Journal of Electronics and Telecommunication System Engineering

Assistive Glove Technology for Individuals with Physical Disabilities

Ravikumar K I, Ravi Rayappa, Anusha K — Sat, 21 Sep 2024 00:00:00 +0000

The smart hand glove serves as an essential tool for individuals with disabilities, enhancing their ability to engage in daily interactions with others. These gloves are especially valuable for those who are deaf or paralyzed, as they address communication challenges by translating hand gestures into written text and pre-recorded audio, thereby eliminating communication barriers. This capability allows users to express themselves clearly while enabling others to understand their messages and respond effectively. Furthermore, the gloves offer practical support in operating household appliances, empowering physically disabled individuals to maintain their independence. Influencing advanced ESP32 technology and flux sensors, the gloves detect and interpret hand movements, with the data processed by the ESP32 microcontroller. This analysis enables the production of corresponding speech output through an integrated speaker while also displaying the message on an LCD screen connected to the ESP32. Through these features, the gloves enhance communication and promote greater autonomy for individuals with physical impairments.

Person Re-Identification Utilizing GLCM, Radon Transform, and LDA based on Generative Adversarial Network

A. Divya, K B Raja — Thu, 03 Oct 2024 00:00:00 +0000

Person re-identification, commonly called Re-id, is an effective non-invasive biometric technique for identifying individuals, validating identities, and monitoring crowds globally. This study proposes a method for human re-identification utilizing a combination of Gray-Level Co-occurrence Matrix (GLCM), Radon Transform, and Linear Discriminant Analysis (LDA) within a framework based on Generative Adversarial Networks (GAN). The GAN model generates output images of the same individual in various new poses. Each original image produces a series of eight predefined poses, resulting in eight unique photos. Texture analysis and a subspace learning approach are utilized to extract features from GAN-generated images using GLCM, Radon Transform, and LDA. For the scaled image size of 128x64, GLCM is employed with matrix dimensions of 4x4, 8x8, 16x16, 32x32, and 64x64 to identify local features within the image dataset. As the dimensionality of the GLCM increased, rank-1 recognition improved significantly, although the GLCM matrix was limited to a dimension of 64x64, totaling 4096 elements. To enhance the accuracy of rank-1 recognition, the GLCM matrix undergoes the Radon Transform, which projects image intensity along radial lines at specific angles. The feature values generated from applying the Radon Transform to the 64x64 GLCM matrix result in a size of 95x180, yielding a total of 17,100, which is quite substantial.

Furthermore, incorporating LDA into the Radon Transform enhances the model's performance, leading to more robust features. To demonstrate the effectiveness of the proposed approach, experiments were conducted and validated using two widely recognized datasets: the smaller CUHK01 and the larger Market-1501 public dataset. The results indicate that the performance of the proposed model surpasses that of existing methods.

Raspberry-Pi & OCR-Based Image-Text to Speech Conversion

Shivkanya V. Dahiphale, S. J. Nandedkar — Mon, 14 Oct 2024 00:00:00 +0000

For individuals with visual impairments, image-to-text conversion into speech represents an essential tool that significantly enhances their ability to navigate the world more skillfully and independently. This research explored the innovative application of image-to-speech technology specifically designed for those with visual impairments. The proposed system is capable of transforming visuals into voice output. It employs a Raspberry Pi 3 B, an earphone, and an 8MP Raspberry Pi camera to achieve this task effectively. The development of this system is realized using the Python programming language in conjunction with powerful libraries such as OpenCV and Pytesseract. After capturing an image, the system processes the visual data and reads any text it has successfully identified aloud. A variety of images have been utilized to test the functionality and performance of the system rigorously. The proposed method demonstrates a significant promise for future development. It has the potential to significantly enhance the overall quality of life for individuals who are blind or visually impaired, allowing them greater access to information and their environment.

Smart Note Taker: A New Era of Digital Handwriting Recognition

Sojwal V. Borale, Amit G. Risodkar, Sanjay P. Satal — Tue, 05 Nov 2024 00:00:00 +0000

The "Smart Note Taker: A New Era of Digital Handwriting Recognition" is an innovative solution tailored for individuals navigating the demands of today's fast-paced, technology-driven environment. This versatile tool revolutionizes note-taking by enabling users to write in the air, with their notes captured and stored in the device's memory chip for convenient later access. Designed for busy professionals, students, and creative alike, the Smart Note Taker effortlessly enhances productivity by transforming handwritten notes into digital formats.

Beyond its user-friendly interface, this powerful device detects and interprets 3D shapes and movements, allowing for dynamic drawing and seamless integration with connected devices. Users can quickly display or broadcast their creations over a network, fostering collaboration and communication in various settings. The Smart Note Taker saves time and enriches the overall note-taking experience, making it an essential tool for anyone looking to enhance their workflow. By merging traditional handwriting with cutting-edge digital recognition technology, the Smart Note Taker represents a significant advancement in how we capture, share, and interact with information in the modern world.

Enhancing Traffic Monitoring with a Python-Based Helmet and Triple Seat Detection System

Payal Wagh, Sonali Patil, Sanika Karande, Vidya Khandekar, M.S. Kumbhar — Tue, 26 Nov 2024 00:00:00 +0000

As traffic congestion escalates, the need for effective monitoring of motorcycle safety practices becomes increasingly urgent, especially in addressing the risks associated with triple seating and helmet non-compliance. This study presents an innovative detection framework that uses advanced deep learning techniques, specifically YOLOv8, to identify non-helmeted bike riders traveling with additional passengers. By utilizing real-time video analytics, our system effectively monitors traffic conditions, highlights critical patterns in violations, and provides valuable insights into rider behavior. The findings underscore the need for enhanced traffic enforcement mechanisms and advocate for targeted safety initiatives to reduce motorcycle-related accidents. This research contributes to the discourse on traffic safety and emphasizes the role of technology in promoting public health and safety on the roads.