Font Size: a A A

Study On Image Fusion Algorithm

Posted on:2011-09-07Degree:MasterType:Thesis
Country:ChinaCandidate:Y X QinFull Text:PDF
GTID:2178360305965022Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
The image fusion technology is fully using the image sensor with complementary and redundancy of multiple source images, by overall treatment, the fusion image is more clearly, more complete and reliable than any single sensor generated, to improve the performance of the system. Multi-sensor image fusion technology can be widely used in depth imaging of digital cameras, the pattern of medical image display, the remote sensing image classification, robot vision, virtual reality,3d image reconstruction and terrorist safe inspection, etc. As discussion and research on the classification problem, based on image fusion algorithm of multi-resolution and fusion image quality evaluation method, the evaluation indexes is used in image fusion process, put forward concerning the image fusion of new ideas and methods.According to the different source image, the source image fusion can be divided into:image fusion, multi-sensor image fusion, remote sensing image fusion multi-source and time series (dynamic) image fusion. People in the process of image fusion research found it difficult to design an image fusion algorithm to adapt to all kinds of source images, so in the practical application different image fusion algorithm is designed according to different source images. Based on the analysis of the image and medical image characteristics, on the basis of a general image evaluation index, a method suitable for both image and medical image fusion is proposed and proves that this is effective.Make the image fusion evaluation information to join parameter selection and fusion rules of process, can make full use of the information provided by source, an image fusion model with information feedback is formed, the results will get more than open-loop image fusion process better effect. Multi-scale multi-resolution methods of image fusion is the most commonly used in a way, because through the scale and (or) direction decomposition, providing more representative features of input image and superior fusion performance. In order to give objective adaptive fusion method, this paper employs the concept of objective evaluation index in the system of multi-scale multi-resolution fusion; the fusion algorithm adaptively and widely can be used in various places.
Keywords/Search Tags:image fusion, multiresolution decomposition, wavelet transform, evaluation criteria
PDF Full Text Request
Related items