Determine distances area of traffic lights using image processing algorithms

• Published in: AIP conference proceedings Vol. 2547; no. 1
• Main Authors: Touma, Teeba A., Abbas, Heba Kh, Saleh, Anwar H. Al, Mohamad, Haidar J., Al-Zuky, Ali A.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 02.12.2022
• Subjects: Algorithms; Color; Image processing; Light; Mathematical analysis; Object recognition; Pixels; Traffic signals
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0113085

Traffic lights recognition and detection considered important step in the advanced driver assistance techniques. The suggested system designed to detect the color and area of the traffic light at different distance. The detection of the traffic color based on designed algorithms to separate colors and calculate area of the signals. The Three colors (red, green, and yellow) distinguished by threshold for each color red, blue, green, and grayscale to give accurate results. The distance between the camera (car) and the traffic light is determined by scale factor (SCF) equation. The real dimension of the traffic measured manually and converted to pixel to calculate the area of each color. The area of each light within the traffic estimated by morphology method. Thus, the light traffic and the area is determined, and the distance between the traffic light and car is determined. The results show that the behavior of the area and distance of the traffic with the video frame shows logic data in pixel domain. Moreover, the light separation is accurate 100% according to the suggested algorithm.