A three-dimensional image segmentation method based on a subjective threshold value and a three-dimensional label technology

Abstract

The invention discloses a three-dimensional image segmentation method based on a subjective threshold value and a three-dimensional label technology. Based on a subjective threshold segmentation method and a three-dimensional point diffusion label technology, the offset and false detection missing detection generated in a three-dimensional image in a traditional two-dimensional segmentation methodare solved, the connection between the two-dimensional images is well reserved, and the problem generated by image registration is avoided; A window adjusting function is added to preprocess the image, the contrast ratio and the brightness of the separated target area are adjusted, the gray value difference between the target area and redundant image information is increased, the threshold segmentation accuracy is remarkably improved, unnecessary information is removed in time through a three-dimensional label technology, and the integrity and the unity of the segmented image are guaranteed.Compared with a traditional two-dimensional segmentation technology, a reasonable threshold value can be found more effectively, three-dimensional areas of organs can be separated more conveniently and quickly, a good foundation is provided for later three-dimensional reconstruction, and the method is generally suitable for various organs of a human body of CT pictures with different image qualities.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a three-dimensional image segmentation method based on subjective threshold and three-dimensional label technology. Background technique [0002] Nowadays, the traditional two-dimensional image processing technology continues to mature, but medical images such as CT and MRI images are all three-dimensional medical image data. Accurate and fast segmentation of three-dimensional medical images has always been the difficulty and focus of three-dimensional organ reconstruction. Segmentation methods such as detection, threshold value extraction, and region growing all have certain limitations, and direct processing of 3D images cannot achieve ideal results. Therefore, 3D organ extraction from medical images has become one of the key issues studied by researchers in the field of medical images. one. [0003] In the segmentation of 3D medical images, the segmentation method of edge dete...

AI Technical Summary

Problems solved by technology

[0004] The edge detection segmentation method is highly targeted, and it needs a lot of improvement in the algorithm for different organs, and if there is an error, it is difficult to correct it again, and the edge loss or distortion often occurs in practical applications, which is difficult to be used as the gold standard in the evaluation system. standard to use