Water flow based vessel detection in retinal images

Liu, Xin U.; Nixon, Mark S.

doi:10.1049/cp:20060554

Cited by 6 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the salt and pepper noise-corrupted images are illustrated in Figure 20. According to a study on the effects of noise, the suggested intensity when using Gaussian and speckle noise should be 0 mean and 10 −3 standard deviation [63][64], and salt and pepper noise should affect 5% of the retinal images [65]. The corrupted images are retrained and tested, and the results are detailed in Table IV To the best of the authors' knowledge, the results of available works are summarized in Table V. The accuracy is higher than 89% (two-class) and 73.8% (four-class) in all available works.…”

Section: Performance Under Noise Conditionsmentioning

confidence: 99%

Hierarchical method for cataract grading based on retinal images using improved Haar wavelet

Cao

Zhang

et al. 2020

Information Fusion

View full text Add to dashboard Cite

Cataracts, which are lenticular opacities that may occur at different lens locations, are the leading cause of visual impairment worldwide. Accurate and timely diagnosis can improve the quality of life of cataract patients. In this paper, a feature extraction-based method for grading cataract severity using retinal images is proposed. To obtain more appropriate features for the automatic grading, the Haar wavelet is improved according to the characteristics of retinal images. Retinal images of non-cataract, as well as mild, moderate, and severe cataracts, are automatically recognized using the improved Haar wavelet. A hierarchical strategy is used to transform the four-class classification problem into three adjacent two-class classification problems. Three sets of two-class classifiers based on a neural network are trained individually and integrated together to establish a complete classification system. The accuracies of the two-class classification (cataract and non-cataract) and four-class classification are 94.83% and 85.98%, respectively.The performance analysis demonstrates that the improved Haar wavelet feature achieves higher accuracy than the original Haar wavelet feature, and the fusion of three sets of two-class classifiers is superior to a simple four-class classifier. The discussion indicates that the retinal image-based method offers significant potential for cataract detection.

show abstract

Section: Performance Under Noise Conditionsmentioning

confidence: 99%

Hierarchical method for cataract grading based on retinal images using improved Haar wavelet

Cao

Zhang

et al. 2020

Information Fusion

View full text Add to dashboard Cite

show abstract

“…The noise models typically found in biomedical images are Gaussian, salt and Pepper and speckle multiplicative noise. Salt and pepper noise affecting the retina image [22], the pixel can have value 0 or 255, and such pixels are considered as noisy pixels. Usually image such as natural, digital and medical images use three channels that are red, green and blue.…”

Section: Extended Median Filter Algorithmmentioning

confidence: 99%

Restoration of retina images using extended median filter algorithm

Subbuthai

Muruganand

2015

2015 2nd International Conference on Signal Processing and Integrated Networks (SPIN)

View full text Add to dashboard Cite

Diabetic Retinopathy is one of the complicated diseases and it is caused by the changes in the blood vessels of the retina. Extraction of retina image through Fundus camera sometimes produces poor contrast and noises; detection of blood vessels in the retina is very complicated. So preprocessing is needed, in this paper a novel algorithm is implemented to remove the noisy pixel in the retina image. The proposed algorithm is Extended Median Filter and it is applied to green channel of the retina, because green channel vessels are brighter than the background. Proposed extended median filter is compared with the existing standard median filter by performance metrics such as Peak Signal-to-Noise Ratio, Mean Square Error and Root Mean Square Error. Experimental results show that the proposed Extended Median Filter algorithm gives better result than the existing standard median filter in terms of noise suppression and detail preservation.

show abstract

“…If an image pixel is flooded by water, the statistics of the two areas (water and non-water) will change and are given by equation (9). The derivation has been shown in [13,14]. Edge-based forces provide a good localization of the contour near the real boundaries but have limited capture range whilst region-based forces have a large basin of attraction and relatively low detection accuracy.…”

Section: Image Forcesmentioning

confidence: 99%

“…Water is chosen because features like fluidity and surface tension can lead to topological adaptability and geometrical flexibility, as well as contour smoothness. We completely redefined the basis of our previous water-flow based segmentation approaches [12,13] by adopting the force filed theory which has been used in feature extractions [14]. The method shows decent segmentation performance in quantitative and qualitative assessments.…”

Section: Introductionmentioning

confidence: 99%

Image and Volume Segmentation by Water Flow

Liu

Nixon

Advances in Visual Computing

View full text Add to dashboard Cite

Abstract.A general framework for image segmentation is presented in this paper, based on the paradigm of water flow. The major water flow attributes like water pressure, surface tension and capillary force are defined in the context of force field generation and make the model adaptable to topological and geometrical changes. A flow-stopping image functional combining edge-and region-based forces is introduced to produce capability for both range and accuracy. The method is assessed qualitatively and quantitatively on synthetic and natural images. It is shown that the new approach can segment objects with complex shapes or weak-contrasted boundaries, and has good immunity to noise. The operator is also extended to 3-D, and is successfully applied to medical volume segmentation.

show abstract

Water flow based vessel detection in retinal images

Cited by 6 publications

References 14 publications

Hierarchical method for cataract grading based on retinal images using improved Haar wavelet

Hierarchical method for cataract grading based on retinal images using improved Haar wavelet

Restoration of retina images using extended median filter algorithm

Image and Volume Segmentation by Water Flow

Contact Info

Product

Resources

About