Complex Shock Filtering applied to Retinal Image Enhancement
01 Jan 2013-pp 900-903
TL;DR: A complex diffusion based shock filter is used for image smoothing and contrast enhancement that outperforms existing image enhancement methods and also acts as an effective adaptive contrast enhancement scheme.
Abstract: Due to several patient dependent and camera dependent factors which are difficult to control in a color retinal imaging setup, non-uniform luminosity and contrast variability is observed within and across images Most algorithms for lesion detection are highly dependent on the local image intensity and contrast and hence tend to suffer if there is significant variability Therefore, image preprocessing for illumination correction and contrast enhancement is a crucial step in any scheme for automated detection of diabetic retinopathy in retinal images Noise at the pixel level is also a major problem as it gets amplified in a contrast stretch operation This necessitates an image smoothing operation Here, we used a complex diffusion based shock filter for image smoothing and contrast enhancement Complex shock filter is a non-linear forward-backward diffusion based approach for image enhancement proposed by Gilboa et al Shock filtering the image results in removal of speckle noise, reduces JPEG compression artifacts and also acts as an effective adaptive contrast enhancement scheme The application of our method on the DIARETDB1 database, and a local database shows that it outperforms existing image enhancement methods
Citations
More filters
TL;DR: The results show that the proposed retinal fundus image enhancement method can directly enhance color images prominently and is different from some other fundu image enhancement methods.
Abstract: Retinal fundus image plays an important role in the diagnosis of retinal related diseases. The detailed information of the retinal fundus image such as small vessels, microaneurysms, and exudates may be in low contrast, and retinal image enhancement usually gives help to analyze diseases related to retinal fundus image. Current image enhancement methods may lead to artificial boundaries, abrupt changes in color levels, and the loss of image detail. In order to avoid these side effects, a new retinal fundus image enhancement method is proposed. First, the original retinal fundus image was processed by the normalized convolution algorithm with a domain transform to obtain an image with the basic information of the background. Then, the image with the basic information of the background was fused with the original retinal fundus image to obtain an enhanced fundus image. Lastly, the fused image was denoised by a two-stage denoising method including the fourth order PDEs and the relaxed median filter. The retinal image databases, including the DRIVE database, the STARE database, and the DIARETDB1 database, were used to evaluate image enhancement effects. The results show that the method can enhance the retinal fundus image prominently. And, different from some other fundus image enhancement methods, the proposed method can directly enhance color images.
59 citations
01 Jan 2019
TL;DR: This chapter reviews preprocessing, enhancement, and registration techniques for the modalities of fundus and tomographic imaging of the human eye.
Abstract: Preprocessing and enhancement is a prerequisite for a wide range of retinal image analysis methods. The goals of such tasks are to improve images and facilitate their subsequent analysis. Registration of retinal images enables the generation of images of higher definition retinal mosaics and facilitates the comparison of images from different examinations. The above processes contribute significantly to the screening and diagnosis of a wide range of diseases. This chapter reviews preprocessing, enhancement, and registration techniques for the modalities of fundus and tomographic imaging of the human eye.
13 citations
TL;DR: Experimental results demonstrate that this algorithm can enhance important details of retinal image data effectively, affording an opportunity for better medical interpretation and subsequent processing.
Abstract: As a common ocular complication for diabetic patients, diabetic retinopathy has become an important public health problem in the world. Early diagnosis and early treatment with the help of fundus imaging technology is an effective control method. In this paper, a robust inverse diffusion equation combining a self-similarity filtering is presented to detect and evaluate diabetic retinopathy using retinal image enhancement. A flux corrected transport technique is used to control diffusion flux adaptively, which eliminates overshoots inherent in the Laplacian operation. Feature preserving denoising by the self-similarity filtering ensures a robust enhancement of noisy and blurry retinal images. Experimental results demonstrate that this algorithm can enhance important details of retinal image data effectively, affording an opportunity for better medical interpretation and subsequent processing.
9 citations
Cites methods from "Complex Shock Filtering applied to ..."
