High-similarity aircraft connecting piece visual identification device and method

By setting standard sampling locations and establishing feature databases, using multiple light sources and camera image acquisition templates for comparison, the problem of high-similar aircraft connector recognition is solved, efficient and accurate identification of aircraft connectors is achieved, and the identification accuracy is improved and efficiency.

CN118941825AActive Publication Date: 2024-11-12AVIC XIAN AIRCRAFT IND GRP CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410984016.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12
Estimated Expiration
2044-07-22

AI Technical Summary

Technical Problem

The prior art is difficult to effectively identify aircraft connectors with high similarity, resulting in low identification accuracy and efficiency, increasing production costs and quality risks.

Method used

By setting the standard sampling position, the minimum external matrix and shape factor of the aircraft connector are obtained, the characteristic database is established, and image templates are collected using multiple light sources and cameras, the shape factor and image templates are compared, and the aircraft connector to be identified is adjusted to Standard locations are identified.

Benefits of technology

It realizes efficient and accurate identification of aircraft connectors, improves identification accuracy and efficiency, reduces production costs, and reduces labor intensity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941825A_ABST
    Figure CN118941825A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of aircraft connecting piece visual identification, and relates to a high-similarity aircraft connecting piece visual identification device and method.The method comprises the steps that standard aircraft connecting piece feature data are collected by controlling an overlook camera and a left-view camera, and an aircraft connecting piece feature database is established; the method comprises the following steps: extracting a shape factor of a backlight image of a to-be-identified airplane connecting piece, and comparing the shape factor with a standard shape factor to realize primary identification of the airplane connecting piece; the method comprises the following steps: extracting a front light image and a left side front light image of a to-be-recognized aircraft connecting piece to obtain an image template, and calculating the similarity between the image template and a standard image template to realize secondary recognition of the aircraft connecting piece; an image template is obtained by obtaining a front light image and a left side front light image when the to-be-recognized aircraft connecting piece is located at a standard sampling position, and local deviation comparison is performed on the image template and an image template in an aircraft connecting piece feature database, so that three-level recognition of the aircraft connecting piece is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of visual recognition of aircraft connectors, and in particular, relates to a device and method for visual recognition of high-similarity aircraft connectors. Background Art

[0002] Aircraft connectors are an indispensable and important part in aircraft research and development and production. They are mainly used to connect different parts during the assembly of aircraft components, and play a role in transmitting and distributing loads between different parts. At present, aircraft connector identification is mainly carried out by manual identification. Affected by the large quantity, many types and high similarity of aircraft connectors, the manual identification method will make it difficult to ensure the accuracy of aircraft connector identification, which will bring quality risks to the subsequent assembly of aircraft components; on the other hand, it will cause problems such as low work efficiency and high labor intensity, increase the production cost of the enterprise, and be detrimental to the physical and mental health of workers. Therefore, based on machine vision technology and automation technology, it is of great significance to realize the visual identification device and method of aircraft connectors for efficient and accurate identification of high-similarity aircraft connectors.

[0003] In the prior art, machine vision technology has been widely used in parts recognition scenarios, but there is no visual recognition application scenario for aircraft connectors. The high similarity of aircraft connectors is reflected in the symmetry of the structure and the similarity of the shape on the one hand; on the other hand, it is reflected in the slight differences in dimensional features (length, width, height and angle, etc.), which poses a huge challenge to the accurate recognition of aircraft connectors. An existing cross-granular sheet metal parts recognition system and method based on machine vision technology aims to distinguish similar parts of aircraft sheet metal parts. Based on machine vision technology, a visual recognition system consisting of a top-down camera, a side-view camera and a line laser projector is constructed to obtain the shape factors and contour features of sheet metal parts for identification and classification. The above patent does not take into account the differences in multiple orientation images of parts and the slight differences in dimensional features, and does not consider the impact of differences in part placement on the recognition results of similar parts, so it cannot be used for accurate identification of aircraft connectors.

[0004] Therefore, how to more effectively identify aircraft connectors is a problem that needs to be solved. Summary of the invention

[0005] The purpose of the present application is to provide a high-similarity aircraft connection parts visual recognition device and method to solve the problem that the existing aircraft connection parts recognition is difficult to effectively identify similar parts.

