Intelligent control method and system of micropore inner wall defect detection equipment

Through intelligent control methods, including image preprocessing, hole positioning, path planning, point cloud model construction and attitude adjustment, the accuracy and intelligence of micro pore inner wall detection are solved, and efficient and accurate defect detection is achieved.

CN119086598AInactive Publication Date: 2024-12-06SUZHOU DINGZHEN INTELLIGENT TECH CO LTD
View PDF 0 Cites 6 Cited by

Patent Information

Application Number
CN202411188554.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to meet the high-precision and high-efficiency detection requirements of micro-pore inner walls. Manual visual inspection is time-consuming and labor-intensive and prone to missed detection and misjudgment. The stability and accuracy of automated detection equipment under complex lighting conditions and posture changes are affected. .

Method used

Using intelligent control methods, preprocess and hole positioning by obtaining the image information of the device to be detected, divide the movable area, plan the detection path, build an internal point cloud model, adjust the posture in real time, and adjust the light source according to the image quality evaluation to achieve adaptive Regulation.

Benefits of technology

It improves the accuracy and intelligence of the detection of the inner wall of the micro pore, ensures that the detection equipment can efficiently and accurately identify defects under complex conditions, and reduces missed detection and misjudgment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119086598A_ABST
    Figure CN119086598A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent control method and system for micropore inner wall defect detection equipment. The intelligent control method comprises the following steps: acquiring image information of a to-be-detected device, carrying out image preprocessing, identifying a hole site of a target to-be-detected device according to a preprocessing result, and carrying out hole site positioning to obtain hole site positioning information; dividing a movable area of the target defect detection equipment, planning an activity path of the defect detection equipment, and generating an equipment detection control scheme; acquiring defect detection image information, performing contour feature extraction on the defect detection image information, and constructing an internal point cloud model of the to-be-detected device; real-time inner wall detection is carried out, an internal point cloud model is updated, the posture change of the detection device is analyzed, and the posture of the detection device is adjusted; and carrying out image quality evaluation based on the defect detection image, and judging whether light source adjustment and self-adaptive regulation are needed during inner wall defect detection. And the accuracy and intelligence of micropore inner wall detection are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of inner wall defect detection equipment control, and in particular to an intelligent control method and system for micro-hole inner wall defect detection equipment. Background Art

[0002] In the process of industrial manufacturing and quality control, defect detection of the inner wall of micropores is an important part of ensuring product quality and reliability. Micropores are usually found in fields such as precision machinery, aerospace, and medical equipment. The quality of the inner wall of these holes directly affects the performance and life of the product. However, due to the extremely small size of micropores and the fact that the inner wall is usually difficult to observe directly, traditional detection methods are difficult to meet the requirements of high-precision and high-efficiency detection. Manual visual inspection is not only time-consuming and labor-intensive, but also limited by the experience and skill level of the inspectors, and is prone to missed detection and misjudgment.

[0003] In recent years, with the development of automation and intelligent technology, vision-based automatic detection equipment has gradually become the main means of detecting defects on the inner wall of micro-holes. These devices are usually equipped with high-precision cameras and light sources. They insert the camera into the micro-hole to capture the inner wall image, and then use image processing algorithms to identify defects on the inner wall, such as cracks, holes, and wear. However, the inner wall surface of the micro-hole is complex, the lighting conditions vary greatly, and the detection equipment may experience posture changes and image distortion during the detection process, which affects the stability and accuracy of the detection.

[0004] Therefore, how to improve the accuracy and intelligence of micropore inner wall detection is an urgent problem to be solved. Summary of the invention

[0005] The present invention overcomes the defects of the prior art and provides an intelligent control method and system for micro-hole inner wall defect detection equipment, the important purpose of which is to improve the accuracy and intelligence of micro-hole inner wall detection.

[0006] To achieve the above-mentioned purpose, the first aspect of the present invention provides an intelligent control method for a micro-hole inner wall defect detection device, comprising:

[0007] Acquire image information of the device to be inspected and perform image preprocessing, identify the hole position of the target device to be inspected according to the preprocessing result, perform hole position positioning, and obtain hole position positioning information;

[0008] Divide the movable area of ​​the target defect detection equipment, plan the defect detection equipment activity path according to the hole position positioning information, and generate an equipment detection control plan;

[0009] Perform defect detection based on the equipment detection control scheme to obtain defect detection image information, extract contour features from the defect detection image information, and construct an internal point cloud model of the device to be detected;

[0010] Performing real-time inner wall detection on the target device to be detected and updating the internal point cloud model, analyzing the posture change of the detection device during the detection process, and adjusting the posture of the detection device;

[0011] Image quality evaluation is performed based on the defect detection image to determine whether light source adjustment is required when performing inner wall defect detection and to perform adaptive control.

