Search

This paper in proper resource allocation to deal with pandemic issue in the world COVID-19. This issue affected many countries in the World, the first case was identified in Wuhan, the capital of Hubei, China and since these diseases spread globally, has been recognised as a pandemic by WHO on March 11, 2020. The emergence spread of the disease in China compared to the world before it recognized as a pandemic.

Protection of critical loads from power quality issues is inadequate due to its negative impact on modem domestic and industrial equipment. The mitigation of power quality problems may be distributed by the employment of Custom power devices (CPD) at grid. "Dynamic voltage restorer" (DVR) is employed for the protection of vital loads against voltage connected power quality issue. DVR can detect and compensate for long-term drops in the applied AC voltage, thereby isolating the entire group from these power supply reliability issues. This paper is focused on comparative study between two different control schemes for DVR such as proportional integral controller (PI) and Artificial neural network (ANN). Both the control schemes are simulated using MATLAB/ SIMULINK and the results depict the comparison of these two control schemes.

The importance of agriculture is known to mankind at the start of early human civilization, whereby farming and use of domestic animals created food supplies to people which helped cities to develop. Due to immense growth of population and industrialization, the demand for fruits and vegetables is more than its supply. Considering the wastage factor of modern agriculture, early detection of diseased fruits and vegetables by image processing and its techniques can hugely benefit in preventing rotting and its spread to neighbouring crops. Generally, these diseased fruits and vegetables are plucked by manual inspection. The same work is done by the autonomous robot with pattern and machine learning-based detection and recognition of diseased fruits and vegetables which covers much greater distance in a short period of time. The objective is divided into three stages. Firstly, after some basic pre-processing steps, the local and global features of diseases in fruits and vegetables are extracted using well-known algorithms of pattern recognition in feature extraction phase. Then the techniques like supervised and unsupervised learning are used to categorize them effectively and stored the same in training database. Then secondly, validating is done using k-NN classification and k-NN regression which is a type of instance-based learning. Lastly, after identification of the diseased fruit or vegetable, the robot plucks the target infected object and this cycle continues until it reaches the end of the agricultural land. This is beneficial for people like farmers, gardeners, home makers who have little or no information about the diseased crops which are growing and this invention saves their time and health by increasing the yield of healthy fruits and vegetables. Instead of having an expert agronomist to find out and study the disease every time, the robot can identify it by placing certain algorithms for disease detection. This also reduces the cost for people who cannot invest a large amount money and time for such work.

Face recognition from a large video database involves more search time. This article proposes a symmetric uncertainty based search space reduction (SUSSR) methodology that facilitates faster face recognition in video, making it viable for real time surveillance and authentication applications. The proposed methodology employs symmetric uncertainty based feature subset selection to obtain significant features. Further, Fuzzy C-Means clustering is applied to restrict the search to nearest possible cluster, thus speeding up the recognition process. Kullback Leibler's divergence based similarity measure is employed to recognize the query face in video by matching the query frame with that of stored features in the database. The proposed search space reduction methodology is tested upon benchmark video face datasets namely FJU, YouTube celebrities and synthetic datasets namely MIT-Dataset-I and MIT-Dataset-II. Experimental results demonstrate the effectiveness of the proposed methodology with a 10 increase in recognition accuracy and 35 reduction in recognition time.

The blood glucose monitoring system was developed to monitor the diabetes, this system consist of sensor unit to get the blood glucose level and then the information unit consists of GPRS which is primarily used to exchange the values to the IOT, ANN and user. ANN is basically used to obtain the projected blood glucose level’s approximation.

Multi swarm Particle swarm optimization (MPSO) is a variant of Particle Swarm Optimization (PSO) where various sub-swarms involved instead of single swarm, thereby balancing the exploitation and exploration. The progression of gbest is achieved by passing the best fitness value obtained from the child swarm and further progression of gbest is achieved by using gbest of parent swarm. TSP is an NP hard problems and its objectives is to find the minimum distance for a given cities and the constraints is to visit all the cities exactly ones to reach the final destination. In this paper we proposed MPSO approach to solve combinatorial optimization problem using BVA techniques comprising (Region_bc(Rg), Replicate_bc(Rp) and Evade_bc(Ev)), the performance of the algorithm is evaluated in terms of computation time, optimal fitness, error rate, convergence rate, convergence diversity and average convergence diversity for particle 30, the proposed techniques outperformance well with minimum computation time for optimal fitness value and error rate. In future we plan to implement the proposed technique in cloud computing environment for job scheduling problems.

An automatic speech recognition system aims to enable communication between human and computer. Here a continuous Malayalam speech recognition system is developed and realized using Kaldi toolkit. For building the proposed system, MFCC features and its transformations such as LDA and MLLT are extracted from Malayalam speech sentences. Then, the system has been trained to estimate the HMM parameters for speech sound units. Performance of both monophone and triphone models using bi-gram language model is compared in terms of word error rate(WER). A substantial drop in WER is obtained for triphone models.