Composite material CT image segmentation method based on improved watershed algorithm

Abstract

The invention discloses a composite material CT image segmentation processing method based on an improved watershed algorithm and morphological evaluation, and the method comprises: carrying out the preprocessing of an original image, strengthening the local features of the original image, obtaining an initial mark point through morphological processing and distance transformation, and carrying out the pre-segmentation of the global image through employing a self-adaptive h value selection algorithm and the watershed algorithm; and carrying out effectiveness evaluation on each connected regionin the segmented marks through region effectiveness indexes, carrying out local h value adaptive selection and watershed algorithm segmentation on regions with effectiveness below a set standard, andcarrying out iteration until almost all connected regions meet the requirements of the effectiveness indexes, thereby obtaining a final algorithm segmentation result. Aiming at the characteristic oflow local contrast, the accuracy of edge detection and instance segmentation is improved through a local adaptive iterative segmentation strategy of an h value; and an effectiveness index is established for the intrinsic morphological characteristics of the research object, so the under-segmentation region is accurately identified, and whether to continuously implement the local segmentation algorithm or not is judged.

Description

technical field [0001] The invention relates to a technology in the field of image processing, in particular to a composite material CT image segmentation method based on an improved watershed algorithm. Background technique [0002] Image segmentation processing refers to the effective separation of regions of interest or individual regions of different categories in the image to be processed, so that the basic components of the image can be effectively divided, thereby effectively improving the pertinence and accuracy of the subsequent image feature extraction process. [0003] Composite materials can obtain their CT cross-sectional images through XCT scanning, so as to intuitively describe the spatial distribution of their internal fibers. However, due to the limitation of scanning accuracy and image resolution, simple image morphology processing cannot effectively divide the fiber filaments in the image. , resulting in the phenomenon of adhesion between different individ...

AI Technical Summary

Problems solved by technology

In view of the low resolution and low contrast between individuals in composite CT images, existing methods are difficult to separate different individuals and lack applicability to composite CT images, resulting in under-segmentation of images.

Images