[0012] In this solution, the acquisition of image information of the device to be detected and image preprocessing, identification of the hole position of the target device to be detected according to the preprocessing result, hole position positioning, and obtaining hole position positioning information specifically include:

[0013] Acquire image information of the device to be detected, use a Gaussian filter to smooth the image information of the device to be detected, and perform grayscale processing to obtain preprocessed image information;

[0014] Calculating a grayscale histogram according to the preprocessed image information and performing normalization processing, and obtaining an average grayscale of the target image according to the calculated grayscale histogram;

[0015] According to the average grayscale of the target image, the inter-class variance of the target image is calculated by the maximum inter-class variance method, and the maximum value of the inter-class variance is obtained by cyclic iteration;

[0016] Using the Canny operator to perform edge detection, taking the maximum inter-class variance as the high threshold of the Canny operator, and half of the maximum inter-class variance as the low threshold of the Canny operator to perform edge detection on the pre-processed image information, to obtain edge detection information;

[0017] According to the edge detection information, the least square method is used to perform circle fitting to obtain an initial fitting circle, the edge points at the whole pixel level are obtained through the edge detection information to perform sub-pixel edge optimization, and the hole position is located according to the optimization result to obtain the hole position positioning information.

[0018] In this solution, the movable area of ​​the target defect detection equipment is divided, the defect detection equipment activity path is planned according to the hole position positioning information, and the equipment detection control plan is generated, which specifically includes:

[0019] Acquire hole location information, perform feature extraction on the hole location information, acquire shape features of target holes and the center point of the target holes, and obtain hole feature information;

[0020] Acquire the movable area information of the detection device, rasterize the movable area of ​​the detection device in combination with the hole feature information, and rasterize the movable area of ​​the target detection device according to the hole shape feature of the target device to be detected;

[0021] A reference coordinate system is constructed based on the gridded target detection device's active area, and the target detection device's real-time position is defined as the initial starting point, and the hole position to be detected is defined as the target end point, to optimize the active path;

[0022] The fast random tree expansion algorithm is used for path planning. Random sampling is performed according to the set initial starting point to obtain a new node. The step size is preset, and the next node is expanded in the direction of the new node. The RTT tree is updated according to the next node and iterative optimization is performed to obtain the activity path planning information.

[0023] The initial posture information of the target detection device is obtained, the expected posture is calculated according to the initial posture information combined with the hole feature information, and the device detection control scheme is generated in combination with the activity path planning information to control the target detection device.

[0024] In this solution, the defect detection based on the equipment detection control solution is performed to obtain defect detection image information, contour feature extraction is performed on the defect detection image information, and an internal point cloud model of the device to be detected is constructed, specifically including:

[0025] Controlling the target detection device to perform defect detection on the device to be detected according to the device detection control scheme to obtain defect detection image information, and performing image preprocessing on the defect detection image information;

[0026] According to the preprocessed defect detection image information, contour features are extracted, an image coordinate system is constructed and the image center point is set, the grayscale maximum value point in the image is calculated and the corresponding image coordinate value is obtained;

[0027] The angle search method is used to obtain other maximum gray value points to obtain several maximum gray value points, and the maximum gray value points are screened. According to the screening results, the corresponding image coordinate values ​​are obtained to form the contour features of the target device to be detected, and the contour feature information is obtained;

[0028] A three-dimensional coordinate transformation is performed according to the contour feature information to obtain three-dimensional coordinates, and an internal point cloud model of the device to be detected is constructed by combining the obtained three-dimensional coordinates with a three-dimensional modeling technology.

[0029] In this solution, the real-time inner wall detection of the target device to be detected and the updating of the internal point cloud model, the posture change of the detection device during the detection process are analyzed, and the posture adjustment of the detection device are performed, which specifically includes:

[0030] Set reference feature points, perform real-time inner wall detection on the target device to be detected, update the internal point cloud model based on the image sequence formed by the acquired real-time detection images, and extract the position features of the updated reference feature points;

[0031] Perform motion posture analysis of the detection device of the target detection equipment according to the position characteristics of the updated reference feature points, perform motion accumulation according to the initial position characteristics and the updated position characteristics, and form a posture change sequence of the detection device;

[0032] Constructing a posture change prediction model based on a random forest algorithm, and inputting a posture change sequence into the posture change prediction model to predict the posture change of the detection device during the detection process;

[0033] Extract features from the input posture change sequence, build multiple decision trees and randomly select several features for splitting until the stopping condition is met, obtain the posture change prediction results of each decision tree for integrated voting, and obtain posture change prediction information;

[0034] Calculate the deviation from the initial position feature according to the posture change prediction information, compare the calculated deviation value with the preset threshold, and analyze whether the posture change during the detection process affects the detection quality;

[0035] If the detection quality is affected, an attitude adjustment warning will be issued, the deviation between the real-time position feature and the initial position feature will be calculated as the attitude adjustment parameter, and an attitude adjustment plan will be generated for adjustment.

