Skip to main content

Über dieses Buch

This book presents selected papers from the International Conference on Emerging Research in Computing, Information, Communication and Applications, ERCICA 2018. The conference provided an interdisciplinary forum for researchers, professional engineers and scientists, educators, and technologists to discuss, debate and promote research and technology in the emerging areas of computing, information, communication and their applications. The book discusses these research areas, providing a valuable resource for researchers and practicing engineers alike.



Use of Blockchain for Smart T-Shirt Design Ownership

This study aims to provide a reliable solution for the T-shirt designer to gain ownership of his design using the blockchain technology. As blockchain is a decentralized technology, it will provide authentication for the ownership of the artwork among various non-trusting members. In this research paper, we have explained how to resolve this issue for the designer as well as the customer. Using this method, a novel distributed application can be created.

Ashley Alexsius D’Souza, Okstynn Rodrigues

A Feasibility Study and Simulation of 450 kW Grid Connected Solar PV System at NMIT, Bangalore

The consumption of energy can be reduced by efficiently using the available resources and effectively energy bill is reduced by considering photovoltaic system, which is most promising nowadays. In this paper, a feasibility study and simulation model on MATLAB/SIMULINK of 450-kW grid-connected solar PV system is considered for NMIT campus. The energy consumption at the campus is studied, and the number of billed units in kWh is considered for the last two years. The modelling of PV array, their integration with MPPT in SIMULINK environment are described. The deployment of available energy resources along with the incoming PV system is studied for effective usage of electricity. The simulation results are shown, the performance of the incoming PV system and its feasibility is described as obtained.

B. Smitha, N. Samanvita, H. M. Ravikumar

Prediction of a Dam’s Hazard Level

A Case Study from South Africa

South Africa has a vast infrastructure of dams. Since the country receives very little rainfall, these dams assume prime importance in storing water and sustaining agriculture, industry, household, etc. Thus prediction of their multiple hazard levels (in this case, three) is of prime importance. In addition, South Africa lacks skilled personnels to classify these dam’s hazards. Under such a framework, this work is an application of single and ensemble decision trees in a multi-class supervised learning framework to predict the hazard level of a dam. The result obtained is highly promising and at is above 94 $$\%$$ . With the implementation of the algorithm, we expect to address the problem of paucity of skilled personnels.

Urna Kundu, Srabanti Ghosh, Satyakama Paul

Elemental Racing

This paper presents a 3D motion sensor racing game for Android/iOS devices called Elemental Racing, in which players can race a car against opponent racers connected through local area network. There are many popular racing games for mobile phones like Asphalt 8, Real Racing 3, and NFS et al. These have dominated the Android racing game market with their extensive graphics and detail, but they are lacking when it comes to newer gameplay mechanisms, the objective of this game is to provide a fun yet intelligent experience to the player. To guarantee these objectives, we have consolidated multiple strategic upgrading mechanisms in the game which makes use of the concepts of elemental powers like Fire, Water, Wind, Lighting, and Earth.

Lingala Siva Karthik Reddy, Karthik Koka, Amiya Kumar Dash, Manjusha Pandey

A Survey on Existing Convolutional Neural Networks and Waste Management Techniques and an Approach to Solve Waste Classification Problem Using Neural Networks

In India, waste management has become one of the major crises with population explosion, coupled with improved lifestyle of people, results in increased generation of solid wastes in urban as well as rural areas of the country. It is well known that waste management policies, as they exist now, are not sustainable in the long term. Thus, waste management is undergoing drastic change to offer more options that are more sustainable. Most of the landfills are becoming full of waste in which most part is reusable and leading to spreading of disease damaging human body and leading to unpleasant air and only 5% of whole waste is actual waste. The government of Karnataka mandated system of 2 BIN 1 BAG to be adapted at every households in Bangalore, and 2 BIN 1 BAG is a color-coded system consisting of green bin which holds garden waste, and the wastes that are compostable, reject waste can be thrown in red bin, and finally a big category called as reusable bag which holds recyclable waste. Segregation of waste at source is best solution and should be done properly. Types of waste need to be remembered by members of home in order to put them to proper bins, and this may lead to human error. So our solution can answer this in good way, what if you just click picture of waste material and application says to which category it belongs. A convolutional neural network is trained with images of waste materials, and model can be inferred by giving waste-material image as input and get the perfect category of waste material in a second. This helps society in dealing with prime problem of segregating waste materials at source.

Tejashwini Hiremath, S. Rajarajeswari

L1-Regulated Feature Selection in Microarray Cancer Data and Classification Using Random Forest Tree

Microarray cancer data are characterized by high dimensionality, small sample size, noisy data, and an imbalanced number of samples among classes. To alleviate this challenge, several machine learning-oriented techniques are proposed by authors from several disciplines such as computer science, computational biology, statistics, and pattern recognition. In this work, we propose L1-regulated feature selection method and classification of microarray cancer data using Random Forest tree classifier. The experiment is conducted on eight standard microarray cancer datasets. We explore the learning curve of the model, which indicates the learning capability of the classifier from a different portion of the training samples. To overcome the overfitting problem, feature scaling is carried out before the actual training takes place and the learning curve is explored using fivefold cross-validation method during the actual training time. Comparative analysis is carried out with state-of-the-art work, and the proposed method outperforms many of the recently published works in the domain. Evaluation of the proposed method is carried out using several performance evaluation techniques such as classification accuracy, recall, precision, f-measure, area under the curve, and confusion matrix.

B. H. Shekar, Guesh Dagnew

Automated Delineation of Costophrenic Recesses on Chest Radiographs

The lung image segmentation using a model-based approach is a challenge owing to the sheer complexity and variability of the lung shape in a given data set. As a part of our effort to segment the lungs, we report a method to delineate the costophrenic (CP) recess without the human intervention. Active shape model (ASM) is used to point to the probable area of the CP recess, and a prior knowledge-based processing delineates the CP recess and hence determines the angle. The proposed method is fast and shows satisfactory results. It is intended to be used as a preprocessing step in segmenting the lungs’ contour. The proposed method also can be used to initialize the model contour in any other ASM-based lung segmentation algorithms. The algorithm was tested on 45 non-nodule lung images from the JSRT database. An average accuracy of 87.02% is achieved. A comparison of the results of proposed method and gold standard which is obtained by manual delineation is given.

