparison With State-of-the-artmentioning

confidence: 85%

“…The diameter of the blood vessels is bigger at the origin (Optic Disc) and slowly decreases outward. For extracting the size, the width, and the orientations of the retinal vessels in retinal images, multi-scale methods are better options as these methods can analyze the shape and intensity of the blood vessels at various scales [24], [25]. In [25], the authors used Gabor wavelet and multi-scale line detector for retinal vessel segmentation.…”

Section: Related Workmentioning

confidence: 99%

An Unsupervised Retinal Vessel Segmentation Using Hessian and Intensity Based Approach

2020

View full text Add to dashboard Cite

The structure of blood vessels play a crucial role in diagnoses of the various vision threatening diseases including Glaucoma and Diabetic Retinopathy (DR). The correct segmentation of retinal blood vessels is a crucial step in the study of retinal fundus images. We proposed a simple unsupervised approach by using a combination of Hessian based approach and intensity transformation approach. We have applied CLAHE for enhancing the contrast of the retinal fundus images. An enhanced version of PSO algorithm is applied for contextual region tuning of CLAHE. Morphological filter and Wiener filter are used to denoise the enhanced image. The eigenvalues are obtained from the Hessian matrix at two different scales to extract thick and thin vessel enhanced images separately. The intensity transformation approach is separately applied to the enhanced image to maximize the vessel details. Global Otsu thresholding is applied on intensity transformed image and thick vessel enhanced image whereas ISODATA local thresholding is applied on thin vessel enhanced image. Finally, a simple post-processing step based on the region parameters such as area, eccentricity, and solidity is used. The region parameters are obtained for each connected component in input binary images. The threshold values of region parameters are empirically investigated and applied to each of the three binary images to remove the non-vessel components. The thresholded images are combined by applying logical OR operator, which resulted in the final segmented binary image. We assessed our developed framework on the open-access CHASE_DB1 and DRIVE datasets, achieving a sensitivity of 0.7776 and 0.7851, and an accuracy of 0.9505 and 0.9559 respectively. These results outperform several state-of-the-art unsupervised methods. The reduced computational complexity and significantly improved evaluation metrics advocates for its use in the automated diagnostic systems for retinal image analysis.

show abstract

Detection of diabetic retinopathy and related retinal disorders using fundus images based on deep learning and image processing techniques: A comprehensive review

Lalithadevi

Krishnaveni²

2022

Concurrency and Computation

View full text Add to dashboard Cite

Diabetes mellitus is a chronic disorder disease in which a person's body fails to adhere insulin produced by their pancreas or unable to segregate enough insulin due to harmonic imbalance. Diabetic people are suffering from eye disorders like diabetic retinopathy (DR), glaucoma and various diseases such as neuropathy, nephropathy, cardiomyopathy over long intervals. One of the most prevalent diabetic consequence is DR. Detecting the morphological variations in retina is difficult and requires an effective automated detection system. DR can be predicted in earlier stage using tremendous development of deep learning models and image processing techniques. Recently, many research articles have been published in DR diagnosis system. This article shows a comprehensive review of automated diagnostic methods for DR detection and other related eye disorders from several points: Causes for DR, publicly available datasets, image preprocessing, segmentation of various DR lesions, feature optimization, various deep learning models, and open research challenges. The study offers a thorough overview of DR detection techniques, which delivers valuable information for researchers, medical professionals, and DR affected patients.

show abstract

Unsupervised Method for Retinal Vessel Segmentation Based on Gabor Wavelet and Multiscale Line Detector

Cited by 38 publications

References 29 publications

RFARN: Retinal vessel segmentation based on reverse fusion attention residual network

RFARN: Retinal vessel segmentation based on reverse fusion attention residual network

An Unsupervised Retinal Vessel Segmentation Using Hessian and Intensity Based Approach

Detection of diabetic retinopathy and related retinal disorders using fundus images based on deep learning and image processing techniques: A comprehensive review

Contact Info

Product

Resources

About