Robust speed-limit traffic sign detection and recognition method

Abstract

The present invention discloses a robust speed-limit traffic sign detection and recognition method. The method comprises the steps of firstly, establishing a multi-feature fusion type saliency model, and subjecting each layer of the multi-feature fusion type significance model to updating and iteration to obtain a level saliency map; secondly, solving a multilayer saliency map to obtain an optimal saliency map, obtaining an ROI from the optimal saliency map, loading the obtained ROI into a super-pixel-based pre-trained CNN model to classify the ROI, and obtaining a recognition result. According to the technical scheme of the invention, traffic signs on both sides can be better highlighted through the saliency model based on a-priori position and boundary features. Meanwhile, the structural information of an image is effectively utilized through the multi-layer fusion type saliency map. Moreover, multiple small-scale detail information in a circular sign is maintained, so that a target is more complete and uniform. Therefore, the recognition efficiency and the recognition accuracy are improved.

Description

technical field [0001] The invention belongs to the field of computer vision and relates to an image recognition method, in particular to a robust speed limit traffic sign detection and recognition method. Background technique [0002] With the development of economy and technology, automobiles play an increasingly important role in people's daily life, and people's demand for automobiles is also increasing, including the gradual emergence of various safety assistance driving technologies, such as adaptive cruise control systems, Collision avoidance system, anticipatory perception collision system, parking space recognition system and night vision system, etc. When science and technology develop to a certain level, safe and efficient vehicle driverless technology will eventually be realized, that is, the realization of driverless vehicles. According to statistics, about 1 million people die in traffic accidents every year, and the fatal accidents caused by human operation e...

AI Technical Summary

Problems solved by technology

Because traffic signs are affected by the service life and the external environment, they are prone to problems such as staining, fading, distortion, and reflection. The above-mentioned recognition methods only analyze a certain feature of the target image, and the effective information in the target image is insufficiently utilized. As a result, the accuracy of target detection and recognition is relatively low, and the actual detection and recognition effect is not good.

Images