...Many different methods have been put forth for retinal image denoising and enhancement [7, 9, 18, 19], such as the Gamma transformation [18], histogram equalization [20, 21], sharpening by the Laplacian operation [22], filtering methods in transformation fields [19], variational methods and partial differential equations (PDEs) [13, 23, 24]....
[...]
TL;DR: A novel contrast enhancement method to make the low-contrast retinal images more qualitative such that the detailed information will be clearer and the proposed method can directly enhance color images.
Abstract: Retinal fundus images play an important role in the diagnosis of Diabetic Retinopathy. The detailed information of retinal images like Retinal Vessels, Exudates and microaneurysms may be in low contrast and retinal image enhancement helps in the accurate diagnosis of retinal images related diseases. This paper proposes a novel contrast enhancement method to make the low-contrast retinal images more qualitative such that the detailed information will be clearer. For this purpose, the proposed mechanism considers the spatial mutual relationships between the gray-levels of image and makes the gray-levels in the output image not only linear to the gray-levels of input image but also related to the neighboring gray-levels. The proposed approach enhances the image both in global and local fashion. Spatial mutual entropy based contrast enhancement is accomplished for global contrast enhancement and Greedy contrast enhancement is accomplished for local contrast enhancement. Extensive simulations carried out over various low-contrast retinal images of High Resolution Fundus (HRF) image database shows the outstanding performance of proposed approach. And, different from some other fundus image enhancement methods, the proposed method can directly enhance color images.
3 citations
Cites methods from "Complex Shock Filtering applied to ..."
...[23] used a complex diffusionbased shock filter for retinal image smoothing and contrast enhancement....
[...]
TL;DR: Experimental results show that the proposed SCCE method improves the contrast of retinal images in a better manner than several existing methods including CLAHE and Normalized Convolution (NC).
Abstract: : Retinal images have a significant role in the diagnosis of Diabetic Retinopathy. However, the main objects of retinal images are Retinal Vessels, Exudates, and Microaneurysms which are of low contrast. Hence, this paper proposes a new contrast enhancement mechanism called as Spatial Collaborative Contrast Enhancement (SCCE). SECE utilizes the mutual relation and spatial spread of gray levels over the retinal image and enhances the contrast. SCCE allocates one rank for every pair of gray levels and ensures a perfect gap between consecutive gray levels that result in an increased contrast in the output retinal image. Several experiments are performed on two standard datasets namely HRF and DRIVE. Experimental results show that the proposed SCCE method improves the contrast of retinal images in a better manner than several existing methods including CLAHE and Normalized Convolution (NC). On an average, the SECE gained an improvement of 14% and 3% in SSIM from CLAHE and NC respectively.
2 citations
References
More filters
TL;DR: A new definition of scale-space is suggested, and a class of algorithms used to realize a diffusion process is introduced, chosen to vary spatially in such a way as to encourage intra Region smoothing rather than interregion smoothing.
Abstract: A new definition of scale-space is suggested, and a class of algorithms used to realize a diffusion process is introduced. The diffusion coefficient is chosen to vary spatially in such a way as to encourage intraregion smoothing rather than interregion smoothing. It is shown that the 'no new maxima should be generated at coarse scales' property of conventional scale space is preserved. As the region boundaries in the approach remain sharp, a high-quality edge detector which successfully exploits global information is obtained. Experimental results are shown on a number of images. Parallel hardware implementations are made feasible because the algorithm involves elementary, local operations replicated over the image. >
12,560 citations
TL;DR: In this article, shock filters for image enhancement are developed, which use new nonlinear time dependent partial differential equations and their discretizations, which satisfy a maximum principle and the total variation of the solution for any fixed fixed $t > 0$ is the same as that of the initial data.
Abstract: Shock filters for image enhancement are developed. The filters use new nonlinear time dependent partial differential equations and their discretizations. The evolution of the initial image $u_0 (x,y)$ as $t \to \infty $ into a steady state solution $u_\infty (x,y)$ through $u(x,y,t)$, $t > 0$, is the filtering process. The partial differential equations have solutions which satisfy a maximum principle. Moreover the total variation of the solution for any fixed $t > 0$ is the same as that of the initial data. The processed image is piecewise smooth, nonoscillatory, and the jumps occur across zeros of an elliptic operator (edge detector). The algorithm is relatively fast and easy to program.
