WO2022110043A1

WO2022110043A1 - Erosion detection method and apparatus, and computer-readable medium

Info

Publication number: WO2022110043A1
Application number: PCT/CN2020/132353
Authority: WO
Inventors: 董晓滨; 赵杰; 许晓东
Original assignee: 西门子股份公司; 西门子（中国）有限公司
Priority date: 2020-11-27
Filing date: 2020-11-27
Publication date: 2022-06-02

Abstract

An erosion detection method and apparatus, and a computer-readable medium. The erosion detection method comprises: acquiring a first detection image and a second detection image of a component to be detected (101), wherein the first detection image and the second detection image are respectively collected by two cameras that are comprised in a binocular camera; determining depth values of at least two target points on said component according to the parallax of the first detection image and the second detection image (102), wherein the depth value of any target point on said component is used for representing the distance between the target point and a connecting line of center points of two lenses that are comprised in the binocular camera; and determining an erosion degree of said component according to the depth value of each target point on said component (103). By means of the method, the efficiency of erosion detection can be improved.

Description

Erosion detection method, apparatus and computer readable medium

technical field

The present invention relates to the technical field of non-destructive testing, and in particular, to an erosion detection method, device and computer-readable medium.

Background technique

Ceramic heat shield (CHS) is located on the outer layer of the thermal protection structure on the inner surface of the gas turbine and has excellent ablation resistance. Defects such as damage and falling blocks, when the ceramic insulation tile has defects such as cracks, damage and falling blocks, the high temperature will cause damage to the metal casing of the combustion chamber. Therefore, the erosion detection of ceramic insulation tiles is an important link in the routine inspection process of gas turbines. According to the results of erosion detection, it can be determined whether the ceramic insulation tiles need to be replaced.

At present, when testing the erosion of ceramic thermal insulation tiles in gas turbines, the staff usually enters the combustion chamber from the manhole, and conducts visual inspection directly with the naked eye. Replacement ceramic insulation tiles result in less efficient erosion detection of ceramic insulation tiles.

SUMMARY OF THE INVENTION

In view of this, the erosion detection method, device and computer readable medium provided by the present invention can improve the efficiency of erosion detection.

In a first aspect, an embodiment of the present invention provides an erosion detection method, including:

acquiring a first detection image and a second detection image of the component to be detected, wherein the first detection image and the second detection image are respectively collected by two cameras included in the binocular camera;

According to the parallax of the first detection image and the second detection image, the depth values of at least two target points on the component to be detected are determined, wherein the depth value of any target point on the component to be detected is used to represent The distance between the target point and the center point of the two lenses included in the binocular camera;

According to the depth value of each of the target points on the component to be inspected, the degree of erosion of the component to be inspected is determined.

In a second aspect, an embodiment of the present invention further provides an erosion detection device, including:

an acquisition module, configured to acquire a first detection image and a second detection image of the component to be detected, wherein the first detection image and the second detection image are respectively collected by two cameras included in the binocular camera;

An image processing module for determining at least two target points on the component to be detected according to the parallax of the first detection image acquired by the acquisition module and the second detection image acquired by the acquisition module , wherein the depth value of any target point on the component to be detected is used to represent the distance between the target point and the center point of the two lenses included in the binocular camera;

An erosion judging module is used for determining the degree of erosion of the part to be inspected according to the depth value of each of the target points on the part to be inspected determined by the image processing module.

In a third aspect, an embodiment of the present invention further provides another erosion detection apparatus, including: at least one memory and at least one processor;

the at least one memory for storing a machine-readable program;

The at least one processor is configured to invoke the machine-readable program to execute the method provided in the first aspect.

In a fourth aspect, an embodiment of the present invention further provides a computer-readable medium, where computer instructions are stored on the computer-readable medium, and when executed by a processor, the computer instructions cause the processor to execute the above-mentioned first method provided by the aspect.

According to the technical solutions provided in the above first to fourth aspects, the first detection image and the second detection image of the component to be detected respectively collected by the two cameras included in the binocular camera are acquired, according to the first detection image and the second detection image. 2. Detect the parallax of the image, determine the depth value of at least two target points on the component to be detected, and then determine the degree of erosion of the component to be detected according to the depth value of each target point, wherein the depth value of any target point on each component to be detected It is used to characterize the distance between the target point and the center point of the two lenses included in the binocular camera. It can be seen that the distance between each target point on the component to be detected and the line connecting the center points of the two lenses of the binocular camera, that is, the depth value, can be obtained by using the parallax between the two images captured by the binocular camera. The depth value of each target point can determine the degree of corrosion of the component to be tested, so that the degree of corrosion can be indirectly reflected by the depth value, without the need for staff to perform visual inspection one by one, and then rely on work experience to determine whether the component to be tested needs to be replaced. Thereby, the efficiency of performing erosion detection is improved.

In a first possible implementation manner, in combination with any of the above-mentioned aspects, when determining the depth value of each target point on the component to be detected according to the parallax of the first detection image and the second detection image, it is possible to first determine that the target point is in pixel points corresponding to the first detection image and the second detection image, and then determine the parallax of the first detection image and the second detection image to determine the depth value of the target point. Specifically, the depth value can be determined in the following ways:

For each of the target points on the component to be detected, execute:

Determine the first pixel point corresponding to the target point in the first detection image, and determine the second pixel point corresponding to the target point in the second detection image;

determining the target parallax between the first pixel point and the second pixel point, wherein the target parallax is used to represent the relative position of the first pixel point in the first detection image and the second pixel point The difference between the relative positions of the pixel points in the second detection image;

The depth value of the target point is determined according to the target parallax, the focal length of the camera in the binocular camera, and the distance between the optical centers of the two cameras in the binocular camera.

In the embodiment of the present invention, for each target point on the component to be detected, by determining the corresponding pixel points of the target point in the first detection image and the second detection image respectively, the coordinates between the two pixel points can be determined according to the coordinates between the two pixel points. Difference, determine the difference between the relative positions of the target point in the two detection images, that is, the target parallax of the target point in the two detection images, and then pass the target parallax, the focal length of the camera in the binocular camera and the two cameras. The distance between the optical centers obtains the depth value of the target point. Since the depth value is the distance from the target point on the component to be detected to the plane of the binocular camera, it can indirectly reflect the erosion information on the upper surface of the component to be detected, and then can treat Test components for objective and reasonable evaluation. At the same time, there is no need for staff to perform visual inspection based on experience, thereby further improving the efficiency of erosion inspection.

