PREPARE@u® | FOSET Conferences

Application of Machine Learning for Atmospheric Characterization Using University of Calcutta ST Radar

Arkadev Kundu — 2022-09-22

An active phased-array Stratosphere-Troposphere (ST) VHF Doppler Radar operating at 53 MHz is being installed at Ionosphere Field Station (22.94°N, 88.51°E, and 34°N geomagnetic latitude) of University of Calcutta in the eastern part of India adjoining northern Bay of Bengal. This radar is unique, being the only one operational at this frequency in the entire eastern and north-eastern part of the country and also in the south-east Asian longitude sector. It is designed to measure atmospheric winds, which could be very severe particularly during tropical cyclones which often affect this region. In this paper, efforts have been made to develop machine learning solutions that are customized to the needs of weather and climate modelling. With this and other radars in India, models of various atmospheric parameters may be developed using measured data over a period of time as training dataset. Applying this model, it can predict the values of the atmospheric parameters for a period outside the training interval. So, the model outputs may help override huge infrastructure and resources necessary for establishing radars.

Elimination of Anionic Dye from Wastewater through Coagulation/Flocculation process

Sambarna Ghosal — 2022-09-22

Effluents discharged from various textile industries are responsible for one of the major environmental pollution. Textile wastewater contains a variety of components some of which are exceedingly toxic to the environment. Several wastewater treatment technologies have been used from time to time by the researchers. Coagulation-flocculation has always been proven an effective treatment technology for removing dye pollutants. This research was conducted to study the maximum removal efficiency of the anionic dye from synthetic wastewater for the best coagulant or coagulant-flocculant combination. Comparing the results among the different coagulants and flocculants with same dosages it was observed that at 400 ppm concentration and at pH 7, polyaluminium chloride alone gave the best possible colour removal efficiency of 99.7%. Although both sodium alginate and sodium aluminate showed significantly good results when used as flocculants. Colour removal efficiency was also investigated for pH variation. Congo Red being an anionic dye showed maximum colour removal efficiency at acidic to neutral pH i.e., pH 5-7.

The Flip Side of Renewable Energy-A Critical Review

Sampa Chakrabarti — 2022-09-22

The ever-increasing world energy demand cannot be satisfied with fossil fuels; because conventional energy sources are finite, and they are exhaustible. Alternatives are required to stop climate collapse and wars for natural resources. Solving the energy crisis is challenging for scientists and engineers. Clean and inexhaustible renewable energy sources like solar, wind, geothermal, biomass, fuel cell,or hydro energy are the sources of alternativeenergy.However, there are a few issues associated with each of the renewable energy sources. Jumping into such alternatives without being aware of their inherent problems may cause serious concerns in the future. The present review compiles the problems associated with renewable energies and their possible solutions and directions of research.

Combined Economic and Environmental load dispatch using multi objective Tunicate Swarm Algorithm

Soumyadip Roy — 2022-10-11

Economic utilization and environmental impact are contradictory objectives in the problem of power generation scheduling where sustainable development of a country is partially depended on these two. The objective of this paper is to schedule the output of committed generating units where both cost and emission of electric power generation have been minimized without shedding any load and satisfying all units and system equality and inequality constraints. In this paper, a new swarm behaviour-based meta-heuristic technique named Tunicate Swarm Algorithm (TSA) has been implemented to solve this problem and the efficacy of this algorithm has also been compared. A six generators system connected with four buses has been considered in this paper as a test model.

Insects Sound Classification with Acoustic Features and k-Nearest Algorithm

Sandip Basak — 2022-10-11

Monitoring particular species and the health of the entire ecosystem necessities determining the existence and number of insects. Several insects are most easily detected by their sounds, and thus, passive acoustic monitoring is highly appropriate. However, practical constraints including the requirement for a human setting, dependency on example sound libraries, low accuracy, low robustness, and limited ability to generalize to novel acoustic situations frequently prevent the advancement of acoustic monitoring. Here, we report outcomes from a collaborative data challenge. The study has utilized improved acoustic monitoring datasets, summarizes the machine learning methods put forth by challenge teams, and carry out extensive performance analysis. The study includes different machine learning models and the study has found 85.4% accuracy from K nearest neighbor method. In the future, this study will be extended to remote monitoring projects. The study also needs to validate more sound features with the help of modern artificial intelligence models.