850 citations
"Complex Shock Filtering applied to ..." refers background in this paper
...Shock filters like the one proposed by Osher and Rudin, [16], on the other hand, enhance edges but do not smooth the image....
[...]
TL;DR: In this study the optic disc, blood vessels, and fovea were accurately detected and the identification of the normal components of the retinal image will aid the future detection of diseases in these regions.
Abstract: Aim—To recognise automatically the main components of the fundus on digital colour images. Methods—The main features of a fundus retinal image were defined as the optic disc, fovea, and blood vessels. Methods are described for their automatic recognition and location. 112 retinal images were preprocessed via adaptive, local, contrast enhancement. The optic discs were located by identifying the area with the highest variation in intensity of adjacent pixels. Blood vessels were identified by means of a multilayer perceptron neural net, for which the inputs were derived from a principal component analysis (PCA) of the image and edge detection of the first component of PCA. The foveas were identified using matching correlation together with characteristics typical of a fovea—for example, darkest area in the neighbourhood of the optic disc. The main components of the image were identified by an experienced ophthalmologist for comparison with computerised methods. Results—The sensitivity and specificity of the recognition of each retinal main component was as follows:99.1% and 99.1% for the optic disc; 83.3% and 91.0% for blood vessels; 80.4% and 99.1% for the fovea. Conclusions—In this study the optic disc, blood vessels, and fovea were accurately detected. The identification of the normal components of the retinal image will aid the future detection of diseases in these regions. In diabetic retinopathy, for example,an image could be analysed for retinopathy with reference to sight threatening complications such as disc neovascularisation, vascular changes, or foveal exudation. (Br J Ophthalmol 1999;83:902‐910)
846 citations
"Complex Shock Filtering applied to ..." refers methods in this paper
...al, [5] transformed the original retinal image to an intensity-hue-saturation representation and then normalized the intensity channel without affecting the hue by applying a point-wise transformation to each pixel in the intensity image....
[...]
TL;DR: An automated method to locate the optic nerve in images of the ocular fundus using a novel algorithm the authors call fuzzy convergence to determine the origination of the blood vessel network is described.
Abstract: We describe an automated method to locate the optic nerve in images of the ocular fundus. Our method uses a novel algorithm we call fuzzy convergence to determine the origination of the blood vessel network. We evaluate our method using 31 images of healthy retinas and 50 images of diseased retinas, containing such diverse symptoms as tortuous vessels, choroidal neovascularization, and hemorrhages that completely obscure the actual nerve. On this difficult data set, our method achieved 89% correct detection. We also compare our method against three simpler methods, demonstrating the performance improvement. All our images and data are freely available for other researchers to use in evaluating related methods.
756 citations
TL;DR: Results of these experiments show that for this particular diagnostic task, there was no significant difference in the ability of the two methods to depict luminance contrast; thus, further evaluation of AHE using controlled clinical trials is indicated.
Abstract: Adaptive histogram equalization (AHE) and intensity windowing have been compared using psychophysical observer studies Experienced radiologists were shown clinical CT (computerized tomographic) images of the chest Into some of the images, appropriate artificial lesions were introduced; the physicians were then shown the images processed with both AHE and intensity windowing They were asked to assess the probability that a given image contained the artificial lesion, and their accuracy was measured The results of these experiments show that for this particular diagnostic task, there was no significant difference in the ability of the two methods to depict luminance contrast; thus, further evaluation of AHE using controlled clinical trials is indicated >
347 citations
"Complex Shock Filtering applied to ..." refers background in this paper
...A locally adaptive version of GHE, contrast limited adaptive histogram equalization (CLAHE), has less noise amplification but like GHE has the disadvantage that it does not correct for nonuniform illumination [3]....
[...]