In the second possible implementation manner, in combination with the implementation manner of any of the above-mentioned aspects, when determining the degree of erosion of the component to be detected according to the depth value of each target point on the component to be detected, you can first determine the degree of erosion of the component to be detected according to the edge of the component to be detected. The depth value of each target point ensures that the surface of the part to be inspected is parallel to the imaging plane of the binocular camera, thereby determining the degree of erosion of the part to be inspected. Specifically, the corrosion degree of the component to be tested can be determined by the following methods:

Judging whether the depth value of each target point located at the edge of the component to be detected is within a preset threshold range;

If the depth value of each of the target points on the edge of the component to be detected is within the preset threshold range, determine the depth of the component to be detected according to the depth value of each of the target points on the component to be detected degree of erosion;

If the depth values of the target points located at the edge of the component to be detected are not all within the preset threshold range, execute:

Determine the angle between the imaging plane of the binocular camera and the surface of the component to be detected according to the depth value of each of the target points on the edge of the component to be detected;

calibrating the depth value of each of the target points on the component to be detected according to the included angle;

The degree of erosion of the part to be inspected is determined according to the calibrated depth value of each of the target points on the part to be inspected.

In the embodiment of the present invention, by judging whether the depth value of each target point located at the edge of the component to be detected is within a preset threshold range, it can be determined whether there is an included angle between the imaging plane of the binocular camera and the surface of the component to be detected, If the depth value of each target point located at the edge of the part to be inspected is within the preset threshold range, then there is no included angle (ie parallel) between the imaging plane of the binocular camera and the surface of the part to be inspected, and the determined The depth value of the target point determines the degree of erosion of the component to be detected; if the depth values of each target point located on the edge of the component to be detected are not all within the preset threshold range, it needs to be determined according to the depth value of each target point on the edge of the component to be detected. The angle between the imaging plane of the binocular camera and the surface of the part to be inspected (that is, not parallel), and the depth value of each target point on the part to be inspected is calibrated according to the angle, and then according to the depth value of each target after calibration Determine the degree of erosion of the part to be inspected. Since the imaging plane of the binocular camera is parallel to the surface of the part to be inspected, the determined depth value is more sufficient to describe the real erosion situation on the surface of the part to be inspected, so that the accuracy of the obtained depth value can be ensured, and the erosion can be improved. Accuracy and reliability of detection.

In a third possible implementation manner, combined with the implementation manner of any of the above aspects, when determining the degree of erosion of the component to be detected according to the depth value of each target point on the component to be detected, the effective detection of the component to be detected may be determined first. area, and then determine the erosion degree of the component to be inspected according to the depth value of each target point in the effective inspection area. Specifically, the corrosion degree of the component to be tested can be determined by the following methods:

According to the depth value of each target point and the distribution of each target point on the component to be detected, an effective detection area on the component to be detected is determined, wherein the effective detection area includes at least two the target point, and the depth value of each of the target points in the effective detection area is within the set range;

According to the depth value of each of the target points in the effective detection area, the degree of erosion of the component to be detected is determined.

In the embodiment of the present invention, the effective detection area is determined according to the depth value of each target point and the distribution of each target point on the component to be detected, so that the target points far away from the center of the component to be detected, with a small number and a large error can be removed, Only the area where 90% of the target points concentrated on the parts to be inspected are reserved as the effective detection area, so as to determine the degree of erosion of the parts to be inspected according to the depth value of each target point in the effective inspection area, so as to realize the detection of each target point. Noise reduction processing of depth values further improves the accuracy and reliability of erosion detection.

In a fourth possible implementation manner, in combination with any of the above aspects or any of the possible implementation manners, when determining the degree of erosion of the component to be detected according to the depth value of each target point on the component to be detected, the maximum depth of The difference between the value and the minimum depth value further determines the degree of erosion of the component to be inspected. Specifically, the corrosion degree of the component to be tested can be determined by the following methods:

Determine a first target depth value and a second target depth value, wherein the first target depth value is the maximum value among the depth values of each of the target points on the component to be detected, and the second target depth value is the minimum value among the depth values of each of the target points on the component to be detected;

performing a difference operation on the first target depth value and the second target depth value to obtain a target difference value;

judging whether the absolute value of the target difference is greater than a preset threshold;

If the absolute value of the target difference value is greater than the preset threshold value, it is determined that the component to be inspected has an erosion degree that needs to be replaced.

In the embodiment of the present invention, the maximum depth value and the minimum depth value are respectively determined from the depth values of each target point on the component to be detected, and the target difference value is compared by calculating the target difference value between the maximum depth value and the minimum depth value. With the preset threshold, if the target difference is not greater than the preset threshold, it is determined that the component to be detected is an erosion degree that does not need to be replaced temporarily; if the target difference is greater than the preset threshold, it is determined that the component to be detected is The level of erosion that needs to be replaced. Among them, the target difference is the actual maximum erosion depth of the surface of the component to be detected. Therefore, when the target difference exceeds the set standard value (ie, the preset threshold), the component to be detected needs to be replaced. Detection of corrosion levels of components. This process eliminates the need for personnel to perform visual inspections with experience, increasing the efficiency of erosion inspections.

Description of drawings

1 is a flowchart of an erosion detection method provided by an embodiment of the present invention;

2 is a flowchart of a method for acquiring a depth value provided by an embodiment of the present invention;

3 is a flowchart of a method for determining an erosion degree provided by an embodiment of the present invention;

FIG. 4 is a flowchart of another method for determining the degree of erosion provided by an embodiment of the present invention;

5 is a flowchart of another method for determining an erosion degree provided by an embodiment of the present invention;

6 is a flowchart of another erosion detection method provided by an embodiment of the present invention;

7 is a schematic diagram of an erosion detection device provided by an embodiment of the present invention;

FIG. 8 is a schematic diagram of another erosion detection device provided by an embodiment of the present invention;

FIG. 9 is a schematic diagram of another erosion detection device provided by an embodiment of the present invention;

10 is a schematic diagram of still another erosion detection device provided by an embodiment of the present invention;

11 is a schematic diagram of still another erosion detection device provided by an embodiment of the present invention;

FIG. 12 is a schematic diagram of an erosion detection apparatus including a memory and a processor provided by an embodiment of the present invention.

