For the purpose of improving the coding efficiency of lossy compressed films, this study presents a groundbreaking deep learning approach. More specifically, it makes use of a Variable-Filter-Size Residue-Learning Convolutional Neural Network with Encoders. Through the utilization of this cuttingedge method, common distortions and artifacts like as blocking, blurring, and ringing are addressed. By surpassing the limitations of the High Efficiency Video Coding (HEVC) standard, our model is able to make significant improvements in both the quality of the video and the efficiency with which it compresses the video. In addition, we present a novel approach to the problem of picture super-resolution by making use of Convolutional Neural Networks (CNNs) and auto encoders. A deep convolutional neural network (CNN) model is trained in conjunction with an auto encoder architecture through the utilization of paired datasets consisting of high-resolution and low-resolution pictures using our methods. Throughout the training process, the CNN is able to extract high-level features from low-resolution photos, while the auto encoder is able to learn how to properly rebuild high-resolution images, therefore capturing delicate details and textures. During the inference phase, our trained model takes an input image with a low resolution and creates an output image with a high resolution that corresponds to the same input image.

Keywords : Encoder, HEVC, CNN, High resolution image

Author : Amritha Mishra, Sai Nikitha Bogisam, Tarun.M, Gayathri Putti

Title : Image/Video Super Resolution Using CNN And Auto encoders

Volume/Issue : 2024;06(03)

Page No : 1-5

This abstract introduces an innovative paradigm for computer interaction that substitutes traditional input modalities such as physical mice and touch displays with Hand Tracking and Gestures. Our system, which was constructed using the Open CV library and Python, implements computer vision techniques in order to monitor hand movements that are captured by a webcam. Sophisticated gesture recognition algorithms are utilized to convert these hand movements into mouse controls, thereby operations in an intuitive manner. Additionally, our system expands its practicality beyond the conventional capabilities of the mouse by integrating hand gestures for volume control, thus enhancing the overall user experience. By leveraging sophisticated computer vision methods, this strategy integrates gestural input into computing environments in a seamless manner, resulting in a moreuserfriendly and effective interaction paradigm. Through the utilization of gesture recognition and hand tracking, our system showcases the capacity to fundamentally transform paradigms of human-computer interaction, especially in situations where conventional input techniques may prove to be laborious or unfeasible.

Keywords : Hand Tracking, Gestures, Virtual Mouse Interface, Computer Vision, Open CV, Python.

Author : G. Hindu , Dr. G.D.K Kishore , J. Mohan Sai , P. Chinna Kasim , S. Saketh

Title : Virtual Mouse Operations Using Webcam

Volume/Issue : 2024;06(03)

Page No : 6-9

The goal of this project is to create a comprehensive and reliable system that is capable of properly diagnosing a wide range of skin illnesses. This aim is what drives this research. Leveraging a huge and diverse dataset supplied from Kaggle, which encompasses a comprehensive collection of photos depict- ing various dermatological disorders like as Acne, Melanoma, Psoriasis, and many more, the initiative leverages state-of-the- art deep learning algorithms. Through the skillful use of Convolutional Neural Networks (CNNs), well-known VGG (Visual Geometry Group) networks, and ResNet (Residual Networks) architectures, the project in- tends to attain levels of precision in illness detection that have never been obtained before. Through the use of these cutting- edge models, the system attempts to painstakingly evaluate and categorize photos of skin diseases. As a result, dermatologists are provided with essential information about the diagnosis of diseases and the planning of treatments. The ultimate objective of this attempt is to supply der- matologists with a categorization tool that is both automatic and dependable, which will complement their experience and enhance their diagnostic capabilities. The goal of the system is to transform the area of dermatology by enabling improved efficiency, accuracy, and efficacy in disease detection and patient treatment. This will be accomplished by integrating modern deep learning technologies into clinical practice in a seamless manner.

Keywords : Skin Disease Classification, Deep Learning, Convolutional Neural Networks (CNN), VGG Networks, ResNet, Dermatology, Image Classification

Author : M. Tanmaya, Dr.N.NeelimaPriyanka, V.Mahesh Reddy, J.Hemanth, Sk.Rumana

Title : Enhanced Deep Learning Approaches for Classifying Skin Disorders

Volume/Issue : 2024;06(03)

Page No : 10-15

In the era of cloud services, the number of data consumers has increased due to the rise in cloud storage services. This has also led to an increase in the number of data owners who store their encrypted data in the cloud. To address this issue, a hybrid algorithm combining Blowfish and AES encryption techniques was employed. The hybrid algorithm uses both Blowfish and AES encryption techniques to encrypt and decrypt datasets. This approach ensures enhanced security and improved performance during data retrieval. When a user enters a specific keyword and decrypts a file, the performance of the system is significantly improved. The performance depends on factors such as Recall, Ranking Privacy, Precision, Searching Speed. The proposed algorithm efficiently retrieves the required data maintaining the privacy and identifies the relevant data at high accuracy and optimal speed. Performance of the system is enhanced; the encrypted file is stored by both the user and the server. The data remains secure and accessible even in the event of a server failure. The hybrid algorithm searches for the keywords in the encrypted dataset using an algorithm that combines the strengths of both Blowfish and AES encryption techniques. The research work ensures optimal performance at a rate of 28.57% and also improved security standard of 33.33% compared to the traditional algorithms.