[0006] The technical solution of the present application is: a method for visually identifying high-similarity aircraft connectors, comprising:

[0007] Setting standard sampling positions of different standard aircraft connectors, first placing any standard aircraft connector on a placement table, obtaining the minimum circumscribed matrix of the standard aircraft connector, and then adjusting the position of the corresponding standard aircraft connector according to the deviation between the position information of the corresponding minimum circumscribed matrix and the standard sampling position, until the standard aircraft connector is adjusted to the standard sampling position;

[0008] Establish an aircraft connector feature database, control the bottom parallel surface light source, the right parallel surface light source, the strip light source and the high-angle ring light source for exposure imaging, obtain the shape factor and image template of the standard aircraft connector, and store them in the aircraft connector feature database; then place other standard aircraft connectors on the placement table, obtain the shape factors and image templates of different standard aircraft connectors, and store them in the aircraft connector feature database;

[0009] The aircraft connector to be identified is placed on the placement table. The light source controller is used to control the bottom parallel surface light source for exposure imaging to obtain the minimum external matrix. Then the XYR motion platform is adjusted by the minimum external matrix to adjust the backlight image of the aircraft connector to be identified to the center of the field of view.

[0010] The light source controller then controls the right parallel surface light source to expose and image, and obtains the shape factor and image template of the backlight image of the aircraft connector to be identified by shooting with the overhead camera and the left camera, and then compares them with the shape factor and image template of the standard aircraft connector in the aircraft connector feature database. When the deviation of the shape factor and the image template are both less than the set value, the identification is judged to be successful; otherwise, it is judged that the aircraft connector does not exist; by repeatedly obtaining the shape factors and image templates of the backlight images of different aircraft connectors to be identified, and performing comparison and identification, the identification of all aircraft connectors to be identified is completed.

[0011] Preferably, the specific method of adjusting the standard aircraft connector to the standard sampling position is:

[0012] The standard aircraft connector is placed in the rectangular frame marked on the placement table. The control center controller issues instructions to the visual processor to control the overhead camera to capture the backlight image of the standard aircraft connector. The minimum external matrix of the standard aircraft connector is obtained by extracting the data of the outer contour of the backlight image. The position information of the minimum external matrix corresponding to the standard aircraft connector is calculated, including the center coordinates (x i ,y i ) and attitude angle value θ i ;

[0013] The position coordinates of the corresponding standard aircraft connector when placed at the exact center of the XYR motion platform are set as the standard sampling position; the control center controller sends instructions to the motion controller 12 to control the operation of the XYR motion platform, and calculates the deviation (Δx i ,Δy i ) and Δθ i , according to the deviation (Δx i ,Δy i ) and Δθ i The backlit image of the standard aircraft connector is adjusted to the exact center of the field of view, and its outer contour is a regular rectangular contour.

[0014] Preferably, the specific method of adjusting the backlight image of the standard aircraft connector to the center of the field of view is: controlling the XYR motion platform to move a distance Δx along the X axis i , the movement distance along the Y axis is Δy i , the distance of movement along the R axis is Δθ i , and obtain the standard sampling position (x i ′,y i ′) and θ i ′, where the formula is as follows:

[0015]

[0016] Preferably, each standard aircraft connection component is numbered, and after the minimum circumscribed matrix of the aircraft connection component is obtained, the minimum circumscribed matrix of each identification standard aircraft connection component is stored separately by number information.

[0017] Preferably, the specific method for storing the shape factor or image template of the standard aircraft connector is:

[0018] The control center controller issues instructions to the light source controller and the visual processor respectively, wherein the light source controller controls the bottom parallel surface light source and the right parallel surface light source to expose and image respectively; the visual processor controls the top-view camera to obtain the bottom backlight image of the standard aircraft connector, and controls the left-view camera to obtain the left backlight image of the standard aircraft connector, thereby obtaining the backlight images of the standard aircraft connector in two directions;

