动态光照条件下足球机器人视觉目标的识别  

Visual Target Recognition of Soccer Robot in Dynamic Lighting Conditions

作　　者：卢艳[1] 徐雪松[1] 

机构地区：[1]华东交通大学电气与电子工程学院,江西南昌330013

出　　处：《计算机仿真》2013年第3期376-379,共4页Computer Simulation

基　　金：国家自然科学基金项目(61165005);江西省教育厅科技研究项目(GJJ11436)

摘　　要：赛场光照条件的波动会对足球机器人视觉目标识别产生较大的影响。为了使目标识别算法在变化的光照条件下保持较高的识别稳定性,本文根据实时采集的图像提供的信息,通过构造随光照条件动态变化的动态背景阈值、亮度阈值,并对有效色彩信息进行动态的饱和度调整,有效地降低了光照条件对图像识别的干扰。实验结果表明:该算法在动态变化的光照条件下,仍然能够保持较高的识别准确性。Fluctuation in the stadium light condition is a greater impact on the visual recognition of soccer Robot. In order to improve the recognition stability of Target Recognition Algorithm in the light changing conditions, accord- ing to the information provided by real-time images, this paper constructed the dynamic background thresholds and brightness threshold according to the dynamic changes of light conditions, and adjusted the saturation of valid color information dynamically, which effectively reduced the interference of the light on the image recognition. The experi- mental results show that the algorithm in dynamical lighting conditions is still able to maintain a high level of recogni- tion accuracy.

关 键 词：足球机器人 背景 动态背景阈值 目标识别 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]