Prashant A. Athavale, P. S. Puttaswamy

Using Location-Based Service for Interaction

Traffic is one of the biggest global factors, affecting over 100 million people. People who live in places with large content of pollutants in air have a 15–25% higher death risk from diseases like lung cancer than people who live in less polluted areas. This paper describes how Google Maps API can be used together with Google Cloud Console for location-based interaction.

K. M. Deepika, Piyush Chaterjee, Sourav Kishor Singh, Writtek Dey, Yatharth Kundra

Modeling Implementation of Big Data Analytics in Oil and Gas Industries in India

Stack holders have been anxious about the quality of performance in oil and gas industries in India and recommending technology intervention to drastically improve its performance. This investigation aims to analyze existing level of information and communication technology integration in oil and gas industries in India. All such industries which generate massive revenue are preparing to leverage big data analytics (BDA) to build efficiencies and improve productivity by removing non-value adding activities. This paper also investigates to identify ways and means of applying BDA tools.

Dilip Motwani, G. T. Thampi

Practical Market Indicators for Algorithmic Stock Market Trading: Machine Learning Techniques and Grid Strategy

In this paper, market indicators from three different approaches for algorithmic trading are analysed (moving average convergence divergence (MACD) crossovers, machine learning (ML) label-based indicators, and grid investing strategy). Market indicators are used by traders in the stock market, to define entry and exit points of a trade. These indicators are also useful to compare different trading strategies. We take a practical stand for the approaches mentioned above, where the same data feed from the exchange is preprocessed to remove redundant or anomalous content. Furthermore, use of correlation data between different stocks is analysed. (i) MACD crossovers are dealt in two dimensions of variability, the dimensions being frequency of trades and length of trading intervals. (ii) The outputs of different algorithms are passed through a voting classifier to get the best possible accuracy in the ML label-based approach. Precision/Recall analysis is done to qualify the algorithms for skewed data. (iii) Finally, a grid-based trading strategy is analysed. We conclude with a trading strategy, proposed using results of indicators based on the three approaches.

Ajithkumar Sreekumar, Prabhasa Kalkur, Mohammed Moiz

A Review on Feature Selection Algorithms

A large number of data are increasing in multiple fields such as social media, bioinformatics and health care. These data contain redundant, irrelevant or noisy data which causes high dimensionality. Feature selection is generally used in data mining to define the tools and techniques available for reducing inputs to a controllable size for processing and analysis. Feature selection is also used for dimension reduction, machine learning and other data mining applications. A survey of different feature selection methods are presented in this paper for obtaining relevant features. It also introduces feature selection algorithm called genetic algorithm for detection and diagnosis of biological problems. Genetic algorithm is mainly focused in the field of medicines which can be beneficial for physicians to solve complex problems. Finally, this paper concludes with various challenges and applications in feature selection.

Savina Colaco, Sujit Kumar, Amrita Tamang, Vinai George Biju

A Review on Ensembles-Based Approach to Overcome Class Imbalance Problem

Predictive analytics incorporate various statistical techniques from predictive modelling, machine learning and data mining to analyse large database for future prediction. Data mining is a powerful technology to help organization to concentrate on most important data by extracting useful information from large database. With the improvement in technology day by day large amount of data are collected in raw form and as a result necessity of using data mining techniques in various domains are increasing. Class imbalance is an open challenge problem in data mining and machine learning. It occurs due to imbalanced data set. A data set is considered as imbalanced when a data set contains number of instance in one class vastly outnumber the number of instances in other class. When traditional data mining algorithms trained with imbalanced data sets, it gives suboptimal classification model. Recently class imbalance problem have gain significance attention from data mining and machine learning researcher community due to its presence in many real world problem such as remote-sensing, pollution detection, risk management, fraud detection and medical diagnosis. Several methods have been proposed to overcome the problem of class imbalance problem. In this paper, our goal is to review various methods which are proposed to overcome the effect of imbalance data on classification learning algorithms.

Sujit Kumar, J. N. Madhuri, Mausumi Goswami

“College Explorer” An Authentication Guaranteed Information Display and Management System

The confusion and dilemma that arises out of the unorganized plethora of information on the Internet can never help school pass outs to reach any conclusion of which college to join for higher studies. Apart from college information, internal environment and feedback from students currently studying in the college is essential part to know about an institution and its administration. This paper stresses about the need for a Web application like college explorer through the novel contributions, system model, and advantages of the Web application developed. It enables the general public to view information such as placement details, admission details, course details, etc., about the colleges. In addition, the features like class notes sharing, notice publication (separate for students as well as for faculties of respective departments of the college), application of leave facility for both faculties as well as students are developed to manage the leave application even in emergency cases for smooth internal administration. The results confirm that Web applications have the potential to address various problem statements stated using Web technology efficiently.

Sonali Majumdar, K. M. Monika Patel, Arushi Gupta, M. N. Thippeswamy

Activity-Based Music Classifier: A Supervised Machine Learning Approach for Curating Activity-Based Playlists

Classification of musical tracks and creation of playlists to match four primary activities such as Sleep, Party, Dinner and Workout, using concepts of machine learning (ML) and musical information retrieval (MIR), is proposed in this paper. A data set of songs using features extracted through digital signal processing (DSP) is developed for training. In this work, several prominent and distinguishing features of individual musical tracks are employed. The ML algorithms used to classify the data set are: super vector machine (SVM), kth nearest neighbour, neural networks and voting classifiers. The results show that the highest accuracy can be attained when classification is performed using the voting classifier compared to other algorithms. The increase in accuracy can be attributed to the voting classifier’s ability to improve the individual classes’ accuracy by utilizing multiple classifier outputs.

B. P. Aniruddha Achar, N. D. Aiyappa, B. Akshaj, M. N. Thippeswamy, N. Pillay

New Password Embedding Technique Using Elliptic Curve Over Finite Field

In the present sophisticated digital era, safe communication of user password from one source to the other is quite difficult in client/server system. Also storing the password as it appears increases the potential risk of the security. Protection of the password is at most important in group communications to avoid the access of the illegal person to group resources. In addition, a roaming user who uses the network from different client terminals requires access to the private key. The present paper explains secure communication of password from one entity to the other. Here the password is encrypted using elliptic curve over finite field, embedded in a large random text at different selected positions, and communicated to the receiver via public channel.