List of reference numbers:

101: Acquire a first inspection image and a second inspection image of the component to be inspected

102: Determine depth values of at least two target points on the component to be detected according to the parallax of the first detected image and the second detected image

103: Determine the degree of erosion of the part to be inspected according to the depth value of each target point on the part to be inspected

201: For each target point on the component to be detected, determine the first pixel point corresponding to the target point in the first detection image, and determine the second pixel point corresponding to the target point in the second detection image

202: Determine the target parallax between the first pixel point and the second pixel point

203: Determine the depth value of the target point according to the target parallax, the focal length of the camera in the binocular camera, and the distance between the optical centers of the two cameras in the binocular camera

301: Determine whether the depth value of each target point located at the edge of the component to be detected is within a preset threshold range

302: Determine the degree of erosion of the component to be detected according to the depth value of each target point on the component to be detected, and end the current process

303: Determine the angle between the imaging plane of the binocular camera and the surface of the component to be detected according to the depth value of each target point on the edge of the component to be detected

304: Calibrate the depth value of each target point on the part to be detected according to the included angle

305: Determine the degree of erosion of the part to be inspected according to the calibrated depth values of each target point on the part to be inspected, and end the current process

401: Determine the effective inspection area on the part to be inspected

402: Determine the degree of erosion of the component to be detected according to the depth value of each target point in the effective detection area

501: Determine the first target depth value and the second target depth value

502: Perform a difference operation on the first target depth value and the second target depth value to obtain a target difference value

503: Determine whether the absolute value of the target difference is greater than the preset threshold

504: Determine the corrosion degree of the component to be inspected that needs to be replaced, and end the current process

505: Determine the corrosion degree of the component to be inspected that does not need to be replaced, and end the current process

601: Acquire a first inspection image and a second inspection image of the component to be inspected

602: Determine the target disparity of the target point

603: Determine the depth value of the target point

604: Determine the effective detection area on the component to be detected

605: Determine whether the depth value of each target point on the edge of the component to be detected is within a preset threshold range

606: Calibrate the depth value of each target point on the part to be detected

607: Determine the first target depth value and the second target depth value

608: Perform a difference operation on the first target depth value and the second target depth value

609: Determining the degree of erosion of the part to be inspected

701: Acquisition module 702: Image processing module 703: Erosion judgment module

704: Memory 705: Processor 7021: Pixel acquisition unit

7022: Parallax acquisition unit 7023: Depth value acquisition unit 7031: First judgment unit

7032: The first relationship determination unit 7033: The second relationship determination unit 7034: The area determination unit

7035: Second erosion determination unit 7036: Depth value filtering unit 7037: Operation unit

7038: The second judgment unit 7039: The erosion degree determination unit

Detailed ways

As mentioned above, in the current corrosion detection of ceramic thermal insulation tiles in gas turbines, the staff usually enters the combustion chamber through the manhole, and directly visually inspects them with the naked eye, and determines the position, direction and length of the cracks against the ceramic defect template. Etc., based on work experience to determine the crack depth of the ceramic thermal insulation tile, and then determine the ceramic thermal insulation tile that needs to be replaced. This method is labor-intensive, slow in detection and susceptible to subjective factors, resulting in low efficiency in corrosion detection of ceramic insulation tiles.

In the embodiment of the present invention, the first detection image and the second detection image of the component to be detected respectively collected by the two cameras included in the binocular camera are acquired, and according to the target points corresponding to the first detection image and the second detection image, respectively According to the coordinate difference between the two pixel points, the target parallax of the target point in the first detection image and the second detection image is determined, and according to the target parallax, the distance between the optical centers of the two cameras in the binocular camera The distance between and the focal length of the camera determine the depth value of the target point. In this way, the depth value of each target point in the component to be detected can be obtained, and the maximum depth value and the minimum depth value are respectively determined from the depth values of each target point. When the target difference between the maximum depth value and the minimum depth value is greater than the preset value When the threshold value is reached, it can be determined that the component to be detected is the degree of corrosion that needs to be replaced, so as to realize the detection of the degree of corrosion of the component to be detected. This process eliminates the need for personnel to perform visual inspections with experience, increasing the efficiency of erosion inspections.

It should be noted that the first detection image and the second detection image may also be collected by two cameras of the binocular camera, the two cameras in the binocular camera have the same focal length, and the two cameras in the binocular camera have the same focal length. . The depth value of any target point on the component to be detected is used to represent the distance between the target point and the center point of the two lenses included in the binocular camera. The target parallax is used to represent the difference between the relative position of the first pixel point in the first detection image and the relative position of the second pixel point in the second detection image.

The erosion detection method and device provided by the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

As shown in FIG. 1, an embodiment of the present invention provides an erosion detection method, and the method may include the following steps:

Step 101: Acquire a first detection image and a second detection image of the component to be detected, wherein the first detection image and the second detection image are respectively collected by two cameras included in the binocular camera;

Step 102: Determine the depth values of at least two target points on the component to be detected according to the parallax of the first detection image and the second detection image, wherein the depth value of any target point on the component to be detected is used to characterize the target point and the dual target point. The distance between the center points of the two lenses included in the eye camera;

Step 103: Determine the degree of erosion of the component to be detected according to the depth values of each target point on the component to be detected.

In the embodiment of the present invention, the first detection image and the second detection image of the component to be detected that are respectively collected by two cameras included in the binocular camera are acquired, and the to-be-detected image is determined according to the parallax of the first detection image and the second detection image. The depth values of at least two target points on the component, and then determine the degree of erosion of the component to be detected according to the depth value of each target point, wherein the depth value of any target point on each component to be detected is used to characterize the target point and the The distance between the center points of the two lenses included in the binocular camera. It can be seen that the distance between each target point on the component to be detected and the line connecting the center points of the two lenses of the binocular camera, that is, the depth value, can be obtained by using the parallax between the two images captured by the binocular camera. The depth value of each target point can determine the degree of corrosion of the component to be tested, so that the degree of corrosion can be indirectly reflected by the depth value, without the need for staff to perform visual inspection one by one, and then rely on work experience to determine whether the component to be tested needs to be replaced. Non-destructive testing of components to be tested is achieved, thereby improving the efficiency of corrosion testing.

In this embodiment of the present invention, the left and right cameras included in the binocular camera simultaneously capture images of the component to be detected, wherein the distance from the binocular camera to the surface of the component to be detected can be determined according to user requirements. For the part to be inspected, the distance from the binocular camera to the surface of the part to be inspected can be adjusted according to the size of the part to be inspected. For example, if the size of the ceramic thermal insulation tile inside the gas turbine generator is 20cm*20cm, the distance between the binocular camera plane and the ceramic thermal insulation tile is 30-50cm. If the detection image includes multiple components to be detected, the distance from the binocular camera to the surface of the components to be detected can be adjusted according to the number of components to be detected. Among them, the two lenses included in the binocular camera should be placed directly above the surface to be inspected when collecting inspection images. Auxiliary lighting can be used to maintain a good background light source to obtain inspection images with clear image quality, thereby improving the accuracy of erosion detection. and reliability.

In the embodiment of the present invention, there is a deviation between the positions of the pixels imaged by the two cameras in the same scene, that is, parallax. Based on the parallax, according to the relative positional relationship between the left and right cameras and the parameter information of the camera itself, the spatial distribution of the target can be obtained. Depth information, which is used to represent the distance of a certain point on the target in the scene from the camera. Therefore, by means of the parallax of the first detection image and the second detection image collected by the binocular camera, the depth value of the target point of the component to be detected can be determined.

