Research On Image Enhancement Algorithm In Infrared Imaging System

With the rapid development of infrared imaging technology,infrared imaging systems are broadly used in military,industrial,and civilian fields.Due to the influence of factors such as the external environment of the infrared imaging system and device performance,infrared images show problems such as inconspicuous edge details,low contrast,and blurred human visual effects.These problems are not conducive to subsequent advanced processing such as target recognition and detection.Therefore,in order to improve the quality of infrared images,it is very important to study infrared image enhancement algorithms.This paper proposes two improved algorithms based on the research of traditional infrared image enhancement algorithms.The main research work is as follows:First,through a large amount of literature review,the research status of infrared image enhancement algorithms is summarized;then,starting from the basic characteristics of infrared images,the merits and Shortcomings of traditional enhancement algorithms are analyzed and compared;furthermore,based on the current technology and practical applications,the future research direction of infrared image enhancement algorithms is pointed out.Then,aiming at the deficiencies of traditional unsharp masking algorithms,an infrared image enhancement algorithm based on improved unsharp masking is proposed.The algorithm first preprocesses the original infrared image;then uses an adaptive bilateral filter to replace the traditional Gaussian filter,and at the same time adds an adaptive edge compensation technology that does not affect the layered detailed image on the basis of the filter to perform edge detail compensation;then perform adaptive gain processing on the layered detail layer,and perform adaptive platform histogram equalization processing on the base layer;finally,merge the processed base layer and detail layer to obtain an infrared enhanced image.The improved algorithm can well contain the noise in the infrared image,enhance the image contrast and brightness,enrich the edge detail information,and solve the problems such as noise amplification that occurs when the traditional unsharp masking algorithm enhances the infrared image,and the strong edges are excessively smoothed or excessively sharpened.Finally,aiming at the shortcomings of the traditional multi-scale Retinex algorithm,combined with the human visual characteristics,an improved multi-scale Retinex infrared image enhancement algorithm is proposed.The algorithm first improves the traditional multi-scale Retinex algorithm,replaces the Gaussian filter in the traditional multi-scale Retinex algorithm with an adaptive guided filter,and uses the optimal scale parameter in the filter to adaptively calculate the final weight of the multi-scale Retinex algorithm to make the algorithm adaptive;then the light component obtained by the algorithm is weighted histogram processing,and the reflection component is first processed with adaptive S-curve and then combined with human visual characteristics for further enhancement processing;Finally,the enhanced infrared image is obtained by combining the light component and the reflection component through a multiplication operation.The improved algorithm effectively enriches the detailed information and contrast in the image,enhances the definition of the infrared image.It solves the problems such as low contrast,local halo and poor visual effects for human eyes.In this paper,the indoor and outdoor infrared images are collected by the infrared imaging system built in the laboratory,and the simulation experiments of the above two improved algorithms are carried out in different scenes.The experiment indicates that the improved algorithm has certain advantages in both subjective visual evaluation and objective quantitative evaluation.The enhanced infrared image has more detailed information,higher contrast and clearer human visual perception.