D. Sravana Kumar, C. H. Suneetha, P. Sirisha

Performance of Wind Energy Conversion System During Fault Condition and Power Quality Improvement of Grid-Connected WECS by FACTS (UPFC)

The demand for the power generation from wind is constantly growing. This situation forces the revision of the grid codes requirements, to remain connected during grid faults. Immediately, the voltage level will drop below 80% when fault occurs at PCC (point of common coupling) and the rotor speed of IG (induction generators) becomes unstable. In this work, UPFC are used under fault condition to improve the low voltage ride-through (LVRT) of wind energy conversion system (WECS) and damping of rotor speed oscillations of IG. Furthermore, after the fault UPFC acts as virtual inductor and leads to increase in terminal voltage of WECS. WECS with DFIG-based system is considered for analysis here. By simulating DFIG-based WECS with UPFC indicates the improvement in LVRT and remains and WTGs continues to operate with grid at certain voltage fluctuations, near grid. Also, indicates voltage improvement at PCC under fault conduction, and voltage is recovered easily to 1 pu at PCC.

Sudeep Shetty, H. L. Suresh, M. Sharanappa, C. H. Venkat Ramesh

Comprehensive Survey on Hadoop Security

The new emerging technologies have provided a way for a large amount of data generation. Secure storage of such a huge data is of prime importance. Hadoop is a tool used to store big data, where security of it is not assured. In this paper, we have considered a survey on various approaches which helps in providing secure storage of files in Hadoop. Hadoop framework is developed for the support of processing and storage of Bigdata in a distributed computing environment. Usage of Bigdata has become a key factor for the companies as they can increase their operating margin. Bigdata contains user-sensitive information and bring forth many privacy issues. Bigdata is a larger and a more complex datasets obtained from a variety of network resources. These datasets are beyond the ability of traditionally used data processing software to capture, manage, and process the data within the given time frame. These massive volumes of data are used by many of the organizations to tackle the problem that could not be done before. Since the data holds a lot of valuable information, these data need to be processed in short span of time by which companies can boost their scale and generate more revenue, traditional system resources are not sufficient for processing and storing, and this is where Hadoop comes into picture. The main objective of Hadoop is running of application of bigdata. Hadoop being a great tool for data processing, it was initially designed for internal use (i.e., within local cluster) without any security perimeter of organization, so they were easily hackable and exposed to threats.

Maria Martis, Namratha V. Pai, R. S. Pragathi, S. Rakshatha, Sunanda Dixit

Descriptive Data Analysis of Real Estate Using Cube Technology

With the progress and application of data analysis and processing expertise, the technologies such as data warehouse and cube technology had become the research spot of each business area. Online analytical processing technology has applications in the areas of merchandising system, college decision support system, enterprise marketing management system, knowledge data warehouse, and to analyze the curriculum chosen by students and many more. If we consider one business domain that is real estate, it is observed that online transactional processing technology has been applied. This paper focuses on application of online analytical processing and specifically descriptive data analytics, to extract more information from the traditional real estate datasets.

Gursimran Kaur, Harkiran Kaur

Temporal Information Retrieval and Its Application: A Survey

With an advent of the Web, a tremendous amount of information is available online. Information can be organized and explored in the time dimension. This temporal information has to be distilled out, so as to extract the temporal entities such as temporal expressions and temporal relations out of it. Temporal information processing is an ongoing field of research that deals with natural language text, temporal relations, events or temporal queries. This paper presents a detailed analysis of the work carried out under temporal information retrieval (TIR) highlighting its subtasks like information extraction, indexing, ranking, query processing, clustering and classification. Also, it presents various challenges while dealing with temporal information. To the end, various application areas are elaborated such as temporal summarization, exploration and future event retrieval.

Rakshita Bansal, Monika Rani, Harish Kumar, Sakshi Kaushal

A Survey on Multi-resolution Methods for De-noising Medical Images

Processing of medical images is important to improve their visibility and quality to facilitate computer-aided analysis and diagnosis in medical science. Such images are usually tainted by noise due to impediments in image capturing devices, unsupportive environment or during transmission over the network. Multi-resolution is a profound technique for decomposing the images into multiple scales and is widely used for image analysis in detail. This paper describes various multi-resolution techniques such as discrete wavelet transform, multi-wavelet transform, and Laplacian pyramid to reduce a wide variety of noise in images. Also, an image de-noising algorithm based on multi-resolution analysis for noise reduction has been described.

G. Bharath, A. E. Manjunath, K. S. Swarnalatha

Performance Analysis of Vedic Multiplier with Different Square Root BK Adders

Multiplication is the main basic operation used by many of the digital signal processor (DSP) and vector processors. DSP application repeatedly performs the operations like signal processing, filtering, processing of discrete signal data, and radar signal processing and use intensive fast Fourier transform (FFT) operations. FFT computation uses butterfly structures, where multiplication is the basic operation. DSPs have to execute a large number of instructions per second, which in turn uses so many FFT computations, and hence, the multiplication operation decides the performance of DSP. Designing a high-performance multiplier improves the overall performance of the processor. Many multiplier architectures have been proposed in the past few decades with the attractive performance, power consumption, delay, area, throughput, etc., and the most acceptable multiplier among them is the Vedic multiplier. When high performance is necessary, Vedic multiplier will be the best choice. Operation of Vedic multiplier is based on ancient Vedic mathematics. This earlier multiplier has been modified to improve the performance. There are 16 sutras for the multiplication operation in this method. These sutras are used to solve large range of multiplication problems in a natural way. This method of multiplication is based on Urdhva Triyagbhyam sutra, which means horizontal and cross-wire technique of multiplication operation. This method uses partial product generation in parallel and eliminates the unwanted steps with zero. Urdhva Triyagbhyam sutra is an efficient sutra which enhances the execution speed of the multiplier by minimizing the delay. This work describes the overall performance of the Vedic multiplier with different high-speed adders like regular square root BK adder (RSRBKA), Modified square root BK adder (MSRBKA) and proposed optimized square root BK adder (OSR-BK-A). The proposed designs are simulated and synthesized in Xilinx ISE 14.7, and the results are tabulated.

