VOLUME 12, ISSUE 2, FEBRUARY 2023

Abstract: Cybersecurity is a critical issue for higher education institutions, as they store and manage a vast amount of sensitive data, including student records, research findings, and financial information. Higher education institutions are also increasingly reliant on technology for teaching, learning, and administration, which means that they are vulnerable to cyber threats such as hacking, phishing, and ransomware attacks. One of the key challenges in protecting higher education institutions from cyber threats is the fact that they often have complex and decentralized networks. This can make it difficult to implement and enforce consistent security measures across the entire organization. Additionally, higher education institutions often have many stakeholders, including students, faculty, staff, and external partners, which requires efforts to coordinate and manage information security.

Keywords: Cyber Security, Cyber security awareness, Information security.  

Abstract: Nowadays, the insulated gate bipolar transistor have been more useful in many applications as in HVDC systems, in FACTS systems and in many devices of power electronics. Comparing to this device to power MOSFET and bipolar transistors, it has the advantages to operate at much higher current density and higher blocking capability. This paper presented inverter losses calculation, a LC filter design for Three Phase Full Bridge IGBT based inverter. The filter is used while simulating in Matlab simulink a Three Phase Full Bridge IGBT based inverter to evaluate the Total Harmonic Distortion using the Fast Fourier Transform analysis and inverter loss calculation. Before filtering the THD is 48.62% and after filtering the THD is equal to 0.70%. The inverter LC filter parameters designed are the cutoff frequency of 400 Hz, the inductor of 3.105 mH and the capacitor of 3.22 10^-5 F. The efficiency of the filter is very high as it reduces the total harmonic distortion of the inverter to less than 5%.

Keywords: IGBT inverter, THD, FFT analysis, LC Filter, IGBT losses.

Abstract: Conversations between those who have trouble speaking and the normal person have always been a difficult task. Gestures and sign language are considered as the most natural expressive way for communications between dumb and normal people. By adopting tactics that involve glove-based hard of hearing correspondence systems, this smart glove will allow vocally handicapped persons to communicate with normal people using their sign language. Every glove contains a signal within. The gesture module adjusts the flex sensor and accelerometer resistance relative to each unique gesture. This device is controlled by a microcontroller that has been Python-programmed. The structure also incorporates a text-to-talk converter that converts the intended actions, such text-to-voice output through a speaker, and also outputs the same in the phone over Bluetooth. With automatic message notification to the concerned, this prototype also has features for health monitoring and detecting human body temperature and heart rate.

Keywords: Flex sensors, Bluetooth Device, micro-controller, Internet of Things, Health Monitoring, Raspberry Pico, LCD Display, APR 9600, Temperature Sensor, Heart Beat Sensor, Speaker, Analog to Digital Convertor, MEMS acceleration censors

Abstract: One of the environmentally friendly methods of intelligent agricultural production is the greenhouse. It is viewed as a different approach to dealing with the food issue brought on by accelerating population expansion, climate change, and environmental pollution. Off-season crops can be supported indoors using this method, even in regions with extreme climates. The crop parameters in a greenhouse must be accurately and safely controlled and maintained. There are already sophisticated methods for automating greenhouse farming practices like managing the internal atmosphere, controlling watering, and monitoring plants.

Keywords: Greenhouse, off season crops, microcontroller, protected environment, Sensors.

Abstract: Artificial intelligence (AI) is crucial in promoting Industry 4.0 worldwide. AI has the potential to revolutionize the engineering and construction industry by automating tasks, improving project efficiency and accuracy, and enabling new capabilities. One application of AI in engineering and construction is in the design and planning phase of projects. AI algorithms can analyse data from previous projects and make recommendations for optimal designs, materials, and construction methods. This can lead to cost savings and improved project outcomes. AI can also be used in the construction phase to assist with surveying, quality control, and equipment maintenance tasks. For example, drones equipped with AI can survey construction sites and generate accurate 3D models, which can be used for progress tracking and identifying potential issues. Another area where AI can have a significant impact is in operation and maintenance of buildings. AI-powered building management systems can optimize energy usage, detect and diagnose equipment malfunctions, and predict maintenance needs. Overall, integrating AI into engineering and construction can improve project efficiency, reduce costs, and increase the safety and reliability of projects. However, it is essential to consider the ethical implications of using AI in the industry, such as potential job displacement and the need for proper training and oversight.

Keywords: AI, Construction engineering, Supply chain, Automation, Quality control, ROI

Abstract: People now commonly suffer from chronic kidney disease (CKD). By detecting and treating those who are at risk for this condition as soon as feasible, a variety of serious problems, such as end-stage renal disease, elevated risk, and cardiovascular disease, may be prevented. Medical researchers can get a lot of help from the machine learning algorithm in accurately diagnosing the disease at the very beginning. Algorithms for machine learning and Big Data platforms have recently been combined to improve healthcare. This work presents hybrid machine learning methods that integrate extraction of the feature strategies and various algorithms of machine learning under classification technique related to massive data platforms to identify chronic kidney disease (CKD). In this study, logistic regression (LR), random forest (RF), decision tree (DT), support vector machine (SVM), Naive Bayes (NB), and gradient boosted trees were employed as six ensemble learning strategies for machine learning classification tasks (GBT Classifier). The results were validated using four evaluation techniques: accuracy, precision, recall, and F1-measure. The results demonstrated that the chosen features had helped SVM, DT, and GBT Classifiers operate at their peak levels.

Keywords: Chronic kidney, Naive Bayes (NB), decision tree (DT), logistic regression (LR), Gradient- Boosted Trees (GBT Classifier) and Random Forest (RF).

