Traffic sign Detection using CNN
K. Mirunalini1, Vasantha Kalyani David2

1K.Mirunalini*, Research Scholar, Avinashilingam Institute for Home Science and Higher Education For Women, Coimbatore.
2Dr.Vasantha Kalyani David, HOD, Professor, Avinashilingam Institute for Home Science and Higher Education For Women, Coimbatore. 

Manuscript received on February 15, 2021. | Revised Manuscript received on February 19, 2021. | Manuscript published on February 28, 2021. | PP: 129-135 | Volume-10 Issue-3, February 2021. | Retrieval Number: 100.1/ijeat.C22450210321 | DOI: 10.35940/ijeat.C2245.0210321
Open Access | Ethics and Policies | Cite | Mendeley
© 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: Lane Detection and Traffic sign detection are the essential components in ADAS .Although there has been significant quantity of analysis dedicated to the detection of lane detection and sign detection in the past, there is still need robustness in the system. An important challenge in the current algorithm is to cope with the bad weather and illumination. In this paper proposes an improved Hough transform algorithm in order to achieve detection of straight line while for the detection of curved sections, the tracking algorithm is studied. The proposed method uses Hybrid KSVD for removing the noise and Hybrid Lane Detection Algorithm is used for identifying the lanes and CNN based approach is used for the Traffic sign Detection. The proposed method offers better Peak Signal to Noise Ratio (PSNR) and Root Mean Square (RMS) in contrast to the existing methods. 
Keywords: Traffic-Sign, Convolution Neural Network (CNN),Hybrid KSVD, Lane Detection, Hybrid Lane Detection Algorithm (HLDA)