数字图像增强算法的研究与实现

东北大学本科生毕业设计（论文）AbstractABSTRACTWhen human beings acquire and exchange information,image,as a carrier,can play animportant role.Relevant studies show that in the external information perceived by humanbeings,the information obtained from the visual channel accounts for more than 80%.Therefore,image processing can be fully used in all walks of life.The analysis shows thatthe so-called image enhancement refers to the classification of image information into twocategories according to the relevant needs,one is useless information;One is usefulinformation,and then the former is weakened or eliminated,and the latter is strengthenedand highlighted.The image obtained in this way is easier to recognize by machine andmore consistent with human visual characteristics.It can be found that image enhancementtechnology has been fully used in many industries,such as people photography andmedical imaging.In various image processing algorithms,such as target tracking algorithmor image fusion algorithm,image enhancement can play a great role as a preprocessingalgorithm.In image processing,image enhancement in preprocessing is of great significance andvalue.It mainly deals with the low-level image processing,and plays the role of pre andpost links.In order to achieve good results in high-level image processing,we must first doa good job in image enhancement,so that the image visual effect can be optimized andimproved,It is easier to recognize it by machine,and it is more consistent with thecharacteristics of human vision.People can quickly get the required data information byobserving such images.This research is based on the image enhancement theory of frequency domain method.Firstly,the low-pass /high pass filter of frequency domain method is studied.However,thesimulation of Butterworth filter and other related filters is completed with the help ofMATLAB.It is found that whether the filter belongs to low-pass filter or high pass filter,they can optimize and improve the image quality.Keywords:image enhancement;Digital image,contrast enhancement algorithm,frequency domain;filteringⅢ