Indian Vegetable Image Classification Using Convolutional Neural Network

Projjal Sarkar — 2022-10-11

– This paper represents a method of recognition and classification of vegetable images based on deep learning. Using the open source deep learning framework , the improved CNN network model was used to train the vegetable image data set. The classification accuracy of the CNN network is improved due to the hyperparameter tuning and layer addition. This study introduce a new, high-quality, dataset of images containing vegetables. This study also present the results of some numerical experiments for training a neural network to detect vegetables. The methods surveyed in this paper can distinguish among different kinds of vegetables in terms of their color and texture. In the future same experiment is going to be done and we will predict different types of vegetable’s disease. Along with local feature detection techniques, computer vision and pattern recognition are developing quickly.In this study, we extracted and learned the object to train the Deep Neural Network (DNN) for object category recognition. In order to recognize vegetable objects, we used deep learning and investigated the convolutional neural network (CNN). According to the evaluation findings, 3 million iterations were sufficient for the CNN vegetable identification learning process..The recognition rate was 97.50 %.

Classification of Indian Herbal Leaf with Random Forest Classifier

Shobnom Dasgupta — 2022-10-11

Herbal Leaf is normally used for preparing medicines. This includes images taken from the locality of West Bengal mainly Kolkata, with white background. By taking into account this problem of image classification, this research tries to identify herbal leaves based on the images taken on white background. Therefore, this research includes an image processing algorithm and Otsu segmentation. Then, the leaf image features are identified, based on the characteristics of the shape and texture. Herbal leaf shape and color features produce high accuracy only when both are applied at the same time. In this research, morphological features were used for shape feature extraction. The contribution of this study is using the proposed image enhancement and segmentation algorithm so that the features of the image can be extracted based on shape and color descriptors. The classification accuracy in this study has reached 97.8% with Random Forest Classifier.

Review of an efficient scientific calculator

Chandrima Ghosh — 2022-10-11

Scientific calculators are being used widely in our modernized education system. It has made our daily life effortless and also acts as the principal key for calculations. With the acceptance of mathematics in science and technology, it required the calculation to be speedy and accurate. Scientific calculators are multi-purpose; it is functional in every field of science. So, every human must be proficient at computing skills. This research has described the successful design and development of a calculator program. This report outlines the design and development of a computer software system with the help of Visual Studio code. The program is written in C language. Basically, all basic operations including all trigonometric functions can be done in this calculator.

DivyaPay: Bridging the Digital Divide

SOURAV BISWAS — 2022-10-11

The last two decades have witnessed remarkable technological inventions and advancements for solving the problems of humanity, thereby boosting the economy of the nation. Unfortunately, the avenue that deals with mitigating the hassles and challenges faced by people with disabilities or divyangjan, particularly in the financial field, is less explored. Nowadays, online transactions have become a necessity with the introduction of the Unified Payment Interface System, but people with physical disabilities lack the privilege to enjoy such a facility. This work proposes a system for divyangjans named DivyaPay, which is a Smartphone application for online payment built using Google’s Flutter Framework and written using Dart Language. The App has primarily two interfaces for two different categories of disabilities: Biometric Authentication for a Visually Impaired person and FaceID Authentication for an Amputee. The presence of a haptic feedback system provides a better user experience. The smart voice navigation facility helps the visually impaired person to navigate through the App easily. Various payment APIs like Razorpay, Stripe, etc. can be integrated for a faster, smoother, and secured payment transaction. Thus, the proposed system, which targets to fulfill United Nations Sustainable Development Goals of Decent Work, Economic Growth and Reduced Inequalities, promotes the vision of Digital India amidst the physically disabled community by aiding them in performing cashless transactions safely and in a secured manner.

