Sri Ramakrishna Engineering College

About: Sri Ramakrishna Engineering College is a based out in . It is known for research contribution in the topics: Computer science & Control theory. The organization has 1030 authors who have published 843 publications receiving 3822 citations.

Transfer Learning Approach for Diabetic Retinopathy Detection using Residual Network

Abstract: Diabetic retinopathy is a disease caused due to Diabetes Mellitus in the diabetic patients. It is the major cause of blindness for several countries. The longer the diabetic retinopathy affected the high chance that the person can go blind. 90% of cases reported that the cause of blindness due to the diabetic retinopathy occur due to the lack of proper treatment and monitoring the retinopathy before severe stage [2]. The doctors can't find a cure to the severe stage of diabetic retinopathy but can be detected in early stage and prevent it. Hence automated computer diagnosis will assist the doctors in finding the Diabetic Retinopathy at early stage with less cost and time. Deep learning is playing a major role in health informatics. Residual Network, a deep learning approach is used to detect the DR and its stages. With the transfer learning approach, Residual Network model detects the stages of diabetic retinopathy with good accuracy.

In situ growth of hexagonal-shaped α-Fe 2 O 3 nanostructures over few layered graphene by hydrothermal method and their electrochemical performance

Abstract: A simple in situ hydrothermal method was developed to anchor α-Fe2O3 (hematite) on reduced graphene oxide (rGO) surfaces The impact of functional groups of graphene oxide (GO) and the coordination between metal and carbonyl group was studied by Fourier transform infrared spectroscopy The crystallographic structure of the hybrid composite material was examined by X-ray diffraction (XRD) analysis The microscopic images indicate the decoration of Fe2O3 nanostructures with an average size of 35–40 nm were encapsulated in rGO sheets The XPS and Raman spectroscopy results further corroborate that α-Fe2O3 is successfully formed on the rGO matrix, which is in good accordance with XRD results The presence of α-Fe2O3 nanostructures between the graphene layers avoids the restacking efficiently and increases the surface area accessibility α-Fe2O3/rGO nanocomposites modified electrode exhibits 894 F g−1 specific capacitance at a current density of 05 A g−1 in 1 M H2SO4 electrolyte Furthermore, these trouble-free and cost effectively-prepared hierarchical materials showed high rate capability and excellent cycle stability

ANN-Based Controllers for Improved Performance of BLDC Motor Drives

Abstract: This paper discusses the development and performance analysis of ANN-based reference model controller and ANN-based self-tuned PID controller for BLDC motor drives As the BLDC motor drives are nonlinear due to its parameter and load variations, there is a need to develop ANN-based controllers to overcome the problems arising due to nonlinearity in BLDC motor drives In this paper, ANN-based self-tuned PID controller is developed for speed control of BLDC motor drives and its performance is compared with the standard ANN-based reference model-controlled BLDC motor drives The unique feature of ANN-based self-tuned PID controller is that it can dynamically change the PID controller gains to provide optimum performance under changing dynamics of BLDC motor drive Experimental results show that ANN-based self-tuned PID-controlled BLDC motor drives can effectively deal with speed tracking, load variations, and parameter variations

Palm pattern recognition using scale invariant feature transform

Abstract: In this research paper, we propose an efficient scale invariant feature transform (SIFT) for palm pattern recognition. A fingerprint recognition which is efficient for individual authentication based on fingerprint pattern. This method leads to fraudulent because it could be extracted easily from individuals. The SIFT method based on feature detection overcomes the above problem and is a combination of fast key point detector and visual descriptor. Using SIFT method contactless palm pattern images can be acquired, matched, recognised, authenticated and their matching performance are simulated using OpenCV. The experimental results show that SIFT method provides significantly fast and improved performance than the conventional methods like oriented FAST and rotated BRIEF (ORB).