基于NSCT域FAST角点检测的电气设备红外与可见光图像配准
|
摘要
针对电气设备在线监测系统中红外与可见光图像配准问题,提出了基于非下采样轮廓波变换(Non-Subsampled Contourlet Transform,NSCT)域改进的加速分割检测特征(Features from Accelerated Segment Test,FAST)提取的图像配准方法。首先采用灰度均衡技术对原始图像进行图像增强;再通过一级NSCT变换得到红外与可见光的低频子带图像;然后对低频子带图像利用FAST角点检测以及局域强度不变特征描述符(Partially Intensity In-variant Feature Descriptor,PIIFD)得到特征描述点对;最后,利用最近邻距离比率法及随机抽样一致性算法(Random Sample Consensus,RANSAC)得到精匹配点对,进而计算仿射变换参数。实验结果表明,相比传统的PIIFD配准算法,该方法的特征点正确匹配率提高了2. 52%,配准均方根误差降低11. 58%。
Infrared and visible image registration has become a big challenge in online monitoring system for high voltage electric equipment. Aiming at this problem,an efficient registration method based on features from accelerated segment test( FAST) in non-subsample contourlet transform( NSCT) domain is proposed in this paper. Firstly,the two images are preprocessed by gray-level equalization. After that,NSCT decomposition is applied to obtain low-frequency subband images. And corner points are detected from the low-frequency images as interest points by FAST. Their descriptors are calculated by partially intensity invariant feature descriptor( PIIFD) and used to make rough matching through best bin distance ratio. Finally,random sample consensus( RANSAC) is further utilized to refine the matched interest point pairs and affine transformation parameters between infrared and visible images are determined. Objective evaluation on experimental results shows that,compared to the traditional registration method based on PIIFD,the correct matching ratio of the proposed registration method is improved by 2. 52% and the root mean square error is reduced by 11. 58%.
Infrared and visible image registration has become a big challenge in online monitoring system for high voltage electric equipment. Aiming at this problem,an efficient registration method based on features from accelerated segment test( FAST) in non-subsample contourlet transform( NSCT) domain is proposed in this paper. Firstly,the two images are preprocessed by gray-level equalization. After that,NSCT decomposition is applied to obtain low-frequency subband images. And corner points are detected from the low-frequency images as interest points by FAST. Their descriptors are calculated by partially intensity invariant feature descriptor( PIIFD) and used to make rough matching through best bin distance ratio. Finally,random sample consensus( RANSAC) is further utilized to refine the matched interest point pairs and affine transformation parameters between infrared and visible images are determined. Objective evaluation on experimental results shows that,compared to the traditional registration method based on PIIFD,the correct matching ratio of the proposed registration method is improved by 2. 52% and the root mean square error is reduced by 11. 58%.
引文
[1]Xiong Z,Zhang Y.A critical review of image registration methods[J].International Journal of Image and Data Fusion,2010,1(2):137-158.
[2]张璐,赵建.红外成像技术在电力系统后备电源热故障监测中的应用研究[J].电测与仪表,2012,49(8):93-96.Zhang Lu,Zhao Jian.Application research of infrared imaging technology on thermal fault monitoring for electrical power system backup power[J].Electrical Measurement&Instrumentation,2012,49(8):93-96.
[3]李伟,王伯雄,王雪,等.一种基于安瓿检测的快速图像配准方法的实现[J].电测与仪表,2010,47(7):5-8.Li Wei,Wang Boxiong,Wang Xue,et al.A fast image registration method applied in ampoule detection[J].Electrical Measurement&Instrumentation,2010,47(7):5-8.
[4]吴炬卓,牛海清,张煌,等.一种应用小波系数GSM模型的混合傅里叶-小波电缆瓷套终端红外图像去噪方法[J].电测与仪表,2018,55(4):113-117.Wu Juzhuo,Niu Haiqing,Zhang Huang,et al.A hybrid Fourier-wavelet denoising method for infrared image of porcelain sleeve cable terminal using GSM model for wavelet coefficients[J].Electrical Measurement&Instrumentation,2018,55(4):113-117.