In the embodiment of the present invention, the three-dimensional position information of the target in space can be obtained according to the depth value information. Therefore, according to the depth value of each target point on the part to be inspected, the three-dimensional position information of the part to be inspected in space can be determined, and then the three-dimensional position information of the part to be inspected can be determined. The degree of corrosion of the parts, to achieve non-destructive testing of the parts to be tested.

Optionally, on the basis of the erosion detection method shown in FIG. 1, when determining the depth value of each target point on the component to be detected according to the parallax of the first detection image and the second detection image, it can be determined that the target point is in the first detection image. The pixel points corresponding to the first detection image and the second detection image are further determined to determine the parallax of the first detection image and the second detection image to determine the depth value of the target point. As shown in Figure 2, determining the depth value according to the disparity can be implemented as follows:

Step 201: for each target point on the component to be detected, determine the first pixel point corresponding to the target point in the first detection image, and determine the second pixel point corresponding to the target point in the second detection image;

Step 202: Determine the target parallax between the first pixel point and the second pixel point, wherein the target parallax is used to represent the relative position of the first pixel point in the first detection image and the position of the second pixel point in the second detection image. difference in relative position;

Step 203: Determine the depth value of the target point according to the target parallax, the focal length of the camera in the binocular camera, and the distance between the optical centers of the two cameras in the binocular camera.

In the embodiment of the present invention, for each target point on the component to be detected, the corresponding pixel points of the target point in the first detection image and the second detection image are determined, and according to the coordinates of the pixel point, it is determined that the target point is in the first detection image and the second detection image. The difference between the relative positions of the two detection images is the target parallax of the target point, and then the depth of the target point is obtained through parameters such as the target parallax, the focal length of the camera in the binocular camera, and the distance between the optical centers of the two cameras. value.

In this embodiment of the present invention, the first detection image may be used as a reference, and for each target point, the pixel coordinates of the first pixel point corresponding to the target point in the reference image may be used as a template, and the second detection image may be searched for the For the same or similar target points, the pixel coordinates of the second pixel point are determined, and the disparity is determined by the positional deviation between the pixel coordinates of the two pixel points. Because the binocular cameras are usually placed horizontally, the positional deviation is generally reflected in the horizontal direction. For example, the coordinate of point X in the scene in the horizontal direction of the left camera is x, and the coordinate of the image in the horizontal direction of the right camera is x+p, then p is the parallax of point X. Based on the obtained parallax of each target point, a parallax map of the image to be detected can be obtained. The parallax map can be based on the first detected image, its size is the size of the first detected image, and the element value is the parallax of each target point. , so that the disparity of each target point can be intuitively obtained through the disparity map, which is convenient for subsequent acquisition of the depth value of each target point.

In this embodiment of the present invention, the depth value of each target point is determined by the following formula:

h=b*c/p

Among them, h is used to characterize the depth value of each target point in the part to be detected, b is used to characterize the distance between the optical centers of the two cameras in the binocular camera, c is used to characterize the focal length of the camera in the binocular camera, p is used to characterize The parallax of each target point between the first detection image and the second detection image is characterized.

In the embodiment of the present invention, the depth value is used to represent the distance between the target point and the center point of the two lenses included in the binocular camera, because the erosion information on the upper surface of the component to be detected can be reflected by the depth value, thereby enabling The objective and reasonable evaluation of the parts to be tested is carried out, the non-destructive testing of the parts to be tested is realized, and the visual inspection is avoided, thereby further improving the efficiency of corrosion testing.

Optionally, on the basis of the erosion detection method shown in FIG. 1 , when determining the degree of erosion of the component to be detected according to the depth value of each target point on the component to be detected, you can first determine the degree of erosion of the component to be detected according to the depth of each target point on the edge of the component to be detected. The depth value ensures that the surface of the part to be inspected is parallel to the imaging plane of the binocular camera, thereby determining the degree of erosion of the part to be inspected. As shown in Figure 3, determining the degree of erosion of the component to be inspected can be achieved as follows:

Step 301: determine whether the depth value of each target point located at the edge of the component to be detected is within the preset threshold range, if it is Y, go to step 302, if not N, go to step 303;

Step 302: According to the depth value of each target point on the component to be detected, determine the degree of erosion of the component to be detected, and end the current process;

Step 303: Determine the angle between the imaging plane of the binocular camera and the surface of the component to be detected according to the depth value of each target point on the edge of the component to be detected;

Step 304: Calibrate the depth value of each target point on the component to be detected according to the included angle;

Step 305: Determine the degree of erosion of the component to be detected according to the calibrated depth values of each target point on the component to be detected, and end the current process.

In the embodiment of the present invention, by judging whether the depth value of each target point located at the edge of the component to be detected is within a preset threshold range, it can be determined whether there is an included angle between the imaging plane of the binocular camera and the surface of the component to be detected, If the depth value of each target point located at the edge of the component to be detected is within the preset threshold range, then there is no included angle (that is, parallel) between the imaging plane of the binocular camera and the surface of the component to be detected, and the determined The depth value of the target point determines the degree of erosion of the component to be detected; if the depth values of each target point located on the edge of the component to be detected are not all within the preset threshold range, it needs to be determined according to the depth value of each target point on the edge of the component to be detected. The angle between the imaging plane of the binocular camera and the surface of the part to be inspected (that is, not parallel), and the depth value of each target point on the part to be inspected is calibrated according to the angle, and then according to the depth value of each target after calibration Determine the degree of erosion of the part to be inspected. Since the imaging plane of the binocular camera is parallel to the surface of the part to be inspected, the determined depth value is more sufficient to describe the real erosion situation on the surface of the part to be inspected, so that the accuracy of the obtained depth value can be ensured, thereby improving erosion detection. accuracy and reliability.

In the embodiment of the present invention, since there is an included angle between the imaging plane of the binocular camera and the surface of the component to be detected, there will be a fixed value error in the obtained depth value. Therefore, in order to ensure the authenticity and accuracy of the depth value , it is necessary to detect whether there is an included angle between the imaging plane of the binocular camera and the surface of the component to be detected to reduce errors.

In the embodiment of the present invention, it is determined whether the depth value of each target point located at the edge of the component to be detected is within the preset threshold range. The threshold range is set to 50cm±0.5cm. If the depth values of each target point on the edge of the component to be detected are 49.5cm, 49.6cm, 50.3cm, 49.8cm, 49.5cm, 50cm, 49.7cm, 50.1cm and 50.5cm respectively, the It is determined that the depth values of each target point are all within the preset threshold range, that is, the imaging plane of the binocular camera is parallel to the surface of the component to be detected.