[0036] In this solution, the image quality evaluation based on the defect detection image is performed to determine whether light source adjustment is required and adaptive control is performed when performing inner wall defect detection, specifically including:

[0037] Acquire defect detection image information of the target device to be inspected, and grayscale the defect detection image information to obtain a grayscale defect detection image;

[0038] The grayscale space is equally divided into grayscale value intervals of the same interval width, all pixel points of the grayscale defect detection image are traversed, and the grayscale value corresponding to each pixel point is matched to the corresponding grayscale value interval according to the grayscale value corresponding to each pixel point;

[0039] Preset the image target value, count the number of pixels in each gray value interval, take the gray value interval with the largest number of pixels as the image feature value of the target defect detection image, and calculate the deviation value from the image target value as the image quality evaluation index;

[0040] A deviation threshold is preset, and the calculated deviation value is compared with the deviation threshold. If the deviation value is less than the deviation threshold, no light source adjustment is required; if the deviation value is less than the deviation threshold, the detection light source is adaptively adjusted;

[0041] The PID control algorithm is used to calculate the detection light source adjustment parameters according to the deviation value, and a control signal is generated to adjust the target detection equipment, and the adjusted defect detection image is obtained, and the image quality evaluation is performed based on the adjusted defect detection image;

[0042] Obtain the image quality evaluation value of the adjusted defect detection image to determine whether it is less than the preset deviation threshold. If it is still less than the preset deviation threshold, continue to adjust the target detection light source according to the image quality evaluation value until the adjustment times are met or the image quality evaluation standard is reached.

[0043] A second aspect of the present invention provides an intelligent control system for a micro-pore inner wall defect detection device, the system comprising: a memory, a processor, the memory containing an intelligent control method program for the micro-pore inner wall defect detection device, the intelligent control method program for the micro-pore inner wall defect detection device being executed by the processor to implement the following steps:

[0044] Acquire image information of the device to be inspected and perform image preprocessing, identify the hole position of the target device to be inspected according to the preprocessing result, perform hole position positioning, and obtain hole position positioning information;

[0045] Divide the movable area of ​​the target defect detection equipment, plan the defect detection equipment activity path according to the hole position positioning information, and generate an equipment detection control plan;

[0046] Perform defect detection based on the equipment detection control scheme to obtain defect detection image information, extract contour features from the defect detection image information, and construct an internal point cloud model of the device to be detected;

[0047] Performing real-time inner wall detection on the target device to be detected and updating the internal point cloud model, analyzing the posture change of the detection device during the detection process, and adjusting the posture of the detection device;

[0048] Image quality evaluation is performed based on the defect detection image to determine whether light source adjustment is required when performing inner wall defect detection and to perform adaptive control.

[0049] The present invention discloses an intelligent control method and system for micro-hole inner wall defect detection equipment, including: obtaining image information of a device to be detected and performing image preprocessing, identifying the hole position of the target device to be detected according to the preprocessing result, performing hole position positioning, and obtaining hole position positioning information; dividing the movable area of ​​the target defect detection device, performing defect detection device activity path planning, and generating an equipment detection control scheme; obtaining defect detection image information, performing contour feature extraction on the defect detection image information, and constructing an internal point cloud model of the device to be detected; performing real-time inner wall detection and updating the internal point cloud model, analyzing the posture change of the detection device, and adjusting the posture of the detection device; performing image quality evaluation based on the defect detection image, judging whether light source adjustment is required when performing inner wall defect detection, and performing adaptive regulation. The accuracy and intelligence of micro-hole inner wall detection are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments or exemplary embodiments of the present invention, the drawings required for use in the embodiments or exemplary descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained according to the drawings without paying creative work.

[0051] Figure 1 A flow chart of an intelligent control method for a micro-hole inner wall defect detection device provided by one embodiment of the present invention;

[0052] Figure 2 A flow chart of a control method in a defect detection process provided by an embodiment of the present invention;

[0053] Figure 3 A block diagram of an intelligent control system for a micro-hole inner wall defect detection device provided by one embodiment of the present invention;

[0054] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0055] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0056] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0057] Figure 1 A flow chart of an intelligent control method for a micro-hole inner wall defect detection device provided by one embodiment of the present invention;

[0058] like Figure 1 As shown, the present invention provides a flow chart of an intelligent control method for a micro-hole inner wall defect detection device, comprising:

[0059] S102, acquiring image information of the device to be inspected and performing image preprocessing, identifying the hole position of the target device to be inspected according to the preprocessing result, performing hole position positioning, and obtaining hole position positioning information;

[0060] S104, dividing the movable area of ​​the target defect detection equipment, planning the defect detection equipment movable path according to the hole position positioning information, and generating an equipment detection control plan;

[0061] S106, performing defect detection based on the equipment detection control solution to obtain defect detection image information, extracting contour features from the defect detection image information, and constructing an internal point cloud model of the device to be detected;

[0062] S108, performing real-time inner wall detection on the target device to be detected and updating the internal point cloud model, analyzing the posture change of the detection device during the detection process, and adjusting the posture of the detection device;

[0063] S110, performing image quality evaluation based on the defect detection image, determining whether light source adjustment is required when performing inner wall defect detection, and performing adaptive control.

[0064] It should be noted that the present invention provides an intelligent control method and system for micro-hole inner wall defect detection equipment, which realizes hole position positioning by identifying the hole position of the target device to be detected. According to the hole position positioning result, the activity path of the defect detection equipment is planned, and the equipment detection control plan is generated to achieve precise control of the defect detection equipment in the initial stage of defect detection. Then, the internal point cloud model of the device to be detected is constructed using the defect detection image obtained during defect detection, and the internal point cloud model is further updated according to the defect detection image obtained in real time, so as to judge the posture change of the detection device during detection, and adjust the posture to avoid collision, while improving the detection quality during defect detection. Finally, based on the image quality evaluation of the detection image, it is analyzed whether the detection light source during defect detection needs to be adjusted, and adaptive regulation is performed to improve the quality and effectiveness of the detection image.

