Outdoor traffic sign identification method under low-illumination scene

Abstract

The invention discloses an outdoor traffic sign identification method under a low-illumination scene. The method comprises the following steps: allowing a low-illumination scene image containing traffic signs to be subjected to HSV conversion, calculating an integration histogram of the low-illumination scene image and detecting a traffic sign interest area through a threshold value method and a mathematical morphology method; adopting an SIFT algorithm to calculate descriptors of scale-changeless characteristic points of the traffic sign interest area and matching the descriptors with an image in a standard library so as to obtain matched point pairs; taking the coordinates of the scale-changeless characteristic points as image center spots to obtain subimages; respectively calculating the color-improved quantification histograms, which are fused with color tone, saturation and brightness, of the subimages, as well as the normalized color-improved histograms; calculating an included angle of two histograms of two matched points of the matched point pairs; removing the matched point pairs with big-different included angle; removing the matched point without a matching point; and obtaining the identification result of the traffic signs of the low-illumination scene image according to the number of the matched points. The method provided by the invention has the advantages of stronger robustness, and higher accuracy rate.

Description

technical field [0001] The invention belongs to the application field of digital image processing, and in particular relates to an outdoor traffic sign recognition method in a low-illuminance scene. Background technique [0002] Traffic sign detection and recognition is a hotspot in intelligent transportation systems, which has received more and more attention in recent years. The situation encountered by the actual traffic road is very complicated, and the traffic sign pictures collected by the sensor often show different degrees of degradation. Evenly, due to the uncertainty of the shooting angle, traffic signs will also be cut horizontally or vertically at a certain angle, and there are also problems such as traffic signs being blocked by trees, advertisements, etc. These problems make the detection and recognition of traffic signs become difficult. quite difficult. [0003] Some researchers have proposed some outdoor traffic sign recognition methods, and they divide th...

AI Technical Summary

Problems solved by technology

Therefore, the existing traffic sign detection and recognition methods inevitably encounter difficulties

Images