CN108846856B

CN108846856B - Picture feature point tracking method and tracking device

Info

Publication number: CN108846856B
Application number: CN201810600445.1A
Authority: CN
Inventors: 罗汉杰
Original assignee: Guangzhou Shiyuan Electronics Thecnology Co Ltd
Current assignee: Guangzhou Shiyuan Electronics Thecnology Co Ltd
Priority date: 2018-06-12
Filing date: 2018-06-12
Publication date: 2020-11-03
Anticipated expiration: 2038-06-12
Also published as: CN108846856A

Abstract

The invention relates to a method and a device for tracking picture characteristic points, wherein the method comprises the following steps: determining a gray gradient value of the target characteristic point in the normal vector direction according to the gray value of the target characteristic point of the original image; acquiring a gray value of an initial tracking point of a target image; and tracking the position of the target characteristic point in the target image along the normal vector direction according to the gray value and the gray gradient value of the target characteristic point and the gray value of the initial tracking point. The method can track the position of the target feature point in the target image along the normal vector direction when the target feature point is tracked, improves the robustness and stability of tracking the target feature point, is beneficial to increasing the matching degree of tracking the target feature point in the target image, shortens the time of tracking the target feature point, improves the tracking efficiency, and is beneficial to providing more effective data support for subsequent image information processing such as machine vision work.

Description

Picture feature point tracking method and tracking device

Technical Field

The present invention relates to the field of image processing technologies, and in particular, to a method and an apparatus for tracking picture feature points, a computer device, and a computer-readable storage medium.

Background

In computer vision systems, it is often necessary to process images, and in particular to identify and track objects in the images. One of the commonly used implementation methods is to extract some stable and strong-robustness pixel points from each image as feature pixel points of the image, and then perform matching tracking on the feature pixel points among different images by using a target tracking algorithm such as an optical flow method.

However, when the traditional target tracking algorithm tracks the characteristic pixel points of the original image, the matching degree of the characteristic pixel points in the target image is easily low, the robustness of tracking the characteristic pixel points of the original image is reduced, and the tracking time of the traditional mode for the characteristic pixel points is too long, so that the tracking efficiency is also reduced.

Disclosure of Invention

Therefore, it is necessary to provide a method, an apparatus, a computer device, and a computer-readable storage medium for tracking picture feature points, in order to solve the problem of low robustness in tracking feature pixel points of an original image in the conventional technology.

A method for tracking picture feature points comprises the following steps:

determining a gray gradient value of a target characteristic point of an original image in a normal vector direction according to the gray value of the target characteristic point; the normal vector is a vector indicating the direction in which the gray scale of the target feature point changes most rapidly;

acquiring a gray value of an initial tracking point of a target image;

and tracking the position of the target feature point in the target image along the normal vector direction according to the gray value and the gray gradient value of the target feature point and the gray value of the initial tracking point.

The image characteristic point tracking method determines the gray gradient value of the target characteristic point in the direction of the normal vector according to the gray value of the target characteristic point of the original image, obtains the gray value of the initial tracking point in the target image, tracking the position of the target feature point in the target image along the direction of the normal vector according to the gray value and the gray gradient value in the direction of the normal vector of the target feature point and the gray value of the initial tracking point, when the target characteristic points are tracked, the positions of the target characteristic points can be tracked in the target image along the normal vector direction, the robustness and the stability of tracking the target characteristic points are improved, the matching degree of tracking the target characteristic points in the target image is increased, and the scheme also shortens the time for tracking the target characteristic points, improves the tracking efficiency and is beneficial to providing more effective data support for subsequent image information processing such as machine vision work.

In one embodiment, the step of tracking the position of the target feature point in the target image along the normal vector direction according to the gray value of the target feature point, the gray gradient value and the gray value of the initial tracking point comprises:

determining the position deviation of the initial tracking point in the normal vector direction according to the gray value and the gray gradient value of the target feature point of the original image and the gray value of the initial tracking point; determining the position of a starting tracking point of the target image; and determining the position of the target characteristic point in the target image according to the position of the starting tracking point and the position deviation.

In one embodiment, the step of determining the position deviation of the initial tracking point in the normal vector direction according to the gray scale value, the gray scale gradient value and the gray scale value of the target feature point of the original image comprises:

the gray value of the target characteristic point is differenced with the gray value of the initial tracking point to obtain a gray deviation value of the target characteristic point and the initial tracking point; and acquiring the position deviation of the initial tracking point in the normal vector direction according to the gray scale deviation value and the gray scale gradient value of the target feature point in the normal vector direction.

In one embodiment, the step of obtaining the position deviation of the start tracking point in the normal vector direction according to the gray scale deviation value and the gray scale gradient value of the target feature point in the normal vector direction includes:

performing square operation on the gray scale gradient value of the target feature point in the normal vector direction to obtain a spatial gradient value of the target feature point in the normal vector direction; obtaining an image deviation value according to the product of the gray gradient value of the target feature point in the normal vector direction and the gray deviation value; calculating the ratio of the image deviation value to the spatial gradient value of the target feature point in the normal vector direction; and determining the position deviation of the initial tracking point in the normal vector direction according to the ratio and the normal vector of the target feature point.

In one embodiment, the step of determining the gray scale gradient value of the target feature point in the normal vector direction according to the gray scale value of the target feature point of the original image includes:

acquiring a gray value of a target feature point of the original image; calculating gray gradient values of the target feature points in the transverse direction and the longitudinal direction according to the gray values; constructing a gray gradient matrix of the target characteristic points according to the gray gradient values of the target characteristic points in the transverse direction and the longitudinal direction; determining the normal vector of the target feature point according to the gray gradient values of the target feature point in the transverse direction and the longitudinal direction; and obtaining the gray gradient value of the target characteristic point in the normal vector direction according to the product of the gray gradient matrix and the normal vector.

In one embodiment, the step of determining the normal vector of the target feature point according to the gray scale gradient values of the target feature point in the transverse and longitudinal directions comprises:

constructing a spatial gradient matrix of the target feature points according to the gray gradient values of the target feature points in the transverse and longitudinal directions; acquiring a feature vector of the spatial gradient matrix; and determining the normal vector of the target characteristic point according to the characteristic vector of the spatial gradient matrix.

In one embodiment, the method further comprises the steps of:

generating a rectangular pixel window with the target feature point as the center in the original image; acquiring the gray value of each pixel point in the rectangular pixel window; calculating gray gradient values of the target characteristic points in the transverse and longitudinal directions of the rectangular pixel window according to the gray values of the pixel points; calculating a spatial gradient matrix of the target feature point in the rectangular pixel window according to the gray gradient values of the target feature point in the transverse and longitudinal directions of the rectangular pixel window; obtaining a characteristic value of the spatial gradient matrix; determining the type of the target characteristic point according to the characteristic value; the types of the target feature points comprise corner points and edge points.