[0019] The shape factors of the backlit images of the standard aircraft connectors in two directions are extracted respectively and stored in the aircraft connector feature database; the shape factors are the aircraft connector features, including length, width, perimeter, area, aspect ratio and circularity, which constitute the aircraft connector feature set L j =(L j1 ,L j2 ,L j3 ,…,L j6), j represents the backlight image of the aircraft connector at the jth (j=1,2) position, j=1 represents the backlight image of the aircraft connector obtained by the top-view camera, and j=2 represents the backlight image of the standard aircraft connector obtained by the left-view camera;

[0020] Calibrate the directions of four groups of strip light sources and the high-angle ring light source at the left camera; the light source controller controls the four groups of strip light sources to expose and image respectively, and the visual processor controls the top-view camera to obtain four strip front light images in four directions of the standard aircraft connector; the light source controller controls the high-angle ring light source to expose and image, and the visual processor controls the left camera to obtain the front light image on the left side of the aircraft connector;

[0021] After performing highlight area elimination and image enhancement processing on the four strip frontal light images of the standard aircraft connector, the processed four frontal light images are synthesized to obtain a strip frontal light image of the standard aircraft connector; after performing highlight area elimination and image enhancement processing on the left side frontal light image of the aircraft connector, the left side frontal light image of the standard aircraft connector is obtained; the standard aircraft connector frontal light image and the left side frontal light image are used as image templates and stored in a specific folder;

[0022] Place various standard aircraft connectors on the placement table in turn, obtain the shape factors and image templates of various standard aircraft connectors, store them in the aircraft connector feature database, and establish a database of all aircraft connector features i represents the i-th standard aircraft connector, j represents the j-th backlit image of the standard aircraft connector, L jk(k=1,2,3,…,6) represents the shape factor of the backlit image of the jth orientation of the standard aircraft connector, δ j Represents the j-th orientation image template of a standard aircraft connector.

[0023] Preferably, the method of adjusting the backlit image of the aircraft connection part to be identified to the center of the field of view is:

[0024] The aircraft connector to be identified is placed on the placement table, and then the control center controller issues corresponding control instructions to the light source controller, visual processor and motion controller respectively;

[0025] The light source controller controls the bottom parallel surface light source to expose and image, and the visual processor controls the top-down camera to obtain the bottom backlight image of the aircraft connector to be identified. The minimum circumscribed matrix of the image outer contour is extracted through the backlight image of the aircraft connector to be identified obtained by the top-down camera, and the center coordinates (x i ′,y i ′) and attitude angle value θ i ′;

[0026] Calculate the deviation (Δx) between the center coordinates of the minimum circumscribed matrix of the aircraft connector to be identified, the attitude angle value and the standard sampling position i ′,Δy i ′) and Δθ i ′, control the XYR motion platform to move along the X axis as Δx i ′, movement along the Y axis is Δy i ′, the movement along the R axis is Δθ i ′, adjust the backlight image of the aircraft connector to be identified to be located in the center of the field of view, and its outer contour is a regular rectangular contour.

[0027] Preferably, the method for comparing the shape factor and image template of the backlit image of the aircraft connection part to be identified with the shape factor and image template of the standard aircraft connection part is:

[0028] The light source controller controls the right parallel surface light source to expose and image; the visual processor controls the downward-looking camera and the left-looking camera to obtain the bottom backlight image of the aircraft connector to be identified, as well as the front light image and the left front light image;

[0029] Extract the shape factor of the backlit image of the aircraft connector to be identified obtained by the top-down camera and compare it with the aircraft connector feature database If the deviation of each shape factor is less than the first threshold, the feature recognition is judged to be successful and the next step is entered; otherwise, the system judges that the aircraft connector does not exist;

[0030] Extract the shape factor of the backlit image of the aircraft connector to be identified obtained by the left-view camera, and the aircraft connector feature database If the deviation of each shape factor is less than the first threshold, the feature recognition is judged to be successful and the next step is entered; otherwise, the system judges that the aircraft connector does not exist;

[0031] Extract the left frontal light image of the aircraft connector to be identified, obtain the image template, and compare it with the aircraft connector feature database The image template is compared to calculate the similarity value and the local deviation. If the similarity value is less than the second threshold value and the local deviation is less than the third threshold value, the feature recognition is judged to be successful and the next step is entered. Otherwise, the system judges that the aircraft connector does not exist.