Ranjith B. Gowda, R. M. Banakar, Basavaprasad

Analysis of Traffic Characteristics of Skype Video Calls Over the Internet

Skype is an important application of real-time systems (RTS). It uses Transmission Control Protocol (TCP) for connection establishment and User Datagram Protocol (UDP) port for transfer of audio and video data. In spite of the popularity of Skype, relatively little is known about its traffic characteristics. In this paper, the sender is sending the audio visual data using UDP port at a constant bit rate (CBR). The destination receives the data and checks with its buffer threshold values. The minimum and maximum threshold values are fixed based on the receiver buffer. When the sender data is higher than the threshold value, the destination asks source to reduce the flow rate by sending an explicit control packet. When the sender rate is lower than the threshold value, the destination can ask the source to increase the sending rate. The introduction of feedback system using a control message overcomes the congestion at the receiver, minimizes the data loss, increases optimal utilization of resources, and enhances the quality of expectation. In Skype application, an ordinary node acts one time as a sender and other time as a receiver. The proposed technique is to be built at both the end nodes. Important quality parameters such as packet loss due to congestion, one-way packet delay, effect of queuing delay on sender performance of feedback are analyzed using graphs and statistical data. Mathematical models are used to analyze the Skype performance. MATLAB software is used to simulate the system and for model authentication.

Gulshan Kumar, N. G. Goudru

Smart Tourist Guide (Touristo)

Individuals travelling, frequently think that it’s hard to seek places and find nearby amenities, and this issue even looks greater when we cannot talk the neighbourhood dialect. Additionally while travelling in groups, individuals like exploring different places and may get lost, which again wind up troublesome for their companions individuals to find different individuals from the group. Touristo is a project about building up an android-based application in the field of travel and tourism. Android is a Google-developed programming language for mobiles and tablets. Additionally, Firebase is a real-time database which is utilized for information stockpiling and handling. Hence using the features provided, we intend to develop the application. By examining the above issues and different others, we are taking this venture with the goal to develop such application to overcome above issues and serve clients better.

M. R. Sowmya, Shashi Prakash, Shubham K. Singh, Sushent Maloo, Sachindra Yadav

Thefted Vehicle Identification System and Smart Ambulance System in VANETs

VANETs act as crucial component of intelligent transportation system (ITS). VANETs are capable of providing connectionless communication between mobile nodes (vehicles) and static nodes such as RSU and BSU to enhance safety and comfort of vehicles on the highways or in urban environments. There is no system in place for finding stolen vehicles and providing faster movement of ambulances on heavy traffic lanes. In this paper, we discuss about automated system for traffic management such as for finding stolen vehicles and providing faster movement of ambulances on heavy traffic lanes, and we are coming up with two systems namely thefted vehicle identification system (TVIS) and smart ambulance system (SAS), respectively.

S. R. Nagaraja, N. Nalini, B. A. Mohan, Afroz Pasha

Performance Study of OpenMP and Hybrid Programming Models on CPU–GPU Cluster

Optimizing complex code of scientific and engineering applications is a challenging area of research. There are many parallel and distributed programming frameworks which efficiently optimize the code for the performance. In this study, we did a comparison study of the performance of parallel computing models. We have used irregular graph algorithms such as Floyd’s algorithm (shortest path problems) and Kruskal’s algorithm (minimum spanning tree problems). We have considered OpenMP and hybrid [OpenMP + MPI] on CPU cluster and MPI + CUDA programming strategies on the GPU cluster to improve the performance on shared–distributed memory architecture by minimizing communication and computation overlap overhead between individual nodes. A single MPI process per node is used to launch small chunks of large irregular graph algorithm on various nodes on the cluster. CUDA is used to distribute the work between the different GPU cores within a cluster node. Results show that from the performance perspective GPU, implementation of graph algorithms is effective than the CPU implementation. Results also show that hybrid [MPI + CUDA] parallel programming framework for Floyd’s algorithm on GPU cluster yields an average speedup of 19.03 when compared to the OpenMP and a speedup of 15.96 is observed against CPU cluster with hybrid [MPI + OpenMP] frameworks. For Kruskal’s algorithm, average speedup of 27.26 is observed when compared against OpenMP and a speedup of 20.74 is observed against CPU’s cluster with hybrid [MPI + OpenMP] frameworks.

B. N. Chandrashekhar, H. A. Sanjay

Signature Analysis for Forgery Detection

Forgery of signature has become very common, and the need for identification and verification is vital in security and resource access control. There are three types of forgery: random forgery, simple or casual forgery, expert or skilled or simulated forgery. The main aim of signature verification is to extract the characteristics of the signature and determine whether it is genuine or forgery. There are two types of signature verification: static or offline and dynamic or online. In our proposed solution, we use offline signature analysis for forgery detection which is carried out by first acquiring the signature and then using image pre-processing techniques to enhance the image. Feature extraction algorithms are further used to extract the relevant features. These features are used as input parameters to the machine learning algorithm which analyses the signature and detects for forgery. Performance evaluation is then carried out to check the accuracy of the output.

Dinesh Rao Adithya, V. L. Anagha, M. R. Niharika, N. Srilakshmi, Shastry K. Aditya

Optimal Sensor Deployment and Battery Life Enhancement Strategies to Employ Smart Irrigation Solutions for Indian Agricultural Sector

In this fast-growing technology, agriculture sector in India is one of the domains where we have observed slow adoption of Internet of Things (IoT) solutions. Unavailability, expensive seasonal labour and inadequate water resources are one of the major problems faced by Indian agriculture. Use of costly industrial standard sensors, energy utilization and placement of sensors also poses a greater problem for adoption of IoT solutions. This paper proposes an IoT-based framework and a strategy for placement of the optimal number of sensors and optimal utilization of battery to address the mentioned issues. Framework adopts NodeMCU and moisture sensor that addresses communication and water scarcity problems. We are using ThingSpeak cloud to store and process the sensor data. Results demonstrate the effectiveness of our strategies and proposed IoT framework.

M. K. Ajay, H. A. Sanjay, Sai Jeevan, T. K. Harshitha, M. Farhana mobin, K. Aditya Shastry

Smart Waste Monitoring Using Wireless Sensor Networks

In today’s technology, waste disposal and management is becoming a very big issue for the people, as it is the main cause for the unhygienic environment. This leads to various diseases and human illness. To avoid this situation, we are going to implement a system with the help of Python and IoT. The concept is based on smart waste monitoring system. This will help us to maintain a clean environment in our city. We can manage the waste disposal in various areas of the city. IoT is a concept in which we can operate various devices without any user intervention. We can able to manage all the devices with the help of IoT. Sitting at one place we can able to monitor our system and keep an eye over the city. We are making use of Raspberry pi to operate Raspbian OS to enable device connectivity. Different IR sensors are used to detect the level of the dustbin camera which is also fixed in the area to capture the image of the dust in the area. The information is send to the authorized person and we can take the immediate action related to that.

