Research Of Images Registration Base On Multiple X-ray Mammography Images

With the breast cancer occurrence still growing, the breast cancer has become the most common form of cancer in the female population. Early detection, early diagnosis and early treatment are the most important methods in diagnosis. Because of a high spatial resolution and sensitivity for the breast mass and calcification, mammography is the first imaging choice. On the basis of the mass detection, this thesis focus on the registration problem of X-ray mammography, after that, this paper fuse the images of different views or different times into one image, in order to reduce the workload of doctors, provide a valuable reference for doctors. This article can be divided into three parts.Mass detection:The mass detection algorithms in this article are divided into three parts. Firstly, we presented improved dynamic programming segmentation methods to get a precise mass contour. Secondly, we fit the mass contour and obtain the RBST(Rubber band straightening transform) image of the mass. In the RBST image, we extract the texture feature based on the gray level co-occurrence matrix. Finally, we used an adaptive SVM to classify and identify the breast mass.Registration:In order to maintain the topology of the image, for intensity-based elastic registration problem, Gaussian smoothing is used to constrain the transformation to be smooth and thus preserve the topology of image. Aiming at the insufficiency of the uniform Gaussian filtering of the deformation fields, an automatic and accurate elastic image registration method based on B-splines approximation is proposed in this paper. In this approach, the regularization strategy is adopted by using multi-level B-splines approximation to regularize the displacement fields in a coarse-to-fine manner. Moreover, it assigns the different weights to the estimated displacements in terms of their reliability. In this way, the level of regularity can be adapted locally, so the estimated transformation is restricted to deformation satisfying the real-world property of matter.Fusion:After the image registration, this paper use the method of TOET to fuse the X-ray images of different views, in order to reduce the workload of doctors, provide a valuable reference for doctors.