In one embodiment, a method for tracking picture feature points is further provided, which includes the steps of:

a. respectively establishing an image pyramid for the original image and the target image; wherein the image pyramid comprises a multi-layered image;

b. determining the position of the target feature point in the current layer of image of the image pyramid of the original image;

c. tracking the target feature point by using the tracking method of the picture feature point according to the position of the target feature point, so as to obtain a tracking point matched with the position of the target feature point in the image of the current layer of the image pyramid of the target image;

d. taking the tracking point as the initial tracking point of the next layer image of the current layer image of the target image;

e. and repeating the steps b to d until the tracking point is the tracking point of the bottom layer image of the image pyramid of the target image, and obtaining the position of the target feature point in the target image.

The method for tracking the picture feature points provided by the above embodiment establishes an image pyramid for the original image and the target image respectively, determining the position of a target feature point in the current-layer image of the image pyramid of the original image, tracking the target feature point by using the tracking method of the picture feature point according to the position of the target feature point to obtain a tracking point matched with the target feature point in the current-layer image of the image pyramid of the target image, repeating the steps b to d until the obtained tracking point is the tracking point of the bottom layer image of the image pyramid of the target image, taking the position of the tracking point as the position of the target characteristic point in the target image, according to the scheme, the characteristic points of the picture are tracked in a mode of combining the image pyramid with the tracking method of the characteristic points of the picture in any embodiment, so that the stability and robustness of tracking the characteristic points of the picture among different pictures are further improved.

In one embodiment, an apparatus for tracking feature points of a picture is provided, including:

the determining module is used for determining a gray gradient value of the target characteristic point in the normal vector direction according to the gray value of the target characteristic point of the original image; the normal vector is a vector indicating the direction in which the gray scale of the target feature point changes most rapidly;

the acquisition module is used for acquiring the gray value of the initial tracking point of the target image;

and the tracking module is used for tracking the position of the target characteristic point in the target image along the normal vector direction according to the gray value, the gray gradient value and the gray value of the initial tracking point of the target characteristic point.

The image feature point tracking device determines the gray gradient value of the target feature point in the direction of the normal vector through the determining module according to the gray value of the target feature point of the original image, acquires the gray value of the initial tracking point in the target image through the acquiring module, tracks the position of the target feature point in the target image along the direction of the normal vector through the tracking module according to the gray value of the target feature point, the gray gradient value in the direction of the normal vector and the gray value of the initial tracking point along the direction of the normal vector, so that when the target feature point is tracked, the position of the target feature point in the target image along the direction of the normal vector can be tracked, the robustness and the stability of tracking the target feature point are improved, the matching degree of tracking the target feature point in the target image is increased, and the time of tracking the target feature point is shortened through the scheme, the tracking efficiency is improved, and more effective data support is provided for subsequent image information processing such as machine vision work.

In one embodiment, the tracking module comprises:

the deviation determining module is used for determining the position deviation of the initial tracking point in the normal vector direction according to the gray value and the gray gradient value of the target feature point of the original image and the gray value of the initial tracking point; the position determining module is used for determining the position of the initial tracking point of the target image; and determining the position of the target characteristic point in the target image according to the position of the starting tracking point and the position deviation.

In one embodiment, the deviation determination module comprises:

the difference operation module is used for carrying out difference on the gray value of the target characteristic point and the gray value of the initial tracking point to obtain a gray deviation value of the target characteristic point and the initial tracking point; and the deviation obtaining module is used for obtaining the position deviation of the initial tracking point in the normal vector direction according to the gray level deviation value and the gray level gradient value of the target characteristic point in the normal vector direction.

In one embodiment, the deviation obtaining module includes:

the square operation module is used for carrying out square operation on the gray gradient value of the target characteristic point in the normal vector direction to obtain a spatial gradient value of the target characteristic point in the normal vector direction; the deviation value calculation module is used for obtaining an image deviation value according to the product of the gray gradient value of the target feature point in the normal vector direction and the gray deviation value; the product operation module is used for calculating the ratio of the image deviation value to the spatial gradient value of the target feature point in the normal vector direction; and determining the position deviation of the initial tracking point in the normal vector direction according to the ratio and the normal vector of the target feature point.

In one embodiment, the determining module comprises:

the gray level obtaining module is used for obtaining the gray level value of the target characteristic point of the original image; calculating gray gradient values of the target feature points in the transverse direction and the longitudinal direction according to the gray values; the first construction module is used for constructing a gray gradient matrix of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse direction and the longitudinal direction; the vector determination module is used for determining the normal vector of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse direction and the longitudinal direction; and the gradient determining module is used for obtaining the gray gradient value of the target characteristic point in the normal vector direction according to the product of the gray gradient matrix and the normal vector.

In one embodiment, the vector determination module comprises:

the second construction module is used for constructing a spatial gradient matrix of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse and longitudinal directions; the vector acquisition module is used for acquiring the characteristic vector of the spatial gradient matrix; and determining the normal vector of the target characteristic point according to the characteristic vector of the spatial gradient matrix.

In one embodiment, further comprising:

the window generation module is used for generating a rectangular pixel window taking the target feature point as the center in the original image; the gradient calculation module is used for acquiring the gray value of each pixel point in the rectangular pixel window; calculating gray gradient values of the target characteristic points in the transverse and longitudinal directions of the rectangular pixel window according to the gray values of the pixel points; the matrix calculation module is used for calculating a spatial gradient matrix of the target characteristic point in the rectangular pixel window according to the gray gradient values of the target characteristic point in the transverse and longitudinal directions of the rectangular pixel window; the type determining module is used for acquiring the eigenvalue of the spatial gradient matrix; determining the type of the target characteristic point according to the characteristic value; the types of the target feature points comprise corner points and edge points.

In one embodiment, there is also provided an apparatus for tracking picture feature points, including:

the image pyramid establishing module is used for executing the step a and respectively establishing an image pyramid for the original image and the target image; wherein the image pyramid comprises a multi-layered image;

b, determining the position of the target feature point in the current layer of image of the image pyramid of the original image;

a tracking point determining module, configured to perform step c, track the target feature point according to the position of the target feature point by using the tracking method for picture feature points according to any one of the above embodiments, and obtain a tracking point, in the current-layer image of the image pyramid of the target image, that is matched with the position of the target feature point;

a tracking point selecting module, configured to perform step d, using the tracking point as the initial tracking point of the next layer image of the current layer image of the target image;

and e, repeating the steps b to d until the tracking point is the tracking point of the bottom layer image of the image pyramid of the target image, and obtaining the position of the target feature point in the target image.

The tracking device for the picture feature points provided by the above embodiment tracks the feature points of the picture by combining the image pyramid with the tracking method for the picture feature points of any one of the above embodiments, thereby further improving the stability and robustness of tracking the picture feature points among different pictures.

In one embodiment, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the steps of the tracking method of picture feature points according to any one of the above embodiments when executing the computer program.

According to the computer equipment, the position of the target feature point in the target image is tracked along the normal vector direction through the computer program running on the processor, the robustness and the stability of tracking the target feature point are improved, the matching degree of tracking the target feature point in the target image is increased, the time for tracking the target feature point is shortened, the tracking efficiency is improved, and more effective data support is provided for subsequent image information processing such as machine vision work.

In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method for tracking picture feature points as described in any one of the above embodiments.

