Freely differences calculus and deformable contour outline extracting system

Abstract

The invention comprises: a image memory 1, a component for executing the high level vision 2, a basic polygon transform component 3, an orthogonal location transform component 4, a free difference operation component 5, a target outline extracting component 6, and an energy function minimizing component 7.

Description

technical field

[0001] The invention relates to image processing, especially a method for implementing edge detection in image processing and a method for implementing deformable contour extraction in the field of image segmentation. Background technique

[0002] In various image applications, as long as it involves the extraction and measurement of image objects, image segmentation is inseparable. Image segmentation refers to extracting the target region or region of interest in the image by using some features in the image information. Usually, the target region or region of interest is called the foreground region, and the rest is called the background region.

[0003] Image segmentation is a key step from image processing to image analysis. As far as specific image segmentation algorithms are concerned, there are thousands of related research literatures. These studies mainly adopt the following two modes: the first one is "bottom-up" processing, including several sta...

Examples

Embodiment 1

[0031] Embodiment 1: Extracting the outline of a human face.

[0032] Next, the operation of the first embodiment of the present invention will be described. Fig. 3 is a flow chart showing the operation of image segmentation according to the first embodiment. In order to facilitate understanding of the operation of the first embodiment, FIG. 4 shows an image of a human face. The size of the face image is 92×112, and the gray scale is 256 levels. The purpose of this embodiment is to extract the contour of the human face. This means that the present invention can be applied to a system for extracting human facial contours. It mainly includes the following steps:

[0033] Step 1: read in the original data of the two-dimensional image;

[0034] Step 2: Preprocessing the image;

[0035] Step 3: The user inputs the initial point column;

[0036] Step 4: Initialization of parameters of the deformable contour extraction system;

[0037] Step 5: Construct basic polygons;

[00...

Embodiment 2

[0091] Embodiment 2: Extracting the outline of a human face.

[0092] The only difference between the operation of the second embodiment of the present invention and the operation of the first embodiment is that the absolute difference of the gradient change is taken in step 4. FIG. 11 shows the extracted facial contour curve. Through the comparison of Embodiment 1 and Embodiment 2, it is not difficult to see that in the deformable contour extraction system provided by the present invention, when calculating pixel gray level changes based on free difference operation, the calculation mode can be flexibly selected according to the characteristics of the image.

Embodiment 3

[0093] Embodiment 3: Extracting the edge of the human brain.

[0094] This embodiment provides an implementation of the present invention with constraints provided according to the characteristics of the target contour. Figure 12 shows the MR image of a human head, the image size is 253×275, and the gray scale is 256 levels. The initial contour is a straight line between two points A and B, and the selection range of points on the perpendicular line is limited to the rectangular area formed by four points A, C, B and D. The calculation method of the pixel gray level change is that the gradient direction is from the inside to the outside, and the relative difference is taken. The evolution process is that the initial contour expands to both sides respectively and splits into two curves. These two curves remain connected at two points A and B to form a closed curve; then, starting from any point on the curve, proceed clockwise Select a point every certain step in the direction...

technical field

[0001] The invention relates to image processing, especially a method for implementing edge detection in image processing and a method for implementing deformable contour extraction in the field of image segmentation. Background technique

[0002] In various image applications, as long as it involves the extraction and measurement of image objects, image segmentation is inseparable. Image segmentation refers to extracting the target region or region of interest in the image by using some features in the image information. Usually, the target region or region of interest is called the foreground region, and the rest is called the background region.

[0003] Image segmentation is a key step from image processing to image analysis. As far as specific image segmentation algorithms are concerned, there are thousands of related research literatures. These studies mainly adopt the following two modes: the first one is "bottom-up" processing, including several sta...

Examples

Embodiment 1

[0031] Embodiment 1: Extracting the outline of a human face.

[0032] Next, the operation of the first embodiment of the present invention will be described. Fig. 3 is a flow chart showing the operation of image segmentation according to the first embodiment. In order to facilitate understanding of the operation of the first embodiment, FIG. 4 shows an image of a human face. The size of the face image is 92×112, and the gray scale is 256 levels. The purpose of this embodiment is to extract the contour of the human face. This means that the present invention can be applied to a system for extracting human facial contours. It mainly includes the following steps:

[0033] Step 1: read in the original data of the two-dimensional image;

[0034] Step 2: Preprocessing the image;

[0035] Step 3: The user inputs the initial point column;

[0036] Step 4: Initialization of parameters of the deformable contour extraction system;

[0037] Step 5: Construct basic polygons;

[00...

Embodiment 2

[0091] Embodiment 2: Extracting the outline of a human face.

[0092] The only difference between the operation of the second embodiment of the present invention and the operation of the first embodiment is that the absolute difference of the gradient change is taken in step 4. FIG. 11 shows the extracted facial contour curve. Through the comparison of Embodiment 1 and Embodiment 2, it is not difficult to see that in the deformable contour extraction system provided by the present invention, when calculating pixel gray level changes based on free difference operation, the calculation mode can be flexibly selected according to the characteristics of the image.

Embodiment 3

[0093] Embodiment 3: Extracting the edge of the human brain.

[0094] This embodiment provides an implementation of the present invention with constraints provided according to the characteristics of the target contour. Figure 12 shows the MR image of a human head, the image size is 253×275, and the gray scale is 256 levels. The initial contour is a straight line between two points A and B, and the selection range of points on the perpendicular line is limited to the rectangular area formed by four points A, C, B and D. The calculation method of the pixel gray level change is that the gradient direction is from the inside to the outside, and the relative difference is taken. The evolution process is that the initial contour expands to both sides respectively and splits into two curves. These two curves remain connected at two points A and B to form a closed curve; then, starting from any point on the curve, proceed clockwise Select a point every certain step in the direction...

Images