Pandemic Survival

Siddhanta Debnath — 2022-10-11

Abstract - Games are usually considered as a trite activity for the youth. However, the successful researches in game design and development over the last few decades have revealed that they can prove to be valuable tools for growth and connection. Exploring the ways digital games can be used as tools for connection has gained increased attention in recent times as a result of the COVID-19 pandemic. This paper proposes a two-dimensional Augmented Reality game titled “Pandemic Survival” that helps in spreading awareness about the effects of Coronavirus on the physical and mental health of an individual. The primary aim of the proposed work is to educate people about safety precautions to be taken to deal with the pandemic while staying indoors or travelling outdoors. Through various gameplays and element design, the utility of maintaining social distancing, using masks and sanitizers, taking dosages of vaccinations etc have been implemented so that players, especially youth and children tend to learn and acquire knowledge of fighting back with the virus attack in a fun-filled manner.

Optimizing the fuel required by a vehicle using Taguchi

Bikash Banerjee — 2022-10-11

The requirement of fuel is essential for everybody to maintain their daily lifestyle. Petrol being a major area of concern in recent times due to its scarcity soon, it is the responsibility of all to use this in an optimized manner. A broad transport network within India results in huge demand for petrol due to which optimizing the use of petrol without giving up on the need is the goal. This goal can be brought through this fuel optimization technique. The main aim of this research was to find out the fuel economy of a vehicle by optimizing the factors influencing the fuel economy. Factors that were considered were the speed at which the vehicle will move, the load that is the number of people on the vehicle, and the tire air pressure of the vehicle. By differing the factors according to the requirement and investigating with the help of S/N ratio signal to noise ratio and regression analysis we have tried to optimize the fuel requirement without sacrificing any of our needs which in turn will also help us economically in a long run.

Sustainability Assessment and Reuse Feasibility of Stabilized Legacy Waste in West Bengal, India: A Circular Economy Transition Perspective

Nabanita Ghosh — 2022-10-11

Present solid waste management policies around the world demand higher recycling and reuse rates. Thousands of uncontrolled dumpsites existing worldwide either under operating or closed condition are deemed not only to be the major sources of environmental pollution and nuisance but also occupying prime land that could be used for other purposes while containing useful materials. Reclamation of the dumpsites containing massive share of residual garbage is therefore an environmentally sensitive issue that continues to be confronted by developing nations, particularly by local metropolitan authorities in India. To address this problem, one scientific technique, i.e., the biomining approach, which refers to digging and shifting legacy garbage from dumpsites and recycling or creating energy from recovered materials, reclaiming land space, and rehabilitating/redeveloping contaminated sites could be explored. The article presents a review on feasibility studies carried on dumpsite mining operation specifically to recover legacy waste and its reuse. Assessment of sustainability of biomining process in terms of not only the direct costs and revenues but also social benefits is discussed in details under the purview of the present study.

Design Approach of Human Thought Controlled Empathy Device with Brain Signal

Susmita Das — 2022-10-11

The main objective of the human thought controlled empathy device is to capture and store different emotional variations of the brain and to feel the same by another person. Human emotion is very difficult to determine just by looking at the face and also the behavior of a person. This study aims at EEG (ElectroEncephalogram) signals of relations of human emotions. The main objective of this recognition is to develop empathy sensing device. The research methodology is divided into four categories such as both visibility and EEG data to extract at the same time from the respondent, the process of complete data record including the capture of images using the camera and EEG, pre-processing, classification and feature extraction. These are done at the same time to gather information about the features extracted to be classified using emotional intelligence techniques from emotional faces.

Prototype Development of Swarm Bot for Defense and Medical application

Susmita Das — 2022-10-11

Swarm bot is an approach to the coordination of multiple bots as a system which consist of large numbers of mostly simple physical bots. It is supposed that a desired collective behavior emerges from the interactions between the bots and interactions of bots with the environment. This approach emerged on the field of artificial swarm intelligence, as well as the biological studies of insects, ants and other fields in nature, where swarm behaviour occurs.

