Research On Image Quality Evaluation Of Optical Surcey And Control System

Since the 20 th century,with the acceleration of the pace of human exploration in space,aerospace related fields have been developed by leaps and bounds,and practical development needs have been put forward for aerospace optical measurement and control.For modern space optical observation systems,the evaluation of the actual performance of optical measurement and control equipment and the analysis of key factors in image quality of remote sensors all require the evaluation of image quality.However,the current main test methods still have problems such as experimental conditions and differences from the actual requirements of the aerospace measurement and control optical system.Based on this,this paper mainly studies how to get rid of experimental conditions and be able to independently and efficiently realize the image quality evaluation method.Then,the comprehensive performance of the space measurement and control optical system was evaluated,and the following work was completed:First of all,according to the characteristics of the space optical measurement and control system,the subjective image quality evaluation and objective quality evaluation methods are analyzed,and from the three aspects of full reference,partial reference and no reference.The application of peak signal-to-noise ratio method,structural similarity method,natural field method and differential sensory valve limit method in aerospace optical measurement and control system is discussed.Secondly,on the basis of detailed analysis of HVS function characteristics and wavelet transform related content,using local band contrast,the CSF characteristics of the human eye vision system are proposed,and a new HVSNRC evaluation method without reference images is constructed,and experimental verification is carried out.The experimental results are more satisfactory.Thirdly,on the basis of low pass filtering,DCT transform,and similarity degree of computational structure,a new method for evaluating the quality of non-reference images based on sharpness is presented,and experimental verification is carried out.The results show that the similarity with human eye recognition is higher.Finally,according to the characteristics of the space launch measurement and control mission,the relationship between the structure characteristic parameters,mean and variance,spectrum vectors,Q quality parameters,etc.and the modulation and transfer function of the image is compared and analyzed.In order to solve the problem of slow convergence speed of BP neural network,an improved algorithm is put forward,and various configuration parameters such as the number of layers of network hidden,the number of nodes,and the incentive function are determined.And the network is trained and tested experimentally.