ABSTRACT
The accurate registration of insulator multispectral images is an important prerequisite for high-quality online evaluation of the status of transmission line insulators using multi-lens cameras carried by drones. An insulator multispectral image registration model based on SuperPoint feature point extraction algorithm and SuperGlue feature matching algorithm was constructed in this paper. Then, a multi-lens multispectral camera was used to collect multispectral image information of insulators from different angles, lighting conditions, and spectral bands, and comprehensive mutual information was introduced as an index to analyze and compare common traditional image registration models and the model constructed in this paper. Finally, a decision model for insulator multispectral image matching was constructed, providing a key prerequisite for evaluating insulator status using multispectral images.
- Chen Lilin, Zang Haiyang, Wei Zhinong and Sun Guoqiang, "Regional level ultra short term photovoltaic power prediction considering multispectral satellite remote sensing," Proceedings of the CSEE 42(20), 7451-7465 (2022).Google Scholar
- Wang Liuwang, "Overview of the application of machine vision technology in power safety monitoring," Zhejiang Electric Power 41(10), 16-26 (2022).Google Scholar
- Chen Cheng, "Application of multispectral UAV in agricultural remote sensing," Modernizing Agriculture 2021(10), 61-62 (2021).Google Scholar
- Chen Jianwei, Gong Hui and Yuan Jing, "Multispectral imaging technology and its application in biomedicine," Laser & Optoelectronics Progress 58(04), 9-20 (2021).Google Scholar
- Li Yunhong, Liu Yudong, Su Xueping, Luo Xuemin and Yao Lan, "Overview of research on infrared and visible light image registration technology," Infrared Technology 44(07), 641-651 (2022).Google Scholar
- Ma Jiayi, Jiang Xingyu, Fan Aoxiang, Jiang Junjun and Yan Junchi, "Image matching from handcrafted to deep features: a survey," International Journal of Computer Vision, 23-79 (2021).Google Scholar
- DeTone D, Malisiewicz T and Rabinovich A, "SuperPoint: Self-Supervised Interest Point Detection and Description," 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 337-33712 (2018).Google Scholar
- Sarlin P-E, DeTone D, Malisiewicz T and Rabinovich A, "SuperGlue: Learning Feature Matching with Graph Neural Networks," 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 4937-4946 (2020).Google Scholar
- Yang Guangxi, "Application of SIFT image registration algorithm in oilfield drone technology," Information System Engineering 2023(06), 60-63 (2023).Google Scholar
- He Ling, "Research on visual automatic tracking technology for monocular unmanned aerial vehicles based on ORB-SLAM," Agricultural Engineering and Equipment 50(01), 42-44 (2023).Google Scholar
- Zhang Qian, Wang Jian, Chu Ruibo and Chen Huanhuan, "A fast registration method for unmanned aerial vehicle images based on BRISK-BEBLID features," Laser Journal 44(06), 92-98 (2023).Google Scholar
- Chen Yong, Wang Zhen and Lu Chentao, "Improved AKAZE algorithm for feature matching of high speed rail contact network image," Progress in Laser and Optoelectronics 59(10), 130-138 (2022).Google Scholar
- He Kang, Ren Shaojun and Si Fengqi, "Application of automatic clustering algorithm based on mutual information in fault diagnosis process," Thermal Power Engineering 38(04), 172-180 (2023).Google Scholar
Recommendations
An Unsupervised Multispectral Image Registration Network for Skin Diseases
Medical Image Computing and Computer Assisted Intervention – MICCAI 2023AbstractMultispectral imaging has a broad, promising and advantageous application prospect in the diagnosis of skin diseases. However, there are inherent deviations such as rigid or non-rigid deformation among multispectral images (MSI), which makes ...
Unregistered Hyperspectral and Multispectral Image Fusion with Synchronous Nonnegative Matrix Factorization
Pattern Recognition and Computer VisionAbstractRecently, many methods have been proposed to generate a high spatial resolution (HR) hyperspectral image (HSI) by fusing HSI and multispectral image (MSI). Most methods need a precondition that HSI and MSI are well registered. However, in practice,...
Feature and Intensity Based Medical Image Registration Using Particle Swarm Optimization
Image registration is an important aspect in medical image analysis, and kinds use in a variety of medical applications. Examples include diagnosis, pre/post surgery guidance, comparing/merging/integrating images from multi-modal like Magnetic Resonance ...
Comments