摘要
在车辆颜色识别的过程中,车辆图像中主要颜色区域的准确分割、排除非颜色干扰区域始终是个问题。因此提出一种基于显著性区域检测的抗干扰车辆颜色识别算法,针对车辆颜色区域分割不准确问题进行一定程度的改善,去除车辆颜色干扰区域并使用自适应k近邻算法(KNN)进行颜色分类。实验结果表明,该方法能有效分割车辆主要颜色区域,并且能达到比较好的分类识别效果。
This paper proposes an anti-interference vehicle color algorithm based on saliency region detection .which improves the inaccuracy of vehicle color region segmentation to some extent,and removes the vehicle color interference region, and uses the adaptive k-nearest neighbor algorithm (KNN) for color classification.
出处
《工业控制计算机》
2019年第5期95-96,共2页
Industrial Control Computer
关键词
车辆颜色识别
显著性区域检测
自适应k近邻算法
抗干扰
vehicle color recognition
saliency region detection
adaptive k-nearest neighbor algorithm
anti-interference