According to the computer-readable storage medium, the position of the target feature point in the target image is tracked along the normal vector direction through the stored computer program, the robustness and the stability of tracking the target feature point are improved, the target feature point tracking matching degree is favorably increased in the target image, the target feature point tracking time is shortened, the tracking efficiency is improved, and more effective data support is favorably provided for subsequent image information processing such as machine vision work.

Drawings

FIG. 1 is a schematic diagram illustrating an application scenario of a method for tracking feature points of a picture according to an embodiment;

FIG. 2 is a flowchart illustrating a method for tracking feature points of a picture according to an embodiment;

FIG. 3 is a diagram illustrating a change in gray level gradient of feature points of an image according to an embodiment;

FIG. 4 is a diagram illustrating the tracking direction of feature points in an embodiment;

FIG. 5 is a schematic diagram illustrating the gray scale of feature points of an embodiment;

FIG. 6 is a diagram illustrating feature values of feature points of an embodiment of a picture;

FIG. 7 is a diagram illustrating types of feature points of a picture according to an embodiment;

FIG. 8 is a flowchart illustrating a method for tracking feature points of a picture according to another embodiment;

FIG. 9 is a schematic diagram of an image pyramid in one embodiment;

FIG. 10 is a diagram illustrating the results of a method for tracking feature points of an image according to an embodiment;

FIG. 11 is a block diagram showing the structure of a tracking means for picture feature points in one embodiment;

FIG. 12 is a block diagram showing the structure of a tracking means for picture feature points in another embodiment;

FIG. 13 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The method for tracking the picture feature points provided by the invention can be applied to an application scene as shown in fig. 1, and fig. 1 is a schematic view of an application scene of the method for tracking the picture feature points in one embodiment. Fig. 1(a) is a schematic diagram of an original image, where the original image 100a may include a shadow area 200 and a pixel 300 on the shadow area 200, and the pixel 300 may be used as a target feature point in the original image, and the position of the pixel 300 is tracked in a target image 100b shown in fig. 1(b) by using characteristics, such as a gray value, of the pixel 300, where the target image 100b may be an image of a next frame of the original image 100a in a video image.

In an embodiment, a method for tracking a picture feature point is provided, referring to fig. 2, fig. 2 is a schematic flowchart of the method for tracking a picture feature point in an embodiment, where the method for tracking a picture feature point may include the following steps:

step S101, determining the gray gradient value of the target characteristic point in the normal vector direction according to the gray value of the target characteristic point of the original image.

In this step, the target feature point is a key pixel point for identifying the feature of the original image, and according to the gray value of the target feature point of the original image, a gray level change value, that is, a gray level gradient value, of the target feature point in each direction on the plane where the original image is located can be obtained. Determining the gray gradient value of the target characteristic point in the normal vector direction according to the gray value of the target characteristic point; the normal vector is a vector used for indicating the direction in which the gray value of the target feature point changes the fastest.

Taking fig. 3 as an example for illustration, fig. 3 is a schematic diagram of gray level gradient change of a picture feature point in an embodiment, assuming that the picture feature point 220 is located at an edge position of a gray area 210, then according to a gray level value of the picture feature point 220, it can be clarified that a gray level change condition of the picture feature point 220 in directions shown by an arrow 210a and an arrow 210b in fig. 3 is present, for example, along the arrow 210a direction, a gray level value change of the picture feature point 220 in a rectangular window is not large, but a gray level change of the picture feature point 220 in the arrow 210b direction in the rectangular window is large, so a gray level change value of the picture feature point 220 in each direction in the original image can be calculated according to the gray level value of the picture feature point 220, if the gray level change value of the picture feature point 220 in the arrow 210b direction is maximum, the arrow 210b direction can be taken as the normal vector direction of the picture feature point 220, it should be noted that the normal vector direction is generally determined by performing correlation algebraic operation using the gray-level value of the picture feature point 220.

Step S102, obtaining the gray value of the initial tracking point of the target image.

The method mainly comprises the step of obtaining a gray value of an initial tracking point in a target image, wherein the initial tracking point is a pixel point used for tracking a target characteristic point of an original image in the target image, and the target image can be the next frame image of any frame image in a video image sequence. The position of the target feature point in the original image is generally determined, and the position of the initial tracking point in the target image in this step may be selected according to the position of the target feature point in the original image, for example, the position of the target feature point in the original image is used as the position of the initial tracking point in the target image. In order to improve the tracking efficiency of the target feature point, the gray value of the pixel point matched with the initial tracking point can be extracted from the pixel feature information base of each pixel point of the pre-stored target image.

And step S103, tracking the position of the target feature point in the target image along the normal vector direction according to the gray value, the gray gradient value and the gray value of the initial tracking point of the target feature point.

Tracking the position of the target feature point in the original image along the normal vector direction in the target image according to the gray value of the target feature point of the original image, the gray gradient value of the target feature point in the normal vector direction and the gray value of the initial tracking point. In the target image, the position of the target feature point may be tracked along the normal vector direction from a preset initial tracking point position.

To illustrate this step by taking fig. 4 as an example, fig. 4 is a schematic diagram illustrating tracking directions of picture feature points in an embodiment, for example, the original image 310 shown in fig. 4(a) includes 3 shadow regions 311a, 311b and 311c, each of which has a plurality of picture feature points illustrated by solid circles, arrows on each solid circle respectively correspond to normal vector directions of each picture feature point, the target image 320 of fig. 4(b) includes shadow regions 321a, 321b and 321c corresponding to the 3 shadow regions 311a, 311b and 311c of the original image 310, which may respectively correspond to shadow regions after a positioning shift occurs to the shadow regions 311a, 311b and 311c, and the solid circles in the target image 320 illustrate tracking points in the target image 320 matching with each feature point of the original image 310, wherein each feature point of the original image 310 is marked on the target image 320 by a dotted circle, the arrows between the dotted line circles and the solid line circles are used to indicate that, when the present step tracks each feature point of the original image 310, the positions of the target feature points of the original image are tracked in the target image along the normal vector direction according to the gray scale value of the target feature point, the gray scale gradient value in the normal vector direction, and the gray scale value of the initial tracking point, so as to obtain the positions of the target feature points in the target image, which are matched with each other.

The picture characteristic point tracking method determines the gray gradient value of the target characteristic point in the direction of the normal vector according to the gray value of the target characteristic point of the original image, acquires the gray value of the initial tracking point in the target image, tracks the position of the target characteristic point in the target image along the direction of the normal vector according to the gray value of the target characteristic point, the gray gradient value in the direction of the normal vector and the gray value of the initial tracking point, and tracks the position of the target characteristic point in the target image along the direction of the normal vector when tracking the target characteristic point, so that the position of the target characteristic point in the target image can be tracked, the robustness and the stability of tracking the target characteristic point are improved, the matching degree of tracking the target characteristic point in the target image is increased, the scheme is simpler than the existing tracking method, the tracking time of the target characteristic point is shortened, the tracking efficiency is improved, and more effective data support is provided for subsequent image information processing such as machine vision work.

In one embodiment, the step of determining the gray gradient value of the target feature point in the normal vector direction according to the gray value of the target feature point of the original image in step S101 may include:

step S201, obtaining a gray value of a target feature point of an original image; and calculating the gray gradient values of the target characteristic points in the transverse direction and the longitudinal direction according to the gray values.