Keywords : encryption; decryption; data retrieval; cryptography; cloud security; Blowfish

Author : Vignesh Saravanan K , Bowyaa L , Harini Priya S Jelena Jophin J , Subhiksha G

Title : Enhancing Data Security and Privacy using Hybrid Cryptography

Volume/Issue : 2024;06(03)

Page No : 16-26

-This project presents an analysis of the "Dr.Visits" dataset using Python tools and libraries. The aim is to gain insights into the patterns and relationships within the dataset, focusing on factors related to doctor visits and health conditions. The analysis utilizes data visualization techniques and statistical methods to uncover key trends and correlations. Importing the dataset, and exploring its characteristics, including the distribution of variables such as gender, age, income, and illness, forms the initial steps. Through visualizations and statistical analysis, the project examines the impact of variables on doctor visits and health-related activities. Notably, the analysis reveals gender-based differences in reduced activity due to illness, prompting further investigation into potential contributing factors. Overall, this project provides valuable insights into the "Dr.Visits" dataset, shedding light on important aspects of healthcare and patient behavior.

Keywords : Data Analysis, Python Tools and Libraries, Data Visualization, Statistical Analysis, Behavior, Trends.

Author : Mrs. N. Nagalakshmi, A. Sai Vishal

Title : An Unveiling Healthcare Patterns: A Data-Driven Analysis of Doctor Visits Using Python Tools and Libraries

Volume/Issue : 2024;06(03)

Page No : 27-30

In this study finite element analysis of multistorey infilled frame building resting on sloping ground subjected to the seismic force is analyzed considering infill wall. In the present study, the performance of the building resting on slope has been studied for different slopes of 0°, 15°,20°, 25° and 30°. The building resting on slopes are analyzed for presence of with and without masonry infill wall at different sloping ground condition. Analyzing the performance of the building for slope and presence of masonry infill wall, it has been found that, building with infill wall gives better performance compared to bare frame. In the present study it has been noticed that the frame without infill at ground floor shows drastic increase in the deformation, base shear and member forces compared to infill frame with infill at ground floor. With increase in slope of ground the infill frame with short column compared to infill frame with long column at ground floor shows the decrease in the deformation and increase in the base shear and member forces.

Keywords :

Author : Vinay. P. Rokhade, B. P Annapurna

Title : SEISMIC ANALYSIS OF MASONRY INFILLED RC FRAME RESTING ON SLOPING GROUND

Volume/Issue : 2024;06(03)

Page No : 31-43

Alzheimer's disease, a leading cause of dementia, is characterized by memory loss and neurodegenerative disorders. Early diagnosis plays a crucial role in enhancing patient care and treatment outcomes. Traditional approaches for Alzheimer's disease diagnosis have limitations in terms of efficiency and learning time. Deep learning-based approaches, particularly Convolutional Neural Networks (CNNs), have shown promise in the classification of neuroimaging data related to Alzheimer's disease. In this presentation, we explore the use of a 12-layer CNN model trained on our datasets for early detection of Alzheimer's disease. Experimental results highlight the effectiveness of our proposed approach in improving accuracy and efficiency in Alzheimer's disease detection. Our research aims to contribute to the advancement of diagnostic techniques for Alzheimer's disease through the application of deep learning algorithms. Alzheimer's disease (AD) is a neurodegenerative disease and the most common cause of dementia in older adults. The part of brain that gets affected in this disease is hippocampus degeneration. Detection of Alzheimer’s disease at preliminary stage is very important as it can prevent serious damage to the patient’s brain. It becomes dangerous and sometimes fatal in case of people of 65 years of age or above. The main objective of this project is to use machine learning algorithms that is and feature extraction and selection to predict the Alzheimer’s disease and build a useful model. The dataset is taken in the form of images. The proposed approach detects the Alzheimer’s Disease such as moderate-demented and non demented using CNN algorithm.

Keywords :

Author : Ch.V Gopikrishna, D. Nagesh, P. Nithin Goud, S. Deepak, T. Shashi

Title : ALZHEIMER’S DISEASE DETECTION USING DEEP LEARNING

Volume/Issue : 2024;06(03)

Page No : 44-47