[0032] The system displays a prompt that the aircraft connector has been successfully identified, and outputs the aircraft connector drawing number information at the same time. Then, various aircraft connectors to be identified are placed on the placement table in turn and the shape factors and image templates are identified until accurate identification of various aircraft connectors to be identified is achieved.

[0033] Preferably, the first threshold is 5%, the second threshold is 5%, and the third threshold is 1%.

[0034] As a specific implementation, a high-similarity aircraft connector visual recognition device includes a recognition unit, a light source unit and a control unit; the recognition unit includes a left-view camera and a top-view camera; the control unit includes a light source controller, a visual controller, a center controller, a motion controller and an XYR motion platform; the light source unit includes a bottom parallel surface light source, a high-angle annular light source, a strip light source and a right parallel surface light source;

[0035] The top-view camera is vertically arranged at the center of the XYR motion platform, four groups of strip light sources are arranged and distributed on the end surface of the XYR motion platform at 90° around the top-view camera as the center; the left-view camera is arranged on one side of the XYR motion platform, and the optical axis of the left-view camera is parallel to and higher than the XYR motion platform; the high-angle annular light source is installed on the working port of the left-view camera, the bottom parallel surface light source is arranged on the moving seat of the XYR motion platform, and a placing table is arranged on the bottom parallel surface light source; a rectangular frame is provided on the placing table for placing the aircraft connectors to be identified and the standard aircraft connectors;

[0036] The central controller is electrically connected to the light source controller, the vision processor and the motion controller. The vision processor is electrically connected to the left-view camera and the top-view camera. The right-side parallel surface light source is located on the right side of the XYR motion platform and is parallel to the high-angle ring light source. The light source controller is connected to the high-angle ring light source, the bottom parallel surface light source, the strip light source and the right-side parallel surface light source. The motion controller is electrically connected to the XYR motion platform.

[0037] Preferably, the placement table is made of an acrylic plate, which is a rectangular frame with a size of 120 mm×120 mm, and the size of the acrylic plate is the same as the size of the bottom parallel surface light source; the XYR motion platform consists of an X-axis, a Y-axis and an R-axis.

[0038] The high-similarity aircraft connector visual recognition device and method of the present application collect standard aircraft connector feature data and establish an aircraft connector feature database by controlling a top-view camera and a left-view camera; realize primary recognition of aircraft connectors by extracting shape factors of backlight images of aircraft connectors to be identified and comparing them with shape factors in the aircraft connector feature database; realize secondary recognition of aircraft connectors by extracting front light images and left front light images of aircraft connectors to be identified to obtain image templates and calculating their similarity with image templates in the aircraft connector feature database; realize tertiary recognition of aircraft connectors by acquiring front light images and left front light images of aircraft connectors to be identified when they are in standard sampling positions to obtain image templates and performing local deviation comparison between them and image templates in the aircraft connector feature database. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solution provided by the present application, the following is a brief introduction to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of the present application.

[0040] Figure 1 This is a schematic diagram of the overall structure of the high-similarity aircraft connector visual recognition device of this application;

[0041] Figure 2 This is a schematic diagram of the XYR motion platform structure for this application;

[0042] Figure 3 This is a schematic diagram of the structure of four groups of bar light sources in this application.

[0043] 1. Bottom parallel surface light source; 2. Left-view camera; 3. High-angle ring light source; 4. Aircraft connector to be identified; 5. Placement table; 6. Top-down camera; 7. Strip light source; 8. Right parallel surface light source; 9. Light source controller; 10. Vision processor; 11. Center controller; 12. Motion controller; 13. XYR motion platform; 131. X-axis; 132. Y-axis; 133. R-axis; 701. Rotating shaft; 702. Slide plate; 703. Slide; 704. Sliding shaft. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] A high-similarity aircraft connector visual recognition device, such as Figure 1 As shown, it includes a recognition unit, a light source unit and a control unit. The recognition unit includes a left-view camera 2 and a top-view camera 6 for data acquisition; the control unit includes a light source controller 9, a visual controller, a center controller 11, a motion controller 12 and an XYR motion platform 13 for working parameter control. The light source unit includes a bottom parallel surface light source 1, a high-angle ring light source 3, a strip light source 7 and a right parallel surface light source 8.