In this step, the horizontal direction and the vertical direction may be directions of two coordinate axes perpendicular to each other in the original image, for example, an X-Y rectangular coordinate system is established in the original image, and the horizontal direction and the vertical direction correspond to the directions of the X axis and the Y axis, respectively. For the gray value I (x, y) of each pixel point in the original image, the gray value I (x, y) of the pixel point in the original image can be calculatedGray scale gradient value I in X-axis direction_x(x, Y), and a gray gradient value I in the direction of the Y-axis_y(X, Y), the original image can be regarded as a two-dimensional discrete function, the gray gradient value of each pixel point of the image can correspond to the derivative of the two-dimensional discrete function, the way of calculating the derivative is various, and for simple calculation, the gray gradient values of the target feature point in the X-axis and Y-axis directions can be calculated by adopting the following formula:

I_x(x,y)＝[I(x+1,y)-I(x-1,y)]/2

I_y(x,y)＝[I(x,y+1)-I(x,y-1)]/2

wherein, I (X +1, Y) represents the gray value of the next pixel point of the target characteristic point in the X-axis direction, I (X-1, Y) represents the gray value of the previous pixel point of the target characteristic point in the X-axis direction, I (X, Y +1) represents the gray value of the next pixel point of the target characteristic point in the Y-axis direction, I (X, Y-1) represents the gray value of the previous pixel point of the target characteristic point in the Y_x(X, y) represents a gray scale gradient value of the target feature point in the X-axis direction, I_y(x, Y) represents a gradation gradient value of the target feature point in the Y-axis direction.

Step S202, a gray gradient matrix of the target characteristic point is constructed according to the gray gradient values of the target characteristic point in the transverse direction and the longitudinal direction.

In this step, the gray gradient matrix is a gray gradient matrix including information of gray gradient values of the target feature point in the lateral and longitudinal directions, and the following matrix may be employed as the gray gradient matrix of the target feature point:

t(x,y)＝[I_x(x,y) I_y(x,y)]

wherein t (X, y) represents a gray gradient matrix of the target feature point, the matrix including a gray gradient value I of the target feature point in the X-axis direction_xGradation gradient value I in the (x, Y) and Y-axis directions_y(x,y)。

Step S203, determining the normal vector of the target feature point according to the gray gradient values of the target feature point in the transverse direction and the longitudinal direction.

The step is mainly to calculate the normal vector of the target feature point of the original image for indicating the direction in which the gray scale gradient value changes fastest according to the gray scale gradient values of the target feature point in the transverse direction and the longitudinal direction.

And step S204, obtaining the gray gradient value of the target characteristic point in the normal vector direction according to the product of the gray gradient matrix and the normal vector.

The gray gradient matrix t (x, y) may be t (x, y) ═ I_x(x,y) I_y(x,y)]And if the normal vector is n, the gray gradient value of the target feature point in the normal vector direction can be expressed by the following formula: s (x, y) ═ I_x(x,y)I_y(x,y)]n, where s (x, y) represents a gray scale gradient value of the target feature point in a normal vector direction.

According to the technical scheme of the embodiment, the gray gradient value of the target feature point in the normal vector direction is obtained according to the gray value change value of the target feature point of the original image in the horizontal and vertical directions and the normal vector of the target feature point, the step of calculating the gray gradient value of the target feature point in the normal vector direction is simplified, the gray change value of the target feature point in the normal vector direction is accurately reflected, and the accuracy and the efficiency of tracking the target feature point are improved.

In one embodiment, the step of determining the normal vector of the target feature point according to the gray scale gradient values of the target feature point in the transverse and longitudinal directions in step S203 may further include:

constructing a spatial gradient matrix of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse direction and the longitudinal direction; acquiring a feature vector of a spatial gradient matrix; and determining the normal vector of the target characteristic point according to the characteristic vector of the spatial gradient matrix.

The method mainly comprises the steps of establishing a spatial gradient matrix through gray gradient values of target characteristic points in the transverse direction and the longitudinal direction, and determining normal vectors of the target characteristic points by using characteristic vectors of the spatial gradient matrix. The spatial gradient matrix is a matrix in which each matrix element is associated with a gray gradient value of the target feature point in the transverse and longitudinal directions, and the following matrix may be used as the gray gradient matrix of the target feature point:

m (x, y) represents a gray gradient matrix of the target characteristic point in a matrix pixel window W, u and v represent coordinates of each pixel point in the matrix pixel window, if the length of the matrix pixel window W is W and the height of the matrix pixel window W is h, the value range of u is from-W/2 to W/2, the value range of v is from-h/2 to h/2, and an intermediate variable I_x(x,y)²＝I_x·I_x，I_y(x,y)²＝I_y·I_y，I_xy(x,y)＝I_x·I_yAnd g (u, v) represents a gaussian weighting function, and the gaussian weighting function is adopted in the gray gradient matrix to improve the capability of resisting image noise, and the expression of the gaussian weighting function is as follows:

the eigenvalue λ of the gray gradient matrix M (x, y) of the target characteristic point can be calculated₁And λ₂And obtaining eigenvectors V corresponding to the two eigenvalues respectively₁And V₂. The gradient variation of the characteristic points can be seen through the size of the characteristic values, and the characteristic vector V of the target characteristic point can be obtained₂(x, y) is a normal vector n (x, y) of the target feature point.

According to the technical scheme of the embodiment, the normal vector of the target feature point is obtained according to the feature value of the gray gradient matrix of the target feature point, the feature vector corresponding to the feature value capable of accurately reflecting the gradient variation of the target feature point is used as the normal vector of the target feature point, the normal vector of the target feature point is accurately obtained, and the accuracy of tracking the target feature point is further improved.

In one embodiment, the step of tracking the position of the target feature point in the target image along the normal vector direction according to the gray scale value of the target feature point, the gray scale gradient value and the gray scale value of the start tracking point in step S103 may include:

s301, determining the position deviation of the initial tracking point in the normal vector direction according to the gray value and the gray gradient value of the target feature point of the original image and the gray value of the initial tracking point;

the method mainly comprises the step of determining the position deviation amount of an initial tracking point in a normal vector direction according to the gray value of a target feature point, the gray gradient value in the normal vector direction and the gray value of the initial tracking point, wherein the position deviation amount can be used for reflecting the position deviation condition of the initial tracking point and a matching point of a target image, and the matching point is a tracking point which is matched with the position of the target feature point of an original image in the target image.

S302, determining the position of the initial tracking point of the target image; and determining the position of the target characteristic point in the target image according to the position of the initial tracking point and the position deviation.

In this step, the position of the start tracking point of the target image may be determined according to the position of the target feature point of the original image, for example, the position of the target feature point in the original image is set as the position of the start tracking point of the target image. And determining the position of the target characteristic point in the target image according to the position of the initial tracking point and the position deviation amount of the initial tracking point in the normal vector direction.