Agrijaanch

Chirantan Das — 2022-10-11

With the advancement in agricultural technology and the use of Artificial Intelligence in diagnosing plant diseases, it becomes important to make pertinent research for sustainable agricultural development. We have proposed and built a Web-based system to spread awareness amongst Farmers and Agro-companies about various diseases that plants are infected with and also their possible remedies. Different diseases have different causes and thus different solutions and mitigation strategies need to be adopted. If the diseases are misjudged or mismanaged, then they can get spread profusely and wreck havoc on the soil. So we came up with a potential approach for effective, efficient and automated detection of the prevalent diseases during the budding phase of the agricultural products to assist the farmers in employing preventive measure on time. We have developed a set of both Machine Learning Models and Deep Learning Models to detect diseases in the potato plant. The models are deployed on our website and when the image of a Potato plant is uploaded, the result that whether there is any disease in the plant or not will be generated. Amongst all models trained and tested in this work, Convolutional Neural Network yields the best classification accuracy of 99.30%.

DEVELOPMENT OF FACE RECOGNISABLE AI ROBOT

Sanghamitra Layek — 2022-10-21

In today's world, Facial recognition by image processing is one of the most and advanced and recent foundations. Facial recognition is a fundamental step in approaching various security systems, data sets for identification of an individual, law enforcement, personal safety, biometrics, etc. This model has been developed can detect a face (known and unknown) among other objects from an image and recognize the faces by matching it with a previously made database using Neural Network. This system will have a predefined set of data. A face, when detected, will be matched with the dataset to find a resemblance. In this paper, A robot has been designed in such a way that it can recognize and track faces and then mention the person's name and also can interact with people via speeches. It can also show live video footage to the controller person`s device.A system has been designed using Cascade classiﬁcation and Local Binary Pattern Histogram (LBPH) Face Recognizer method based on OpenCV library and Python Language. The movement of the robot is controlled by Nodemcu microcontroller board and is wirelessly connected to controlling person`s smartphone via inbuilt WI-FI modules of these two devices. The controller person can control the robot via a Nodemcu controller Application. It can be used in various fields like surveillance purposes to experiments purposes.

A CRITICAL REVIEW OF DISPLACMENT CURRENT

Soumyadeep Ghosh — 2022-10-21

In the field of physical science, the concept of displacement current of James Clerk Maxwell is considered as one of the most innovative concepts ever through which discovery of the electromagnetic theory leads him. In the Maxwell-Ampere law, the derivation of electric intensity shows that the variation in electric field can generate a magnetic field along with displacement current. During the development of its own concept, along with applause for the scientific content, they would also receive Criticism on it.

High Gain Appendage Loaded Dipole Antenna with Corner Reflector

PRABAL GHOSH — 2022-10-21

A technique for performance enhancement of a simple wire dipole antenna at 2.4 GHz is demonstrated in this paper. Initially the operating bandwidth of the antenna is improved by extending the width followed by addition of appendages to the antenna elements. Thereafter the antenna gain in a specific direction is enhanced with the use of metallic corner reflector with 90⁰. The antenna bandwidth (-10 dB) is improved up to 30.63% with the appendages. The proposed corner reflector is composed of two conducting plates at an angle of 90⁰. The corner reflector is placed at a specific distance from the antenna and the corresponding parametric study is also included in the paper. With the inclusion of the corner reflector the gain is increased up to 32dB in the desired frequency band. Besides, directivity of the antenna is also enhanced. The proposed antenna being a high gain antenna is suitable for various wireless communication applications.