[0065] Further, in a preferred embodiment of the present invention, the step of acquiring image information of the device to be detected and performing image preprocessing, identifying the hole position of the target device to be detected according to the preprocessing result, performing hole position positioning, and obtaining hole position positioning information specifically includes:

[0066] Acquire image information of the device to be detected, use a Gaussian filter to smooth the image information of the device to be detected, and perform grayscale processing to obtain preprocessed image information;

[0067] Calculating a grayscale histogram according to the preprocessed image information and performing normalization processing, and obtaining an average grayscale of the target image according to the calculated grayscale histogram;

[0068] According to the average grayscale of the target image, the inter-class variance of the target image is calculated by the maximum inter-class variance method, and the maximum value of the inter-class variance is obtained by cyclic iteration;

[0069] Using the Canny operator to perform edge detection, taking the maximum inter-class variance as the high threshold of the Canny operator, and half of the maximum inter-class variance as the low threshold of the Canny operator to perform edge detection on the pre-processed image information, to obtain edge detection information;

[0070] According to the edge detection information, the least square method is used to perform circle fitting to obtain an initial fitting circle, the edge points at the whole pixel level are obtained through the edge detection information to perform sub-pixel edge optimization, and the hole position is located according to the optimization result to obtain the hole position positioning information.

[0071] It should be noted that, first, the image information of the device to be detected is obtained, and a Gaussian filter is used for smoothing filtering. The Gaussian filter can effectively reduce unnecessary interference by reducing noise and subtle textures in the image, making the key features more obvious. After the filtering process is completed, grayscale processing is performed to convert the color image into a single-channel grayscale image. Next, the grayscale histogram is calculated based on the preprocessed image and normalized. After obtaining the normalized grayscale histogram, the average grayscale value of the target image is further calculated. In order to accurately segment the target image, the maximum inter-class variance method is used to calculate the inter-class variance of the image. The inter-class variance reflects the degree of difference between different gray levels. Through a cyclic iteration, the maximum value of the inter-class variance is calculated, and then the optimal segmentation threshold is determined. The Canny operator is used for edge detection, and the maximum value of the inter-class variance is used as the high threshold, and half of the maximum value of the inter-class variance is used as the low threshold. The preprocessed image is subjected to edge detection to obtain edge detection information. Finally, the least squares method is used to fit the edge points to a circle based on the edge detection information. The least square method can find the circular contour closest to the edge point through the optimization algorithm. In order to further improve the accuracy of fitting, the detected whole pixel edge points are optimized at the sub-pixel level. The edge features in the image are accurately described through sub-pixel adjustment. The hole position is accurately located according to the optimized results, and the hole position location information is finally obtained.

[0072] Furthermore, in a preferred embodiment of the present invention, the movable area of ​​the target defect detection equipment is divided, the defect detection equipment activity path is planned according to the hole position positioning information, and the equipment detection control scheme is generated, which specifically includes:

[0073] Acquire hole location information, perform feature extraction on the hole location information, acquire shape features of target holes and the center point of the target holes, and obtain hole feature information;

[0074] Acquire the movable area information of the detection device, rasterize the movable area of ​​the detection device in combination with the hole feature information, and rasterize the movable area of ​​the target detection device according to the hole shape feature of the target device to be detected;

[0075] A reference coordinate system is constructed based on the gridded target detection device's active area, and the target detection device's real-time position is defined as the initial starting point, and the hole position to be detected is defined as the target end point, to optimize the active path;

[0076] The fast random tree expansion algorithm is used for path planning. Random sampling is performed according to the set initial starting point to obtain a new node. The step size is preset, and the next node is expanded in the direction of the new node. The RTT tree is updated according to the next node and iterative optimization is performed to obtain the activity path planning information.

[0077] The initial posture information of the target detection device is obtained, the expected posture is calculated according to the initial posture information combined with the hole feature information, and the device detection control scheme is generated in combination with the activity path planning information to control the target detection device.

[0078] It should be noted that, first, feature extraction is performed on the positioning information to obtain the specific shape features of the target hole, including the circularity, diameter, center point, etc. of the hole, and obtain the hole feature information. Next, combined with the hole feature information, the active area of ​​the detection equipment is rasterized, and the continuous space is divided into discrete grids, which helps to perform accurate path planning in complex environments. The active area of ​​the target detection device is divided into corresponding units according to the hole shape features of the target device to be detected, so as to facilitate subsequent motion control and path optimization. After the rasterization is completed, a reference coordinate system is constructed, and the real-time position of the target detection device is defined as the initial starting point, and the center point of the hole to be detected is defined as the target end point. The fast random tree expansion algorithm (RRT) is used for path planning to obtain active path planning information. After the path planning is completed, the initial posture information of the target detection device is obtained, and the hole feature information is used to calculate the expected posture to determine the optimal posture of the device at the beginning of the detection process, so as to facilitate high-precision defect detection. Finally, an equipment detection control plan is generated to control the target detection equipment in real time, ensuring that the equipment can reach the hole position of the device to be detected according to the optimal path and expected posture, completing the initial preparation work for defect detection.