According to the technical scheme of the embodiment, the position of the target feature point in the target image is determined through the position deviation of the initial tracking point in the normal vector direction and the position information of the initial tracking point, the position change of the target feature point in the target image can be accurately reflected, the calculation mode of the position of the target feature point in the target image is flexible and changeable, the position deviation amount can be calculated for multiple times through multiple iterations, and the position deviation amount with higher precision is obtained, so that the position of the target feature point in the target image is accurately calculated.

In one embodiment, the step of determining the position deviation of the initial tracking point in the normal vector direction according to the gray scale value, the gray scale gradient value and the gray scale value of the target feature point of the original image in step S301 may further include:

In this embodiment, the gray value of the target feature point of the original image is subtracted from the gray value of the initial tracking point to obtain a gray deviation value between the target feature point and the initial tracking point, where the gray deviation value may be used to reflect a gray difference between the target feature point and the initial tracking point, and the gray deviation value and the gray gradient value of the target feature point in the normal vector direction are used to calculate a position deviation of the initial tracking point in the target image in the normal vector direction. The technical scheme of the embodiment calculates the position deviation of the initial tracking point in the target image by combining the gray difference condition of the target characteristic point and the initial tracking point and the gray gradient value of the target characteristic point in the normal vector direction, so that the position deviation can accurately reflect the gray difference condition of the target characteristic point and the initial tracking point, and the gray gradient value of the target characteristic point in the normal vector direction is combined to be beneficial to accurately and quickly determining the position matched with the target tracking point from the initial tracking point in the target image.

In one embodiment, the step of obtaining the position deviation of the start tracking point in the normal vector direction according to the gray deviation value and the gray gradient value of the target feature point in the normal vector direction in step S301 may include:

In this embodiment, the gray gradient value of the target feature point in the normal vector direction may be calculated in the rectangular pixel window W, where the target feature point is generally located at the center of the rectangular pixel window W, and the gray gradient value of the target feature point in the normal vector direction may be expressed as:

S(u,v)＝[I_x(p_x+u,p_y+v)I_y(p_x+u,p_y+v)]n

wherein p is_xAbscissa representing target feature point, p_yRepresenting the ordinate of the target feature point, u and v represent the position coordinates of each pixel point in the matrix pixel window relative to the target feature point, n represents a normal vector, if the length of the matrix pixel window W is W and the height is h, the value range of u is from-W/2 to W/2, the value range of v is from-h/2 to h/2, S (u, v) represents the gray gradient value of the target feature point in the normal vector direction, the spatial gradient value of the target feature point in the normal vector direction can be obtained according to the square operation of the gray gradient value, and the spatial gradient value can be calculated by adopting the following formula:

where m represents a spatial gradient value of the target feature point in the normal vector direction.

The image deviation value of the original image and the target image can be calculated by the following formula:

where b denotes the image bias value, S (u, v) denotes a gray scale gradient value of the target feature point in the normal vector direction, [ I (p) ]_x+u,p_y+v)-J(q_x,q_y)]As the gray scale deviation value, I (p)_x+u,p_y+ v) represents the point (p) in the original image I_x+u,p_y+ v) gray value, J (q)_x,q_y) Representing the point (q) in the target image J_x,q_y) The gray value of (a). The image deviation value is mainly used for representing image deviation information of a characteristic point of an original image and an initial tracking point of a target image and comprises gray level deviation informationAnd (4) information.

The ratio of the image deviation value to the spatial gradient value of the target feature point in the normal vector direction may be

Product with normal vector n of target feature point

As a positional deviation of the start tracking point in the normal vector direction.

In the embodiment, the spatial gradient value of the target feature point in the normal vector direction and the image deviation value of the target feature point and the initial tracking point are calculated, the position deviation of the initial tracking point in the normal vector direction is determined according to the ratio of the image deviation value to the spatial gradient value, the calculation mode of the position deviation is further refined, the image deviation condition of the target feature point and the initial tracking point is comprehensively considered, the position deviation of the initial tracking point in the normal vector direction, which is obtained when the image deviation value is zero, is also zero, and the position of the target feature point in the target image can be accurately reflected.

In one embodiment, the method further comprises the following steps:

in step S310, a rectangular pixel window centered on the target feature point is generated in the original image.

In the step, the target feature point is taken as a geometric center to generate a rectangular pixel window W, the length of the rectangular pixel window W is W, the height of the rectangular pixel window W is h, the value range of the relative position coordinates of each pixel point in the rectangular pixel window W is, for example, the value range of the abscissa u can be from-W/2 to W/2, and the value range of the ordinate v can be from-h/2 to h/2.

Step S320, acquiring the gray value of each pixel point in the rectangular pixel window; and calculating the gray gradient values of the target characteristic points in the transverse and longitudinal directions of the rectangular pixel window according to the gray values of the pixel points.

In this step, the gray scale gradient value of the target feature point in the transverse direction of the rectangular pixel window can be expressed as:I_x(x, y), the gray gradient values in the longitudinal direction can be expressed as: i is_y(x, y). For the gray value I (X, Y) of each pixel point in the rectangular pixel window, for simple calculation, the gray gradient values of the target feature point in the X-axis and Y-axis directions of the rectangular pixel window may be calculated by the following formula:

I_x(x,y)＝[I(x+1,y)-I(x-1,y)]/2

I_y(x,y)＝[I(x,y+1)-I(x,y-1)]/2

wherein, I (X +1, Y) represents the gray value of the next pixel point of the target feature point in the X-axis direction, I (X-1, Y) represents the gray value of the previous pixel point of the target feature point in the X-axis direction, I (X, Y +1) represents the gray value of the next pixel point of the target feature point in the Y-axis direction, and I (X, Y-1) represents the gray value of the previous pixel point of the target feature point in the Y-axis direction.

And step S330, calculating a spatial gradient matrix of the target characteristic point in the rectangular pixel window according to the gray gradient values of the target characteristic point in the transverse and longitudinal directions of the rectangular pixel window.

This step may adopt the following matrix as the gray gradient matrix of the target feature point:

wherein M (x, y) represents the gray gradient matrix of the target feature point in the matrix pixel window W, and the intermediate variable I_x(x,y)²＝I_x·I_x，I_y(x,y)²＝I_y·I_yAnd I_xy(x,y)＝I_x·I_yAnd g (u, v) represents a gaussian weighting function, and the gaussian weighting function is adopted in the gray gradient matrix to improve the capability of resisting image noise, and the expression of the gaussian weighting function is as follows:

referring to fig. 5, fig. 5 is a schematic diagram of the gray scale of the feature points of the picture in an embodiment, wherein fig. 5(a) corresponds to the original image, fig. 5(b) andFIG. 5(c) shows the gray scale gradient values I in the X-axis direction of the pixels of the original image_xAnd a gray gradient value I in the Y-axis direction_yFIG. 5(d), FIG. 5(e) and FIG. 5(f) correspond to the intermediate variable I, respectively_x·I_x、I_y·I_yAnd I_x·I_yFIG. 5(g), FIG. 5(h) and FIG. 5(I) correspond to the intermediate variable I_x·I_x、I_y·I_yAnd I_x·I_yAnd (5) carrying out Gaussian weighting function processing on the gray-scale image.

Step S340, obtaining the eigenvalue of the space gradient matrix; and determining the type of the target characteristic point according to the characteristic value.

