Search

In ophthalmological field, automated segmentation of vasculatures is an arduous job for the analysis and monitoring of eye threatening diseases like hypertension, diabetic retinopathy, and glaucoma. This paper presents an unsupervised technique for the extraction of vascular structures from ophthalmoscope images. At first from the ophthalmoscope images, extraction of green channel takes place and then the vessels are enhanced by the matched filter technique. Then gray level co-occurence matrix is used to extract the textural information of the image. Finally, the post-processing method is implemented to get the final segmented image. Evaluations are attended on the newly freely accessible IOSTAR dataset. The average accuracy of this proposed method implemented on the new dataset is 0.943.

A very crucial question in sparse approximation-based image compression is the choice of the dictionary. An effective dictionary can lead to excellent representation of an image with least number of dictionary atoms and lead to excellent image compression. The performance of block-based intra-image compression scheme depends on how well the residuals obtained from intra-prediction are encoded. Transform-based intra-compression method uses fixed DCT dictionary for encoding of prediction residuals. However, DCT dictionary is not suited for efficient encoding of prediction residuals with complex and non-periodic characteristics. This paper presents an algorithm to design an over-complete residual dictionary suitable for encoding prediction residuals. Simulation results demonstrate that the proposed residual dictionary yields superior performance as compared to other standard methods.

An image is called visually rich when its brightness and contrast are properly adjusted. But it was reported in different applications that the contrast of an image was affected by the external interferences. Due to the above phenomena, the image quality degrades. To improve the quality of the image, contrast enhancement method has been used. This technique adjusts the intensity distribution of the low-quality image to solve the problem. But the old techniques were not suitable for all kinds of images. To overcome the limitations, this paper presents a novel variational mode decomposition (VMD)-based enhancement technique associated with singular value decomposition (SVD). Initially, the reference images and the processed image were decomposed into various modes using variational mode decomposition (VMD). Then in the second step, the maximum value of the singular matrix was calculated for the selected mode of each image. An image-dependent correction factor was calculated using the singular value matrix (SVM). Finally, the image is reconstructed by simply adding the β-corrected mode image with the unprocessed modes of the reference image. To avoid the over-enhancement, a new weighted factor was applied to the original as well as to the enhanced image. To validate the algorithm, the proposed method was tested on different types of publicly available low contrast images. The experimental result shows that the proposed algorithm gives significant enhancement in terms of contrast over other state-of-the-art methods.

This paper describes about the Load Frequency Control (LFC) of an interconnected linear/nonlinear power system having two degree freedom of PID controller (2DOF-PID). The reheat thermal applied on a two-area system is considered for frequency control analysis by applying an abrupt load perturbation of 0.1 pu in area 1. Two systems are scrutinized for studying the load frequency control namely a linear system and a nonlinear system. The nonlinearities of the system are included by implementing governor with dead band (GDB) in each of the areas. The parameters of 2DOF-PID controllers are tuned on basis of Cuckoo Search Algorithm (CSA) considering time domain based fitness function, i.e., ITAE (Integral Time Absolute Error). For both types of system the dominance of 2DOF-PID controller is proved over conventional PID controller in terms of responses like minimum undershoot, settling time and peak overshoot. The supremacy of the recommended methodology is also established over some pre-published results.