Abstract:  Food is a fundamental requirement that gives our bodies the nutrition they need. Given that food grains are being adulterated at an increasingly rapid rate, food quality is the most important thing to be examined. The current quality assessment process is laborious and prone to human mistake (unknowingly or intentionally). This will have an impact on the farmer who depends on the farm for his daily sustenance because they don't receive a fair price for their years of combined labour. Additionally, Manual Assessment promotes the adulteration of food grains, misleading consumers by combining inferior grains or compounds that mimic grains while generating high margin profits.

Keywords: Grain adulteration, Pre processing of image, Brightness equalization, Edge detection, Image segmentation, Feature extraction of image, Color feature extraction, Classification of Extracted image,Support Vector Machine (SVM), k-Nearest Neighbor (k-NN)  

Abstract: Methodologies for their proof of identity have been researched, and an automated system for bird species reputation has been created. Invariably trying to identify bird calls without human intervention has proven to be a challenging and time-consuming task for extensive research in ornithology's taxonomy and various other subfields. An identity process at the level is hired for this venture. The first step involved constructing a perfect dataset with all of the sound recordings from different chicken species. The final step involved applying a variety of sound preprocessing techniques to the audio clips, including-emphasis, framing, removal of peace, and re-construction. For each and every constructed audio clip, spectrograms were produced. The second process entails establishing a neural community with the spectrograms as its input. based

Keywords: Deep Learning, Neural Networks, Image Processing, Convolution Neural Network

Abstract: There are numerous online services available in almost every business thanks to contemporary web technologies. Every major sector is transforming and building a digital face for all of its core activities to stay competitive in the expanding digital sector. The globe today has an extremely rapid information flow, therefore adopting duplicate methods won't help the individual or the organization. Internet connectivity is a must for all contemporary enterprises that want to operate effectively. One of these fields where intelligence should be swiftly and efficiently digitised is healthcare. This study focuses on that particular topic and paves the way for the creation of software that makes the switch from notebook paper to electronic documents easier. The paper proposes Advance Healthcare System, that would enhance patient operational effectiveness, physician routing efficiency, and provide ubiquitous access to patient-related information throughout the hospital. It also defines a theory for a browser framework that would replace the requirement for paper prescription medications in hospitals. Advance healthcare System alters the standard clinical structure in a comprehensive approach, improving the quality, accessibility, and individualization of healthcare. In this study, we basically identify the key technologies that enable the concept of the advanced healthcare system. We speak of the issues with Advance healthcare System that are currently present and make some suggestions for remedies. Finally, we consider the prospects for Advance healthcare in the future. Keywords: Advance healthcare System, Database, Webpage, Interface, Hospital Information Systems, Environment-Shaping Factors, Healthcare Intelligence Services.

Abstract: The major objective of our project is to lessen movie piracy. The most common method of film piracy is to record a movie being seen in a theatre with a camera, edit the footage, and produce a better image. In order to reduce and stop movie piracy, the goal of our initiative is to stop people from recording videos of films that are being viewed in theatres. Even if we are unable to completely eradicate it, we can lessen the costs brought on by piracy. We employ infrared radiations for this purpose since IR rays have the ability to be recognized by cameras but not by human eyes. Therefore, we employ this characteristic of infrared to stop the camera from recording the film. Given that direct infrared is harmful to humans, In the upper and lower portions of the bandwidth, we use the near-infrared spectrum.

Keywords: Infrared light; RFID; GSM; GPS;

Abstract: The counting of blood cells plays a very important role in the health sector. The old conventional method used in hospital laboratories involves the manual counting of blood cells using a device called a hemocytometer. But this process is extremely monotonous and time-consuming, which leads to inaccurate results. In order to overcome these complications, this project presents an automated software solution, enriched with image processing and machine learning techniques, to detect and count the number of RBC, WBC, and platelet cells in the blood sample images and to classify diverse types of leukemia. This approach identifies various color feature statistics with geographical measures for machine learning centered on supervised learning.

Keywords: Counting of blood cell, Health sector, image processing, RBC, WBC, Platelets cells.

Abstract: Intrusion detection and prevention systems (IDPS) are an essential tool in protecting modern networks against cyber threats. They monitor network traffic and detect malicious activity, such as malware infections, unauthorized access, and network intrusions. When an IDPS detects such activity, it can take a variety of actions to prevent or mitigate the threat. IDPS systems can be configured to operate at various points within a network, including at the perimeter, at key servers, or on individual devices, allowing for a multi-layered approach to security. There are several types of IDPS technologies available, including signature-based systems and anomalybased systems, which use machine learning to identify deviations from normal network behavior. IDPS systems are an important part of a comprehensive network security strategy, but they are not foolproof and must be carefully configured and maintained to ensure optimal performance. It is also important to use IDPS in conjunction with other security measures, such as firewalls and user training.

Keywords: SOC (Security Operation Center), IDPS (Intrusion Detection and Prevention System), IDS (Intrusion Detection System), NIC (Network Interface Controller), G-IDS (Generative adversarial network – IDS), CPS (cyber-physical system)

Abstract: Since its inception by Google, Federated Learning (FL) has been instrumental in improving the performance of a wide range of applications. Android's Gboard for predictive text and Google Assistant are two of the most well-known and widely used FL-powered applications. FL is a configuration that enables on-device, collaborative Machine Learning. A diverse body of literature has investigated FL technical considerations, frameworks, and limitations, with several works presenting a survey of the prominent FL literature. Prior surveys, however, have focused on FL's technical considerations and challenges, and there has been a limitation in more recent work that presents a comprehensive overview of FL's status and future trends in applications and markets. We introduce the fundamentals of FL in this review, describing its underlying implementation of technologies, pros and cons, and recommendations along with privacy-preserving methods. More importantly, this work contributes to the understanding of a wide range of FL current applications and future trends in technology and markets today.