The eigenvalue λ of the gray gradient matrix M (x, y) of the target characteristic point constructed in step S330 can be calculated₁And λ₂The type of the target feature point can be determined by the two feature values, including the corner point and the edge point.

Referring to fig. 6, fig. 6 is a diagram illustrating feature values of feature points of a picture according to an embodiment, where an arrow denoted by 330a in fig. 6 indicates λ₁Increasing direction, arrow 330b indicates λ₂Increasing direction, and 330c denotes a planar area, λ₁And λ₂Are small, 330d and 330f both indicate the area where the edge point is located, λ is in the area indicated by 330d₁Much less than λ₂λ in the region indicated by 330f₂Much less than λ₁And the region indicated by 330e is the corner region, λ₁And λ₂Are all larger, and λ₁≈λ₂. The type of the target feature point can be determined from the feature value. As shown in fig. 7, fig. 7 is a schematic diagram illustrating types of feature points of a picture according to an embodiment, and each of the original images 340 includes a shaded area 350 having a first feature point 351, a second feature point 352, and a third feature point 353. The gray level change of the first feature point 351 in the direction indicated by each arrow is not obvious, that is, the first feature point corresponds to the planar region 330c in fig. 6, the gray level change of the second feature point 352 in the direction indicated by the arrow 352a is not obvious, and the gray level change in the direction indicated by the arrow 352b is obvious, which may be similar to that in fig. 6The corner regions of 330d and 330f correspond to each other, and the change in the gray level of the third feature point 353 in the direction indicated by each arrow is significant, and may correspond to the corner region of 330e in fig. 6.

In a conventional manner, a corner point of a picture is generally selected as a key point, because a technician generally considers that a corner point has good characteristics: it does not change with large deformations of the image and is also insensitive to small deformations of the image. However, in actual use, stable corner points that can be extracted from an image are very limited, and if too few corner points are used as key points, fewer corner points can be successfully tracked, which greatly affects subsequent work. In order to increase the number of key points in the picture, edge points in the picture, such as straight edge points, can also be used as the key points, and the scheme has the advantages of extractable number so as to increase the number of key points of the original image, and can improve the stability and robustness of tracking.

Based on this, for convenience of description, it may be assumed that the eigenvalue λ of M (x, y) of each pixel point of the original image₁Are all less than lambda₂For satisfying λ₂(x, y) is greater than α × Max (λ)₂) And λ₁(x, y) is less than β × Max (λ)₁) The conditional point (x, y) may be an edge point in the original image. Wherein, alpha and beta are preset threshold values, generally, 0 < alpha, beta < 1, Max (lambda)₁) Represents the maximum lambda in the original image₁Value, Max (λ)₂) Represents the maximum lambda in the original image₂The value is obtained. Optionally, a minimum distance d between edge points may be set, the obtained edge points are screened to obtain an edge point set, and a feature vector V corresponding to each edge point (x, y)₂(x, y) is the normal vector n (x, y) of the edge point as V₂(x,y)。

In an embodiment, a method for tracking a picture feature point is further provided, as shown in fig. 8, fig. 8 is a schematic flowchart of a method for tracking a picture feature point in another embodiment, where the method for tracking a picture feature point may include the following steps:

step S401, a, respectively establishing an image pyramid for the original image and the target image.

The method comprises the steps of establishing an image pyramid (IL) for an original image I and a target image J respectively_L＝0...LmAnd { JL }_L＝0...LmThe image pyramid may comprise a plurality of layers, L_mRepresenting a given pyramid level, typically 3, as shown in fig. 9, fig. 9 is a schematic diagram of an image pyramid in one embodiment, and image pyramiding typically includes two steps: firstly, low-pass filtering is carried out on an original image for smoothing, and then 1/2 sampling processing is carried out on pixel points of the original image in the horizontal and vertical directions, so that a series of images with reduced scales are obtained. When L is 0, the size and resolution of the original image decrease when the original image moves to the upper layer of the pyramid, and the amount of detail decreases accordingly. The target characteristic points can be tracked from the top layer, a rough result is obtained firstly, then the rough result is used as the initial point of the next layer to be tracked, and iteration is continuously carried out until the 0 th layer is reached, so that the rough-to-fine analysis strategy is adopted.

And S402, b, determining the position of the target feature point in the image of the current layer of the image pyramid of the original image.

The layer image refers to the current image layer for tracking the target feature point, and if the layer 3 is tracking the target feature point, the layer image is the layer 3 image. The step is mainly to confirm the position of the target characteristic point in the image of the layer. Target feature points such as edge points u can be set in the L-th pyramid layer image I of the original image I^LIn the position of

Wherein p is_xAnd p_yThe coordinates of the edge point u can be represented.

Step S403, c, tracking the target feature point by using the tracking method of the picture feature point according to the position of the target feature point, so as to obtain a tracking point matched with the position of the target feature point in the image of the current layer of the image pyramid of the target image;

in this step, the gray of the target feature point in the normal vector direction can be calculated in the matrix pixel window WDegree gradient value: s (u, v) ═ I_x(p_x+u,p_y+v)I_y(p_x+u,p_y+v)]n；

U and v represent position coordinates of each pixel point in a matrix pixel window W relative to a target feature point, if the length of the matrix pixel window W is W and the height of the matrix pixel window W is h, the value range of u is from-W/2 to W/2, the value range of v is from-h/2 to h/2, and I is_x(X, y) represents the gray scale value of the image IL in the X-axis direction at the (X, y) position, I_y(x, Y) represents a gray scale gradient value in the Y-axis direction, n represents a normal vector, S (u, v) represents a gray scale gradient value of the target feature point in the normal vector direction, and a spatial gradient value of the target feature point in the L-th layer is calculated from the gray scale gradient values

This step can initialize the position iteration parameter gamma⁰＝[0 0]^TFor in the graph J^LPerforming iterative processing.

Assume the starting tracking point is in target image J^LIs at a position of

Wherein the content of the first and second substances,

and

for searching for preset offset position

Wherein

Can be preset to [ 00 ]]^TFrom 1 to K, according to a preset variable K, the following iterative approach is adopted:

calculating an image deviation value:

updating location iteration parameters

And continuing to perform iterative operation on the iterative parameter in the image of the current layer when k is equal to k +1, and obtaining the final tracking offset in the pyramid of the L layer as follows: d^L＝γ^k。

Step S404, d, using the tracking point as the initial tracking point of the next layer image of the current layer image of the target image.

This step initializes the tracking offset position g of the pyramid image of the next layer of the present layer image^L-1＝2(g^L+d^L)。

And S405, e, repeating the steps b to d until the tracking point is the tracking point of the bottom layer image of the image pyramid of the target image, and obtaining the position of the target feature point in the target image.