[0079] Further, in a preferred embodiment of the present invention, the defect detection based on the equipment detection control scheme is performed to obtain defect detection image information, contour feature extraction is performed on the defect detection image information, and an internal point cloud model of the device to be detected is constructed, which specifically includes:

[0080] Controlling the target detection device to perform defect detection on the device to be detected according to the device detection control scheme to obtain defect detection image information, and performing image preprocessing on the defect detection image information;

[0081] According to the preprocessed defect detection image information, contour features are extracted, an image coordinate system is constructed and the image center point is set, the grayscale maximum value point in the image is calculated and the corresponding image coordinate value is obtained;

[0082] The angle search method is used to obtain other maximum gray value points to obtain several maximum gray value points, and the maximum gray value points are screened. According to the screening results, the corresponding image coordinate values ​​are obtained to form the contour features of the target device to be detected, and the contour feature information is obtained;

[0083] A three-dimensional coordinate transformation is performed according to the contour feature information to obtain three-dimensional coordinates, and an internal point cloud model of the device to be detected is constructed by combining the obtained three-dimensional coordinates with a three-dimensional modeling technology.

[0084] It should be noted that, in the process of defect detection, the posture control of the defect detection device is particularly important, that is, the control of the detection camera. If the detection device performs inner wall defect detection in a bad posture, it will not only lead to low quality of the detection image, but also may cause the detection device to contact the inner wall and cause damage. Construct an image coordinate system, calculate the grayscale maximum point in the image and obtain the corresponding image coordinate value according to the detection image obtained during detection, use the angle search method to perform a comprehensive grayscale value search, obtain several maximum grayscale value points, and screen these points to eliminate noise points or irrelevant points, and the corresponding image coordinate values, which represent the contour features of the target device to be detected, and finally form complete contour feature information. Among them, when searching for the maximum grayscale value, if there are three consecutive maximum grayscale values, the coordinate value of the fourth pixel point is taken as the coordinate of the maximum grayscale value point; if there are four consecutive maximum grayscale values, the coordinate average of the third and fourth two pixel points is taken as the coordinate of the maximum grayscale value point. Perform three-dimensional coordinate conversion according to the acquired contour feature information, and convert the two-dimensional coordinates in the image into three-dimensional coordinates in real space. Using 3D modeling technology and combining the acquired 3D coordinate data, an internal point cloud model of the device to be tested is constructed. The point cloud model is a geometric representation method based on 3D coordinates. It accurately describes the 3D structure of an object through a large number of points, providing a basis for subsequent posture change analysis.

[0085] Furthermore, in a preferred embodiment of the present invention, the real-time inner wall detection of the target device to be detected and the updating of the internal point cloud model, the posture change of the detection device during the detection process, and the posture adjustment of the detection device specifically include:

[0086] Set reference feature points, perform real-time inner wall detection on the target device to be detected, update the internal point cloud model based on the image sequence formed by the acquired real-time detection images, and extract the position features of the updated reference feature points;

[0087] Perform motion posture analysis of the detection device of the target detection equipment according to the position characteristics of the updated reference feature points, perform motion accumulation according to the initial position characteristics and the updated position characteristics, and form a posture change sequence of the detection device;

[0088] Constructing a posture change prediction model based on a random forest algorithm, and inputting a posture change sequence into the posture change prediction model to predict the posture change of the detection device during the detection process;

[0089] Extract features from the input posture change sequence, build multiple decision trees and randomly select several features for splitting until the stopping condition is met, obtain the posture change prediction results of each decision tree for integrated voting, and obtain posture change prediction information;

[0090] Calculate the deviation from the initial position feature according to the posture change prediction information, compare the calculated deviation value with the preset threshold, and analyze whether the posture change during the detection process affects the detection quality;

[0091] If the detection quality is affected, an attitude adjustment warning will be issued, the deviation between the real-time position feature and the initial position feature will be calculated as the attitude adjustment parameter, and an attitude adjustment plan will be generated for adjustment.

[0092] It should be noted that, first, the reference feature points are set as a reference for subsequent posture analysis. In the process of real-time inner wall detection of the target device to be detected, an image sequence is formed based on the acquired real-time detection image, and the internal point cloud model is dynamically updated using these image sequences. The position features of the reference feature points are extracted through the updated point cloud model. These position features reflect the movement and change of the detection device during the detection process. Next, the initial position features of the reference feature points are compared with the updated position features, and the cumulative amount of movement is calculated to form a posture change sequence of the detection device to characterize the posture change of the detection device during the detection process. A posture change prediction model is constructed based on the random forest algorithm. The posture change sequence is input into the prediction model to predict the future posture change of the detection device during the detection process, and the posture change prediction information is obtained. The deviation between the predicted position and the initial position at different times is calculated according to the posture change prediction information, and compared with the preset threshold value to determine whether the posture change during the detection process will affect the detection quality. If the analysis results show that the posture change will affect the quality of the detection image, the posture adjustment warning mechanism will be triggered. At this time, the deviation between the real-time position feature and the initial position feature is calculated as the parameter for posture adjustment, and a specific posture adjustment plan is generated based on these parameters to ensure that the detection device can complete the detection task in the optimal posture.