Keywords: Federated Learning, Cybersecurity, IoT, Decentralized networks, NIST.

Abstract: In opposition to crime and criminals. The police are becoming less willing to respond to crime scenes unless there is visible confirmation, either by manned patrols or by electronic images from the surveillance cameras. The current systems do not classify routine and abnormal events.The proposed work is used for a variety of reasons, including live tracking, monitoring, classifying weaponry, and surveillance. In this work, real time image processing techniques are used to extract live surveillance footage from monitoring and identifying unusual events.The proposed project contains three processing modules. The first processing module uses Convolutional Neural Networks (CNN) for object identification, the second processing module handles the classification of weapons, and the third processing module handles monitoring and alert functions. A circular area will be monitored by CCTV, which will operate and be managed automatically. Before being implemented in such an environment, shape detection algorithms and object detection algorithms have been tested for accuracy in detection and analysis of processing time. The results provide the best accuracy in matching weapon and object types with names and shapes in predefined databases like ALEXNET. The proposed work will significantly lower crime rates, increase security in some regions, and shorten the time it takes to apprehend offenders.

Keywords: Convolution Neural Network (CNN), Video Surveillance, Voice Assistance, Weapon Detection, Faster Region based Convolution Neural Network (RCNN),picture segmentation.

Abstract: Researcher’s efforts to create a clever and intelligent directing mechanism that can function in both indoor and outdoor settings for persons with intelligent eyes have been hampered over the past few decades by advancements in the field of steer and routing devices. The equipment’s is utilized to recognize items. We also use an ultrasonic sensor mounted on a servomotor to gauge the distance between objects and the persons wearing the prothetic eye. When compared to earlier systems, the given research recognizes the barrier with a noticeably high efficiency while just required straightforward computations for its implementation. The goal of this study is the creation of a tool that will aid stone-blind people and provide an efficient answer. Independent mobility and navigation are difficult for the blind. Daily tasks are hampered.

Keywords: Include Bluetooth, Cane module, Digital compass, IR ranging sensor, PIC microcontroller, Voice chip, Smart system, visual losses, biomedical sensor, tensorflow, Viola Jones, Ultra sensors, deep learning, obstacle detection and object recognition.

Abstract: Depression is a major illness that is found in people of all the ages. It can be caused due to a various of reasons. This kind of illness needs constant supervision which is a major problem. This tends to cause multiple chronic issues that needs to be taken care of with constant monitoring of the patient. In this paper we have proposed a system which monitors the depression levels of the patients using the application that is being developed which records the video clips of the patient periodically and are processed using image processing, Deep Learning and Neural networks by this it concludes the depression levels. The ECG (Electrocardiogram) system constantly monitors the ECG graph and reads the patient’s heart condition and BPM (Blood Pressure Monitoring) sensor will monitor the heartbeat rate. The location of the patient can be traced via GPS in emergency situations.

Keywords: ECG (Electrocardiogram), BPM (Blood Pressure Monitoring), GPS (Global Positioning System), Deep Learning, Neural Networks, Image Processing

Abstract: From a technological standpoint, Blockchain creates intriguing study topics since it allows for confidentiality, anonymity, and data integrity without the need for a third party to oversee transactions. A large number of companies now use blockchain, a decentralised transaction and data management system that were initially created for the Bitcoin cryptocurrency. In this article, we'll focus on the uses and contributions of blockchain technology in finance generally, as well as on places where the technology might have a bigger impact on payment systems. In addition to providing a complete analysis of blockchain technology and cryptocurrencies, the authors look at the successful uses of blockchain technology in a number of financial sectors, including cryptocurrency. The technical studies on the price behaviour of bitcoin are carefully examined by the authors. As the first effective implementation of a blockchain, cryptocurrencies.

Keywords: Blockchain, Integrity, Confidentiality, Anonymity Cryptocurrency, Technology.

Abstract: Human survival depends heavily on plants since they give us air, food, clothing, fuel, medicine, gums, and environmental protection. Numerous plants have a high therapeutic potential and are rich in active components. There are numerous useful plant species now going extinct and destruction occurring as a result of reasons like climate change, population growth, professional secrecy, a lack of governance for research activities and a lack of understanding of medicinal plants. The process of manually identifying therapeutic herbs takes a lot of hours, so professional assistance is required. Automatic classification and medicinal plants identification are required to solve this issue for the benefit of humanity as a whole. Image processing research is currently focused on the computerized grading and naming of therapeutic plants. The primary processes in the identification of medicinal plants and the classification process, which have an impact on the overall accuracy of the classification system, are feature extraction and classification.

Keywords: Medicinal leaves, Convolution Neural Networks (CNN), Ayurveda, image processing, Machine learning

Abstract: This article addresses the use of extraction of characteristics for facial expressions in conjunction along a neural network to identify various sentiments (happy, sad, angry, fear, surprised, neutral etc..). The rapid advancement of artificial intelligence has made significant contributions to the field of technology. One of the most crucial aspects of human communication that aids in our understanding of what the other person is attempting to say is the expression on their face. Only one-third of the message is understood vocally, and the other two-thirds are understood nonverbally. There are numerous face emotion recognition (FER) systems in use today, however they are ineffective in real-world situations. Despite the fact that many assert that their system is nearly perfect. The testing data then examines the information and its classification report names the testing data and indicates how accurately it was classified. For better data categorization, many strategies are used, modifying the images using the Histogram of Oriented Gradients (HOG) and Discrete Wavelet Transform or passing the training images through a Gabor filter (DWT).The training images are first run through The Histogram of Oriented Gradients (HOG) produces the best results to date, with an average precision of 92%.