T. V. Chandan, R. Chaitra Kumari, Renu Tekam, B. V. Shruti

Digital Filter Technique Used in Signal Processing for Analysing of ECG Signal

The area of signal processing holds high significance in biomedical engineering, acoustics and sonar fields. The main finding of coronary heart illnesses is done utilizing ECG. It demonstrates the bio-physiology of cardiac muscles and modifications like arrhythmia and also conduction surrenders. ECG in flag handling is a prime zone of study in bio-signal processing. Present-day advancements in personal computer equipment and computerized channel approach in flag preparing have made correspondence with personal computers through ECG signals suitable. Efficient determination of ECG is a mechanical test. This study exhibits a far-reaching review of computerized sifting strategies to adapt to the clamour curios in ECG flag. The goal of this paper is to separate noteworthy components of ECG utilizing signal preparing methods. Methodologies of different computerized channels for ECG in flag preparing are analysed. Noteworthiness of flag preparing gives off an impression of being with no noticeable indication of immersion in today’s world.

A. E. Manjunath, M. V. Vijay Kumar, K. S. Swarnalatha

GeoFencing-Based Accident Avoidance Notification for Road Safety

The aim of the project is to attempt the reduction of occurrence of accidents by providing effective and precautionary notifications to the user on the progressive Web app, if the user is approaching the accident-prone zone. In the past, due to road accidents, over ten lakhs people had been killed and 50 lakhs had been severely injured, in India. The project provides notifications about approaching vehicles in and around accident-prone areas. Notifications are voice-based so that user could focus on driving and need not constantly view the phone for the precautionary warnings. The project uses global positioning system (GPS) to specify a virtual boundary called the geofence and open-source Google APIs for setting the accident-prone areas on Google Maps. Firebase is used as a backend to store the data on real-time databases for providing a warning on approaching vehicles. The geofenced areas are marked at a sufficient distance from the actual accident-prone zones so as to provide notifications in advance. Necessary notifications are provided on the app, only if the user is in the geofenced area. The prototype also demonstrates temporary geofencing, to provide warnings about the conditions like a roadblock.

Bhavyashree Nayak, Priyanka S. Mugali, B. Raksha Rao, Saloni Sindhava, D. N. Disha, K. S. Swarnalatha

An Innovative IoT- and Middleware-Based Architecture for Real-Time Patient Health Monitoring

WSN is deployed in every sector of human life, an instance of which is in medical device network. With the increasing requirement of handling large patient data and faster data conversion in the hospital and medical diagnostic centres, there is a need for innovative and low-cost ways of interconnecting medical equipment, middleware and network support. Hospitals and medical diagnostic centres require cost-effective sensor network for handling large patient data and fast conversion. The high cost of medical devices, equipment and lack of interoperability and portability necessitates new approaches. This paper presents low-cost and portable approach for medical data transmission from devices to middleware using IoT support and conversion to universal standard called Health Level-7 (HL7) and a method to store in cloud.

B. A. Mohan, H. Sarojadevi

Travelling Salesman Problem: An Empirical Comparison Between ACO, PSO, ABC, FA and GA

Travelling salesman problem (TSP) is one of the optimization problems which has been studied with a large number of heuristic and metaheuristic algorithms, wherein swarm and evolutionary algorithms have provided effective solutions to TSP even with a large number of cities. In this paper, our objective is to solve some of the benchmark TSPs using ant colony optimization (ACO), particle swarm optimization (PSO), artificial bee colony (ABC), firefly algorithm (FA) and genetic algorithm (GA). The empirical comparisons of the experimental outcomes show that ACO and GA outperform to ABC, PSO and FA for the given TSP.

Kinjal Chaudhari, Ankit Thakkar

Twitter Data Sentiment Analysis on a Malayalam Dataset Using Rule-Based Approach

Opinion characterization is nowadays a potential and intense research focus because of the hasty growth of social media such as blogs and social networking sites, where individuals put in freely their perspectives on different themes. Researches prove that people find it comfortable to opinionate in their mother tongue, be it verbal or written. Given that now almost all social platforms support most of the popular languages, the requirement to mine the sentiments in various dialects is on the rise. However, not all data may be relevant; some may not have any impact on the end result and some may have similar meanings. A preprocessing phase is hence required to help make the dataset concise. In this paper, the authors focus on finding out the polarity of the words input by various users through their reviews exhibited using the South Indian language, Malayalam. Malayalam like the other languages in the Dravidian family exhibits the characteristics of an agglutinative language. The preprocessing process consists of cleaning the data, tokenization, stopword removal, etc. In this paper, authors are focusing on the document-based polarity calculation of the Malayalam reviews. The overall polarity of the corpus is calculated based on the positivity and negativity values of individual documents. It is found that negativity value is higher for the user reviews in our corpus which shows their negative attitude toward the news thread with the classifier accuracy of 89.33%.

Deepa Mary Mathews, Sajimon Abraham

An IoT-Based Smart Water Microgrid and Smart Water Tank Management System

Water is most important resource which needs to be managed smartly. Managing house water supply in a society consisting of water tanks, motors, and pumps automatically is an important task for efficient consumption of water. In this paper, we propose a smart solution for leakage detection in the tank using its dimensions and sensor data. The data from each house is stored on the cloud for analyzing the water consumption of each house in a society and main water supply, through GSM/GPRS 900a module. A hybrid application, Smart Water Grid, is responsible for monitoring the water level in the tank continuously, to control the motor automatically, and it consists of an inspection mode to detect the leakage in the tank and its dimension.

Shubham Kumar, Sushmita Yadav, H. M. Yashaswini, Sanket Salvi

Smart Sensing for Vehicular Approach

Every day around the world, a humongous amount of people die from road accident and the subsequent injuries. There are many problems which are largely prevalent in the everyday life of a driver around the globe. Some of the techniques that are available in the market are too expensive to implement on a common vehicle. If we take a look around the common household in an Indian society, most of the people are using average cost vehicles and they are not able to afford the existing techniques which can detect the obstacle to prevent from the road accident. The survey has been conducted on the problems which are being faced by the driver at the time of driving and we have proposed a suitable and less expensive ways to implement the solutions of, not all the problems, but few of them to detect the causes of road accident by using some sensors like ultrasonic sensor, ldr sensor, ir sensor and prevention from collision. As smart-driver assistance system, invisibility problem is our main focus in this project. The concept is that it assists the driver with information and actions. In our proposed work, the smart-driver assistant system will provide the information after analyzing results of various sensors existing in the system and then if the driver is unable with actions necessary to ensure the driver’s safety. Invisibility in fog is one of the major reasons of road accidents, various approaches have been made to counter this problem. We have found that ultrasonic sensor can be used to counter this problem. The sensed information is provided to the driver who takes appropriate action depending on the information. However, there are cases where the driver is incapacitated or unable or there are cases where the driver actually needs to drive faster for some urgency. In such cases, the smart-driver assistant system comes in play and slows down the vehicle for the drive, which changes their direction. If unable, the system slows the vehicle itself and if still not stopped, it stops the vehicle at 20 cm away from the obstacle. The proposed work has been tested with four parameters and found to be a better solution.

