Computing and Informatics

Traffic light recognition in urban environments is crucial for vehicle control. Many studies have been devoted to recognizing traffic lights. However, existing recognition methods still face many challenges in terms of accuracy, runtime and size. This paper presents a novel robust traffic light recognition approach that takes into account these three aspects based on image processing and deep learning. The proposed approach adopts a two-stage architecture, first performing detection and then classification. In the detection, the perspective relationship and the fractal dimension are both considered to dramatically reduce the number of invalid candidate boxes, i.e. region proposals. In the classification, the candidate boxes are classified by SqueezeNet. Finally, the recognized traffic light boxes are reshaped by postprocessing. Compared with several reference models, this approach is significantly competitive in terms of accuracy and runtime. We show that our approach is lightweight, easy to implement, and applicable to smart terminals, mobile devices or embedded devices in practice.