In this step, the above steps b to d may be repeated with L ═ L-1 until the obtained tracking point is the tracking point of the bottom layer image of the image pyramid of the target image, and then the position of the matching point v of the target feature point, such as the edge point u, in the image J is v ═ u + g⁰+d⁰。

The target feature point is tracked by the technical scheme provided by the above embodiment, and the tracking result can refer to fig. 10, fig. 10 is a schematic diagram of a tracking method result of the picture feature point in an embodiment, fig. 10(a) is a schematic diagram of a result obtained by a conventional feature point tracking method, 340a and 340b respectively represent an original image and a target image, fig. 10(b) is a schematic diagram of a result obtained by a picture feature point tracking method provided by the embodiment of the present invention, and 340c and 340d respectively represent an original image and a target image, it can be clearly seen that the technical scheme provided by the embodiment of the present invention successfully matches more than one time of feature points compared with the conventional method, and is simpler than the existing method, has stronger robustness, can greatly increase the number of successfully matched key points, provides better data support for subsequent machine vision work, and adopts a mode of combining an image pyramid and the tracking method of the picture feature point of any one embodiment to track the picture The method further improves the stability and robustness of tracking the image characteristic points among different images.

In an embodiment, an apparatus for tracking a picture feature point is provided, as shown in fig. 11, where fig. 11 is a block diagram of a structure of the apparatus for tracking a picture feature point in an embodiment, the apparatus for tracking a picture feature point may include: a determination module 101, an acquisition module 102, and a tracking module 103, wherein:

the determining module 101 is configured to determine a gray scale gradient value of a target feature point in a normal vector direction according to a gray scale value of the target feature point of the original image; the normal vector is a vector indicating the direction in which the gray scale of the target feature point changes most rapidly;

an obtaining module 102, configured to obtain a gray value of an initial tracking point of a target image;

and the tracking module 103 is configured to track the position of the target feature point in the target image along the normal vector direction according to the gray value, the gray gradient value, and the gray value of the initial tracking point of the target feature point.

The image feature point tracking device determines the gray gradient value of the target feature point in the direction of the normal vector according to the gray value of the target feature point of the original image through the determining module 101, acquires the gray value of the initial tracking point in the target image through the acquiring module 102, tracks the position of the target feature point in the target image along the direction of the normal vector through the tracking module 103 according to the gray value of the target feature point and the gray gradient value in the direction of the normal vector and the gray value of the initial tracking point along the direction of the normal vector, so that when the target feature point is tracked, the position of the target feature point in the target image along the direction of the normal vector can be tracked, the robustness and the stability of tracking the target feature point are improved, the matching degree of tracking the target feature point in the target image is increased, and the time for tracking the target feature point is shortened, the tracking efficiency is improved, and more effective data support is provided for subsequent image information processing such as machine vision work.

In one embodiment, the tracking module 101 includes:

In one embodiment, the deviation determination module includes:

In one embodiment, the deviation acquisition module includes:

In one embodiment, the determining module 101 includes:

In one embodiment, the vector determination module comprises:

In one embodiment, further comprising:

For specific limitations of the tracking device for the picture feature points, reference may be made to the above limitations on the tracking method for the picture feature points, which are not described herein again. The modules in the tracking device of the picture feature points can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In an embodiment, there is further provided a tracking apparatus for picture feature points, with reference to fig. 12, where fig. 12 is a block diagram of a tracking apparatus for picture feature points in another embodiment, the tracking apparatus for picture feature points may include: an image pyramid establishing module 401, a feature point position determining module 402, a tracking point determining module 403, a tracking point selecting module 404 and a target position determining module 405; wherein the content of the first and second substances,

an image pyramid establishing module 401, configured to perform step a. respectively establish an image pyramid for the original image and the target image; wherein the image pyramid comprises a multi-layered image;

a feature point position determining module 402, configured to execute step b, determine a position of the target feature point in the current-layer image of the image pyramid of the original image;

a tracking point determining module 403, configured to perform step c, track the target feature point according to the position of the target feature point by using the tracking method for picture feature points according to any one of the above embodiments, and obtain a tracking point in the current-layer image of the image pyramid of the target image, where the tracking point is matched with the position of the target feature point;

a tracking point selecting module 404, configured to execute step d, use the tracking point as the initial tracking point of the next layer image of the current layer image of the target image;

a target position determining module 405, configured to execute step e, repeat steps b to d until the tracking point is a tracking point of a bottom layer image of an image pyramid of the target image, and obtain a position of the target feature point in the target image.

In one embodiment, a computer device is provided, the computer device may be a server, the internal structure of which may be as shown in fig. 13, fig. 13 is an internal structure of the computer device in one embodiment. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing data required by the tracking method of the image edge points. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of tracking feature points of a picture.

Those skilled in the art will appreciate that the architecture shown in fig. 13 is merely a block diagram of some of the structures associated with the inventive arrangements and is not intended to limit the computing devices to which the inventive arrangements may be applied, as a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:

determining a gray gradient value of a target characteristic point of an original image in a normal vector direction according to the gray value of the target characteristic point; acquiring a gray value of an initial tracking point of a target image; and tracking the position of the target feature point in the target image along the normal vector direction according to the gray value and the gray gradient value of the target feature point and the gray value of the initial tracking point.

In one embodiment, the processor, when executing the computer program, further performs the steps of:

In one embodiment, there is also provided a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:

a. respectively establishing an image pyramid for the original image and the target image; b. determining the position of the target feature point in the current layer of image of the image pyramid of the original image; c. tracking the target feature point by using the tracking method of the picture feature point according to the position of the target feature point, so as to obtain a tracking point matched with the position of the target feature point in the image of the current layer of the image pyramid of the target image; d. taking the tracking point as the initial tracking point of the next layer image of the current layer image of the target image; e. and repeating the steps b to d until the tracking point is the tracking point of the bottom layer image of the image pyramid of the target image, and obtaining the position of the target feature point in the target image.

The computer device according to any one of the embodiments above, through the computer program running on the processor, realizes tracking of the position of the target feature point in the target image along the normal vector direction, improves robustness and stability of tracking of the target feature point, is beneficial to increase of matching degree of tracking of the target feature point in the target image, also shortens time of tracking of the target feature point, improves tracking efficiency, and is beneficial to providing more effective data support for subsequent image information processing such as machine vision work.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, there is also provided a computer readable storage medium having a computer program stored thereon, the computer program when executed by a processor implementing the steps of:

The computer-readable storage medium in any of the above embodiments, through the computer program stored therein, realizes tracking of the position of the target feature point in the target image along the normal vector direction, improves robustness and stability of tracking of the target feature point, is beneficial to increasing matching degree of tracking of the target feature point in the target image, also shortens time of tracking of the target feature point, improves tracking efficiency, and is beneficial to providing more effective data support for subsequent image information processing, such as machine vision work.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method for tracking picture feature points is characterized by comprising the following steps:

acquiring a gray value of an initial tracking point of a target image;

2. The method for tracking the image feature point according to claim 1, wherein the step of tracking the position of the target feature point in the target image along the normal vector direction according to the gray scale value of the target feature point, the gray scale gradient value and the gray scale value of the initial tracking point comprises:

determining the position deviation of the initial tracking point in the normal vector direction according to the gray value and the gray gradient value of the target feature point of the original image and the gray value of the initial tracking point;

determining the position of a starting tracking point of the target image; and determining the position of the target characteristic point in the target image according to the position of the starting tracking point and the position deviation.

