Font Size: a A A

Biological Cell Image Processing Using PCNN And Rough Set Theory

Posted on:2010-03-22Degree:MasterType:Thesis
Country:ChinaCandidate:D M LinFull Text:PDF
GTID:2178360275996231Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
Combining Pulse Coupled Neural Network (PCNN) and Rough set theory can get good achievements in digital image processing because of PCNN's biology background and Rough set theory's excellent ability to categorize difference things according to their characteristics. This thesis focused on the exploring of some achievements of their combination research in biological cell image processing such as enhancement, fusion and watermarking. First, the feasibility of the combination of PCNN and rough set theory to apply to image processing was analyzed. Then for the problems of biological cell image processing, carried out research to try to complete the following tasks.1. In some occasions, the background belongs to the bright region of an image, while the object belongs to the dark region. And sometimes, the image is corrupted by noise, or the contrast of the image is very low. Considering these images, a new algorithm for image enhancement based on PCNN time matrix and Rough set theory was proposed, and was used in the enhancement of biological cell images. From the experimental results we can conclude that the algorithm can effectively improve the image contrast, and can obtain good enhancement effect. At the same time, it can also suppress noise and retain image details. The enhancement effect is better than that of traditional methods.2. For the defects of imaging equipments and imaging principle, a multi-focus image fusion method for plant cell images was put forward by using PCNN and Rough set theory. Experimental results show that the proposed method is superior to some other fusion approaches, both in visual effect and objective evaluation criteria. And good anti-noise performance of this algorithm was also illustrated. It can effectively preserve image details such as edges and textures, producing clear images. Furthermore, the new algorithm is easy to achieve, having an obvious significance in practical applications.3. After giving the analysis of the need for biological cell images to be copyright protected, a new spatial-frequency-domain-based double watermarking algorithm was presented for biological cell images. Experimental results indicate that the algorithm has good invisibility and strong robustness. It is a feasible and effective digital image watermarking algorithm.
Keywords/Search Tags:Digital image processing, Pulse Coupled Neural Network, Rough set theory, Image enhancement, Multi-focus image fusion, Multiple digital watermarking
PDF Full Text Request
Related items