Research On Focusing Algorithm Of Infrared Thermal Imager Based On Image Processing

The infrared thermal imager has the function of imaging infrared images and measuring temperature,and performs imaging and temperature measurement by detecting radiation energy density distribution in a specific wave band.This paper studies the focusing algorithm that based on the working principle of infrared thermal imaging camera.The purpose of this paper is to design a set of focusing scheme for it,and then it can clearly obtain images and accurately measure temperature.This paper,firstly,introduced radiation law used commonly,from the infrared radiation theory,and on this basis,the physical model of the thermal imager is analyzed,after theoretical derivation,the process of temperature inversion is given.The quantitative relationship between the output response value of the infrared imager and the temperature of object was established by radiation calibration method.Then,the optical characteristics of the defocusing process of the optical imaging system are analyzed,and the relationship between the image sharpness and the focal length of the lens is reached.This paper introduces three kinds of algorithms commonly used in image sharpness discrimination,such as grayscale gradient distribution,frequency domain distribution and grayscale statistics.Because of the traditional sharpness discrimination algorithm only judges the focusing degree of the system from the perspective of image processing,the experiment shows that the imaging quality has a great influence on the output value of the thermal imager signal response.Therefore,the method of combining the improved contour extraction with the response output value of the thermal imager is used to determine the degree of focus,and the principle is analyzed and explained.In the end,the principles and characteristics of the four commonly used search methods—method of exhaustion,Fibonacci algorithm,curve fitting and hill climbing method were explained.After the analysis,the search is combined with the exhaustive search method and the hill-climbing method.Finally,the infrared image that collected by the thermal imager was used to test the above algorithms.Among them,four quantitative indicators—sharpness comparison,local extremum,gentle fluctuation and sensibility—were used to evaluate and analysis the focus curve.The results showed that the algorithm with better test results was: improved edge contour extraction algorithm and improved SMD,Tenenbaum gradient algorithm.And based on this,by using C# and MATLAB mixed programming to achieve the focusing algorithm,and finally using Visual Studio tools to achieve the development of focusing software,so that it has function of statistics,three-dimensional display and other functions besides focusing.