Keywords: Facial Expressions, Face Emotion Recognition (FER), Histogram of Oriented Gradients (HOG), Discrete Wavelet Transform (DWT).

Keywords: Agriculture, Farming, Crops, Climate, Deep Learning, Neural network, Image Processing.

Abstract: One of the oldest human food sources in the world is agriculture. Weeds are a concern because they crowd out valuable crops and take up space, water, and nutrients. accurate recognition of the undesirable vegetation is necessary for the development of a successful weed removal method. The Arduino Uno, the device that acts as the foundation for the robot's operation, is supplied with software that controls every movement of the robot.. The robot's primary function is to recognise weeds, chop them down using a cutter, and spray pesticides only in areas where weeds are present.

Keywords: Agbot, Weed Detection, Pesticide, CNN.

Abstract: The scope and severity of network concerns about information security have increased over time. Nowadays, most hacking methods target technology from start to finish while also exploiting human frailties. Pharming, phishing, and social engineering are just a few of these methods. One of the aspects of these Using damaging URRs to fool consumers is considered an assault (URLs). This is why identifying malicious Content is a hot issue right now. A variety of academic research has demonstrated several ways to identify malicious URLs using machine learning and deep learning technologies. Based on our hypothesized URL behaviours and characteristics, we provide a machine learning-based solution for detecting malicious URLs in this work. Furthermore, big data technology is applied to enhance the ability to appreciate fraudulent URLs based on aberrant activity. The proposed detection approach consists of a novel set of URL attributes and behaviours, a machine learning algorithm, and big data technologies. The results of the experiment indicate that the stated URL characteristics and behaviour can improve overall ability to identify risky URLs. This implies that consumers may successfully detect risky URLs using the suggested methods.

Keywords: phishing, machine learning, malicious URL detection.

Abstract: Many references support the non-invasive detection of aberrant speech using machine learning function descriptors and classifiers. Deep learning with feature descriptors and time frequency images is a better option. The majority of deep learning frameworks for speech-language pathology use a binary classification model. A network that can identify accurate medical conditions is required to construct a hardware system. It is essential to do a serious examination of time-frequency analysis using advanced deep learning algorithms. Current research is focused on creating a non-invasive, dependable, and computationally expensive architecture for detecting multiclass laryngeal lesions. In a realistic scenario application, compare the performance of a fully linked network versus a completely collapsed deep learning voice denoiser network. Three alternative time- frequency picture corpora are generated in the noise reduction training example.For applying in a realistic environment, the capability of a fully-connected network and a fully convolutional deep-learning voice denoiser network is initially investigated. Denoised training samples are used to create three different time-frequency image corpuses. These multivariate image datasets are used to train three upgraded forms of the state-of-the-art convolution neuron network model using a 3D convolution kernel.

Keywords: Deep neural network , pathology classification , non invasive , random forest algorithm , speech language pathology , CNN algorithm , binary classification model voice dataset , pre processing , vocal issues , detection or identification.

Abstract: Electronic voting or e-voting, has been used in various forms from the 90s. Increase efficiency and reduce errors. However, wide adoption of such systems remains a challenge, especially in terms of improving resilience to potential failures. Blockchain is today's disruptive technology and promises to improve universal flexibility of electronic polling systems. It represents an attempt to exploit the advantages of blockchain for security. The proposed system fulfills the basic requirements of electronic voting systems and achieves persistent verifiability. This wrapper describes the details about proposed electronic voting system and its fulfilment using a multi-chain manifesto. This paper contains a detailed estimation of the system and it demonstrates successfully about the effectiveness in achieving an ongoing auditable electronic voting system.

Keywords: Cryptography, peer-to-peer, deep learning, security, distributed ledger technology, EVM.

Abstract: India is developing nation so government knows to increase economic growth they need to develop startup culture in country and they are taking actions regarding that. Indian government is serious in promoting entrepreneurship at the startup level and has taken a Number of initiatives to make sure appropriate support. There is no any one online platform where we can found multiple startups at one place. Every startup has their own platform or some startup doesn’t have there any online platform. So it is online platform for those who want to start new business as well as for those who want to promote their startup online. It is platform where multiple businesses found at on place. Also using these platform new entrepreneurs can connect with each other. If any investor wants to invest in any startup then this platform will give recommendations about best investments. Any startup wants any professional support then platform will connect them with professionals which has good knowledge of market. This platform will be beneficial for new entrepreneurs. Multiple entrepreneurs can connect with each other using this platform. This platform connect small retailer with wholesaler so it will beneficial for small retailers.so this platform will help any new entrepreneurs to start their business and further help will be provided at one place in platform.

Keywords: Startup culture, Economic growth, Entrepreneurship, Investments, Professionals.

Abstract: Obstructive sleep apnea (OSA) is one of the most significant nap disarray. Person suffering against OSA can not be conscious of  such that it’s air duct is gum up together with it had difficulty in breathing. Due to this , actual-time nap monitoring  is important in everyday existence. In the course of this need, I would like to introduce an unobtrusive wireless sleep monitoring system called Sleep Smart.  Sleep Smart is a smart mattress pad equipped with textile pressure sensors that allow people to monitor their sleep custom with breathing tariff in actual time in its bed along with IoT capabilities. The all-inclusive analysis comprise of three climax: (1) conspiring the mattress pad, (2) carryout gesture survey on the stress data to fish out the live rate, and (3) building an IoT architecture to supply helping hand. The present study shows the favourable current investigation outcome and the goal for time ahead climax.

