Automated retinal blood vessels segmentation based on simplified PCNN and fast 2D-Otsu algorithm

Abstract

According to the characteristics of dynamic firing in pulse coupled neural network (PCNN) and regional configuration in retinal blood vessel network, a new method combined with simplified PCNN and fast 2D-Otsu algorithm was proposed for automated retinal blood vessels segmentation. Firstly, 2D Gaussian matched filter was used to enhance the retinal images and simplified PCNN was employed to segment the blood vessels by firing neighborhood neurons. Then, fast 2D-Otsu algorithm was introduced to search the best segmentation results and iteration times with less computation time. Finally, the whole vessel network was obtained via analyzing the regional connectivity. Experiments implemented on the public Hoover database indicate that this new method gets a 0.803 5 true positive rate and a 0.028 0 false positive rate on an average. According to the test results, compared with Hoover algorithm and method of PCNN and 1D-Otsu, the proposed method shows much better performance.

Diabetic retinopathy through retinal image analysis: A review

“…The kernel is designed to model some feature in the image at some unknown position and orientation, and the matched filter response (MFR) indicates the presence of the feature. Various approaches [71][72][73] are proposed in earlier based on matched filtering. Al-Rawiet.al., [72] improved Chaudhuriet.al.,'s [71] matched filter (MF) by using an exhaustive search optimization procedure to find the best parameters for matched filter size, the standard deviation and threshold value.…”

“…The improved MF outperforms Chaudhuri's classical parameter matched filter. Yao and Chen [73] uses a 2-D Gaussian matched filter for retinal vessel enhancement and then a simplified pulse coupled neural network is employed to segment the blood vessels by firing neighborhood neurons. Next, a fast 2-D-Otsu algorithm is used to search the best segmentation results.…”

Estimating Leaf Area Density of Individual Trees Using the Point Cloud Segmentation of Terrestrial LiDAR Data and a Voxel-Based Model

“…The leaf point cloud is determined using the magnitude of ∆ n (p, r) as the threshold. The Otsu algorithm was applied to estimate the threshold [32]. The normal difference results of the point cloud data were viewed as the grey values of images.…”

Methods for digital analysis of human vascular system. Literature review

“…Изменение фона изо-бражения и присутствие различных патологических артефактов также увеличивает процент ложных сраба-тываний фильтра, так как патологии могут иметь па-раметры, схожие с параметрами сосудов. Использова-ние согласованной фильтрации является оправданной в комбинации с другими методами обработки изобра-жений [49,[77][78][79][80][81].…”

“…Изменение фона изо-бражения и присутствие различных патологических артефактов также увеличивает процент ложных сраба-тываний фильтра, так как патологии могут иметь па-раметры, схожие с параметрами сосудов. Использова-ние согласованной фильтрации является оправданной в комбинации с другими методами обработки изобра-жений [49,[77][78][79][80][81].В [26, 82] авторы предложили использовать для сегментации сосудов двумерное ядро фильтра на ос-нове функции Гаусса или её производной. Профиль фильтра проектируется таким образом, чтобы наи-лучшим образом соответствовать профилю сосуда, который обычно имеет форму функции Гаусса или её производной.…”