In the embodiment of the present invention, if the depth values of each target point located at the edge of the component to be detected are not all within the preset threshold range, the depth value of each target point in the component to be detected can be calibrated in the following manner:

S1: Determine the minimum depth value (A) and the maximum depth value (B) from the depth values of each target point on the edge of the component to be detected;

S2: According to the minimum depth value (A) and the maximum depth value B), the depth value of each target point in the component to be detected is calibrated by the following formula:

Among them, h _i is used to represent the depth value of the i target point in the component to be detected after calibration processing, θ is used to represent the angle between the imaging plane of the binocular camera and the surface of the component to be detected, h _b is used to represent the maximum The depth value (B), _{ha is used to represent the minimum depth value (A), x b} _is used to represent the abscissa of the pixel corresponding to the B target point corresponding to the maximum depth value (B) in the first detection image, x _a is used to represent the abscissa of the pixel point corresponding to the A target point corresponding to the minimum depth value (A) in the first detection image, and h _i ¹ is used to represent the i target point in the component to be detected before the calibration process. Depth value; x _i is used to represent the abscissa of the pixel point corresponding to the i target point in the first detection image.

In the embodiment of the present invention, the component to be detected may be other components in the gas turbine, such as a blade, but the surface of the blade is an arc surface. Therefore, during the erosion detection process, it is necessary to use the above-mentioned depth value of each target point in the component to be detected for calibration way to calibrate the depth value of the blade.

Optionally, on the basis of the erosion detection method shown in FIG. 1, when determining the degree of erosion of the component to be detected according to the depth value of each target point on the component to be detected, the effective detection area of the component to be detected can be determined first, and then The erosion degree of the component to be inspected is determined according to the depth value of each target point in the effective inspection area. As shown in Figure 4, determining the degree of erosion of the component to be inspected can be achieved in the following ways:

Step 401: According to the depth value of each target point and the distribution of each target point on the component to be detected, determine the effective detection area on the component to be detected, wherein, the effective detection area includes at least two target points, and the effective detection area The depth value of each target point is within the set range;

Step 402: Determine the degree of erosion of the component to be detected according to the depth value of each target point in the effective detection area.

In the embodiment of the present invention, the effective detection area is determined according to the depth value of each target point and the distribution of each target point on the component to be detected, so that the target points far away from the center of the component to be detected, with a small number and a large error can be removed, Only the area where 90% of the target points concentrated on the parts to be inspected are reserved as the effective detection area, so as to determine the degree of erosion of the parts to be inspected according to the depth value of each target point in the effective inspection area, so as to realize the detection of each target point. The noise reduction processing of depth value further improves the accuracy and reliability of erosion detection.

In the embodiment of the present invention, a polymorphic filter may be used to remove abnormal depth values, so as to realize noise reduction processing on the depth values of each target point and reduce errors in the erosion detection result.

Optionally, on the basis of the erosion detection method shown in FIG. 1 or the depth value acquisition method shown in FIG. 2 or the erosion degree determination method shown in FIG. 3 or the erosion degree determination method shown in FIG. The depth value of the target point, when determining the degree of erosion of the component to be inspected, the degree of erosion of the component to be inspected may be further determined by the difference between the maximum depth value and the minimum depth value. As shown in FIG. 5 , the method for determining the degree of erosion of the component to be inspected includes the following steps:

Step 501: Determine a first target depth value and a second target depth value, wherein the first target depth value is the maximum value among the depth values of each target point on the component to be detected, and the second target depth value is the depth value of each target point on the component to be detected. The minimum value among the depth values of the target point;

Step 502: Perform a difference operation on the first target depth value and the second target depth value to obtain a target difference value;

Step 503: Determine whether the absolute value of the target difference is greater than the preset threshold, if it is Y, go to step 504, if not N, go to step 505;

Step 504: Determine the corrosion degree of the component to be detected that needs to be replaced, and end the current process;

Step 505 : Determine the corrosion degree of the component to be inspected that does not need to be replaced, and end the current process.

In the embodiment of the present invention, when the target difference between the maximum depth value and the minimum depth value among the depth values of each target point on the component to be detected is greater than a preset threshold, it is determined that the component to be detected is an erosion degree that needs to be replaced , and output the target difference, which is the actual erosion depth. Based on all the output target difference values and the parts to be inspected, an inspection report for this inspection can also be automatically generated. The inspection report can include the location of the inspected parts that need to be replaced, the erosion condition, the erosion depth value, and the maintenance date. and other statistical data, thereby helping to complete the erosion detection work faster and improve the efficiency of erosion detection.

In the embodiment of the present invention, based on the obtained parallax of each target point, a parallax map of the image to be detected can be obtained, and the depth value of each target point on the component to be detected is converted to the parallax map to obtain a depth map, a depth map It can be a grayscale image or a pseudo-color image, but the pixel value of the depth map is the actual distance between the target point and the center point of the two lenses included in the binocular camera. The color in the depth map represents the depth. A three-dimensional image corresponding to the part to be inspected can be obtained according to the depth map, wherein the three-dimensional image includes the three-dimensional position information and depth value of each target point of the part to be inspected in space, and the three-dimensional image of the part to be inspected can be visually seen through the three-dimensional image. The topography information is more visualized, and it is convenient to directly judge the two-dimensional topography and erosion depth of the surface of the component to be tested.

The following takes the ceramic heat insulating tile in the gas turbine as the component to be detected as an example to further describe the erosion detection method provided by the embodiment of the present invention. As shown in FIG. 6 , the method may include the following steps:

Step 601: Acquire a first inspection image and a second inspection image of the component to be inspected.

In this embodiment of the present invention, the first detection image and the second detection image may be collected by two cameras of the binocular camera or two cameras in the binocular camera, and the first detection image and the second detection image may include a to-be-detected image. The image of the inspection part may also include images of a plurality of parts to be inspected.

For example, for a piece of ceramic thermal insulation tile with a size of 20cm*20cm, a binocular camera is used, and the binocular camera is placed on the surface of the ceramic thermal insulation tile to be detected at a position where the distance between the binocular camera plane and the ceramic thermal insulation tile is 50cm. Right above, using auxiliary lighting, a first inspection image and a second inspection image of the ceramic thermal insulation tile are collected.

Step 602: Determine the target disparity of the target point.

In the embodiment of the present invention, for each target point on the component to be detected, determine the first pixel point and the second pixel point corresponding to the target point in the first detection image and the second detection image respectively, and according to the The difference between the relative positions of one pixel point and the second pixel point determines the target parallax of the target point in the first detection image and the second detection image.

For example, as described in the previous example, for the X point on the ceramic heat insulation tile, first determine the first pixel point X1 (x1, y1) corresponding to the X point in the first detection image, and determine the first pixel point X1 (x1, y1) in the second detection image. The corresponding second pixel point X2 (x2, y2), according to the difference in the relative position of X1 and X2 in the horizontal direction, determine the target parallax of point X in the first detection image and the second detection image is x1-x2 absolute. value.

