Method for identifying traffic signal lamps based on multiple classifiers

Abstract

The invention relates to a method for identifying traffic signal lamps based on multiple classifiers, which belongs to the technical field of image processing. First, a region of interest of an imageis positioned. Then, a candidate region is extracted according to the red / green color characteristic of traffic signal lamps, the candidate region of the traffic signal lamps is filtered according tothe morphological features, and the HOG features and color features of the candidate region are extracted in turn. Next, multiple binary classification SVM models are trained by making use of sample pictures of traffic signals to construct multiple classifiers based on SVM. The disadvantages of slow calculation and low identification accuracy for identification of multiple targets based on a single classifier can be overcome.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a traffic signal lamp recognition method based on a multi-classifier. Background technique [0002] With the development of society, economy and technology, there are more and more vehicles on the traffic roads, and the road environment is becoming more and more complex, and the subsequent traffic accidents also occur frequently. In order to reduce the occurrence of such situations, intelligent transportation comes into being In intelligent transportation, intelligent vehicles are a key part of intelligent transportation. The emergence of intelligent vehicles helps drivers drive, reduces the frequency of traffic accidents, and makes it possible for people with red and green color blindness to drive. Therefore, for Traffic light recognition is of great significance [0003] To achieve this goal, we must first prepare to find the location of the traffic lights from the comp...

AI Technical Summary

Problems solved by technology

At present, this method is accurate, but not widely applicable.

Images