3. The method according to claim 2, wherein the step of determining the position deviation of the initial tracking point in the normal vector direction according to the gray scale value, the gray scale gradient value and the gray scale value of the target feature point of the original image comprises:

the gray value of the target characteristic point is differenced with the gray value of the initial tracking point to obtain a gray deviation value of the target characteristic point and the initial tracking point;

and acquiring the position deviation of the initial tracking point in the normal vector direction according to the gray scale deviation value and the gray scale gradient value of the target feature point in the normal vector direction.

4. The method according to claim 3, wherein the step of obtaining the position deviation of the initial tracking point in the normal vector direction according to the gray scale deviation value and the gray scale gradient value of the target feature point in the normal vector direction comprises:

performing square operation on the gray scale gradient value of the target feature point in the normal vector direction to obtain a spatial gradient value of the target feature point in the normal vector direction;

obtaining an image deviation value according to the product of the gray scale gradient value of the target feature point in the normal vector direction and the gray scale deviation value;

calculating the ratio of the image deviation value to the spatial gradient value of the target feature point in the normal vector direction; and determining the position deviation of the initial tracking point in the normal vector direction according to the ratio and the normal vector of the target feature point.

5. The method for tracking image feature points according to claim 1, wherein the step of determining the gray gradient values of the target feature points in the normal vector direction according to the gray values of the target feature points of the original image comprises:

acquiring a gray value of a target feature point of the original image; calculating gray gradient values of the target feature points in the transverse direction and the longitudinal direction according to the gray values;

constructing a gray gradient matrix of the target characteristic points according to the gray gradient values of the target characteristic points in the transverse direction and the longitudinal direction;

determining the normal vector of the target feature point according to the gray gradient values of the target feature point in the transverse direction and the longitudinal direction;

and obtaining the gray gradient value of the target characteristic point in the normal vector direction according to the product of the gray gradient matrix and the normal vector.

6. The method for tracking image feature points according to claim 5, wherein the step of determining the normal vector of the target feature point according to the gray scale gradient values of the target feature point in the transverse and longitudinal directions comprises:

constructing a spatial gradient matrix of the target feature points according to the gray gradient values of the target feature points in the transverse and longitudinal directions;

acquiring a feature vector of the spatial gradient matrix; and determining the normal vector of the target characteristic point according to the characteristic vector of the spatial gradient matrix.

7. The method for tracking picture feature points according to any one of claims 1 to 6, further comprising the steps of:

generating a rectangular pixel window with the target feature point as the center in the original image;

acquiring the gray value of each pixel point in the rectangular pixel window; calculating gray gradient values of the target characteristic points in the transverse and longitudinal directions of the rectangular pixel window according to the gray values of the pixel points;

calculating a spatial gradient matrix of the target feature point in the rectangular pixel window according to the gray gradient values of the target feature point in the transverse and longitudinal directions of the rectangular pixel window;

obtaining a characteristic value of the spatial gradient matrix; determining the type of the target characteristic point according to the characteristic value; the types of the target feature points comprise corner points and edge points.

8. A method for tracking picture feature points is characterized by comprising the following steps:

c. tracking the target feature point according to the position of the target feature point by using the tracking method of the picture feature point as claimed in any one of claims 1 to 7 to obtain a tracking point matched with the position of the target feature point in the current layer image of the image pyramid of the target image;

9. An apparatus for tracking feature points of a picture, comprising:

10. The apparatus for tracking picture feature points according to claim 9, wherein the tracking module comprises:

the deviation determining module is used for determining the position deviation of the initial tracking point in the normal vector direction according to the gray value and the gray gradient value of the target feature point of the original image and the gray value of the initial tracking point;

the position determining module is used for determining the position of the initial tracking point of the target image; and determining the position of the target characteristic point in the target image according to the position of the starting tracking point and the position deviation.

11. The apparatus for tracking picture feature points according to claim 10, wherein the deviation determining module comprises:

the difference operation module is used for carrying out difference on the gray value of the target characteristic point and the gray value of the initial tracking point to obtain a gray deviation value of the target characteristic point and the initial tracking point;

and the deviation obtaining module is used for obtaining the position deviation of the initial tracking point in the normal vector direction according to the gray level deviation value and the gray level gradient value of the target characteristic point in the normal vector direction.

12. The apparatus for tracking picture feature points according to claim 11, wherein the deviation obtaining module comprises:

the square operation module is used for carrying out square operation on the gray gradient value of the target characteristic point in the normal vector direction to obtain a spatial gradient value of the target characteristic point in the normal vector direction;

the deviation value calculation module is used for obtaining an image deviation value according to the product of the gray gradient value of the target feature point in the normal vector direction and the gray deviation value;

the product operation module is used for calculating the ratio of the image deviation value to the spatial gradient value of the target feature point in the normal vector direction; and determining the position deviation of the initial tracking point in the normal vector direction according to the ratio and the normal vector of the target feature point.

13. The apparatus for tracking picture feature points according to claim 9, wherein the determining module comprises:

the gray level obtaining module is used for obtaining the gray level value of the target characteristic point of the original image; calculating gray gradient values of the target feature points in the transverse direction and the longitudinal direction according to the gray values;

the first construction module is used for constructing a gray gradient matrix of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse direction and the longitudinal direction;

the vector determination module is used for determining the normal vector of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse direction and the longitudinal direction;

and the gradient determining module is used for obtaining the gray gradient value of the target characteristic point in the normal vector direction according to the product of the gray gradient matrix and the normal vector.

14. The apparatus for tracking picture feature points of claim 13, wherein the vector determination module comprises:

the second construction module is used for constructing a spatial gradient matrix of the target characteristic point according to the gray gradient values of the target characteristic point in the transverse and longitudinal directions;

the vector acquisition module is used for acquiring the characteristic vector of the spatial gradient matrix; and determining the normal vector of the target characteristic point according to the characteristic vector of the spatial gradient matrix.

15. The apparatus for tracking picture feature points according to any one of claims 9 to 14, further comprising:

the window generation module is used for generating a rectangular pixel window taking the target feature point as the center in the original image;

the gradient calculation module is used for acquiring the gray value of each pixel point in the rectangular pixel window; calculating gray gradient values of the target characteristic points in the transverse and longitudinal directions of the rectangular pixel window according to the gray values of the pixel points;

the matrix calculation module is used for calculating a spatial gradient matrix of the target characteristic point in the rectangular pixel window according to the gray gradient values of the target characteristic point in the transverse and longitudinal directions of the rectangular pixel window;

the type determining module is used for acquiring the eigenvalue of the spatial gradient matrix; determining the type of the target characteristic point according to the characteristic value; the types of the target feature points comprise corner points and edge points.

16. An apparatus for tracking feature points of a picture, comprising:

a tracking point determining module, configured to perform step c, track the target feature point according to the position of the target feature point by using the tracking method for the picture feature point according to any one of claims 1 to 7, so as to obtain a tracking point, in the current-layer image of the image pyramid of the target image, that matches the position of the target feature point;

17. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method for tracking picture feature points according to any one of claims 1 to 8 when executing the computer program.

18. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of tracking picture feature points of any one of claims 1 to 8.