Mukesh Chandra Sah, Chandan Kumar Sah, Shuhaib Akhter Ansari, Anjit Subedi, A. C. Ramachandra, P. Ushashree

Android Malware Detection Techniques

Importance of personal data has increased along with the evolution of technology. To steal and misuse this data, malicious programs and software are written to exploit the vulnerabilities of the current system. These programs are referred to as malware. Malware harasses the users until their intentions are fulfilled. Earlier malware was major threats to the personal computers. However, now there is a lateral shift in interest toward Android operating system, which has a large market share in smartphones. Day by day, malware is getting stronger and new type of malware is being written so that they are undetected by the present software. Security parameters must be changed to cope up with the changes happening around the world. In this paper, we discuss the different types of malware analysis techniques which are proposed till date to detect the malware in Android platform. Moreover, it also analyzes and concludes about the suitable techniques applicable to the different type of malware.

Shreya Khemani, Darshil Jain, Gaurav Prasad

A Comparative Study of Machine Learning Techniques for Emotion Recognition

Humans share emotions which they exhibit through facial expressions. Automatic human emotion recognition algorithm in images and videos aims at detection, extraction, and evaluation of these facial expressions. This paper provides a comparison between various multi-class prediction algorithms employed on the Cohn-Kanade dataset (Lucey in The extended Cohn-Kanade dataset (CK+): a complete dataset for action unit and emotion-specified expression, pp. 94–101, 2010 [1]). The different machine learning algorithms can be used to provide emotion recognition task. We have compared the performance of K-nearest neighbors, Support Vector Machine, and neural network.

Rhea Sharma, Harshit Rajvaidya, Preksha Pareek, Ankit Thakkar

IoT-Based Smart Parking System

Post liberalization, Indian cities are growing at an exponential growth rate. The rapid growth of the towns is giving birth to many socioeconomic problems. With the increase in the number of personal vehicles and shrinking parking spaces, the problem of parking vehicles at wrong parking spaces is steadily increasing which causes home and business establishment owners a lot of discontents, time wastage, and unnecessary chaos. By adopting latest technologies, the parking issue can be addressed more smartly. In this work, we have designed, developed, and tested an IoT-based smart parking solution. We have conducted rigorous testing of our solution in real life under various circumstances and observe that our approach provides a practical solution to the wrong parking issue.

G. Abhijith, H. A. Sanjay, Aditya Rajeev, Chidanandan, Rajath, Mohan Murthy

Smart Agricultural Monitoring System Using Internet of Things

India is one of the largest agricultural countries with a population of 1.3 billion. Farming in India is labor intensive and absolute. 70% of India’s residents are dependent on farming, and one-third of nations’ funds come from agriculture. Even after decades of cultivation practice, it is lagging behind in maximizing the yield thereby hampering the progress of the nation. In order to overcome this, there is a need for promoting cultivation practice for high yield of crops. With the availability of IT and internet, Internet of Things is proliferating at an unprecedented rate. The perception of agricultural IoT (Internet of things) utilizes networking equipment in farming construction. The hardware part of this project includes processors with data processing capability and sensors which are used to measure various parameters like temperature, humidity, and water level. In this paper, the sensor node is designed to monitor the environmental conditions that are vital for the proper growth of crops. The collected data received are analyzed for proper monitoring and improving the yield of the crop. The result depicts the data being stored and retrieved on Agri Cloud ( [1]).

H. V. Asha, K. Kavya, S. Keerthana, G. Kruthika, R. Pavithra

Detecting Healthiness of Leaves Using Texture Features

Agriculture is the dominant sector of our economy and contributes in various ways but the yield in the productivity leads to a significant reduction in the farmer’s income. Monitoring crop health is important to increase the quality and quantity of the yield. But this requires manually monitoring the crops and also expertise in the field. Hence, automatic disease detection using image texture features is used for ease and to detect the disease at an early stage. The proposed methodology for the project is to design and implement the algorithm on two sets of databases: firstly, a locally generated leaf database which contains images of leaves and secondly, a standard database which is a common test database. The basic steps for crop disease detection include image acquisition, image preprocessing, image segmentation, feature extraction, and classification using image processing techniques. The acquired leaf images are preprocessed by removing undesired distortion and noise, and then, the processed image is further subjected to K-means-based segmentation. The segmented image is further analyzed using Haar wavelet transform and GLCM based on its texture by extracting feature vector. SVM is used for classification of image. Thus, the presence of diseases in leaf is identified along with all the features values of the leaf. It also calculates the accuracy rate of the prediction made by the system.

Srishti Shetty, Zulaikha Lateef, Sparsha Pole, Vidyadevi G. Biradar, S. Brunda, H. A. Sanjay

A Survey on Different Network Intrusion Detection Systems and CounterMeasure

Recent studies have pulled tons of research in the domain of cloud security and various intrusion detection systems (IDSs). This is because of advancement in the different types of attacks on computer systems. Distributed denial of service (DDoS) attack is one of them wherein the attackers can compromise the cloud system by exploiting vulnerabilities. Initially, during the multi-step exploration, vulnerability with low frequency along with the virtual machine which is identified and compromised are included in DDoS attacks. In this context, various IDSs have been surveyed with different countermeasure techniques including some effective techniques to minimize the malicious activities within end systems or networks. The main aim of IDSs is to detect different attacks within networks and end systems or to be precise against any information systems which are very difficult to maintain in a secure state for a long duration. Some studies have shown that the use of host-based systems and the network-based systems help to improve the attack detection. This paper focuses on the study of various well-known IDS and various techniques to minimize malicious activities within the system.