Keywords: unconstructive nap layoff, electronic element, solid fabric, IoT, implant structure

Abstract: Many nations' economies are based on agriculture, and soil is its most crucial component. There are various types of soil, and each type has various characteristics for various crops. Today, several techniques and models are utilized in this industry to boost the yield of the crops. Therefore, the main goal of this system is to develop a model that aids farmers in determining which crop to plant in a specific type of soil. We are utilizing machine learning techniques in this system to suggest crops based on soil categorization or soil series. The model merely recommends soil types, and depending on those types, it can recommend appropriate crops. Various classifiers are employed in this, and as a result.

Keywords: Machine Learning; Agriculture; Soil types; recommend appropriate crops

Abstract: Lip reading technology uses analysis of lip movement to record the speaker's message. It is used widely in many aspects of daily life. The performance of the entire lip-reading system is impacted by the dataset's quality. As a result, this study investigates the dataset for lipreading. Scikit video is used to extract frames from the source video. Idlib is then used to conduct facial detection. Lip cropping is accomplished by processing the feature points to obtain lip pictures. The dataset is then expanded by doing data augmentation. 33 voices are included in the collection, and each speaker's lips are represented by 7,000 images. A technique for creating datasets is suggested. Prior to decomposing the treated films in the Scikit-Video library.

Keywords: Lip reading, Idlib, Scikit video, lip pictures and lip cropping.

Abstract: Security is among the basic needs of a human being. The requirement for protection has not subdued with time but increased proportional. Because of a lack of resources, effective security cannot be planned. It costs a great deal of money to get sufficient security, which not everyone can afford. The goal of the study was to develop a smart robot that can deliver high-quality safety at a price that, when long-term impacts are taken into account, is significantly less than the current market rate. The plan was to create a rover that would harness contemporary technology to drive independently without human assistance, search the region, and notify the control room as soon as any anomalies have been found. The robot includes GPS.

Abstract: Music classification is a field in the field of Music Recovery and sound signal processing research. Neural Network is a modern way of classifying music. The classification of music using neural networks (NN) has become very successful in past few years. Different song collections, machine learning methods, input formats, and neural network applications are all to varying degrees effective. Spectrograms made from time-slices of songs are input to a neural network in order to classify songs into the appropriate musical genres. The Neural Network (NN) employs spectrograms generated by time song slaves as an entry to classify songs into their numerous genres. The Convolutional Neural Network (CNN) audio signal input system will employ the generated spectrograms. Tasks involving picture pattern recognition are handled by CNNs. Acoustic feature extraction is the most important process while evaluating music. Models are trained using the GTZAN dataset in the suggested system.  

Keywords: Deep learning, spectrogram, music, classification of music genre, Convolution Neural Network (CNN).

Abstract: Driver intoxication and driver tiredness are two of the most common causes of human-centered accidents, which are gradually on the rise. In order to assure safety and lower accidents brought on by drowsiness and alcohol consumption, researchers have recently revealed approaches that can identify fatigue by examining facial expressions. On the other hand, modern gadgets can only inform the driver when they sense sleepiness. The detection technique is typically divided into two parts, such as identifying the driver's facial expressions for indicators of fatigue and educating them further. The existing models are therefore unable to perform any further safety measures to ensure greater safety if the driver is still unable to control the vehicle after sounding an alarm.

Keywords: Drowsiness Detection, Haar Classifier, Hypo-vigilance.

Abstract: Being confronted by ailments such as blindness, hearing loss, or deafness is becoming more common. Individuals have become increasingly dependent on comfort as a result of science and invention, yet there is a group of disadvantaged people working to develop an innovative way to make communication simpler for them. The Globe Health Organization estimates that there are roughly 285 million blind individuals, 300 million deaf people, and 1 million mute persons in the world. The primary forms of communication for individuals with disabilities are sign language and braille. Without a translation, it can be difficult for people with impairments to communicate with others. Because of this, implementing a technology that understands sign language would significantly and favourably affect the social lives of impaired individuals. In this research, we've put forth a visual, marker-free system to recognise Indian Sign Language as a portable communication tool for use with or among impaired individuals.

Keywords: Braille, Sign Language, BOF, SVM.

Abstract: The city must be kept clean by maintaining the sewage system. Drainage obstructions are caused by irregular sewage system monitoring. Flooding and pollution in the sewers are frequently brought on by blocked sewer systems. Due to their lack of knowledge about the conditions inside the manhole, workers run the risk of being hurt in an accident. Utilized in this model include Regulator circuits, sensor driver circuits, microcontrollers, serial communication devices, and IoT modules are all utilised to receive the required output from the IoT module. One of the most common issues with the sewage system is overflowing drains, which gets worse during rainy seasons during which the authorities are not aware of the backed-up drains. It leads to waterlogging, which breeds bugs, and is unhealthy for the nearby population.

Keywords: Internet of Things, sensors, motion detection, garbage overflow detection, manhole monitoring

Keywords: Include Bluetooth, Cane module, Digital compass, IR ranging sensor, PIC microcontroller, Voice chip, Smart system, visual losses, biomedical sensor, tensorflow, Viola Jones, Ultra sensors, deep learning, obstacle detection and object recognition.  

