Anti-interference target laser line extraction method

Abstract

The application discloses an anti-interference target laser line extraction method, comprising the following steps: S1, Gaussian filtering processing is performed on an image; S2, start and end position points of the target laser line are determined: the start and end position points of the target laser line are b and b1 respectively, a difference value of adjacent two pixel gray values is diffval, and the determination method of b and b1 is as follows: the determination method of b is b-a<=diffval and c-b>=diffval; the determination method of b1 is b1-a1>=-diffval and b1-c1<=-diffval; wherein, a and c are respectively two adjacent points on both sides of b in the width direction of the laser line, and a1 and c1 are respectively two adjacent points on both sides of b1 in the width direction of the laser line; S3, the line width of the target laser line is set; S4, a center point of the target laser line is extracted: the gray gravity center method is used to extract the center of the laser line; and S5, the target laser line is extracted; the application can be applied to a 3D laser profiler to extract a correct laser line under the condition of interference, so that serious loss caused by inaccurate extraction of the laser line to industrial production is avoided. The application improves the laser line extraction precision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser line extraction, and particularly to an anti-interference target laser line extraction method. BACKGROUND

[0002] 3D laser profilers are widely used in manufacturing (e.g., workpiece inspection), automotive industry (e.g., part measurement and quality control), construction industry (e.g., structural scanning), medical field (e.g., volume measurement and surgical planning), and robotics (e.g., path planning and environment perception). They are used for high-precision measurement and inspection, helping to improve product quality and production efficiency.

[0003] Commonly used traditional laser stripe center extraction methods include gray-weighted centroid method, threshold method, extreme value method, and Steger algorithm. When a laser line is projected onto a conventional non-reflective workpiece, only one laser line approximately obeying Gaussian distribution is formed, and any of the above methods can be used to extract the target laser line. When the workpiece is transparent or reflective, multiple laser lines may be formed when the laser line is projected onto the measured workpiece, or interference spots and bright spots may appear due to scattered light. At this time, it is difficult to extract the target laser line by using the above traditional methods. In order to solve the influence of interference, the present application proposes an anti-interference target laser line extraction method. SUMMARY

[0004] The present application aims to provide an anti-interference target laser line extraction method, which aims to solve the problem that the correct target laser line cannot be easily extracted when the laser line is interfered.

[0005] The present application is implemented as follows:

[0006] An anti-interference target laser line extraction method, comprising the following steps:

[0007] S1, performing Gaussian filtering processing on the image;

[0008] S2, determining the start and end position points of the target laser line: the start and end position points of the target laser line are b and b1 respectively, the difference value of the gray values of the adjacent two pixels is diffval, and the determination method of b and b1 is as follows:

[0009] The determination method of b is b-a≤diffval and c-b≥diffval, and the determination method of b1 is b1-a1≥-diffval and b1-c1≤-diffval; wherein a and c are respectively two adjacent points on both sides of b in the width direction of the laser line, and a1 and c1 are respectively two adjacent points on both sides of b1 in the width direction of the laser line;

[0010] S3, set the line width of the target laser line: according to step S2, determine the starting position point which may be the target laser line, calculate the line width W = b1-b at each starting and ending position, set the possible line width range W of the target laser line min ~ W max When the line width satisfies W min ≤ W ≤ W max , it may be the target laser line;

[0011] S4, extract the center point of the target laser line: use the gray gravity center method to extract the center of the laser line, and the formula used is as follows:

[0012]

[0013] Where I(i,j) represents the gray value of the image at (i,j), bi and bj represent the pixel positions of the starting position point b, b1i and b1j represent the pixel positions of the ending position point b1;

[0014] S5, extract the target laser line, specifically including the following steps:

[0015] S51, select the brightest laser point in the first row as the target laser line;

[0016] S52, select the target laser line in the second row which is closest to the target laser line in the first row in the x direction;

[0017] S53, select the target laser line in the third row which is closest to the average value of the first row and the second row in the x direction;

[0018] S54, repeat steps S51-S53 until the target laser line in the nth row is closest to the average value of the previous n-1 rows in the x direction;

[0019] S55, determine the target laser line of the subsequent rows.