[0093] Furthermore, in a preferred embodiment of the present invention, the image quality evaluation based on the defect detection image is performed to determine whether light source adjustment is required and adaptive control is performed when performing inner wall defect detection, specifically including:

[0094] Acquire defect detection image information of the target device to be inspected, and grayscale the defect detection image information to obtain a grayscale defect detection image;

[0095] The grayscale space is equally divided into grayscale value intervals of the same interval width, all pixel points of the grayscale defect detection image are traversed, and the grayscale value corresponding to each pixel point is matched to the corresponding grayscale value interval according to the grayscale value corresponding to each pixel point;

[0096] Preset the image target value, count the number of pixels in each gray value interval, take the gray value interval with the largest number of pixels as the image feature value of the target defect detection image, and calculate the deviation value from the image target value as the image quality evaluation index;

[0097] A deviation threshold is preset, and the calculated deviation value is compared with the deviation threshold. If the deviation value is less than the deviation threshold, no light source adjustment is required; if the deviation value is less than the deviation threshold, the detection light source is adaptively adjusted;

[0098] The PID control algorithm is used to calculate the detection light source adjustment parameters according to the deviation value, and a control signal is generated to adjust the target detection equipment, and the adjusted defect detection image is obtained, and the image quality evaluation is performed based on the adjusted defect detection image;

[0099] Obtain the image quality evaluation value of the adjusted defect detection image to determine whether it is less than the preset deviation threshold. If it is still less than the preset deviation threshold, continue to adjust the target detection light source according to the image quality evaluation value until the adjustment times are met or the image quality evaluation standard is reached.

[0100] It should be noted that, during the defect detection process, the posture change of the detection device and the intensity of the light source during detection are both important factors affecting the quality of the detection image. Even if the detection device works in the desired posture, if the light source is not in a suitable state, it will still affect the quality of the corresponding detection image. Based on the above conclusions, the adaptive control of the light source during detection should start from the image quality. By grayscale processing the defect detection image, the grayscale space is equally divided, and the entire grayscale range is divided into several grayscale value intervals with the same width. Then, all pixels of the grayscale defect detection image are traversed, and each pixel is matched to the corresponding grayscale value interval according to its grayscale value. Then, the image target value is preset, and the number of pixels in each grayscale value interval is counted. Then, the grayscale value interval with the largest number of pixels is used as the image feature value of the target defect detection image. The deviation value between this image feature value and the preset image target value is calculated as the evaluation index of image quality. Among them, the image feature value represents the grayscale feature of the image under the current illumination, and the image target value represents the grayscale feature of the image under the optimal illumination. By judging the deviation threshold, it is analyzed whether the light source needs to be adjusted. If adjustment is required, the PID control algorithm is used to calculate the adjustment parameters of the detection light source when adjusting the light source. The PID control algorithm is an algorithm commonly used in automatic control systems. It can dynamically adjust the parameters of the light source according to the deviation value, thereby generating a control signal and accurately adjusting the target detection equipment. At the same time, the adjusted image is evaluated again. If the image quality after adjustment still does not meet the expected standard, the detection light source continues to be adjusted until the set adjustment number limit is reached, or until the image quality evaluation meets the preset standard. This ensures that the image quality obtained during the detection process is always in an ideal state, which helps to accurately identify and analyze inner wall defects.

[0101] Figure 2 A flow chart of a control method in a defect detection process provided by an embodiment of the present invention;

[0102] like Figure 2 As shown, the present invention provides a flow chart of a control method in a defect detection process, including:

[0103] S202, acquiring image information of the device to be inspected, identifying the hole position of the target device to be inspected and locating the hole position, and obtaining hole position locating information;

[0104] S204, planning the activity path of the defect detection equipment according to the hole location information, and generating an equipment detection control plan to perform initial detection control;

[0105] S206, performing real-time inner wall detection on the target device to be detected to generate an internal point cloud model and dynamically update it, and analyzing the posture change of the detection device during the detection process;

[0106] S208, judging whether posture abnormality occurs according to the posture change analysis result, and formulating a posture adjustment plan for adjustment;

[0107] S210, performing image quality evaluation based on the real-time defect detection image, determining whether light source adjustment is required, and performing adaptive control.

[0108] It should be noted that when performing defect detection on the device to be detected, the detection hole position of the device to be detected is identified to achieve pre-detection control of the detection equipment, so that the detection equipment is in the desired state before the detection begins. During the detection process, the posture of the detection device in the hole is analyzed and further adjusted to improve the detection quality and avoid collision between the detection device and the inner wall. At the same time, in order to improve the accuracy of subsequent defect identification, image quality evaluation is performed based on the real-time defect detection image to determine whether the image obtained under the current light source intensity meets the expectations. If not, it is adjusted. Improve the detection quality and accuracy of defects on the inner wall of micro-holes.