Divya Rajput, Ankit Thakkar

Compressed Sensing for Image Compression: Survey of Algorithms

Compressed sensing (CS) is an image acquisition method, where only few random measurements are taken instead of taking all the necessary samples as suggested by Nyquist sampling theorem. It is one of the most active research areas in the past decade. In this age of digital revolution, where we are dealing with humongous amount of digital data, exploring the concepts of compressed sensing and its applications in the field of image processing is very much relevant and necessary. The paper discusses the basic concepts of compressed sensing and advantages of incorporating CS-based algorithms in image compression. The paper also discusses the drawbacks of CS, and conclusion has been made regarding when the CS-based algorithms are effective and appropriate in image compression applications. As an example, reconstruction of an image acquired in compressed sensing way using $$ l_{1} $$ minimization, total variation-based augmented Lagrangian method and Bregman method is presented.

S. K. Gunasheela, H. S. Prasantha

Intrusion Detection System Using Random Forest on the NSL-KDD Dataset

In the modern world of interconnected systems, network security is gaining importance and attracting a lot of new research and study. Intrusion detection systems (IDSs) form an integral part of network security. To enhance the security of a network, machine learning algorithms can be applied to detect and prevent network attacks. Taking advantage of the robust NSL-KDD dataset, we have employed the supervised learning algorithm random forests to train a model to detect various networking attacks. To further increase the classification accuracy of our model, we have employed the use of famous data mining technique of feature selection. Smart feature selection using Gini importance has been employed to reduce the number of features. Experimental results have shown that our model not only runs faster but also performs with a higher accuracy.

Prashil Negandhi, Yash Trivedi, Ramchandra Mangrulkar

Dual-Mode Wide Band Microstrip Bandpass Filter with Tunable Bandwidth and Controlled Center Frequency for C-Band Applications

This paper presents a unique approach for designing dual-mode wide band BPF with tunable bandwidth and controlled center frequency for C-band (4–8 GHz) applications. The proposed filter is designed using radial stub-loaded dual-mode λg/2 resonator to get wide passband. The dual-mode behavior of the resonator, i.e., odd- and even-mode resonance frequencies are realized by inserting a radial stub at the center of the resonator and further the size of filter is reduced by folding the resonator. A modified feed structure which embraces the two arms of the resonator is used to obtain two transmission zeros in upper stop band. By keeping all calculated dimensions of filter fixed and by varying only radial angle θ (in degrees) of radial line stub, FBW is tuned while controlling center frequency. From simulation results, it is observed that the designed filter has very good passband characteristics, a wide 3-dB passband from 4.4 to 7.8 GHz with center frequency at 6 GHz, fractional bandwidth of 56.6%, return loss S11 more than 13 dB, and transmission loss S21 better than 0.3 dB, respectively.

Shobha I. Hugar, Vaishali Mungurwadi, J. S. Baligar

ADHYAYAN—An Innovative Interest Finder and Career Guidance Application

ADHYAYAN is an innovative mobile application which determines a user’s interest in a particular domain and nurtures them effectively so that they can pursue career in the field which they are interested in. The system takes into account social media posts, results of a test and application activity to find out the interest of users in different fields and then assists, guides and evaluates them continuously to improve their skills in these fields. ADHYAYAN is a three-tier system which consists of a front-end, middle layer, and back-end. Front-end is an Android application which provides personalized GUI for each user. Middle layer is Firebase, while back-end is a server hosted on ‘Google Cloud Platform’. An algorithm has been developed for ADHYAYAN which calculates the ratio of user’s interest in different domains and eventually feeds are generated in the same ratio on user’s profile. To cater the increasing need of skilled employees in different fields and promote interest-based learning, ADHYAYAN has been proposed to overcome various limitations and drawbacks of existing solutions.

Akshay Talke, Virendra Patil, Sanyam Raj, Rohit Kr. Singh, Ameya Jawalgekar, Anand Bhosale

Implementation of Cure Clustering Algorithm for Video Summarization and Healthcare Applications in Big Data

The Data Mining Techniques provide useful ways to generate desired patterns from the large data and establish relations between them to solve problems using data analysis. This paper focuses on a data mining algorithm called CURE, and its applications on Health Care and Video data. Big Data consists of large volume, ever growing Datasets with multiple sources. Big Data in Health Care is an emerging area which helps healthcare organizations for their analytics and reporting needs. Data Mining Techniques, predictive analytics, and prescriptive analytics are some of the methods to analyze the healthcare data and derive useful information for several applications. On the other hand, Video Processing is an emerging area of research which gives rise to variety of applications like object tracking, shot detection, Video Summarization, etc. This paper discusses the application of CURE clustering algorithm on Video Processing for generating Video Summary and application of the same algorithm on Big Data Health Care Dataset for deriving disease related information.

Jharna Majumdar, Sumant Udandakar, B. G. Mamatha Bai

Redundancy Management of On-board Computer in Nanosatellites

Satellite bus has subsystems like attitude determination and control system (ADCS); electrical power system (EPS); and communication, command and data handling (C&DH) for operations at different phases of the mission. These subsystems’ requirements are processed and controlled by the on-board computer (OBC) subsystem in the satellite. On-board computer (OBC) subsystem plays a vital role in the functioning of the satellite system; a small malfunction in this system might result in the entire mission failure. For such critical subsystem where on-board manual intervention to repair or replace a failed component is difficult, it is very much essential to have a redundant mechanism. Using redundancy concepts to improve the reliability of systems or subsystems is a well-known principle. This paper describes the architecture of OBC and the redundancy configuration in the nanosatellite. Further, the redundancy management between master and redundant subsystem is explained. This is achieved by first detecting the failure by using an external watchdog timer that monitors master OBC unit along with the redundant. The isolation of the fault signals from the failed unit is controlled by the power control switch.

Shubham, Vikash Kumar, Vishal Pandey, K. Arun Kumar, S. Sandya

A Fault Tolerant Architecture for Software Defined Network

Software defined network is a single-point control architecture where the control plane and data plane are disaggregated. It has a centralized controller, switches and hosts. Here, the OVS switches which act as the data-forwarding plane are connected to the controller having forwarding details. In the above-said architecture, if the controller goes down due to bottleneck problems that arise because of packet injection or any other attacks, then the network experiences performance drawbacks. Hence in this paper, we propose a fault tolerant approach for software defined networks. Once the controller fails, OVS switches are made as controllers by using a switch type by name “UserspaceSwitch in namespace” which separates switches’ namespace from the controller. We have used Mininet an emulator for simulation of software defined networks with POX, a remote controller. The proposed fault tolerant approach for software defined networks is also simulated using the mentioned software. The hosts are made as client and server, and the data traffic is generated between them using UDP which gives different parameters as output for analysing the performance. Further, graphs are plotted considering three-parameter delay, packet loss and throughput for both SDN with centralized control and fault tolerant approach to analyse the performance.