[5]金立军,田治仁,高凯,等.基于红外与可见光图像信息融合的绝缘子污秽等级识别[J].中国电机工程学报,2016,36(1):3682-3691,3389.Jin Lijun,Tian Zhiren,Gao Kai,et al.Discrimination of insulator contamination grades using information fusion of infrared and visible images[J].Proceedings of the CSEE,2016,36(1):3682-3691,3389.
[6]纪利娥,杨风暴,王志杜,等.基于边缘图像和SURF特征的可见光与红外图像的匹配算法[J].红外技术,2012,34(11):629-635.
[7]Chen Y J,Zhang X W,Zhang Y N,et al.Visible and infrared image registration based on region features and edginess[J].Machine Vision and Applications,2018,29(1):113-123.
[8]李寒,王库,刘韶军.基于灰度冗余和SURF算法的电气设备红外与可见光图像配准[J].电力系统保护与控制,2011,39(11):111-115,123.Li Han,Wang Ku,Liu Shaojun.Registration method between infrared and visible images of electrical equipment based on gray-level redundancy and SURF[J].Power System Protection and Control,2011,39(11):111-115,123.
[9]徐军,付天宇,杨健,等.采用显著性分析与改进边缘方向直方图特征的红外与可见光图像配准[J].光学精密工程,2016,24(11):2830-2840.Xu Jun,Fu Tianyu,Yang Jian,et al.Registration of infrared image and visible image based on saliency and EOH feature analysis[J].Optics and Precision Engineering,2016,24(11):2830-2840.
[10]吴一全,王志来.基于小波域改进SURF的遥感图像配准算法[J].天津大学学报(自然科学与工程技术版),2017,50(10):1084-1092.
[11]Fatiha M,Miloud C E M,Nasreddine T,et al.A rigid image registration based on the non-subsample Contourlet transform and genetic algorithms[J].Sensors,2010,10(9):8553-8571.
[12]梁栋,颜普,朱明,等.一种NSCT和SIFT的遥感图像配准算法[J].仪器仪表学报,2011,32(5):1083-1088.Liang Dong,Yan Pu,Zhu Ming,et al.Remote sensing image registration algorithm based on SIFT and NSCT[J].Chinese Journal of Scientific Instrument,2011,32(5):1083-1088.
[13]刘刚,周珩,梁晓庚,等.非下采样轮廓波域红外与可见光图像配准算法[J].计算机科学,2016,43(11):313-316.
[14]Lowe G D.Distinctive image features from scale-invariant key points[J].International Journal of Computer Vision,2004,60(2):91-110.
[15]Bay H,Tuytelaars T,Gool L V.SURF:speeded up robust features[C]//Proceedings of European Conference on Computer Vision,2006:404-417.
[16]金斌,周伟,从瑜,等.一种基于局部不变特征的SAR图像配准新算法[J].哈尔滨工程大学学报,2014,46(11):112-118.Jin Bin,Zhou Wei,Cong Yu,et al.A SAR image registration algorithm based on local variant feature[J].Journal of Harbin Institute of Technology,2014,46(11):112-118.
[17]王辉,王嘉梅,张红燕.基于NSST和FAST的图像配准方法[J].河南大学学报,2016,46(6):683-691.
[18]Rosten E,Drummond T.Faster and better:a machine learning approach to corner detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(2):105-119.
[19]Chen J,Tian J,Lee N,et al.A partial intensity invariant feature descriptor for multimodal retinal image registration[J].IEEE Transactions on Biomedical Engineering,2010,5(7):1707-1718.
[20]许金鑫,李庆武,马云鹏,等.基于斜率一致性的电气设备红外与可见光图像配准方法[J].光电子激光,2017,28(7):794-802.
[2]张璐,赵建.红外成像技术在电力系统后备电源热故障监测中的应用研究[J].电测与仪表,2012,49(8):93-96.Zhang Lu,Zhao Jian.Application research of infrared imaging technology on thermal fault monitoring for electrical power system backup power[J].Electrical Measurement&Instrumentation,2012,49(8):93-96.
[3]李伟,王伯雄,王雪,等.一种基于安瓿检测的快速图像配准方法的实现[J].电测与仪表,2010,47(7):5-8.Li Wei,Wang Boxiong,Wang Xue,et al.A fast image registration method applied in ampoule detection[J].Electrical Measurement&Instrumentation,2010,47(7):5-8.
