A Detection Method of Tree Structure Terminal Point in Digital Image
A tree-like structure and digital image technology, applied in the field of image processing, can solve the problems of unrecognizable three-dimensional terminal points, insufficient robustness of branch direction changes, etc., to achieve the effect of enriching the scope of application, avoiding misjudgment, and accurate classification
Active Publication Date: 2021-06-22
HUNAN UNIV
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Second, the adjacent slice verification algorithm based on the Gaussian model used in step d is not robust enough to changes in branch directions, and cannot identify the three-dimensional terminal points of the tree structure almost perpendicular to the z direction
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0064] The algorithm of the present invention is realized by compiling a computer program, and the view is obtained by using Vaa3D software. The operating environment of the program is windows, programmed with matlab language, the overall flow chart of the method for detecting the three-dimensional end point of the tree structure in the three-dimensional image is as follows figure 1 shown.
[0065] First, a group of two-dimensional slice images is extracted from the three-dimensional image, and the tree-like two-dimensional end points in each two-dimensional slice image are detected by the following steps; the detected two-dimensional end points are used as candidate three-dimensional end points;
[0066] Step 1, first adopt the image enhancement method based on the Hessian matrix to process the two-dimensional image, then detect the two-dimensional high curvature point in the enhanced two-dimensional image, and use it as a candidate two-dimensional end point;
[0067] Step 2....
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for detecting terminal points of a tree-like structure in a digital image. The two-dimensional terminal point detection steps are as follows: step 1, detecting a two-dimensional high-curvature point in a two-dimensional image, and using it as a candidate two-dimensional terminal point; Step 2. Detect the local diameter of the tree structure around each candidate 2D terminal point; Step 3. Detect the real 2D terminal point from the candidate 2D terminal points based on the multi-scale diverging ray model. The three-dimensional end point detection step is as follows: firstly, extract a group of two-dimensional slice images from the three-dimensional image, detect the two-dimensional end point in each two-dimensional slice image, and use the detected two-dimensional end point as a candidate three-dimensional end point; Then, the true 3D endpoints are detected from the candidate 3D endpoints based on the endpoint visual priors. The invention has high detection accuracy.
Description
technical field [0001] The invention belongs to the technical field of image processing, and relates to a method for detecting terminal points of a tree structure in a digital image. Background technique [0002] In biomedical research, the reconstruction of the morphology of dendritic structures such as neurons, retinal vessels, and bronchi plays a crucial role. At present, many tree structure tracking methods rely on suitable seed points, and one of the most ideal seed points is the start point and end point of the tree structure, that is, the terminal point. If the position of the terminal points of the tree structure can be detected automatically and accurately, many reconstruction tools and methods can reconstruct the 2D / 3D shape of the tree structure. [0003] A method for automatic detection of nerve endings in a three-dimensional image stack is disclosed in the prior art [1]. The steps for automatic detection of targets are as follows: [0004] a. Detect two-dimens...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00G06T7/10G06T7/62G06K9/32G06K9/46
CPCG06T7/0012G06T7/62G06T7/10G06V10/25G06V10/464
Inventor 刘敏杨博陈伟迅
Owner HUNAN UNIV