Kongu Engineering College

About: Kongu Engineering College is a based out in . It is known for research contribution in the topics: Computer science & Cluster analysis. The organization has 2001 authors who have published 1978 publications receiving 16923 citations.

Detection of retinal area from scanning laser ophthalmoscope images (SLO) using deep neural network

Abstract: Earlier detection and treatment of the retinal disease are crucial to avoid preventable vision loss. Scanning laser ophthalmoscopes (SLOs) can be used for early detection of retinal diseases and the SLO can image a large part of the retinal image to diagnose it in a better way. Eyelashes and eyelids are also imaged with the retinal image during the image process used to exclude the artefacts of the retinal image. The proposed SLO approach automatically extracts both the true retinal area and artefacts of the image based on image processing and machine learning approach. Superpixel and Deep Neural Network (DNN) are used to reduce the complexity of image processing tasks, the result is being provided with a primitive image pattern. The framework performs the calculations of textural and structural-based information of features and this approach results in effective analysis of retinal area and the artefacts.

Physicochemical and Thermal Properties of Cellulosic Fiber Extracted from the Bark of Albizia Saman

Abstract: The environmental threats posed by manmade composites during their disposal and depletion of fossil fuel resources have necessitated the application of ecofriendly materials for composite making. I...

Adaptive Neuro-Fuzzy Model for Grid-Connected Photovoltaic System

Abstract: This paper proposed an adaptive neuro-fuzzy inference system (ANFIS) model to multilevel inverter for grid-connected photovoltaic (PV) system. The purpose of the proposed controller is to avoid the requirement of any optimal PWM (pulse width-modulated) switching-angle generator and proportional–integral controller. The proposed method strictly prevents the variations present in the output voltage of the cascaded H-bridge multilevel inverter. Here, the ANFIS models have the inputs which are the grid voltage and the difference voltage, and the output target is the control voltage. By means of these parameters, the ANFIS makes the rules and can be tuned perfectly. During the testing time, the ANFIS provides the control voltage according to the different inputs. Then, the ANFIS-based algorithm for multilevel inverter for grid-connected PV system is implemented in the MATLAB/simulink platform, and the effectiveness of the proposed control technique is analyzed by comparing the model’s performances with the neural network, fuzzy logic control, etc.

Throughput stability and flow fairness enhancement of TCP traffic in multi-hop wireless networks

Abstract: In the multi-hop wireless network, transmission control protocol (TCP) throughput stability and flow fairness performances are worsened due to slower flow convergence in the loss recovery phase and flat-rate reduction during the congestion control process. In this article, a combination of network-assisted and window utilization based congestion control approach, known as feedback assisted improved recovery + (FAIR+), is proposed to overcome TCP's limitations under multi-hop wireless networks. The proposed FAIR+ algorithm initiates the congestion control process based on the queue level notification of the relay node and trims down the sending rate based on TCP flow's utilization level. In the congestion recovery phase, the FAIR+ algorithm implements a newer window increment pattern that achieves a faster convergence rate than the existing RFC 6582 implementation. The throughput stability of the FAIR+ algorithm is validated using the duality model and multi-hop wireless simulation. The simulation results convincingly prove that the FAIR+ attains significant improvement in throughput, flow fairness, and end-to-end latency performance over the existing TCP variants (OQS, NRT, and Westwood).