模糊图像中仪表盘修正力Snake模型检测  

Detection of Dashboard in Fuzzy Image by Correction Force Snake Model

作　　者：于舒春[1] 董静宜 YU Shu-chun;DONG Jing-yi(Higher Educational Key Laboratory for Measuring and Control Technology and Instrumentation of Heilongjiang Province, Harbin University of Science and Technology,Harbin 150080,China)

机构地区：[1]哈尔滨理工大学测控技术与仪器黑龙江省高校重点实验室,黑龙江哈尔滨150080

出　　处：《哈尔滨理工大学学报》2018年第2期65-69,共5页Journal of Harbin University of Science and Technology

基　　金：国家自然科学基金(61401126);黑龙江省自然科学基金(QC2015083)

摘　　要：针对模糊图像中仪表盘检测准确率不高的问题,提出了一种基于修正力的改进Snake模型。首先,采用Hough变换确定模糊图像中仪表盘所在的区域。其次,以Hough变换检测到的区域边界为Snake算法的初始边界,引入修正力参数,扩大Snake算法的控制范围,调整能量函数对目标曲线的连续控制,实现模糊图像中仪表盘的精确定位。实验结果表明,基于修正力的改进Snake算法,对于模糊图像中的仪表盘检测准确率,比传统Snake算法提升近20%。In order to solve the problem that the accuracy of the dashboard detection in fuzzy images is not high,an improved Snake model based on the correction force is proposed.First,the Hough transform is used to determine the area of the dashboard in the fuzzy image.Secondly,the boundary area detected by Hough transform is the initial boundary of Snake algorithm.The correction force parameter is introduced to expand the control range of Snake algorithm,adjust the continuous control of energy function to the target curve,and achieve the precise location of the dashboard in the fuzzy image.The experimental results show that the improved Snake algorithm based on the correction force is nearly 20%better than the traditional Snake algorithm for the accuracy of the dashboard detection in the fuzzy image.

关 键 词：模糊图像 仪表盘 SNAKE模型 准确率 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]