###
DOI:
计算机系统应用英文版:2014,23(2):142-145,85
本文二维码信息
码上扫一扫!
电路板红外图像多目标提取算法
(中国民航大学 模式识别与智能系统, 天津 300300)
Multi-Target Extraction Algorithm of Circuit Board Infrared Image
(Pattern recognition and intelligent system, Civil aviation university of China, Tianjin 300300, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1362次   下载 3046
Received:July 08, 2013    Revised:September 09, 2013
中文摘要: 电路板红外图像芯片提取是电路板红外图像故障检测系统中的重要环节,传统的芯片发热区域提取方法或多或少需要人工干预,在芯片较多和热辐射情况复杂电路板中人工参与效果不是很理想。基于电路板红外图像特征提出了一种自适应多目标区域增长算法,将该多目标区域增长算法与GVF-Snake模型相结合用于电路板红外图像芯片提取。利用多目标区域增长算法将每一块芯片的发热区域和辐射区域一并提取,再利用区域增长确定GVF-Snake模型初始轮廓,利用GVF模型进行芯片发热区域提取。实验表明,这种算法能够无人工干预的准确提取电路板红外图像所有芯片发热区域,具有一定的实用性和鲁棒性。
Abstract:Extracting the infrared image chip exothermic area at a circuit board is an important part of the fault detection system. The traditional extraction methods to exothermic area of chip need manual intervention more or less, and the effects of artificial participation aren’t ideal at these circuit boards such as the more chips and complex heat radiation. this algorithm put forward a multi-objective region growing arithmetic according to the features of infrared images of a circuit board, which combine with GVF-Snake model to extract infrared image of circuit board. this algorithm can ensure original outline of GVF-Snake model by using the growing area and draw the exothermic area of chips by GVF model. The experiment results show that we can extract accurately exothermic areas of all circuit board chips without manual intervention .And the algorithm have practical and robust traits.
文章编号:     中图分类号:    文献标志码:
基金项目:中央高校基本科研业务项目(ZXB2011A003);中国民航大学科技基金项目(2010kyE07);中国民航大学校级科研项目号(2011kyE01);中国民航机务维修科研基地资助
引用文本:
王力,曾佩佩,郝建新.电路板红外图像多目标提取算法.计算机系统应用,2014,23(2):142-145,85
WANG Li,ZENG Pei-Pei,HAO Jian-Xin.Multi-Target Extraction Algorithm of Circuit Board Infrared Image.COMPUTER SYSTEMS APPLICATIONS,2014,23(2):142-145,85