Overview of Image cutting (I.)

Overview of Image cutting (I.)

[Email protected]

Http://blog.csdn.net/zouxy09

The so-called image cutting refers to the gray, color, texture and shape characteristics of the image into a number of non-overlapping areas, and make these features in the same region to show similarity, and in different regions show a significant difference between. We first outline the basic image cutting method now, and then do some specific understanding and learning of the individual methods.

1. Cutting method based on threshold value

The basic idea of threshold method is to calculate one or more gray threshold values based on the gray feature of the image, and compare the gray value of each pixel in the image with the threshold value, and then divide the pixels according to the comparison result into the appropriate category. Therefore, one of the most critical steps in this class of methods is to solve the best gray threshold values according to a criterion function.

2, Edge-based cutting method

The so-called Edge refers to the set of successive pixels on the boundary line of two different regions in the image, and is the reflection of the discontinuity of the local feature of the image, which reflects the mutation of the image characteristics such as grayscale, color and texture. In general, edge-based cutting method refers to the edge detection based on gray value, it is based on the edge of Gray value will show a step or roof type changes on the basis of the method.

There are obvious differences between the gray values of the pixel points on the edge of the step type, while the roof shape edge is at the turning point where the gray value rises or falls. Based on this feature, it is possible to use differential operators for edge detection, even with the extremum of the first derivative and the second derivative over 0 points to determine the edge, the detailed implementation can use the image and template convolution to complete.

3, Region-based cutting method

These methods are divided into different regions according to the similarity criterion, including seed region growth method, region splitting merging method and watershed method.

The seed region growth method starts from a group of seed pixels representing different growth regions, then merges the eligible pixels in the seed pixel neighborhood into the growth region represented by the seed pixel, and continues the merge process as a new seed pixel until the new pixels matching the criteria are found. The key of this method is to select the appropriate initial seed pixel and reasonable growth criterion.

The basic idea of the regional division merging method (Gonzalez,2002) is to divide the images randomly into disjoint regions and then divide or merge the areas according to the relevant criteria, and then complete the cutting task. This method is suitable for both grayscale image cutting and texture image cutting.

Watershed method (Meyer,1990) is a kind of cutting method based on topological theory, and its basic idea is to think of images as topological landforms on the Earth, and the gray value of each pixel in the image indicates the altitude of the point. Each local minimum and its affected area is called a catchment basin, while the boundary of the basin forms a watershed. The implementation of this algorithm can simulate the process of flooding, the lowest point of the image is submerged first, and then the water gradually drowns the whole valley. When the water level reaches a certain height will overflow, when the water overflow in the area of the construction of dams, repeated this process until the entire image of the point is submerged, then the establishment of a series of dams will be divided into the watershed of each basin. The watershed algorithm has a good response to the weak edge, but the noise in the image causes the watershed algorithm to cut.

4. Cutting method based on graph theory

Such a method associates the problem of image cutting with the minimum cut (min cut) Problem of the graph. First, the image is mapped to a weighted g=<v,e>, each node N∈V corresponds to each pixel in the image, each edge ∈E is connected to a pair of adjacent pixels , the weighted value of the edge represents the non-negative similarity between neighboring pixels in terms of grayscale, color, or texture. And the image of a cut s is a cut of the graph, the cut of each area C∈s corresponding to a sub-graph in the diagram. The optimal principle of cutting is to keep the divided sub-graphs within the maximum similarity, while the similarity between the sub-graphs remains minimal. The essence of the cutting method based on graph theory is to remove the specific edges and divide the graph into several sub-graphs to realize the cutting. There are graphcut,grabcut and Random Walk , which are based on graph theory.

5. Cutting method based on energy functional

The method mainly refers to the active contour model (activeContourmodel) and the algorithm developed on it, the basic idea is to use continuous curve to express the target edge, and define an energy function to make its independent variable contain the edge curve, Thus the cutting process is transformed into the process of solving the minimum value of the energy functional, which can be solved by solving the function of Euler (Euler). Lagrange) Equation to achieve, the curve position where the energy reaches the hour is the contour of the target. The active contour model can be divided into two categories according to the different curves in the model: the reference active contour model (parametric active Contourmodel) and the geometric Active contour model (geometric active Contour Model).

The reference activity Contour model is based on the Lagrange frame, which expresses the curve directly in the form of the curve, and the most representative is the Snake model proposed by Kasset A1 (1987) . This kind of model has been successfully applied in the field of early bio-image cutting, but it has many disadvantages such as the influence of the initial contour setting and the difficulty to deal with the change of the topological structure of the curve, and its energy function depends only on the selection of curve parameters, which is independent of the geometrical shape of the object, which limits its further application.

The curve motion process of the geometric active contour model is based on the geometric metric parameter of the curve and not the expression of the curve, so it can deal with the change of topological structure well and solve the problem that the reference activity contour model is difficult to solve. The introduction of thelevel set method (Osher,1988) has greatly promoted the development of the geometric active contour model, so the geometric active contour model can also be called the level set method.

Overview of Image cutting (I.)