Font Size: a A A

Algorithm Fusion Method PCNN And Image Block Multi-focus Image Fusion

Posted on:2014-06-15Degree:MasterType:Thesis
Country:ChinaCandidate:K LiFull Text:PDF
GTID:2268330401453225Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
Multi-image fusion is one of the most important fields in image fusion.Some methods can be used to make a new image which is focused on each position based on two or more source images on the same background but with different focusing position. Nowadays, with the in-depth research, scholars have proposed many image fusion methods, such as Pyramid transform method, Wavelet transform method and Principal component analysis method. But according to the performance of image fusion, there is still a possibility to improve the performance.Pulse coupled neural network (Pulse Coupled Neural Network, PCNN) is a neural network based on the biological background. As the third generation of artificial neural network, there is a very wide range of applications in the field of image processing, and it can be used for image denoising, image segmentation, image enhancement, and so on. In the field of image fusion, the PCNN model also has good applicability.Block segment fusion method means a source image is divided into image blocks of equal size, and then some operator is used as the clarity index to calculate definition of each image block as the basis for image fusion.This paper gives a detailed introduction of PCNN model, the basic principles of this model and the realization of method are fully elaborated, and the application of PCNN model in image processing are discussed in detail, and then the paper introduces the blocks of image fusion method based on. Finally, the PCNN model combined with the blocks of image fusion method is applied to the multi-focus image fusion.In this paper, the main research work mainly includes some aspects as follows:Firstly, the structure,principle and characteristics of PCNN model were sum-marized, and then the simplified PCNN model and improved PCNN model were introduced. The application of PCNN model in image processing is also introdu-ced.Secondly, the basic conception of image fusion is introduced briefly, and multi-focus image fusion and some common algorithms of multi-focus image fusion are introduced.Subjective and objective evaluation indicators are described finally.Thirdly, by the idea of block segment of the images and calculation the image clarity of each block, a fusion algorithm is proposed based on simplified PCNN and the method of block segment of the image in which the energy of image gradient is used to measure the clarity of each block. Experimental results show that the performance of this algorithm is better than some common algorithm for multi-image fusion both in the subjective and objective evaluation.Fourthly, since there are too many parameters in the PCNN model, another fusion method is proposed based on the improved PCNN and the method of block segment of the image in which the spatial frequency of the image is used to measure the clarity of each block. The experimental results show that the proposed algorithm is feasible by comparing with conventional methods both in the subjective and objective evaluation.
Keywords/Search Tags:Multi-image fusion, Pulse coupled neural network, Block segment fusionmethod
PDF Full Text Request
Related items