[0109] Figure 3 An intelligent control system 3 of a micro-pore inner wall defect detection device provided in one embodiment of the present invention comprises: a memory 31 and a processor 32. The memory 31 contains an intelligent control method program of the micro-pore inner wall defect detection device. When the intelligent control method program of the micro-pore inner wall defect detection device is executed by the processor 32, the following steps are implemented:

[0110] Acquire image information of the device to be inspected and perform image preprocessing, identify the hole position of the target device to be inspected according to the preprocessing result, perform hole position positioning, and obtain hole position positioning information;

[0111] Divide the movable area of ​​the target defect detection equipment, plan the defect detection equipment activity path according to the hole position positioning information, and generate an equipment detection control plan;

[0112] Perform defect detection based on the equipment detection control scheme to obtain defect detection image information, extract contour features from the defect detection image information, and construct an internal point cloud model of the device to be detected;

[0113] Performing real-time inner wall detection on the target device to be detected and updating the internal point cloud model, analyzing the posture change of the detection device during the detection process, and adjusting the posture of the detection device;

[0114] Image quality evaluation is performed based on the defect detection image to determine whether light source adjustment is required when performing inner wall defect detection and to perform adaptive control.

[0115] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0116] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0117] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0118] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, etc. Various media that can store program codes.

[0119] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0120] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An intelligent control method for micro-hole inner wall defect detection equipment, characterized in that: include: Acquire image information of the device to be detected and perform image preprocessing, identify the hole position of the target device to be detected according to the preprocessing result, perform hole position positioning, and obtain hole position positioning information; Divide the movable area of ​​the target defect detection equipment, plan the defect detection equipment activity path according to the hole position positioning information, and generate an equipment detection control plan; Perform defect detection based on the equipment detection control scheme to obtain defect detection image information, extract contour features from the defect detection image information, and construct an internal point cloud model of the device to be detected; Performing real-time inner wall detection on the target device to be detected and updating the internal point cloud model, analyzing the posture change of the detection device during the detection process, and adjusting the posture of the detection device; Image quality evaluation is performed based on the defect detection image to determine whether light source adjustment is required when performing inner wall defect detection and to perform adaptive control.

2. The intelligent control method of a micro-hole inner wall defect detection device according to claim 1, characterized in that: The method of acquiring image information of the device to be detected and performing image preprocessing, identifying the hole position of the target device to be detected according to the preprocessing result, performing hole position positioning, and obtaining hole position positioning information specifically includes: Acquire image information of the device to be detected, use a Gaussian filter to smooth the image information of the device to be detected, and perform grayscale processing to obtain preprocessed image information; Calculating a grayscale histogram according to the preprocessed image information and performing normalization processing, and obtaining an average grayscale of the target image according to the calculated grayscale histogram; According to the average grayscale of the target image, the inter-class variance of the target image is calculated by the maximum inter-class variance method, and the maximum value of the inter-class variance is obtained by cyclic iteration; Using the Canny operator to perform edge detection, taking the maximum inter-class variance as the high threshold of the Canny operator, and half of the maximum inter-class variance as the low threshold of the Canny operator to perform edge detection on the pre-processed image information, to obtain edge detection information; According to the edge detection information, the least square method is used to perform circle fitting to obtain an initial fitting circle, the edge points at the whole pixel level are obtained through the edge detection information to perform sub-pixel edge optimization, and the hole position is located according to the optimization result to obtain the hole position positioning information.

3. The intelligent control method of a micro-hole inner wall defect detection device according to claim 1 is characterized in that: The method of dividing the movable area of ​​the target defect detection equipment, planning the defect detection equipment movable path according to the hole position positioning information, and generating the equipment detection control plan specifically includes: Acquire hole location information, perform feature extraction on the hole location information, acquire shape features of target holes and the center point of the target holes, and obtain hole feature information; Acquire the movable area information of the detection device, rasterize the movable area of ​​the detection device in combination with the hole feature information, and rasterize the movable area of ​​the target detection device according to the hole shape feature of the target device to be detected; A reference coordinate system is constructed based on the gridded target detection device's active area, and the target detection device's real-time position is defined as the initial starting point, and the hole position to be detected is defined as the target end point, to optimize the active path; The fast random tree expansion algorithm is used for path planning. Random sampling is performed according to the set initial starting point to obtain new nodes. The step size is preset, and the next node is expanded in the direction of the new node. The RTT tree is updated according to the next node and iterative optimization is performed to obtain the activity path planning information. The initial posture information of the target detection device is obtained, the expected posture is calculated according to the initial posture information combined with the hole feature information, and the device detection control scheme is generated in combination with the activity path planning information to control the target detection device.