Abstract: Forest fire is a most seen forest tragedy. Specially in some particular areas, such as high altitude region, once fire comes, it is not only resulting in huge revenue failures, and  it causes more death. So, real-time good vision is useful, and decreasing it into the less task in fire optimizaion. Comparatively  to old temperature-sensed smoke sensor, photo detection has more uses. Used in huge areas , and it can provide photo details. What is more, in huge space, temperature or smoke technology cannot obtain smoke and temperature information faster, so it is impossible in detect fast. Most common  UV and single band IR detectors are applied globally. They are confned by implementation stage. And they are reason for false alertings.Two band IR detection can reduce daylight, light and other sources, and other uses  of dust proof, water proof, and antielectric features. Therefore they are widely used in large areas and below buildings.  The paper has Two band IR fire detection is used on large altitude fire sensing in early time.

Keywords: Deep learning, Internet of Things, Fire detection and Object recognition, Sensors, Zigbee, Micro-controller, Raspberry Pi.

Abstract: Every major city in the world is currently facing a worsening situation in terms of public health and environmental disruption. More than 85% of people in urbanized regions are exposed to high quantities of particulate matter, making poor air quality a big concern. Recent estimates show that the burden of disease brought on by air pollution is now on pace with other major global health issues, such as poor food and cigarette use. The global nature of the challenge necessitates a more comprehensive global response. Despite some noticeable air quality improvements, since the 1990s, the number of fatalities and years of healthy life lost around the world has not changed.                                                                    Accordingly, we have suggested a compact, low-cost system to monitor and visualize air quality. This system is implemented using a Raspberry Pi and a SDS011 sensor module to calculate the amount of Particulate Matter which is majorly responsible for degrading the quality of air. This problem is solved using a simple Air filter used in vehicles which drastically reduces the PM levels thereby providing a low-cost way of measuring and improving the overall quality of Air.

Keywords: Air-Quality, air-quality analysis, particulate matter, green computing, IOT, raspberry pi  

Abstract: This IOT Project is based on monitoring the fish.As fish need lot of care to be taken unlike cats and dogs as they are very sensitive and stay in water.Hence this IOT- based smart aquarium helps to monitor the things going on insid the fish tank.Fishes usually need atmost care,the water in the tank has to be cleaned on regular bases.This system includes automatic feeding system and semi-automatic cleaning.This system will helpto keep fresh water inside the aquarium by keeping it clean and in a good working condition.This Using the mobile application users can track the situation inside hetank. Arduino MEGA and Node MCU controllers are included in this system design.So,there will be a wifi module on the Node MCU used between the smart phoneand user to control the operations.Ph level of water and its temperature will be monitored through LCD display and also on smartphone application.The fishes also requires oxygen for their survival.This oxygen supply is provided within the aquarium by the oxygen pump.This oxygen pump is a device that will provide oxygen within the water.Fish feeding is a automatic process which is done twice a day.The Embedded C coding is done using Arduino Software IDE and BLYNK Software is used for creation of applications.Basically,this project monitorsthe Ph value,Temperature through the smart phone.Due to developed economy, this project makes human life easy with time management.

Keywords: Internet of Things (IoT) ,Smart Aquarium ,pH water monitoring,fish feeders,pH Sensor

Abstract: organ Donation process has remodeled the health sector. We come across many people around the world who are ready to donate their organs while alive, after death or even in situations of brain death. Our aim with this project is to make organ donation procedure more secure and transparent using Blockchain technology. Through this project, we aim at attempting to put a limit to the unlawful access to the personal and medical details of the donor and the patient by making this procedure between the two parties (patient and donor) more secure and private. An intruder in this system wanting to get the details will not be able to make alterations secure and private. An intruder in this system wanting to get the details will not be able to make alterations or get access to the data.

Abstract: Air pollution includes things like dangerous gases and tiny particulate matter (PM2.5) that deteriorate air quality. This has developed into a crucial area for scientific research and a significant social issue that has an impact on the lives of the general public. As a result, multiple experts and academics at various R&D centres, institutions, and elsewhere are conducting extensive research on PM2.5 pollutant predictions. The authors provided a range of machine learning methods, such as linear regression and random forest models, in this scenario to forecast PM2.5 pollutants in polluted cities. This experiment is carried out with Python 3.7.3 and Jupyter Notebook. Observed to be more dependable models are random forest and based on results for the MAE, MAPE, and RMSE metrics among the models.

Keywords: Pollution detection, Pollution prediction, Logistic regression, Linear regression, Auto regression

Abstract: In the proposed system our main aim is to detect the pot holes and automatic levelling of the detected pot holes. When we go back in history, we get to know that India has grown tremendously, due which many numbers of accidents are taking, place which in turn is affecting both people and vehicle. Potholes have become a major reason for road accidents and loss of human lives . According to the survey reports of “ Road Accidents in India “ , almost 1,42,485 people have lost their lives due to this pothole leading to road accidents in 2011 . Nearly 1.5 per cent or nearly 2,200 fatalities are caused due to poor condition of roads . The problem is worth solving as it helps saving people’s lives and prevents road accidents, thereby helps in growth of the nation having pothole free roads. The ultrasonic sensor measures the depth of the potholes & the normal road and compares between them. Once vehicle detects the pothole, the levelling mechanism gets activated and it will level the pot holes with the normal road. The movement of the model can be controlled both manually and automatically by the user through the android application. This system provides the more accurate results and helps in avoiding the accidents.

Keywords: Arduino UNO, Ultra Sonic Sensors, Wi-fi Modules , H Bridge , DC Motors.

