01-06-2024, Saturday

 Joined as a Professor in CSE, PSCMR College of Engineering and Technology, Vijayawada

Thank You ALIET - 29-05-2024

End of the day as Associate Professor & Examination-Incharge,ALIET.

 

International conference of Embracing the Digital Horizon: Pioneering Commerce and Management Strategies for a Transformative Future(EDH 2024)

 Paddlers Drowning Detection System Using Ultralytics-Yolo v8 in Deep Learning

Abstract -This paper presents the development and implementation of a his paper presents the development and implementation of a deep learningbased approach for real-time detection of drowning swimmers using the YOLO v8 (You Only Look Once version 8) object detection framework. Drowning remains a significant cause of mortality worldwide, necessitating effective surveillance and rapid response systems. Leveraging the capabilities of YOLO v8, we train a neural network model on a custom dataset comprising annotated images of drowning scenarios. The dataset encompasses diverse environmental conditions, swimmer orientations, and occlusions to enhance the model's robustness. Our approach utilizes transfer learning to fine-tune the pre-trained YOLO v8 model on the drowning swimmer detection task, achieving high accuracy and efficiency. We evaluate the performance of our proposed method on both synthetic and real-world datasets, demonstrating its effectiveness in detecting drowning swimmers with high precision and recall rates. Furthermore, we conduct comparative experiments with existing drowning detection methods to validate the superiority of our approach in terms of accuracy, speed, and real-time applicability. The proposed system holds promise for enhancing water safety measures by enabling timely detection and intervention in drowning incidents.

FUTURE FORGE FORUM IN ACADEMIA

Abstract -This paper presents the development and implementation of a comprehensive college community platform, termed ForgeNet, built using the MERN (MongoDB, Express.js, React.js, Node.js) stack. ForgeNet aims to facilitate student engagement, academic support, and career development within the college ecosystem. The platform encompasses features such as student networking, buying and selling collegerelated products, accessing exam references, exploring job postings, and receiving mentorship from senior peers. Leveraging modern web technologies, ForgeNet provides a seamless and intuitive interface for students to connect, collaborate, and thrive academically and professionally. The paper outlines the system architecture, implementation details, and evaluation of ForgeNet, demonstrating its efficacy in fostering a vibrant and supportive college community. Through ForgeNet, students can access a centralized hub for academic and career resources, enhancing their overall college experience and readiness for the future Keywords: College Community, MERN Stack, Student Engagement, Exam Resources, Job Postings, Mentorship, Student Connectivity, User Interface, User Experience.

AI SIMULATED FAUX IMAGE DETECTION SYSTEM USING DEEP LEARNING

 Journal of Nonlinear Analysis and Optimization  Vol. 15, Issue. 1, No.12 :  2024  ISSN : 1906-9685

AI SIMULATED FAUX IMAGE DETECTION SYSTEM USING DEEP LEARNING

ABSTRACT –  Although biometric technology is essential for identifying people, thieves are always evolving to avoid being caught. We are using a state-of-the-art method called Deep Texture Features extraction from photos to tackle this problem. Building a Convolutional Neural Network (CNN) is our method for efficiently utilizing this technology. Known as LBPNET, this CNN-based model emphasizes the use of the Local Binary Pattern (LBP) methodology for feature extraction, making it stand out as a cuttingedge approach in the industry. To counter the spread of fake face photos in identification systems, we train the machine learning model to understand LBP descriptors taken from images. Our technique promises to improve the accuracy and dependability of facial recognition systems by integrating CNN technology with LBP descriptors, hence reducing the risk associated with fraudulent modifications to physical and psychological traits. 

KEYWORDS:  Deep textures, CNN, NLBPNet, LBPNet, LBP descriptor images.