基于HSV颜色空间的运动目标识别  

Moving Target Recognition Based on HSV Color Space

作　　者：付长斐 叶宾 李会军 FU Changfei;YE Bin;LI Huijun(School of Information and Control Engineering,China University of Mining and Technology,Xuzhou,Jiangsu 221116,China;Xuzhou Key Laboratory of Artificial Intelligence and Big Data,Xuzhou,Jiangsu 221116,China)

机构地区：[1]中国矿业大学信息与控制工程学院,江苏徐州221116 [2]徐州市人工智能与大数据重点实验室,江苏徐州221116

出　　处：《控制与信息技术》2020年第2期70-74,共5页CONTROL AND INFORMATION TECHNOLOGY

基　　金：徐州市应用基础研究计划项目(KC18069)。

摘　　要：感兴趣区域(region of interest,ROI)图像的提取在运动目标的检测与跟踪等领域有着广泛的应用;HSV(hue saturation value)是根据颜色的直观特性创建的一种颜色空间,用于对颜色进行定量描述。文章将HSV颜色空间用于图像中颜色信息的计算,识别颜色特征。针对RoboMaster全国大学生机器人大赛中云台实时识别和捕捉敌方移动机器人装甲这一基本任务,提出一种基于色彩特征的目标识别算法。首先对图像进行阈值分割,计算二值图像轮廓的形状描述参数,寻找图像中所有的高亮条形物体;再用HSV颜色空间根据ROI颜色判断ROI中物体是否为目标灯柱。在此之前,需要统计符合要求的目标灯柱像素的HSV参数及RGB参数的大致范围,再结合HSV中典型颜色对应参数范围表,确定算法所采用的参数范围。最终识别目标的判定准则是,在初步筛选所得感兴趣区域中,HSV及RGB参数处于经分析所得参数范围内的像素点占整个区域的比例大于某一设定阈值。经测试,该方法对于像素为640×480的一组图片,平均处理速度可达到102.33帧/s,平均识别率约为97.4%。在RoboMaster大赛中,使用该目标识别算法的机器人能实时跟踪移动机器人,验证了该方法的有效性。Extracting of ROI(region of interest)has been widely used in the fields of detecting and tracking moving objects.HSV(hue saturation value)is a color space created according to the visual properties of colors,which can be used to describe colors quantitatively.Based on HSV,we calculate the color information and recognize the color features in images.For the basic task of recognizing and capturing the armors mounted on the opponent robots in real-time in the RoboMaster National University Robot Competition,a detecting algorithm based on the color features was proposed.Firstly,we apply threshold segmentation to the source image,and calculate the shape parameters of the contours to find out all bright light bar in this image.Then HSV color space is used to judge the color of preliminary ROI to confirm whether it is our target or not.Before this procedure,we need analyze the pixels of the right colorful light bar to confirm a pixel range of HSV and RGB parameters.This range is also determined according to a sheet about corresponding pixel parameters of typical colors in HSV and RGB color space.The criterion to judge the right object is the percentage of the pixels whose HSV and RGB parameters are in the range we confirmed before.Experimental results show that the average processing speed can reach 102.33 frame per second and the recognition accuracy is about 97.4%by the proposed method.This algorithm has been successfully applied in the RoboMaster Robot Competition and the results show that robot can effectively track a moving robot in real-time.

关 键 词：运动目标跟踪 感兴趣区域 图像分类 特征检测 HSV颜色空间 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]