Research On Image Processing Algorithm Based On Non-independent Identical Distribution

In recent years,the development of image processing technology is increasingly rapid,because image is the basis of information interaction in daily life,throughout the whole process of information perception,transmission and processing,which has far-reaching significance.Therefore,how to accurately and quickly obtain the main content contained in the image is particularly important.In view of the problems of traditional image segmentation methods for dealing with a single image type,easily to fall into false boundaries,this paper proposes a level set image segmentation algorithm based on non-independent and same distribution.The traditional image segmentation method basically assumes that all samples are independent and have the same distribution.However,since the samples are usually spatially-connected or temporally-correlated with their physically-connected neighbors,they are not IID,which cannot be directly handled by existing methods.Therefore,this paper proposes an image segmentation method based on non-independent identically distributed.Firstly,the super-pixel over segmentation is carried out,and the image to be segmented is divided into many super-pixel blocks.Secondly,the feature vectors are extracted from the pixel blocks by using the method of non-independent and same distribution.Finally,the feature vectors obtained are combined with the CV model of the level set,the energy function of the level set method is changed,and the new energy function is applied to the image segmentation.Experimental results show that the algorithm proposed in this paper can effectively segment all kinds of images,and also has better segmentation results for weak edge images.In view of the accuracy and efficiency of existing target detection methods,by considering the variability of ship spatial distribution in different image scenes and the characteristics of processing scene-specific ship detection tasks,this paper proposes a non-based ship detection method of independent and identical distribution scene classification.Firstly,the non-independent and identical distribution method is used to classify the scene,and the image is divided into two types of scenes on the coast and the sea,and then the ship detection is performed based on the prior knowledge of scene classification.Experimental results prove that our method has a relatively high detection accuracy and a relatively low false alarm rate.Based on pyqt5 platform,a ship detection software is designed and manufactured in this paper.Each function module of the software is designed and implemented,including super-pixel segmentation method and ship detection method.Experiments show that the software achieves good results,including the test results of different functional modules and the overall output of the software.