Real Time Object Detection for Visually Challenged Persons
Anitha.J1, Subalaxmi.A2, Vijayalakshmi.G3
1Anitha.J, Department of Information Technology, Sri Ramakrishna Engineering College, Coimbatore, India.
2Subalaxmi.A, Department of Information Technology, Sri Ramakrishna Engineering College, Coimbatore, India.
3Vijayalakshmi.G, Department of Information Technology, Sri Ramakrishna Engineering College, Coimbatore, India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 312-314 | Volume-8 Issue-8, June 2019 | Retrieval Number: H6339068819/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The visually impaired and blind people face various challenges in their day to day life. The objective of the proposed work is to develop an application for visually challenged persons based on the Android smart phone. It will eliminate the need for dedicated devices and other wearable devices to assist them to recognize objects as they move around. The Android application helps the visually impaired to navigate independently using real-time object detection and identification technology. The application makes use of the image processing technique to detect the object and speech synthesis to produce the voice output. The objective of the system is to detect real time objects which are scanned through the mobile camera and notify the blind persons about the object through audio or vocal information. The detection of images on moving objects has been a significant research area in computer vision which has been highly worked upon, and integrated with residential, commercial and industrial environments. Due to lack of data analysis of the trained data, and dependence of the motion of the objects, inability to differentiate one object from the other has led to various limitations in the existing techniques which include less accuracy and performance. Hence, Fast R-CNN (Region-based Convolutional Neural Networks) algorithm has been implemented to detect the object with high accuracy and processing speed. The detected image information is provided as a voice output using a speech synthesizer to the visually challenged persons to assist them in their mobility.
Keyword: Computer Vision, Convolution Neural Networks, Deep Learning, Mobile Application, Object Detection, Tensorflow
Scope of the Article: Sensor Networks, Actuators for Internet of Things