Volume Rendering Method for Automatic Identification of Overlapping Structures Based on Two-Dimensional Transfer Function

Abstract

The invention discloses a volume rendering method for automatically identifying overlapping structures based on a two-dimensional transfer function. The steps are: (1) reading volume data, and calculating each voxel according to two attribute values ​​required by the two-dimensional transfer function space, And thus construct the transfer function space; (2) In order to obtain the complete structure of interest and eliminate the over-segmentation problem of the watershed algorithm, first set the transfer function classifier component in the transfer function space constructed in step (1) and determine the structure of interest , and then automatically correct the gray value of the original volume data; (3) for the corrected volume data, use the watershed algorithm to obtain the classification result based on its gray value, so that the structure of interest overlapping in the two-dimensional transfer function space identified at the same time. The principle of the invention is simple, easy to understand, and easy to realize in the existing two-bit transfer function space and visualization system.

Description

technical field [0001] The invention belongs to the field of three-dimensional visualization in computer graphics, and in particular relates to a new method for identifying structures of interest in volume data, that is, a volume rendering method for automatically identifying overlapping structures based on two-dimensional transfer functions. Background technique [0002] With the application of new image acquisition technologies (such as CT, MRI and PET, etc.) in clinical medicine, a large amount of medical volume data is generated. In order to better assist doctors in processing and analyzing these three-dimensional data, new data are needed Analysis and visualization methods. Direct volume rendering (Direct Volume Rendering) is a 3D visualization method that can show the internal structure of tissues. Because each pixel value of the final rendering result is generated by the accumulation of the optical property values ​​(color value and opacity value) of all voxels on a ...

AI Technical Summary

Problems solved by technology

[0005] Although the two-dimensional transfer function space has an intuitive interface and better recognition ability, the brightness value of each pixel in the two-dimensional space is determined by the number of voxels containing the attribute pair in the volume data, so when different When the structure of the transfer function space has a similar area, it cannot be identified directly by setting the transfer function

E.g Figure 5 A simulated data in (a) containing 2 spatially connected spherical structures O 1 and O 2 , if they are to be separated by a two-dimensional transfer function, they must correspond to two 'arch' structures in space, but since these two structures have the same attribute information, they are projected to the same position in the transfer function space ( Figure 5 (b)), so it cannot be separated by setting the transfer function

Images