Step 603: Determine the depth value of the target point.

In the embodiment of the present invention, according to the target parallax of each target point, the focal length of the camera in the binocular camera, and the distance between the optical centers of the two cameras in the binocular camera, the depth value of each target point is determined by the following formula:

h=b*c/p

For example, as described in the previous example, for the X point on the ceramic thermal insulation tile, the target parallax p is the absolute value of x1-x2, the distance between the optical centers of the two cameras in the binocular camera is b, and the camera in the binocular camera is the absolute value of x1-x2. The focal length is c, then the depth value h of point X is calculated by the above formula.

Step 604: Determine the effective inspection area on the component to be inspected.

In the embodiment of the present invention, after obtaining the depth value of each target point of the component to be detected, the effective detection area on the component to be detected is determined according to the depth value of each target point and the distribution of each target point on the component to be detected, The noise reduction processing for the depth value of each target point is realized, wherein at least two target points are included in the effective detection area, and the depth value of each target point in the effective detection area is within the set range.

For example, as described in the previous example, for the ceramic thermal insulation tile, after obtaining the depth values of each target point on the ceramic thermal insulation tile, the target points far away from the center of the component to be detected, with a small number and a large error can be removed, and only Retain the area where 90% of the target points concentrated on the ceramic heat insulation tile are located as the effective detection area.

Step 605: Determine whether the depth value of each target point on the edge of the component to be detected is within the preset threshold range, if yes, go to Step 607, if not, go to Step 606.

In the embodiment of the present invention, after obtaining the depth value of each target point of the component to be detected, it is determined whether the depth value of each target point on the edge of the effective detection area of the component to be detected is within the preset threshold range, if The depth value of the target point is within the preset threshold value range, then step 607 is performed for the depth value of each target point on the component to be detected, if the depth value of each target point on the edge is not within the preset threshold value range, then for the component to be detected. Step 606 is performed on the depth value of each upper target point.

For example, as described in the previous example, in the effective detection area of the ceramic heat insulation tile, determine whether the depth value of each target point located on the inner edge of the effective detection area is within 50cm±0.5cm, if the depth of each target point on the edge is within 50cm ± 0.5cm The values are 49.5cm, 60cm, 56.8cm, 54.2cm, 52.5cm, 50cm, 49.7cm, 53.1cm and 56.2cm respectively, then the depth value of each target point on the edge is not within the preset threshold range, and the imaging of the binocular camera If the plane is not parallel to the surface of the component to be inspected, step 606 is executed.

Step 606: Calibrate the depth value of each target point on the component to be detected.

In this embodiment of the present invention, the angle between the imaging plane of the binocular camera and the surface of the component to be detected is determined according to the depth value of each target point on the edge of the component to be detected, and the depth value of each target point on the edge of the component to be detected is determined from the depth value of each target point on the edge of the component to be detected. Determine the minimum depth value (A) and the maximum depth value (B) respectively;

According to the minimum depth value (A) and the maximum depth value B), the depth value of each target point in the component to be detected is calibrated by the following formula:

For example, as described in the previous example, the minimum depth value (A: 50cm) and the maximum depth value (B: 60cm) are respectively determined from the depth values of each target point on the edge of the component to be detected, and the corresponding point A in the first detection image is determined. The pixel coordinates (1, 2) are determined, the pixel coordinates (19, 4) corresponding to point B in the first detection image are determined, and the distance between the imaging plane of the binocular camera and the surface of the component to be detected is determined by the above formula. The angle is arctan5/9, the Y point on the ceramic heat insulation tile, the pixel coordinate in the first detection image is (10,9), and its depth value is 60cm, then the calibration is performed by the above formula to obtain the calibration The depth value of the rear Y point is 55cm (ie 60-5).

Step 607: Determine the first target depth value and the second target depth value.

In the embodiment of the present invention, a first target depth value and a second target depth value are determined, wherein the first target depth value is the maximum value among the depth values of each target point on the component to be detected, and the second target depth value is the value to be detected. Detects the minimum value of the depth values of each target point on the part.

For example, as described in the previous example, after the depth value of the target point on the ceramic heat insulating tile is calibrated, the maximum depth value is Y 55cm, and the minimum depth value is 50cm.

Step 608: Perform a difference operation on the first target depth value and the second target depth value.

In the embodiment of the present invention, a difference value operation is performed on the first target depth value and the second target depth value to obtain a target difference value.

For example, following the previous example, after the depth value of the target point on the ceramic heat insulation tile is calibrated, the maximum depth value is Y 55cm, the minimum depth value is 50cm, and the target difference is 5cm.

Step 609: Determine the erosion degree of the component to be inspected.

In the embodiment of the present invention, when judging whether the absolute value of the target difference is greater than a preset threshold, if so, it is determined that the component to be inspected is an erosion degree that needs to be replaced; if not, it is determined that the component to be inspected is an erosion that does not require replacement. degree.

For example, as described in the previous example, if the preset threshold is 3.5 cm, and the target difference value is 5 cm greater than the preset threshold, then it is determined that the ceramic thermal insulation tile is the degree of erosion that needs to be replaced.

As shown in FIG. 7 , an embodiment of the present invention provides an erosion detection device, including:

An acquisition module 701, configured to acquire the first detection image and the second detection image of the component to be detected, wherein the first detection image and the second detection image are respectively collected by two cameras included in the binocular camera;

An image processing module 702 for determining the depth values of at least two target points on the component to be detected according to the parallax of the first detection image acquired by the acquisition module 701 and the second detection image acquired by the acquisition module, wherein the The depth value of any target point on the component is used to represent the distance between the target point and the center point of the two lenses included in the binocular camera;

An erosion judging module 703 is configured to determine the degree of erosion of the part to be inspected according to the depth values of each target point on the part to be inspected determined by the image processing module 702 .

In this embodiment of the present invention, the acquisition module 701 can be used to perform step 101 in the above method embodiments, the image processing module 702 can be used to perform step 102 in the above method embodiments, and the erosion determination module 703 can be used to perform the above method embodiments. step 103.

Based on the erosion detection device shown in FIG. 7 , as shown in FIG. 8 , the image processing module 702 includes:

A pixel point acquisition unit 7021, for each target point on the component to be detected, to determine the first pixel point corresponding to the target point in the first detection image, and to determine the target point corresponding to the second detection image. The second pixel point of ;

A parallax acquisition unit 7022 for determining the target parallax between the first pixel point and the second pixel point determined by the pixel point acquisition unit 7022, wherein the target parallax is used to represent the position of the first pixel point in the first detection image. the difference between the relative position and the relative position of the second pixel in the second detection image;

A depth value acquisition unit 7023, configured to determine the depth value of the target point according to the focal length of the camera in the binocular camera, the distance between the optical centers of the two cameras in the binocular camera, and the target parallax determined by the parallax acquisition unit 7022.

