DEHAZING ROAD IMAGES FOR DEEP LEARNING BASED TRAFFIC SIGN RECOGNITION
| Author(s) | : | Jameel Ahmed Khan, Hyunchul Shin |
| Institution | : | Division of Electrical Engineering, Hanyang University ERICA, South Korea |
| Published In | : | Vol. 5, Issue 11 — November 2018 |
| Page No. | : | 55-57 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
Features of traffic signs on road images become dull due to low visibility in hazy weather. In this paper, wepropose an algorithm that use image pre-processing before deep learning based traffic sign recognition. We combinedreliability guided fusion schemed for image dehazing with Traffic Sign (TS) detector to recognize the traffic signs fromhazy day road images. Dehazing is applied on input hazy images and then detection algorithm is applied to detect threeclasses of traffic signs. Experimental results show that detection accuracy of TS detector is increased by 6.77% owing tothe dehazing.
Jameel Ahmed Khan, Hyunchul Shin, “DEHAZING ROAD IMAGES FOR DEEP LEARNING BASED TRAFFIC SIGN RECOGNITION”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 5, Issue 11, pp. 55-57, November 2018.