[0020] Further, in step S1, the image is subjected to 5*5 Gaussian filtering.

[0021] Further, in step S55, the target laser line in the current row is closest to the average value of the previous n rows in the x direction.

[0022] Further, in step S55, n can be set to different sizes according to the extraction effect.

[0023] Further, step S5 further includes the following steps:

[0024] S56, the current nth row is invalid, i.e. no laser point is extracted, the (n+1)th row is valid, i.e. a valid point is successfully extracted, and the determination method of the target laser point in the (n+1)th row is the same as that in the first row.

[0025] Compared with the prior art, the present application has the beneficial effect that the present application can be applied in a 3D laser profiler to extract a correct laser line in the presence of interference, thereby avoiding serious losses in industrial production caused by inaccurate extraction of the laser line. The present application improves the extraction accuracy of the laser line. BRIEF DESCRIPTION OF DRAWINGS

[0026] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application and, together with the specific description below, serve to explain the present application, but do not constitute a limitation of the present application. In the drawings, various elements or parts are not necessarily drawn to scale.

[0027] Figure 1 is a Gaussian distribution diagram of the laser line;

[0028] Figure 2 is a diagram of the extracted laser points;

[0029] Figure 3 is a diagram of the extracted target laser line;

[0030] Figure 4 is a diagram of the target laser points determined using the present application. DETAILED DESCRIPTION

[0031] The following drawings describe the specific embodiments of the present application in detail. The embodiments are intended to further explain the present application, rather than limit the present application.

[0032] An anti-interference target laser line extraction method, comprising the following steps:

[0033] S1, performing 5*5 Gaussian filtering processing on the image, and the size of the filter kernel can be increased or decreased according to the actual situation.

[0034] S2, determining the start and end position points of the target laser line: the middle of the laser line is the brightest, and the farther it extends to the surrounding, the weaker the light intensity is. This distribution rule conforms to the Gaussian distribution, as shown in Figure 1 .

[0035] The start and end position points of the target laser line are b and b1 respectively, and the difference between the gray values of the adjacent two pixels is diffval. The determination method of b and b1 is as follows:

[0036] The determination method of b is b-a≤diffval and c-b≥diffval, and the determination method of b1 is b1-a1≥-diffval and b1-c1≤-diffval. Wherein, a and c are respectively two adjacent points on both sides of b in the width direction of the laser line, and a1 and c1 are respectively two adjacent points on both sides of b1 in the width direction of the laser line.

[0037] S3, set the line width of the target laser line: according to step S2, determine the starting position point which may be the target laser line, calculate the line width W = b1-b at each starting and ending position, set the possible line width range W of the target laser line min ~ W max When the line width satisfies W min ≤ W ≤ W max , it may be the target laser line, and the restriction of the line width of the target laser line can filter out some interference points.

[0038] S4, extract the center point of the target laser line: use the gray gravity method to extract the center of the laser line, and the formula used is as follows:

[0039]

[0040] Where I(i,j) represents the gray value of the image at (i,j), bi and bj represent the pixel position of the starting position point b, b1i and b1j represent the pixel position of the ending position point b1.

[0041] Set diffval = 10, W min = 3, W max = 100, and the extraction result is as shown in Figure 2 , where the green points in the figure are the extracted possible laser line center points.

[0042] S5, extract the target laser line, specifically including the following steps:

[0043] S51, the first row selects the brightest laser point as the target laser line.

[0044] S52, the second row selects the one closest to the target laser line of the first row in the x direction as the target laser line.

[0045] S53, the third row selects the one closest to the average value of the x directions of the first and second rows in the x direction as the laser line.