[0046] The XYR motion platform 13 is a precision mobile fine-tuning optical slide and rotation platform, which has a rectangular structure and a moving seat is arranged in the middle. The XYR motion platform 13 is an existing product and will not be described in detail.

[0047] The top-view camera 6 is vertically arranged at the center of the XYR motion platform 13, and four groups of strip light sources 7 are arranged and distributed on the end surface of the XYR motion platform 13 at 90° around the top-view camera 6 as the center. The left-view camera 2 is arranged on one side of the XYR motion platform 13, and the optical axis of the left-view camera 2 is parallel to and higher than the XYR motion platform 13, and the top-view camera 6 is arranged directly above the placement table 5. The high-angle ring light source 3 is installed on the working port of the left-view camera 2, and the bottom parallel surface light source 1 is arranged on the moving seat of the XYR motion platform 13. A placement table 5 is arranged on the bottom parallel surface light source 1, and a rectangular frame is opened on the placement table 5 for placing the aircraft connector 4 to be identified and the standard aircraft connector.

[0048] The central controller 11 is electrically connected to the light source controller 9, the visual processor 10 and the motion controller 12. The visual processor 10 is electrically connected to the left-view camera 2 and the top-view camera 6. The right-side parallel surface light source 8 is located on the right side of the XYR motion platform 13 and is parallel to the high-angle ring light source 3. The light source controller 9 is connected to the high-angle ring light source 3, the bottom parallel surface light source 1, the strip light source 7 and the right-side parallel surface light source 8. The motion controller 12 is electrically connected to the XYR motion platform 13.

[0049] The motion controller 12 is used to control the XYR motion platform 13, driving the XYR motion platform 13 to move linearly along the X and Y directions and to rotate in the R direction; the light source controller 9 is used to control four groups of strip light sources 7, the bottom parallel surface light source 1, the right parallel surface light source 8 and the high-angle ring light source 3; the visual processor 10 is used to collect and process the top view image and the left view image of the aircraft connection to be detected; the central controller 11 is used to issue instructions to the light source controller 9, the motion controller 12 and the visual processor 10 to control the top view camera 6, the left view camera 2, the high-angle ring light source 3, the bottom parallel surface light source 1, the right parallel surface light source 8 and the XYR motion platform 13 to obtain the image information of the aircraft connection 4 to be identified in all directions.

[0050] The placement table 5 is made of an acrylic plate, which is a rectangular frame with a size of 120 mm × 120 mm. The size of the acrylic plate is the same as the size of the bottom parallel surface light source 1, forming an aircraft connector placement table 5. The XYR motion platform 13 is composed of an X-axis 131, a Y-axis 132 and an R-axis 133. Figure 2 .

[0051] The side of each group of strip light sources 7 is connected to the slide plate 702 through the rotating shaft 701 and the sliding shaft 704. The strip light source 7 moves in the slide groove 703 of the slide plate 702 through the sliding shaft 704 to adjust the angle of the strip light source 7 in the vertical plane to control the lighting direction. Figure 3 .

[0052] By controlling the movement of the XYR motion platform 13, the aircraft connector can be adjusted in different directions. Multiple groups of different light sources are set for exposure, and multiple controllers are set for exposure and position control, which solves the problem of acquiring images of aircraft connectors in multiple directions and identifying differences in images in multiple directions.

[0053] As a specific implementation, a method for visually identifying high-similarity aircraft connectors is also provided, comprising:

[0054] Step S100, setting standard sampling positions of different standard aircraft connectors, first placing any standard aircraft connector on the placement table 5, obtaining the minimum circumscribed matrix of the standard aircraft connector, and then adjusting the position of the corresponding standard aircraft connector according to the deviation between the position information of the corresponding minimum circumscribed matrix and the standard sampling position until it is adjusted to the standard sampling position.

