Assistive Frame Work for Physically Debilitated using Convolutional Neural Networks
B. Uday Kumar Veera Manikanta1, Kallakunta Ravi Kumar2, Kagga Koteswaraao3
1B.Uday Kumar Veera Manikanta, B.Tech Student, Department of Electronics & Computer Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram (A.P), India.
2Kallakunta Ravi Kumar, Assistant Professor, Department of Electronics & Computer Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram (A.P), India.
3Kagga Koteswaraao, B.Tech Student, Department of Electronics& Computer Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram (A.P), India.
Manuscript received on 07 March 2019 | Revised Manuscript received on 20 March 2019 | Manuscript published on 30 March 2019 | PP: 963-966 | Volume-8 Issue-5, March 2019 | Retrieval Number: E3089038519/19©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: Extraction of hand developments alongside their always showing signs of change shapes for acknowledgment of communication through signing is viewed as a troublesome issue in PC vision. A nonverbal correspondence includes developments of various body parts to convey a message hand development or different developments of body parts are viewed as signal. There are a lot of utilizations where hand motion acknowledgment can be connected for enhancing control, time, exactness, availability, correspondence and learning. In the work exhibited in this paper we directed trials with different sorts of convolutional neural systems, including our own exclusive model. This will make an extension among hard of hearing and unable to speak will speak with the outside world without need of a mediator openly puts like railroad stations, banks, and so forth. The execution of each model was assessed on the MNIST dataset (Modified National Institute of Standards and Technology dataset). The motivation behind the framework is to enhance the current framework here regarding reaction time and exactness with the utilization of proficient calculations. We accomplished 92% acknowledgment rate contrasted with other classifier models covered the equivalent dataset.
Keyword: Convolutional Neural Networks (CNN), Artificial intelligence (AI), Indian Sign Language (ISL).
Scope of the Article: Neural Information Processing