基于等温线温度识别的示温漆温度自动识别算法  

Automatic recognition algorithm for temperature-sensitive paint’s temperature based on isotherm temperature identification

作　　者：王荣华[1] 杜平安[1] 黄明镜[2] 聂宝林[1] 

机构地区：[1]电子科技大学机械电子工程学院,成都611731 [2]中国燃气涡轮研究院,江油621703

出　　处：《电子测量与仪器学报》2010年第6期542-547,共6页Journal of Electronic Measurement and Instrumentation

摘　　要：本文对目前基于RGB颜色温度特征曲线的示温漆温度自动识别算法的原理及缺陷进行分析,提出一种基于等温线温度识别的示温漆温度自动识别新算法。该算法综合考虑了示温漆测温特点和人工识别新动态,通过定位等温线位置,用K-Means聚类识别等温线附近区域颜色特征来识别示温漆温度,较好地解决了当前算法过分依靠RGB像素点,温度识别可靠性差,建立色温库工程浩大,操作困难等缺点。实验结果表明,该新算法温度识别准确度高,且操作性强。A new automatic recognition algorithm for temperature-sensitive paint （TSP） temperature based on isotherm temperature recognition is proposed after an investigation to the principles and drawbacks of the current RGB-pixel Temperature-Character-Curve based algorithm. This algorithm takes into account the characteristics of TSP’s tempera-ture measurement and new developments of artificial identification, and it can identify the TSP’s temperature by posi-tioning the isotherm and then using K-Means Clustering to identify the nearby color features of the isotherm. This algo-rithm overcomes the following drawbacks of current algorithm： depending too much on RGB-pixel, low reliability, dif-ficult implementation process etc. The experimental results show that this method is easy-to-use and highly effective.

关 键 词：示温漆 温度识别 K-MEANS聚类 

分 类 号：TN751[电子电信—电路与系统]