Bini Y. Baby, B. Karunakara Rai, N. Karthik, Akshith Chandra, R. Dheeraj, S. RaviShankar

Optimal Thresholding in Direct Binary Search Visual Cryptography for Enhanced Bank Locker System

Visual cryptography (VC) is one of the strongest cryptographic method present. The main advantage of this system is that the decryption doesnot need any specific requirements for decoding other than human eyes. Using halftoning techniques binary images are obtained for grayscale and color images, this technique is applied in Halftone VC. In this paper, direct binary search (DBS) is implemented and initial images are modified for better quality of recovered images. The concept is proposed for bank locker systems. Comparison has been made using parameters like PSNR, Correlation, UQI and SSIM.

Sandhya Anne Thomas, Saylee Gharge

Comparison Between the DDFS Implementation Using the Look-up Table Method and the CORDIC Method

An efficient communication system requires synchronization between the transmitter and the receiver, which is achieved by generating the same local carrier frequency. Direct Digital Frequency Synthesizer (DDFS) is one of the methods to generate various frequencies, centered around a reference frequency. This paper presents the comparison between the DDFS implementation using the look-up table (LUT) method and the CORDIC, a multiplier-less algorithm. The implementation has been carried out in Simulink and various parameters have been analyzed.

Anish K. Navalgund, V. Akshara, Ravali Jadhav, Shashank Shankar, S. Sandya

Adding Intelligence to a Car

The automobile business has been globalized from its initial days. Carmakers and innovation firms are investigating every possibility in their joint endeavors to upgrade the execution of keen automobile stages. Mischances are expanding everywhere; pace and different advancements are being utilized to diminish it. Utilizing generally straightforward programming and changes in accordance with existing equipment, we can accomplish an exceedingly secure automobile. This venture builds up a framework that the majority of its activities are controlled by smart programming inside the ARM LPC 2148. It expects to plan and build up a framework which can be controlled from the outside world utilizing Bluetooth and furthermore guarantees the driver well-being by utilizing a contrasting option to air bags with the assistance of rack and pinion framework. At the point when the automobile is being utilized by any unapproved individual, a message containing the automobile area with the assistance of GPS and GSM will achieve the proprietor quickly. The proprietor derives about the security rupture and tries to control and stop the automobile with Bluetooth. Adding to this, when the temperature of the motor is raised past a specific point of confinement, the automobile is made to stop naturally until the point that the temperature is under control.

Komal Suresh, Svati S. Murthy, Usha Nanthini, Shilpa Mondal, P. Raji

Thermal Care and Saline Level Monitoring System for Neonatal Using IoT

A newborn baby usually has a problem to adapt the change in temperature, be it a full-term healthy baby or a preterm baby or low-birth-weight babies. Neonatal usually has little body fat, and they are too immature in handling and regulating the body temperature. A temperature ranging between 37.5° and 36.5° is considered to be normal body temperature; according to the WHO, any newborn whose temperature is below the normal range and drops below 32° is considered as a risk leading to hypothermia condition in the newborn. In such conditions, babies are kept in incubators so that the babies can regulate their body temperature and get adjusted to the environment. Hypothermia in neonates is associated with increased mortality rate. Thermal management of babies is a vital and critical part of neonatal care. And frequent check on the level of saline status is a must when given to any neonatal which cannot be neglected or show inattentiveness which may lead to life-risking condition. With the advancement of technology and IOT in the boom, this paper provides the health monitoring system of neonatal care using the IoT, a system which can monitor and maintain the necessary temperature of the neonates and monitoring of the saline bottle from a distant place.

Huma Kousar Sangreskop

Home Security System Using GSM

In areas where robbery and theft are a major issue, home security becomes a matter of prime importance to the residents of that area. Everyone in the locality is forced to take security measures to prevent their precious belongings from being stolen. It is therefore invincible that a technological solution has to be formulated to ensure the safety of the house. Hence, a security device has been designed to send an alert message to the owner of the house and to the security forces nearby in an attempt to void the theft taking place. The system is designed by interfacing sensor modules with a microcontroller to detect the motion in the house and a GSM module to send alert message to the owner of the house when the house is locked. This system uses low-cost sensors for motion detection and proves to be affordable. The installation of the system is easy and also the sensors and modules require very less space and consume low power when installed.

P. Mahalakshmi, Raunak Singhania, Debabrata Shil, A. Sharmila

Automatic Toll Tax Collection Using GSM

This paper proposes a very novel approach to implement the automatic toll tax collection system on the toll plazas using radio frequency identification (RFID) and global system for mobile (GSM). Nowadays, the cities and highways are bursting with traffic, and very often long queues of vehicles can be seen at various toll plazas so that they can pay the toll and then able to use the road or highway. So a system is proposed wherein the toll tax could be paid via cashless transactions and people wouldn’t have to wait for a long time for the cash payment of the toll tax. This would save people’s money and time simultaneously. It would also eliminate errors in cash transactions and further ease the job of the toll plaza companies. It would definitely bring down any of the corruptions occurring at the toll plazas. Finally, it would make the existing toll tax collection more efficient and ease our lives a bit more.

P. Mahalakshmi, Viraj Pradip Puntambekar, Aayushi Jain, Raunak Singhania

Facial Expression Recognition by Considering Nonuniform Local Binary Patterns

Recognizing a face with an expression has paying attention due to its well-known applications in a broad range of fields like data-driven animation, human–machine interaction, robotics, and driver fatigue detection. People can vary significantly their facial expression; hence, facial expression recognition is not an easy problem. This paper presents a significant contribution for facial expression recognition by deriving a new set of stable transitions of local binary pattern by selecting the significant nonuniform local binary patterns. The proposed patterns are stable, because of the transitions from two or more consecutive ones to two or more consecutive zeros. For better recognition rate, the new set of patterns are combined with uniform patterns of local binary pattern. A distance function is used on proposed texture features for effective facial expression recognition. Preprocessing method is also used to get rid of the effects of illumination changes in facial expression by preserving the significant appearance details that are needed for facial expression recognition. The investigational analysis was done on the popular JAFFE facial expression database and has shown good performance.

K. Srinivasa Reddy, E. Sunil Reddy, N. Baswanth
Weitere Informationen