Font Size: a A A

Research On Microscopic Image Fusion

Posted on:2010-08-20Degree:MasterType:Thesis
Country:ChinaCandidate:G X DiFull Text:PDF
GTID:2178360272482742Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
As a powerful tool of human to explore the microcosm, microscope has been widely used in materials testing, cell histology, geological and archaeological and so on. However, in the microscopic image processing, some problems was often faced: It is well known that optical lenses of microscope normally have high resolving power and suffer from the problem of limited depth of focus. Sincerely a small part of each image slice is in focus, the visibility of the objects in a single image is limited, even for relatively flat specimens with 3D structures. This problem can be solved by microscopic image fusion. Microscopic image fusion is used to produce a completely clear image by fusing the registered images of the same scene with different focus, which are under the same imaging conditions. Therefore, Wavelet based Contourlet is more appropriate for microscopic image fusion tasks and better performance would be expected. The main research work in the dissertation is as follows:1. A systematic overview of image fusion theory and methods is presented. And image fusion evaluation criteria, mainly including subjective evaluation and objective evaluation standard, was introduced in detail.2. Detailed analysis is provided for the advantages of Multiscale Geometric Analysis (MGA) from the view of statistics of natural image and the functionality of human vision system. A systematical research of the development and common analysis tools are also given, including Ridgelet, Curvelet, Beamlet, Bandelet, etc. For Contourlet transform and Wavelet based Contourlet Transform, we provide a detailed analysis to its basic principles, implementation schemes, advantages and disadvantages, which can be seen as the foundation of following image processing applications.3. A method for multi-focus image fusion was proposed based on wavelet based contourlet transform and region statistics. The wavelet based contourlet transform has the advantages of multi-scale and multi-directional. At the same time, the wavelet based contourlet transform reduces the redundancy in the contoulet transform. Multi-resolution analysis of the image results using the transform in several highpass subbands representing efficiently the detailed information of the image. Firstly, the wavelet based contourlet transform is performed on source images to get high-frequency parts and low-frequency part. Then, region activity and regional variance are adopted as fusion rules in high-frequency parts and low-frequency part, respectively. Finally, the fused image is obtained by inverse transform. Extensive fusion experiments have been made on standard images and own images, both visual and quantitative analysis show that comparing with conventional image fusion algorithms, the proposed approach can provide a more satisfactory fusion outcome.4. According to the actual problem, a microscopic image fusion system was developed. The system could finish matching and fusion for any two microscopic images.
Keywords/Search Tags:Wavelet based contourlet transform, Multiscale Geometric Analysis, Multi-focus image, Microscopic image fusion
PDF Full Text Request
Related items