In this embodiment of the present invention, the pixel point acquisition unit 7021 may be used to perform step 201 in the above method embodiments, the parallax acquisition unit 7022 may be used to perform step 202 in the above method embodiments, and the depth value acquisition unit 7023 may be used to perform the above mentioned method embodiments. Step 203 in the method embodiment.

Optionally, based on the erosion detection device shown in FIG. 7 , as shown in FIG. 9 , the erosion judgment module 703 includes:

A first judgment unit 7031 for judging whether the depth value of each target point located at the edge of the component to be detected is within the preset threshold range;

A first relationship determination unit 7032 is used for when the first judgment unit 7031 determines that the depth value of each target point on the edge of the component to be detected is within the preset threshold range, then according to the depth value of each target point on the component to be detected, Determine the degree of erosion of the part to be inspected;

A second relationship determination unit 7033 is used for when the first judgment unit 7031 determines that the depth values of each target point on the edge of the component to be detected are not all within the preset threshold range, according to the depth of each target point on the edge of the component to be detected. The value determines the angle between the imaging plane of the binocular camera and the surface of the part to be inspected, and calibrates the depth value of each target point on the part to be inspected according to the angle, and according to the calibrated depth value of each target point on the part to be inspected The depth value determines the degree of erosion of the part to be inspected.

In this embodiment of the present invention, the first judging unit 7031 may be configured to perform step 301 in the foregoing method embodiments, the first relationship determining unit 7032 may be configured to execute step 302 in the foregoing method embodiments, and the second relationship determining unit 7033 may be configured to perform

Steps

304 and 305 in the above method embodiments are performed.

Optionally, based on the erosion detection device shown in FIG. 9 , as shown in FIG. 10 , the erosion judgment module 703 includes:

An area determination unit 7034 for determining the effective detection area on the component to be detected according to the depth value of each target point and the distribution of each target point on the component to be detected, wherein the effective detection area includes at least two target points , and the depth value of each target point in the effective detection area is within the set range;

A second erosion determination unit 7035 is configured to determine the degree of erosion of the component to be inspected according to the depth values of each target point in the effective inspection area determined by the area determination unit 7034 .

In this embodiment of the present invention, the region determination unit 7034 may be configured to perform step 401 in the above method embodiments, and the second erosion determination unit 7035 may be configured to perform step 402 in the above method embodiments.

Optionally, based on the erosion detection device shown in FIG. 7 , FIG. 8 , FIG. 9 or FIG. 10 , as shown in FIG. 11 , the erosion determination module 703 includes:

A depth value screening unit 7036 for determining a first target depth value and a second target depth value, wherein the first target depth value is the maximum value among the depth values of each target point on the component to be detected, and the second target depth value is is the minimum value among the depth values of each target point on the component to be detected;

an operation unit 7037 for performing a difference operation on the first target depth value determined by the depth value screening unit 7036 and the second target depth value determined by the depth value screening unit 7036 to obtain the target difference value;

A second judging unit 7038 for judging whether the absolute value of the target difference determined by the arithmetic unit 7037 is greater than a preset threshold;

An erosion degree determination unit 7039, configured to determine that the component to be detected is an erosion degree that needs to be replaced when the absolute value of the target difference determined by the second determination unit 7038 is greater than a preset threshold.

In this embodiment of the present invention, the depth value screening unit 7036 may be configured to perform step 501 in the foregoing method embodiments, the computing unit 7037 may be configured to perform step 502 in the foregoing method embodiments, and the second determining unit 7038 may be configured to execute the foregoing method implementations In step 503 in this example, the erosion degree determination unit 7039 may be used to perform

steps

504 and 505 in the above method embodiments.

As shown in FIG. 12, an embodiment of the present invention provides an erosion detection apparatus, including: at least one memory 704 and at least one processor 705;

the at least one memory 704 for storing a machine-readable program;

The at least one processor 705 is configured to invoke the machine-readable program to execute the erosion detection methods provided in the foregoing embodiments.

The present invention also provides a computer-readable medium storing instructions for causing a machine to perform the erosion detection method as described herein. Specifically, it is possible to provide a system or device equipped with a storage medium on which software program codes for implementing the functions of any of the above-described embodiments are stored, and which enables a computer (or CPU or MPU of the system or device) ) to read and execute the program code stored in the storage medium.

In this case, the program code itself read from the storage medium can implement the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code form part of the present invention.

Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (eg CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), Magnetic tapes, non-volatile memory cards and ROMs. Alternatively, the program code may be downloaded from a server computer over a communications network.

In addition, it should be clear that part or all of the actual operations can be implemented not only by executing the program code read out by the computer, but also by the operating system or the like operating on the computer based on the instructions of the program code, so as to realize the above-mentioned embodiments. Function of any one of the embodiments.

In addition, it can be understood that the program code read from the storage medium is written into the memory provided in the expansion board inserted into the computer or into the memory provided in the expansion module connected with the computer, and then based on the program code The instructions cause the CPU or the like installed on the expansion board or expansion module to perform part and all of the actual operations, so as to realize the functions of any one of the above-mentioned embodiments.

It should be noted that not all steps and modules in the above-mentioned processes and system structure diagrams are necessary, and some steps or modules may be omitted according to actual needs. The execution order of each step is not fixed and can be adjusted as required. The system structure described in the above embodiments may be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or may be implemented by multiple physical entities. Some components in separate devices are implemented together.

In the above embodiments, the hardware modules may be implemented mechanically or electrically. For example, a hardware module may include permanent dedicated circuits or logic (eg, dedicated processors, FPGAs or ASICs) to perform corresponding operations. The hardware modules may also include programmable logic or circuits (eg, general-purpose processors or other programmable processors), which may be temporarily set by software to complete corresponding operations. The specific implementation (mechanical, or dedicated permanent circuit, or temporarily provided circuit) can be determined based on cost and time considerations.

The present invention is shown and described in detail above through the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Those skilled in the art can know that the above-mentioned different embodiments can be combined based on the above-mentioned multiple embodiments. More embodiments of the present invention can be obtained by the code review method in the present invention, and these embodiments are also within the protection scope of the present invention.

