Dr. Meghana Madhukar Deshpande

@moderncoe.edu.in

Assistant Professor at Department of Electronics and Telecommunication
PES's Modern College of Engineering

Dr. Meghana Madhukar Deshpande


            Download Compact PDF  
https://researchid.co/megh1986

RESEARCH, TEACHING, or OTHER INTERESTS

Engineering, Artificial Intelligence, Electrical and Electronic Engineering
9

Scopus Publications

Scopus Publications

  • AI Based Smart Traffic Law Enforcement System
    Sahil Shelote, Ritesh Chaudhari, Payal Sirmokadam, Rupali Kamathe, Meghana Deshpande, Vandana Hanchate, Sheetal Borde
    Lecture Notes in Networks and Systems, 2026
  • Development of Real Time Person Identification Using SVM and Resnet-50 Algorithm
    Meghana Deshpande, Rupali Kamathe, Vandana Hanchate, Sheetal Borde, Kalyani Joshi, Purva Marwade, Pratiksha K. Kamble, Priti D. More
    Lecture Notes in Networks and Systems, 2026
  • Automated LPG Gas Detection and Booking System
    Isha Dasharathe, Yash Lohar, Vineet Patil, Rupali Kamathe, Meghana Deshpande, Vandana Hanchate
    3rd International Conference on Intelligent Data Communication Technologies and Internet of Things Idciot 2025, 2025
    In Indian household and commercial settings, LPG gas cylinders are extensively used for cooking and various other purposes. However, the highly flammable nature of LPG means that any leakage can result in dangerous explosions, posing significant risks to both life and property. Leakage can occur due to faulty valves, improper handling, or aging cylinders. This risk is heightened in both residential and commercial environments, where larger quantities of LPG are stored, increasing the potential hazard. Presently, there is no standard method for measuring the remaining gas in the cylinder, often leading to unexpected depletion and inconvenience. Moreover, many users are unaware that the expiry date of the gas cylinder is encoded directly on the cylinder. To tackle these issues, an advanced system using the ESP8266 microcontroller has been developed. This system includes gas sensors to detect leakage and load cells to measure the gas quantity in the cylinder. It employs Internet Protocol (IP) to alert users about gas leakage and cylinder weight, thereby enhancing safety and convenience in LPG usage. In the event of a leakage, a stepper motor mechanism stops the gas flow, preventing potential loss of life or property. The system also utilizes an ESP-32 camera module to identify the cylinder's expiry date, with OCR technology to read and communicate this information via Telegram messages. A Telegram bot updates users on the cylinder's status, including weight, gas level, leakage status, and expiry date. Additionally, when the gas level drops below a certain threshold, the system automatically books a new cylinder. This comprehensive solution ensures efficient management of LPG usage and enhances safety protocols.
  • Soldier Health Monitoring and Position Tracking System
    Rupali Kamathe, Meghana Deshpande, Sheetal Borde, Vedanti Bagade, Vaishnavi Deshmukh, Sahil Naik
    2025 IEEE 1st International Conference on Innovations in Engineering and Next Generation Technologies for Sustainability Icinvents 2025, 2025
    In high-risk and remote environments such as military operations, coal mines, and disaster zones, continuous health and location monitoring of personnel is critical for ensuring timely intervention and life-saving support. Soldiers, in particular, face extreme conditions including avalanches, injuries, and isolation, where immediate communication and medical response can mean the difference between life and death. However, the absence of robust, real-time monitoring systems poses a significant challenge in such scenarios. To address this issue, we propose a Soldier Position Tracking and Health Monitoring System that enables real-time tracking of a soldier's vital health parameters and geolocation. The system comprises a wearable module attached to the soldier and a base station for command and monitoring. It captures key health metrics– such as heart rate, oxygen saturation, body temperature, motion status, and UV exposure–and transmits them along with GPS coordinates to a central base station via a long-range, low-power LoRa communication system. In critical situations, an SOS button allows manual alert initiation. Upon data receipt, the base station forwards this information to designated emergency contacts or services using the Twilio communication platform. This system enhances situational awareness and response time in life-threatening situations, making it a valuable asset for defense, mining, disaster rescue, and healthcare operations.
  • Sensor-Fusion based Autonomous Railway Track Fault Detection Bot using YOLOv5s and Ultrasonic Sensing
    Rupali Kamathe, Meghana Deshpande, Vandana Hanchate, Pranav Pattewar, Soham Gokhale, Sanket Palve
    Proceedings of 5th International Conference on Soft Computing for Security Applications Icscsa 2025, 2025
