Research On Restoration Technology Of Motion Degraded Image

In the process of motion imaging,the relative motion between the imaging device and the object causes the image blur,This phenomenon is called motion blur.This blur which caused by motion has the greatly affects on the image resolution,and makes the acquisition information of image become weak,for further image processing such as feature extraction,object recognition,and analysis aspects of the adverse,and so on,it seriously affects the accurate judgment of things,and it is not conducive to the implementation of management decision.Therefore,it is necessary todeal with it,which in order to improve the identification of information,and for the further study of image matting.In this paper makes the motion blurred image as the research object,firstly,the research background and practical significance of the project are described,and the status at home and abroad are summarized.On this basis,the paper describes the main contents,objectives,specific ideas and implementation.Secondly,the paper study the outside interference of the image,and its elimination,according to the research goal,the paper summarize the advantages and disadvantages about the commonly used restoration algorithms and applications,and select the restoration algorithm which has the good performance for the follow restoration work.Finally,the degradation model of uniform linear motion blur is studied.though the deduction we know that the key of image restoration is the accurate estimation of the spread function PSF.and by analyzing the characteristics of image spectrum and Cepstrum in frequency domain,The experiment proves that the cepstrum method is not limited by length and width of the image,and the paper proposes enhancement filter and histogram equalization preprocessing to the cepstrum figure of the motion blurred image,and get clear cepstrum map,and in the end we completed the research and improvement of the algorithm based on cepstrum algorithm.The main result is that presented the picture cepstrum pretreatment.in the clear cepstrum,according the characteristics that bright fringe and movement have the same direction,we determine the fuzzy direction;and using the symmetry of fuzzy scale,we get the corresponding PSF(point spread function).And then use Lucy-Richardson method for image restoration.Based on the preprocessing cepstrum algorithm,this paper makes the following innovations in terms of image definition,peak signal-to-noise ratio and application value:(1)The other documents and the restoration results show that the algorithm of image edge contour clear,retains more image details,from the objective evaluation level,improve the peak signal-to-noise ratio.(2)Using multiple images,aiming at different scales,different angles,restoration,restoration of the evaluation result proved that using the same preprocessing algorithm can improve the image quality,demonstrate the effectiveness of pretreatment methods however,inverted spectrum preprocessing algorithm based on preprocessing method in the same picture,the recovery of high quality.(3)For the root of our country agriculture,whether the aircraft or other equipment to collect information on field crops,in certain situations,it will seriously affect the decision of the judge of motion blurred image caused by the relative motion,will be presented in this paper for the restoration method of crop diseases and insect pests,weeds,drought and flood,a good foundation Quefei the subsequent feature extraction and target recognition,and provides a new method and a reliable theoretical basis for image restoration.