Claims

Erosion detection methods, including:

acquiring a first detection image and a second detection image of the component to be detected, wherein the first detection image and the second detection image are respectively collected by two cameras included in the binocular camera;

According to the parallax of the first detection image and the second detection image, the depth values of at least two target points on the component to be detected are determined, wherein the depth value of any target point on the component to be detected is used to represent The distance between the target point and the center point of the two lenses included in the binocular camera;

According to the depth value of each of the target points on the component to be inspected, the degree of erosion of the component to be inspected is determined.
The method according to claim 1, wherein the determining the depth values of at least two target points on the component to be detected according to the parallax of the first detection image and the second detection image comprises:

For each of the target points on the component to be detected, execute:

Determine the first pixel point corresponding to the target point in the first detection image, and determine the second pixel point corresponding to the target point in the second detection image;

determining the target parallax between the first pixel point and the second pixel point, wherein the target parallax is used to represent the relative position of the first pixel point in the first detection image and the second pixel point The difference between the relative positions of the pixel points in the second detection image;

The depth value of the target point is determined according to the target parallax, the focal length of the camera in the binocular camera, and the distance between the optical centers of the two cameras in the binocular camera.
The method according to claim 1, wherein the determining the degree of erosion of the part to be inspected according to the depth value of each of the target points on the part to be inspected comprises:

Judging whether the depth value of each target point located at the edge of the component to be detected is within a preset threshold range;

If the depth value of each of the target points on the edge of the component to be detected is within the preset threshold range, determine the depth of the component to be detected according to the depth value of each of the target points on the component to be detected degree of erosion;

If the depth values of the target points located at the edge of the component to be detected are not all within the preset threshold range, execute:

Determine the angle between the imaging plane of the binocular camera and the surface of the component to be detected according to the depth value of each of the target points on the edge of the component to be detected;

calibrating the depth value of each of the target points on the component to be detected according to the included angle;

The degree of erosion of the part to be inspected is determined according to the calibrated depth value of each of the target points on the part to be inspected.
The method according to claim 1, wherein the determining the degree of erosion of the part to be inspected according to the depth value of each of the target points on the part to be inspected comprises:

According to the depth value of each target point and the distribution of each target point on the component to be detected, an effective detection area on the component to be detected is determined, wherein the effective detection area includes at least two the target point, and the depth value of each of the target points in the effective detection area is within the set range;

According to the depth value of each of the target points in the effective detection area, the degree of erosion of the component to be detected is determined.
The method according to any one of claims 1 to 4, wherein the determining the degree of erosion of the part to be inspected according to the depth value of each of the target points on the part to be inspected comprises:

Determine a first target depth value and a second target depth value, wherein the first target depth value is the maximum value among the depth values of each of the target points on the component to be detected, and the second target depth value is the minimum value among the depth values of each of the target points on the component to be detected;

performing a difference operation on the first target depth value and the second target depth value to obtain a target difference value;

judging whether the absolute value of the target difference is greater than a preset threshold;

If the absolute value of the target difference value is greater than the preset threshold value, it is determined that the component to be inspected has an erosion degree that needs to be replaced.
Erosion detection device including:

An acquisition module (701) for acquiring a first detection image and a second detection image of the component to be detected, wherein the first detection image and the second detection image are collected by two cameras included in the binocular camera respectively ;

An image processing module (702) for determining the component to be detected according to the parallax of the first detection image acquired by the acquisition module (701) and the second detection image acquired by the acquisition module The depth values of at least two target points on the above, wherein the depth value of any target point on the component to be detected is used to represent the distance between the target point and the center point of the two lenses included in the binocular camera;

An erosion judging module (703) is configured to determine the degree of erosion of the part to be inspected according to the depth values of the target points on the part to be inspected determined by the image processing module (702).
The apparatus of claim 6, wherein the image processing module (702) comprises:

A pixel point acquisition unit (7021), configured to, for each of the target points on the component to be detected, determine the first pixel point corresponding to the target point in the first detected image, and determine the target point the corresponding second pixel in the second detection image;

A parallax acquisition unit (7022), configured to determine the target parallax between the first pixel point and the second pixel point determined by the pixel point acquisition unit (7021), wherein the target parallax is used to represent The difference between the relative position of the first pixel point in the first detection image and the relative position of the second pixel point in the second detection image;

A depth value acquisition unit (7023), configured to obtain the result determined by the disparity acquisition unit (7022) according to the focal length of the camera in the binocular camera, the distance between the optical centers of the two cameras in the binocular camera, and the parallax acquisition unit (7022). Describe the target disparity, and determine the depth value of the target point.
The apparatus according to claim 6, wherein the erosion determination module (703) comprises:

A first judging unit (7031) for judging whether the depth value of each of the target points located at the edge of the component to be detected is within a preset threshold range;

A first relationship determination unit (7032), configured to, when the first determination unit (7031) determines that the depth value of each of the target points located at the edge of the component to be detected is within the preset threshold range, then Determine the degree of erosion of the component to be detected according to the depth value of each of the target points on the component to be detected;

A second relationship determination unit (7033), used for when the first determination unit (7031) determines that the depth values of each of the target points located at the edge of the component to be detected are not all within the preset threshold range , determine the angle between the imaging plane of the binocular camera and the surface of the component to be detected according to the depth value of each target point on the edge of the component to be detected, and determine the angle between the The depth value of each target point on the detection part is calibrated, and the degree of erosion of the to-be-detected part is determined according to the calibrated depth value of each of the target points on the to-be-detected part.
The apparatus according to claim 6, wherein the erosion determination module (703) comprises:

An area determination unit (7034), configured to determine the effective detection area on the component to be detected according to the depth value of each of the target points and the distribution of each of the target points on the component to be detected, wherein the The effective detection area includes at least two of the target points, and the depth value of each of the target points in the effective detection area is within a set range;

A second erosion determination unit (7035), configured to determine the degree of erosion of the component to be inspected according to the depth values of the target points in the effective inspection area determined by the area determination unit (7034).
The apparatus according to any one of claims 6 to 9, wherein the erosion determination module (703) comprises:

A depth value screening unit (7036), for determining a first target depth value and a second target depth value, wherein the first target depth value is a The maximum value, the second target depth value is the minimum value among the depth values of the target points on the component to be detected;

An operation unit (7037) configured to perform an operation on the first target depth value determined by the depth value screening unit (7036) and the second target depth value determined by the depth value screening unit (7036); Difference operation to obtain the target difference;

a second judging unit (7038) for judging whether the absolute value of the target difference determined by the computing unit (7037) is greater than a preset threshold;

An erosion degree determination unit (7039), configured to determine that the component to be detected is erosion that needs to be replaced when the absolute value of the target difference determined by the second judgment unit (7038) is greater than the preset threshold degree.
an erosion detection apparatus, comprising: at least one memory (704) and at least one processor (705);

the at least one memory (704) for storing a machine-readable program;

The at least one processor (705) is configured to invoke the machine-readable program to execute the method of any one of claims 1-5.
A computer-readable medium having computer instructions stored thereon, the computer instructions, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 5.