[4]吴炬卓,牛海清,张煌,等.一种应用小波系数GSM模型的混合傅里叶-小波电缆瓷套终端红外图像去噪方法[J].电测与仪表,2018,55(4):113-117.Wu Juzhuo,Niu Haiqing,Zhang Huang,et al.A hybrid Fourier-wavelet denoising method for infrared image of porcelain sleeve cable terminal using GSM model for wavelet coefficients[J].Electrical Measurement&Instrumentation,2018,55(4):113-117.
[5]金立军,田治仁,高凯,等.基于红外与可见光图像信息融合的绝缘子污秽等级识别[J].中国电机工程学报,2016,36(1):3682-3691,3389.Jin Lijun,Tian Zhiren,Gao Kai,et al.Discrimination of insulator contamination grades using information fusion of infrared and visible images[J].Proceedings of the CSEE,2016,36(1):3682-3691,3389.
[6]纪利娥,杨风暴,王志杜,等.基于边缘图像和SURF特征的可见光与红外图像的匹配算法[J].红外技术,2012,34(11):629-635.
[7]Chen Y J,Zhang X W,Zhang Y N,et al.Visible and infrared image registration based on region features and edginess[J].Machine Vision and Applications,2018,29(1):113-123.
[8]李寒,王库,刘韶军.基于灰度冗余和SURF算法的电气设备红外与可见光图像配准[J].电力系统保护与控制,2011,39(11):111-115,123.Li Han,Wang Ku,Liu Shaojun.Registration method between infrared and visible images of electrical equipment based on gray-level redundancy and SURF[J].Power System Protection and Control,2011,39(11):111-115,123.
[9]徐军,付天宇,杨健,等.采用显著性分析与改进边缘方向直方图特征的红外与可见光图像配准[J].光学精密工程,2016,24(11):2830-2840.Xu Jun,Fu Tianyu,Yang Jian,et al.Registration of infrared image and visible image based on saliency and EOH feature analysis[J].Optics and Precision Engineering,2016,24(11):2830-2840.
[10]吴一全,王志来.基于小波域改进SURF的遥感图像配准算法[J].天津大学学报(自然科学与工程技术版),2017,50(10):1084-1092.
[11]Fatiha M,Miloud C E M,Nasreddine T,et al.A rigid image registration based on the non-subsample Contourlet transform and genetic algorithms[J].Sensors,2010,10(9):8553-8571.
[12]梁栋,颜普,朱明,等.一种NSCT和SIFT的遥感图像配准算法[J].仪器仪表学报,2011,32(5):1083-1088.Liang Dong,Yan Pu,Zhu Ming,et al.Remote sensing image registration algorithm based on SIFT and NSCT[J].Chinese Journal of Scientific Instrument,2011,32(5):1083-1088.
[13]刘刚,周珩,梁晓庚,等.非下采样轮廓波域红外与可见光图像配准算法[J].计算机科学,2016,43(11):313-316.
[14]Lowe G D.Distinctive image features from scale-invariant key points[J].International Journal of Computer Vision,2004,60(2):91-110.
[15]Bay H,Tuytelaars T,Gool L V.SURF:speeded up robust features[C]//Proceedings of European Conference on Computer Vision,2006:404-417.
[16]金斌,周伟,从瑜,等.一种基于局部不变特征的SAR图像配准新算法[J].哈尔滨工程大学学报,2014,46(11):112-118.Jin Bin,Zhou Wei,Cong Yu,et al.A SAR image registration algorithm based on local variant feature[J].Journal of Harbin Institute of Technology,2014,46(11):112-118.
[17]王辉,王嘉梅,张红燕.基于NSST和FAST的图像配准方法[J].河南大学学报,2016,46(6):683-691.
[18]Rosten E,Drummond T.Faster and better:a machine learning approach to corner detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(2):105-119.
[19]Chen J,Tian J,Lee N,et al.A partial intensity invariant feature descriptor for multimodal retinal image registration[J].IEEE Transactions on Biomedical Engineering,2010,5(7):1707-1718.
[20]许金鑫,李庆武,马云鹏,等.基于斜率一致性的电气设备红外与可见光图像配准方法[J].光电子激光,2017,28(7):794-802.