Smart Vision
Gopika K1, Ani Sunny2
1Gopika K, Department of Computer Science, Mar Athanasius College of Engineering, Kothamangalam, India.
2Ani Sunny, Department of Computer Science, Mar Athanasius College of Engineering, Kothamangalam, India.
Manuscript received on 08 April 2019 | Revised Manuscript received on 15 April 2019 | Manuscript Published on 26 April 2019 | PP: 586-591 | Volume-8 Issue-6S April 2019 | Retrieval Number: F61190486S19/19©BEIESP
Open Access | Editorial and Publishing 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: Vision is the focal point to the world, translating the existence occasions. It says the manner by which see individuals and increase learning by perusing from regular scenes. On the off chance that scratch found on our glasses, it might appear to us that everybody around us has such a large number of scratches. The issue really lies with us, individuals who are visually impaired or outwardly disabled. They has no capacity to realize what are the things be occur around them and exceptionally rely upon others for everything. So a sentiment of weight is emerges. Smart vision is proposed here, which comprises of a few modules for helping the visually impaired individuals in an effective way. The structure includes Human face discovery framework and Text identification which make vision for outwardly tested. The current framework for supporting visually impaired individuals is utilize the gadget raspberry pi. Be that as it may, raspberry pi has little memory estimate. So it is troublesome for preparing and retraining. In my framework I tackled this issue by utilizing the idea quantization. Docker picture is utilized for retraining the entire picture, so it makes the framework in quicker way. The other real favorable position of the undertaking, utilization of constant gaming board UDDO x86 independent of gadget like raspberry pi and so on. The brilliant unit contains an eyeglass gave camera, headphone, receiver, battery and UDDO x86. The camera present at the eyeglass catch the proposed picture of the client and regular scenes as previews and exchange to the UDDOx86 where it gets handled and produce the ideal sound portrayals as yield utilizing headphone.
Keywords: Face Recognition, Text Detection, Tesseract OCR, Eye Aspect Ratio.
Scope of the Article: Computer Science and Its Applications