International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
"Towards Smarter Navigation: A Mobile App for Object and Traffic Sign Detection" Prof. M. P. Shinde 1, Shreyash Dhurupe 2, Viraj Karanjavane 3, Sanna Shaikh4, Abhishek Suryawanshi5 1Professor, Department of Computer Engineering, SKNCOE, Savitribai Phule Pune University, Pune 411041, India 2,3,4,5Undergraduate Students, Department of Computer Engineering, SKNCOE, Savitribai Phule Pune University,
Pune 411041, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – This paper introduces a novel Android
A crucial part of urban navigation for visually impaired people is comprehending and interacting with traffic signs, which is covered in detail in the second aspect. Users can upload images from their gallery for analysis using this feature, unlike real-time object detection. After that, the application finds any traffic signs in these pictures and plays audio descriptions of them. This ability is especially helpful for understanding complicated crossroads, creating route plans, and improving mobility in general when outdoors. The creation of this application marks a substantial advancement in the integration of artificial intelligence, computer vision, and auditory feedback to make the world more approachable and comprehensible for people with visual impairments. With an emphasis on immediate support and in-depth examination of uploaded photos, the application provides a flexible tool that tackles a variety of issues that its users encounter daily. This introduction lays the groundwork for an in-depth examination of the development of the application, covering everything from the technical implementation to the user-centered design and conceptual foundations. It seeks to demonstrate the possibilities for these cuttingedge applications to improve visually impaired people's mobility and safety while also giving them more independence and opportunities to interact with their surroundings.
application that aims to greatly enhance the safety and autonomy of visually impaired people when they navigate their surroundings. The application uses cutting-edge object detection algorithms in conjunction with technologies for recognizing traffic signs to give users auditory feedback in real-time. Users can enjoy a safer and more informed navigation experience. Our aim is to incorporate advanced image captioning capabilities along with improved audio output features, with the goal of offering a more comprehensive and contextualized auditory depiction of the user's environment. This proposed update aims to change how blind people engage with their surroundings by providing a more complex and all-encompassing perception of their immediate visual context. Key Words: Object Detection, Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Image Captioning, YOLO V3, COCO Dataset
1.INTRODUCTION Advances in artificial intelligence (AI) and machine learning have created new opportunities in the rapidly developing field of assistive technologies to improve the quality of life for those who are visually impaired. The development of apps that convert visual data into auditory information has been made possible by these technological advancements, giving visually impaired people more autonomy and security when navigating and understanding their environment. To meet the specific needs of this community, this paper presents an innovative Android application that focuses on realtime object detection and traffic sign recognition from uploaded images. This application's dual-functional strategy is at its core and aims to improve user experience by offering customized help according to need and context. This system's first feature uses state-of-the-art object detection technology to recognize and announce in real-time the presence of objects in the user's immediate surroundings. This feature is essential for both indoor and outdoor navigation as it aids users in avoiding obstacles and navigating areas with greater assurance and safety.
© 2024, IRJET
|
Impact Factor value: 8.226
As we continue to explore the frontiers of assistive technologies, the incorporation of image captioning has great potential to improve visually impaired people's perception. Although the real-time object detection and traffic sign recognition features of our Android application are its primary focus, the exciting prospect of image captioning technology integration in the future will allow us to further enhance its functionality. Thanks to developments in natural language processing and computer vision, image captioning algorithms can now provide comprehensive textual descriptions of visual scenes. These algorithms provide users with a deeper understanding of their surroundings by generating captions that convey important details and relationships within the scene based on their analysis of the content and context of images. The incorporation of image captioning is a compelling future direction for our application.
|
ISO 9001:2008 Certified Journal
|
Page 282