Research On Infrared And Visible Image Registration Technology

In order to further improve information monitoring capabilities and reduce labor costs,multi-dimensional information integration technology is particularly important.As the primary link of multi-dimensional information integration technology,image registration technology has attracted the attention of more and more researchers,especially the registration direction of infrared and visible light images,which has a wide range of application prospects.If low-cost infrared and visible light image registration technology can be developed and applied to the field of information monitoring,it will greatly reduce the labor cost of monitoring and testing.Firstly,this paper studies the four types of image registration methods.In the field of infrared image and visible light image registration,methods based on image features and domain transformation are mostly used.But the method of domain transformation has a large amount of calculation,and the registration operation is related to the state before and after the transformation,and its adaptability is weak.For image features,the distance between the features in the infrared image and the visible light image is large,and it is not easy to perform feature matching.To make up for this shortcoming,this paper has made some improvements to the method based on image features,and proposed a multi-granularity image registration method.First,the traditional distancebased dual-target method is used to coarsen the image under the calibrated granularity.After registration,precise registration based on feature matching is performed at the feature granularity,and the ORB feature matching algorithm is improved to improve the speed and accuracy of feature matching,thereby improving the efficiency of image registration.It mainly completes environment construction,design of power supply reverse connection circuit,camera parameter calibration,image correction,design of calibration board,coarse registration based on depth information under calibration granularity,and fine registration based on improved ORB feature matching under characteristic granularity.Through experiments,the multi-granularity-based image registration method proposed in this paper can achieve precise registration of infrared images and visible light images,and through experiments with recorded video data,the registration operation can also be performed in real time in the video.And through experimental comparison with several methods,it is proved that the infrared and visible image registration method based on multi-granularity proposed in this paper can achieve better registration effect,it can greatly improve the accuracy of image feature matching,realize the accurate registration of infrared and visible light images in the whole process,and have been tested in timeliness,which can meet the needs of daily life and production.