Abstract: False information concerning topics or events, such as COVID-19, is referred to as fake news. Social media juggernauts claimed to take COVID-19-related falsehoods seriously at the same time, yet they were ineffective. Real news data are gathered for this study via information fusion from websites related to news broadcasting, health, and the government, whereas false news data are gathered from social media platforms. Using cutting-edge deep learning models, 39 features were extracted from multimedia texts and utilised to identify bogus news about COVID-19. The accuracy of our model's false news feature extraction increased from 59.20 to 86.12 percent. Our best recall and F1-Measure for fake news were 83% utilising the Gated Recurrent Units (GRU) model, which has an overall high precision of 85%. Similarly, F1-Measure for real, recall, and precision

Keywords: Fake news, Social media, Deep learning, NLP, Mining Emotions

Abstract: Among the most frequent and costliest natural disasters to impact humanity. Therefore, the most effective partial solutions for reducing flood damage and loss are systems for flood forecasting and warning. For a sufficiently lengthy initial time, a reliable flood forecast and warning system is necessary. The goal of the a forementioned project is to establish guidelines for giving enough lead-time warnings of flash floods in urban streams based on rainfall. It also proposes a technique for analysing a tsunami forecasting and warning system. The findings of our study will be applied to modelling for flood disaster forecasts in the future as well as flood control.

Keywords: flash flood, flood forecasting, urban streams, observing system.

Abstract: With the world's population continuing to grow, rubbish management has become the most concerning issue, especially in the world's most populous nations. Overflowing trash cans or overflowing garbage disposal areas can spread a variety of dangerous infections to nearby residents. As a result, "IoT and deep learning based smart bin" are needed for garbage disposal, clearance, and monitoring. This will help to keep the neighbourhood clean and hygienic. The IoT-based smart garbage monitoring and clearing alarm system was proposed in this study. RGB led lights would be mounted to the bins to serve as an indicator of the level of waste inside the bin at any one time, giving users a sense of how much garbage is present inside each bin.After garbage has been disposed of, the sensor in the bin will monitor the amount of trash. When the waste level exceeds the maximum capacity level, the municipality will be notified. If the threshold is not reached but the trash is being removed for more than two days, a clearance alert will also be generated. To communicate alarms from the system's microcontroller to management, an android app is being built. The entire process saves on monitoring's human effort.

Keywords: Internet of Things, load sensors, garbage management, waste detection, and garbage monitoring

Abstract: Children with autism have a developmental problem that gets worse with time. Children with autism have difficulty communicating and interacting with others, and they also exhibit restricted behavior. If their illness is recognized early on and they receive comprehensive care and therapy, children with autism can enjoy happy, full lives. In wealthy countries, it might be difficult to diagnose autistic children until it is too late. Since there are no specific medical tests for autism, a qualified medical professional must make the diagnosis. Given that youngsters need to be closely monitored, medical experts require plenty of time to identify it. In this study, useless to most people images of children were utilized to identify those with autism using artificial intelligence technologies. For the diagnosis of, we have utilized five different algorithms to analyze the prevalence of autism spectrum disorder (ASD) in children: Multilayer Perceptron (MLP), Random Forest (RF), Gradient Boosting Machine (GBM), AdaBoost (AB), and Convolutional Neural Network (CNN). When comparing different algorithms' categorization outcomes, we found that CNN outperformed other traditional Machine Learning (ML) techniques with an accuracy of 92.31%, outperforming all other algorithms. Hence, we suggested a CNN-based prediction model to detect ASD, particularly in youngsters.

Keywords: Convolutional Neural Network (CNN), autism spectrum disorder and Machine Learning

Abstract: The recommended system presents a novel communication protocol for monitoring and managing the home environment that encompasses more than just switching operations and differs from previous systems in that it does not call for a dedicated server computer. This project offers clever fan and lighting control. Here, the system is connected to both light and temperature control.

Keywords: PIR, sensor, camera, microcontroller, GSM modem, Wi-Fi, an Arduino, a home automation system, an Android phone.

Abstract: The usual known fact about hunger in the world is not shortage of food but rather access to the food. The amount of world's food thrown in garbage is one-third of the total food. Right now, world produces enough food to feed every person on the planet. In order to help, we have developed an android-based application that will provide a platform to the people or organizations to donate their leftovers to those in need. The main objective of this project based on android application is to take care of donations and connecting the donators with those in need. This application will construct a shared collaboration medium for hostels, hotels, restaurants and NGO's or volunteers. This system consists of four modules that are admin, NGO‟s, volunteers and donators. The majority of the population today uses smart phones with active internet connectivity, which is the basic requirement for this product to function properly.

Keywords: food, food waste, waste management, agriculture

Abstract: Women's safety is a growing issue in India and beyond. A major problem with the police handling of these cases is their limited ability to respond quickly to emergency calls. These restrictions include not knowing the location of the crime or being completely ignorant of the crime. For victims, it is difficult to contact the police safely and discreetly. To remove these limitations, this white paper introduces a mobile application called WoSApp (Women's Safety App). It provides a reliable way for women to make emergency calls to the police. The user can easily and discreetly trigger the call function by explicitly manipulating her UI of the application by simply shaking the phone or pressing the on-screen panic button.A message containing the user's geographic location and contact details from a pre-selected emergency contact list is immediately sent to the police. This white paper describes the application, its development and technical implementation.

Abstract: online auction system is a one type of web application that helps people to buy and sell the products by attending online sessions. These system holds various types of products on website for sellers and bidders. Thus system allow users to set up there products for selling and bidder are comes with high amount to bid for the products. These is very easy and most useful way to sell and purchase the various products. All the unique types of products are available on these website which are not available locally.

Keywords: Admin, bidder, seller.

Abstract: Two wheelers are widely used than different shape of automobiles due to its low value and ease. maximum of the time rider doesn’t want to put on helmet that could bring about deadly injuries. under the influence of alcohol and trip and rash driving are the principal factors for on because of loss of focus, under the influence of alcohol and riding and no longer sporting helmet. The primary problem of all riders is safety. taking into consideration the safety of shipping boys who paintings for online business tour across regions using two wheelers, wherein safety of motorcycle rider counts.