[0046] S54, repeat steps S51-S53 until the nth row is closest to the average value of the x directions of the previous n-1 rows in the x direction as the target laser line.

[0047] S55, determine the target laser line of the subsequent rows, and the one closest to the average value of the x directions of the previous n rows in the x direction as the target laser line, n can be set to different sizes according to the extraction effect, such as 5 or 10 and the like.

[0048] S56, if the current nth row is invalid, i.e. no laser point is extracted, the (n+1)th row is valid, i.e. a valid point is extracted, and the determination method of the target laser point of the (n+1)th row is the same as that of the first row.

[0049] The extracted target laser line diagram and the target laser point determined by the method for extracting the target laser line are shown in Figure 3 and Figure 4 . Figure 3 In the figure, the red dot is the center point of the extracted target laser line, Figure 4 In the figure, the red dot is the target laser point determined by the method for extracting the target laser line.

[0050] The present application can be applied in 3D laser profilometer to extract correct laser line in the presence of interference, avoiding serious loss caused by inaccurate extraction of laser line in industrial production. In addition, the present application improves the extraction precision of laser line.

[0051] Although the present application has been described in detail according to the previous embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions in the described embodiments, or make equivalent replacements to some or all of the technical features. These modifications or replacements will not change the essence of the related technical solutions, making them beyond the technical scope defined by the various embodiments of the present application. In addition to the technical features described in the specification, they are known to those skilled in the art.

Claims

1. A method for extracting target laser lines with anti-interference capabilities, characterized in that, Includes the following steps: S1. Perform Gaussian filtering on the image; S2. Determine the possible start and end points of the target laser line: The start and end points of the target laser line are b and b1, respectively. The difference between the gray values ​​of two adjacent pixels is diffval. The method for determining b and b1 is as follows: The method for determining b is: ba≤diffval and cb≥diffval; the method for determining b1 is: b1-a1≥-diffval and b1-c1≤-diffval; where a and c are two adjacent points on both sides of b in the laser line width direction, and a1 and c1 are two adjacent points on both sides of b1 in the laser line width direction. S3. Set the linewidth of the target laser line: Based on the possible starting points of the target laser line determined in step S2, calculate the linewidth W = b1 - b at each starting and ending point, and set the possible linewidth range W of the target laser line. min ~W max When the line width satisfies W min ≤W≤W max It may be the target laser line; S4. Extract the center point of the target laser line: Use the gray-scale centroid method to extract the center of the laser line. The formula used is as follows: Where I(i,j) represents the gray value of the image at (i,j), bi and bj represent the pixel positions of the starting point b, and b1i and b1j represent the pixel positions of the ending point b1. S5. Extract the target laser line, which includes the following steps: S51. Select the brightest laser point in the first row as the target laser line; S52. The second row selects the target laser line that is closest to the target laser line in the first row in the x-direction. S53. The laser line is selected from the third row whose x-direction distance is closest to the average x-direction distance of the first and second rows. S54. Repeat steps S51-S53 until the target laser line is the one whose x-direction distance in the nth row is closest to the average x-direction distance of the previous n-1 rows. S55. Determine the target laser line for the subsequent row.

2. The method for extracting anti-interference target laser lines according to claim 1, characterized in that, In step S1, the image is subjected to a 5*5 Gaussian filter.

3. The method for extracting anti-interference target laser lines according to claim 1, characterized in that, In step S55, the line closest to the average value of the previous n rows in the x-direction is the target laser line.

4. The method for extracting anti-interference target laser lines according to claim 3, characterized in that, In step S55, n is set to different sizes depending on the extraction effect.

5. The method for extracting anti-interference target laser lines according to claim 3, characterized in that, Step S5 also includes the following steps: S56. The current nth row is an invalid value, meaning no laser point was extracted. The (n+1)th row is a valid value, meaning a valid point was successfully extracted. The method for determining the target laser point in the (n+1)th row is the same as the method for determining the first row.

Images