4. The intelligent control method of a micro-hole inner wall defect detection device according to claim 1, characterized in that: The defect detection based on the equipment detection control scheme is performed to obtain defect detection image information, contour feature extraction is performed on the defect detection image information, and an internal point cloud model of the device to be detected is constructed, specifically including: Controlling the target detection device to perform defect detection on the device to be detected according to the device detection control scheme to obtain defect detection image information, and performing image preprocessing on the defect detection image information; According to the preprocessed defect detection image information, contour features are extracted, an image coordinate system is constructed and the image center point is set, the grayscale maximum value point in the image is calculated and the corresponding image coordinate value is obtained; The angle search method is used to obtain other maximum gray value points to obtain several maximum gray value points, and the maximum gray value points are screened. The corresponding image coordinate values ​​are obtained according to the screening results to form the contour features of the target device to be detected, and the contour feature information is obtained; A three-dimensional coordinate transformation is performed according to the contour feature information to obtain three-dimensional coordinates, and an internal point cloud model of the device to be detected is constructed by combining the obtained three-dimensional coordinates with a three-dimensional modeling technology.

5. The intelligent control method of a micro-hole inner wall defect detection device according to claim 1, characterized in that: The real-time inner wall detection of the target device to be detected and the updating of the internal point cloud model, the posture change of the detection device during the detection process, and the posture adjustment of the detection device specifically include: Set reference feature points, perform real-time inner wall detection on the target device to be detected, update the internal point cloud model based on the image sequence formed by the acquired real-time detection images, and extract the position features of the updated reference feature points; Perform motion posture analysis of the detection device of the target detection equipment according to the position characteristics of the updated reference feature points, perform motion accumulation according to the initial position characteristics and the updated position characteristics, and form a posture change sequence of the detection device; Constructing a posture change prediction model based on a random forest algorithm, and inputting a posture change sequence into the posture change prediction model to predict the posture change of the detection device during the detection process; Extract features from the input posture change sequence, build multiple decision trees and randomly select several features for splitting until the stopping condition is met, obtain the posture change prediction results of each decision tree for integrated voting, and obtain posture change prediction information; Calculate the deviation from the initial position feature according to the posture change prediction information, compare the calculated deviation value with the preset threshold, and analyze whether the posture change during the detection process affects the detection quality; If the detection quality is affected, an attitude adjustment warning will be issued, the deviation between the real-time position feature and the initial position feature will be calculated as the attitude adjustment parameter, and an attitude adjustment plan will be generated for adjustment.

6. The intelligent control method of a micro-hole inner wall defect detection device according to claim 1, characterized in that: The image quality evaluation based on the defect detection image is performed to determine whether light source adjustment is required and to perform adaptive control when performing inner wall defect detection, specifically including: Acquire defect detection image information of the target device to be inspected, and grayscale the defect detection image information to obtain a grayscale defect detection image; The grayscale space is equally divided into grayscale value intervals of the same interval width, all pixel points of the grayscale defect detection image are traversed, and the grayscale value corresponding to each pixel point is matched to the corresponding grayscale value interval according to the grayscale value corresponding to each pixel point; Preset the image target value, count the number of pixels in each gray value interval, take the gray value interval with the largest number of pixels as the image feature value of the target defect detection image, and calculate the deviation value from the image target value as the image quality evaluation index; A deviation threshold is preset, and the calculated deviation value is compared with the deviation threshold. If the deviation value is less than the deviation threshold, no light source adjustment is required; if the deviation value is less than the deviation threshold, the detection light source is adaptively adjusted; The PID control algorithm is used to calculate the detection light source adjustment parameters according to the deviation value, and a control signal is generated to adjust the target detection equipment, and the adjusted defect detection image is obtained, and the image quality evaluation is performed based on the adjusted defect detection image; Obtain the image quality evaluation value of the adjusted defect detection image to determine whether it is less than the preset deviation threshold. If it is still less than the preset deviation threshold, continue to adjust the target detection light source according to the image quality evaluation value until the adjustment times are met or the image quality evaluation standard is reached.

7. An intelligent control system for micro-hole inner wall defect detection equipment, characterized in that: The system includes: a memory and a processor, wherein the memory contains an intelligent control method program of a micro-pore inner wall defect detection device, and when the intelligent control method program of the micro-pore inner wall defect detection device is executed by the processor, the following steps are implemented: Acquire image information of the device to be detected and perform image preprocessing, identify the hole position of the target device to be detected according to the preprocessing result, perform hole position positioning, and obtain hole position positioning information; Divide the movable area of ​​the target defect detection equipment, plan the defect detection equipment activity path according to the hole position positioning information, and generate an equipment detection control plan; Perform defect detection based on the equipment detection control scheme to obtain defect detection image information, extract contour features from the defect detection image information, and construct an internal point cloud model of the device to be detected; Performing real-time inner wall detection on the target device to be detected and updating the internal point cloud model, analyzing the posture change of the detection device during the detection process, and adjusting the posture of the detection device; Image quality evaluation is performed based on the defect detection image to determine whether light source adjustment is required when performing inner wall defect detection and to perform adaptive control.

Citation Information

Cited By

  • Valve body tapping calibration method and system based on machine vision

    CN119658035A

  • Shield tunneling machine tool bit defect detection system and method

    CN119936039A

  • Hydraulic support oil cylinder barrel inner hole defect detection method and system based on image processing

    CN120070410A

  • A hydraulic support oil cylinder barrel hole defect detection method and system based on image processing

    CN120070410B

  • Gallery defect detection method and related system

    CN120314319A