Keywords: UNI2GO Mobility, Two wheelers, Motorcycle.

Abstract: Brain tumours are mostly produced by aberrant brain cell development, which can harm the brain's structure and eventually progress to dangerous brain cancer. The proper detection of various disorders in the gorgeous MRI pictures is one of the primary obstacles in providing an early opinion to allow decisive therapy utilising a computer-backed opinion (CAD) system. In this study, a novel Deep Convolutional Neural Network (DCNN) framework for accurate diagnosis of glioma, meningioma, and pituitary tumours is suggested together with a three-step preprocessing method to improve the quality of MRI images. For quick training with a high literacy rate and simple initialization of the sub caste weights, the armature employs batch normalisation. The suggested armature is a lightweight computational model with a few convolutional, maximum-- pooling layers and training duplications.  

Keywords: Brain tumors, deep convolutional neural network, image processing, MRI images.  

Keywords: electric vehicles, wireless charging, position detection, photoelectric sensor, IoT, efficiency

Abstract: Smart contracts are programs organized in blockchain surroundings, which manage the performance of accounts inside the Ethereum state. It is the algorithmic explanation of a prescribed business deal procedure that is spontaneously executed and composed of the data delivered by its parties. Solidity is an object-oriented, high-level language for applying smart contracts. It is impacted by Python, C++, and JavaScript, and is intended to aim at the (EVM) Ethereum Virtual Machine. This article mainly emphasizes how a smart contract is written in the programming language Solidity. Smart contracts are an outstanding approach to generating contracts that can be placed out concretely without human sentiments. This article clarifies what smart contracts are and how to write a contract by displaying the Solidity programming language syntax, i.e. known as the smart contract language.  The extension of this article, beginning with the description of smart contracts, also comprises in which regions and in which schemes smart contracts are used. The article also emphasizes the Solidity programming language syntax, which is a statically typed (characteristic of a programming language in which various types are explicitly declared and thus are determined at compile time) programming language considered to generate smart contracts executing on the EVM. The article ends by displaying showing steps to execute a simple solidity smart contract using Remix IDE. Finally.

Keywords: Blockchain, Ethereum, Remix, Solidity, Smart Contracts

Abstract: Current day Blood donation systems fall short in providing the solution for real-time time transfusion of blood, where the systems deal with Information which only responsive not dynamic. Starting from donation to transfusion. In present-day situations, there is no platform for blood transfusion, where blood present in one region is requested from the different regions where blood is scarce, which may lead to the wastage of blood. Lack of transparency and proper blood quality checks have led to several cases of blood infected with a transmitted disease such as HIV, hepatitis (HVB), or hepatitis(HVC) being used for transfusion, In addition to this, this system also deals with the blood mafia problem by providing transparency. This System aims at solving the issues regarding the supply chain. The system provides a facility for the blood donation process to be transparent by tracking the blood passageway and also helps to avoid wasting blood by providing a platform for the exchange of blood between blood banks. For ease of use, a web application is also built for accessing the system.

Keywords: donation, transfusion, responsive not dynamic. scarce, transparency

Keywords: Traffic sign recognition, traffic sign detection, image processing, convolutional neural network

Abstract: The number of aging high-rise civil structures is growing throughout the world, and maximum of them use concrete as a building material and is also very important material. There are high chances of concrete lose its strength due to continuous loading and environmental impacts. There by, damage may occur on the exterior surface of the structure. Whenever these deformities are left without investigated and untouched, the integrity of the structure may be compromised. Therefore, regular maintenance of the structure is very much nessesary. Some of the prior studies have used a drone as a instrument to capture and record the current state of the structure. Later, captured videos and images should analyze all the pictures to determine damage using object classification, localization, and segmentation methods. Sometimes the drones relay the collected data which uses a wireless medium. However, the developed systems are very complicated, time consuming, and requires a very high bandwidth

Keywords: Crack detection, Concrete bridge deck, Machine learning Real Times

Abstract: One of the most popular technologies in the world of image processing nowadays is face recognition across age groups. has become a very prevalent and challenging task in the realm of face recognition. Notwithstanding the numerous contributions made in this sector by professionals and researchers, there is still a substantial gap that has to be filled... Using the appropriate feature extraction and classification techniques is essential in this sector. A Convolutional neural networks combine feature extraction and classification in a single structure for deep learning. Using CNN architecture to recognise facial pictures as a person matures has overcome the issue of ageing. The Extensive experimentation has been used to evaluate the effectiveness of the suggested system.

Keywords: Feature Extraction, CNN, Face Recognition, Deep Learning.

Abstract: The Land Registry system is an essential department for any government system, who maintains the crucial information about the land like ownership, transaction history and many more. There are numerous loopholes in existing system that raise the chances of deceit and disputes. To address this issue, we have come with a solution of implementing blockchain in the land registry system. A secure platform for land registry system using Blockchain has been projected. Blockchain here is used as associate electronic ledger of digital records and transactions that are encrypted using cryptography. This system aims at coming up with a model for secure and steady land registration system supported blockchain technology, which can make easy to get rid of the loopholes in the current land registry system. Our system provides unique identity for each land, registered and verified for the land owner which cannot be tampered or replicated so, this provides secure trading of land. The land owner will be provided with the ID when the land is registered in our system and will be transferred to the other user once he sells his land. So no one can impersonate as other user and there will no chances for fraudulent or scam. Keywords: Include at least 4 keywords or phrases.

Keywords: Security, SMS (Short Message Service), Cloud, Authentication, Cryptography, Confidentiality.