The Infrared Image Processing Research Under Unvisible Conditions

With the rapid development of infrared technology, infrared thermalimaging technology has been widely applied to military and civilian areas, and the heatcontinues to heat up. However, because infrared imaging system has the longerwavelength imaging, coupled with the impact by the detection background and imagepickup device itself the objective conditions, resulting in poor quality infrared images,specifically non-uniformity of the poor, high background, low contrast, smallgrayscale range, poor image resolution, noise, pollution, etc., and thus to observe thetarget object, this type of imaging is far from ideal, so in order to improve the qualityof the infrared image, you need to go through some of the infrared image enhancementprocessing. Typically, in order to improve image quality, people often working onaspects of infrared detector materials, manufacturing processes, electronicscomponents, but current hardware has gradually become perfect, advanced infrareddigital image processing technology can effectively improve imaging performance,which received extensive attention of scholars, and it is now extremely importantinfrared image processing pretreatment technology.The authors start with the characteristics of infrared images, as well as traditionalalgorithms as the theoretical basis for the infrared image contrast, sharpness, and otheraspects of information entropy to study the infrared image enhancement andstrengthening effect, the paper work is as follows:This article first give a brief description of uncooled focal plane imaging system,encountered a detailed discussion and analysis about the non-uniformity of the imagegeneration process,and do some analysis of hardware to improve the non-uniformity, and then through to explore the types of infrared image noise, a method ofinfrared image noise detection was purposed. This article focuses on the infrared image enhancement, for image enhancement,the article firstly descripe the two types of enhancement algorithm in detail, andseveral commonly used key technologies in detail and analyze their strengths andweaknesses, and then aim at histogram equalization algorithms for a detaileddescription and analysis of the principles, on this basis, gradually expanding to theprovisions of the histogram and adaptive histogram equalization algorithm, and itssimulation, respectively, and made for improving the clarity and infrared imagesimproved information entropy histogram equalization algorithm matlab simulation anddiscuss the advantages and disadvantages.Then start from the contrast, this paper introduces the principles and methods ofhomomorphic filtering, analysis of its imaging features, on this basis, an improvedhomomorphic filtering algorithm was purposed, the simulation to discuss theadvantages and disadvantages of this method.Finally, on the basis of the previous text summary, this paper presents acomprehensive enhancement algorithm based on singular value decomposition andcombined wiith divided and homomorphic filtering, aim at image sharpness andcontrast enhancement corresponding improvement in simulation by matlab itsreliability, image infrared enhancement, and accordingly discussed the advantages anddisadvantages of a comprehensive approach.