Modelling Circular Shaped High Gain Patch Antenna for Biomedical Application

Arunava Roy — 2022-10-21

With the rise in healthcare industry and biomedical engineering, biomedical telemetry has enjoyed a rich diversification in research and patents. Implanted Biomedical Devices (IMD) have become immensely popular and these devices play a vital role to monitor the patient with the help of wireless telemetry. While there has been progress and various research works has been carried out in the field of wearable antenna for biomedical purposes, few are convenient and easy to build and fabricate. In our work we will try to find and explore the most convenient and simple process to build a high gain microstrip patch antenna either by using superstrate, metamaterial, FSS or by building a patch of different shape. For building a biomedical antenna, we have to take into account the size, weight, radiation, material of the antenna, and which part of the body it will be mounted on. Taking everything into consideration, our aim is to design a simple, non-complex, convenient high gain patch antenna for biomedical purposes.

Restoration and rejuvenation of water bodies with bio enzymes produced from Citrus Fruit peels

Shaoli Das — 2022-10-21

- In this present article, we attempt to demonstrate how to restore and rejuvenate water bodies using bio-enzymes. We have prepared bio-enzyme from Citrus Fruit peels and tried to explore how it affects water quality parameter indices, such as, hardness, alkalinity, chloride ion concentration etc. Water quality parameters are measured for water samples with and without the bio-enzyme at different time intervals. Our study shows a marked improvement in the water quality parameters that not only reveals the potential use of the bio-enzyme on water treatments but also enables the whole treatment to be cost effective and eco-friendly. Moreover, the organic residue derived during the preparation of bio-enzymes can be further used as good quality manure which can improve the soil quality instead of using harmful chemical fertilizer and thus demonstrates an efficient and effective way towards the management of organic solid wastes across the ecosystem.

Classification of Animal Images with modified CNN model

Nanda Kishor Jana — 2022-10-21

In the age of modern artificial intelligence, new methodologies are evolving day by day for species classification. There is some need for categorization of the animal for the preservation and restoration of species. Due to this, there are several techniques for the identification of animals out of the deep learning methods that are most useful for animal classification for their images. In this study, the modified and improved convolutional neural network (CNN) has been employed. This study is the identification methodologies of the ten animals which are cat, dog, tiger, lion, sheep, rhino, cheetah, elephant, squirrel, and panda. These animals need to be identified by the artificial intelligence-based system for the large-scale preservation system and the accuracy obtained by the modified CNN is this. In the future, this study is going to evolve deep learning for human-less classification systems and this study will maintain the balance between the machine and animals in restricted areas that humans can’t reach there. Animal classification is one of the core problems in Computer vision. A lot of attention has been associated with Deep Learning, specifically neural networks such as CNN. This animal classification model gives an accuracy of 95%.

Exploratory Data Analysis on Healthy Lifestyle & World Happiness Report in Asia & Europe

Karnapali Mukhopadhyay — 2022-10-21

State of mind and lifestyle have been affected due to COVID -19 pandemic. There has been a certain impact on daily lifestyles. There is one the metrics like happiness index, to be one portion to exhibits that each country has an indicator which affects each other. This indicator is required to understand the impact of the happiness index and lifestyles. In the dataset, it is necessary to explore, analyze, and visualize to give knowledge about the citizens of Asia & Europe. Exploratory data analysis (EDA) has been utilized for further investigations. These works explore happiness reports and lifestyles in the Asia & Europe region using a combined dataset of World Happiness Report 2020 & Healthy Lifestyle report 2021. The result describes and discusses the datasets using a table with correlation score, correlation analysis, and heat map-plotting visualization. In the future, this study will further investigate different regions with artificial intelligence models.

Blind o Techmate: A technology for the vision-impaired

Sumana Saha — 2022-11-10

We use and experience technology every day, every moment. Computers and the internet have shaped the way 21st-century will be remembered in the pages of history. Unfortunately, many of us are still unable to access technology independently due to vision impairment. Devices have become more sophisticated and require a proper vision for independent operation. Individuals with vision impairment require help from others to operate devices, which deprives them of privacy and makes them completely dependent. We present this paper with the vision of eliminating the barrier between technology and blind individuals. We have developed an idea of modifying the already existing computers and making them specially built for blind individuals so that they can independently use and explore technology in their way. We intend to present an idea that will bring technology closer to them and enable them to interact with the technology independently. We believe that technology must be accessible to everyone irrespective of any disability.