    This paper presents an efficient, low-cost, and Sensor-Fusion Based Autonomous Railway Track Fault Detection Bot Using YOLOv5s and Ultrasonic Sensing capable of identifying surface anomalies using a hybrid deep learning and sensor-based system. Conventional inspection techniques often require a lot of manual effort and can be difficult to carry out in hard-to-reach areas. Our proposed system integrates ultrasonic sensors for surface fault detection and a Raspberry Pi camera module to capture track images for advanced surface fault detection analysis using the YOLOv5s object detection model. The dataset used for training comprises 1,869 labeled images resized to 416×416 pixels, with a batch size of 4, trained over 50 epochs in 1.456 hours. Testing was performed on 251 unlabeled images using the best.pt weights. The system achieved a mean average precision (mAP@0.5) of 91.4%. The results demonstrate the model’s effectiveness in real-time fault detection with potential for deployment across diverse railway environments.
  • DEVELOPMENT OF AN ENERGY-EFFICIENT CCTV CAMERA SYSTEM FOR REAL-TIME HUMAN DETECTION USING YOLOV8 MODEL
    Meghana Deshpande, Alok Agarwal, Rupali Kamathe
    Eastern European Journal of Enterprise Technologies, 2024
    Human recognition is widely used in variety of fields such as autonomous vehicles, surveillance field, automatons, assisting blind peoples and many more. Many machine learning (ML) and deep learning (DL) algorithms exist for video analysis the main motive of these algorithms is to find human in complicated image. The research presented in this paper focuses on the development of an energy-efficient, smart CCTV camera system for real-time human detection, utilizing the YOLOv8 (You Only Look Once) model. The problem addressed is the need for more advanced, autonomous surveillance systems capable of human detection under various background conditions, overcoming the limitations of traditional CCTV systems, which require constant manual monitoring. The proposed system was trained on the PASCAL VOC 2012 dataset and optimized through hyperparameter tuning, achieving high accuracy and real-time performance. Key results demonstrate that the YOLOv8 model, implemented on the NVIDIA Jetson Nano platform, offers remarkable accuracy, precision, and energy efficiency. It consistently detects human figures in real-time, even in non-ideal conditions like poor lighting or complex backgrounds. This success can be attributed to YOLOv8’s cross-stage partial network (CSPNet) architecture, which enhances its ability to process images quickly and accurately, ensuring it meets the demands of continuous surveillance. The distinguishing features of this system are its energy-efficient design and adaptability to diverse environmental conditions. These characteristics not only solve the challenge of real-time human detection but also make the system a robust and scalable solution for modern security and surveillance applications
  • Skin defect detection of pomegranates using color texture features and DWT
    Meenakshi M. Pawar, Meghana M. Deshpande
    2012 National Conference on Computing and Communication Systems Ncccs 2012 Proceeding, 2012
    Various Skin disorders lower the quality of fruits due to environmental stress such as high temperature and solar radiation some other skin disorders are induced by chemical treatments and pathogens. Skin defect detection is important in the development of automatic grading and sorting system for pomegranate, because manual sorting process is very expensive and time consuming to automate this process skin defect can be identified with the help of color texture feature and discrete wavelet transform. For color texture feature analysis, acquired image is transformed into HSI color space, which is further used for generating SGDM matrix. Total 12 texture features were computed for hue (H), saturation (S) and intensity (I) images from each image samples. Then wavelet transform is used to compute statistical features, Total 3 features were computed for R, G & B components of each image samples. Best features were used as an input to Support Vector Machine (SVM) classifier and tests were performed to identify best classification model. Features showing optimal results were mean (99%), variance (99.80), cluster shade (99.88%), cluster prominence (99.88%), Mean intensity (99.81%).
  • Human detection based on discrete wavelet transform
    M.M. Deshpande, J.G. Rana, M.M. Pawar
    Iet Conference Publications, 2012
    Human detection is an important application where security is the main concern. This paper presents the machine vision approach for detection of human in video using support vector machine. In this paper, in order to improve the efficiency of the machine learning 2D Wavelet transform based features are used which are obtained from red, green and blue layers of sample images which forms the training input to the SVM. This technique is demonstrated using real world video data and SVM classifier gives encouraging detection rate for these features.
  • Real time visual surveillance system for human detection
    Meghana M. Deshpande, Prashant M. Pawar, Jaideep G. Rana
    3rd Nirma University International Conference on Engineering Nuicone 2012, 2012
    This paper proposes a real time visual surveillance system for human detection this interest needs in many applications where human entry is restricted. In order to develop a real time system initially wavelet transform based multiple features obtained from three layers of each image sample which forms training input to the SVM classifier. Finally, we used trained recognizer to identify whether there is somebody broken into object region. If there is, the automatic warning device gives the alarm, which guarantees a real-time surveillance.