Traffic signs recognition using cuckoo search algorithm and Curvelet transform with image processing methods.

Abdullah, Ahmed saadi; Abed, Majida Ali; Ismael, Ahmed Naser

Compliance with traffic signs is one of the most important things to follow to avoid traffic accidents as well as compliance with traffic rules in terms of parking, speed control, and other traffic sings. Progress in different areas, such as self-propelled car manufacturing or the production of devices that help the visually impaired, require values to find a way to determine traffic signals with high precision in this research, The first step is to take a picture of the traffic sign and apply some digital image processing techniques to increase image contrast and eliminate noise in the image, the second step resize of origin image , the third step convert color to(YCbCr, HSB) or stay on RGB, the fourth step image is disassembled using curvelet transform and get coefficients , and the last step using cuckoo search algorithm to recognition sings traffics , the MATLAB (2011b) program was used to implement the proposed algorithm . After applying this method to a set of traffic the percentage of discrimination of traffic signs was yellow 93%, green 94%, blue 94.5%, red 96%.

TRAFFIC signs & signals; SEARCH algorithms; PEOPLE with visual disabilities; NOISE; IMAGE processing

Journal of Qadisiyah Computer Science & Mathematics, 2019, Vol 11, Issue 2, p74

2074-0204

Academic Journal

10.29304/jqcm.2019.11.2.591