[0055] Here are the steps:

[0056] Step 110: Place the standard aircraft connector in the rectangular frame marked on the placement table 5, and the control center controller 11 sends instructions to the visual processor 10, thereby controlling the overhead camera 6 to capture the backlight image of the standard aircraft connector, and obtain the minimum circumscribed matrix of the standard aircraft connector by extracting the data of the outer contour of the backlight image, and calculate the position information of the minimum circumscribed matrix corresponding to the standard aircraft connector, including the center coordinates (x i ,y i ) and attitude angle value θ i .

[0057] For the convenience of identification, each standard aircraft connector is numbered. After the minimum circumscribed matrix of the aircraft connector is obtained, the minimum circumscribed matrix of each identification standard aircraft connector is stored separately according to the numbering information.

[0058] Step 120: Set the position coordinates of the corresponding standard aircraft connector when it is placed at the center of the XYR motion platform 13 as the standard sampling position; control the central controller 11 to send instructions to the motion controller 12, thereby controlling the operation of the XYR motion platform 13, and calculating the deviation (Δx i ,Δy i ) and Δθ i , according to the deviation (Δx i ,Δy i ) and Δθ i The backlit image of the standard aircraft connector is adjusted to be located in the center of the field of view, and its outer contour is a regular rectangular contour.

[0059] The specific method of adjusting the backlight image of the standard aircraft connector to the center of the field of view is: controlling the XYR motion platform 13 to move along the X axis 131 by a distance Δx i , the movement distance along the Y axis 132 is Δy i , the movement distance along the R axis 133 is Δθ i , and obtain the standard sampling position (x i ′,y i ′) and θ i ′, where the formula is as follows;

[0060]

[0061] By extracting the backlight image of the standard aircraft connector, obtaining the central coordinates and attitude angle value of its minimum circumscribed matrix, calculating the deviation between it and the coordinate zero point, and controlling the XYR motion platform 13 to adjust the outer contour of the aircraft connector to a regular rectangular contour, the standard sampling position of the aircraft connector is obtained, and then the standard aircraft connector is identified at the standard sampling position, which solves the problem of the influence of the placement difference of aircraft connectors on the recognition results of similar parts and the recognition of slight differences in the size characteristics of aircraft connectors.

[0062] Step S200, establish an aircraft connector feature database, respectively control the bottom parallel surface light source 1, the right parallel surface light source 8, the strip light source 7 and the high-angle ring light source 3 to expose and image, obtain the shape factor and image template of the standard aircraft connector, and store them in the aircraft connector feature database; then place other standard aircraft connectors on the placement table 5 respectively, obtain the shape factor and image template of the corresponding standard aircraft connector, and store them in the aircraft connector feature database.

[0063] Step 210: firstly establish an aircraft connection part feature database according to the serial number information of different standard aircraft connection parts;

[0064] The control center controller 11 issues instructions to the light source controller 9 and the visual processor 10 respectively, wherein the light source controller 9 controls the bottom parallel surface light source 1 and the right parallel surface light source 8 to expose and image respectively;

[0065] The visual processor 10 controls the top-view camera 6 to obtain the backlit image of the bottom of the standard aircraft connector, and controls the left-view camera 2 to obtain the backlit image of the left side of the standard aircraft connector, thereby obtaining the backlit images of the standard aircraft connector in two directions.

[0066] Step 220: extract the shape factors of the backlit images of the standard aircraft connector in two directions respectively, and store them in the aircraft connector feature database; the shape factors are the aircraft connector features, including length, width, perimeter, area, aspect ratio and circularity, which constitute the aircraft connector feature set L j =(L j1 ,L j2,L j3 ,…,L j6 ), j represents the backlight image of the aircraft connector at the jth (j=1,2) orientation, j=1 represents the backlight image of the aircraft connector acquired by the overhead camera 6, and j=2 represents the standard backlight image of the aircraft connector acquired by the left-view camera 2.

[0067] Preferably, the aircraft connection part feature data in the aircraft connection part feature database are all information about the aircraft connection parts being in standard sampling positions; and the aircraft connection part feature database adopts a MySQL database.

[0068] Step 230: Calibrate the directions of the four groups of strip light sources 7, and calibrate the high-angle ring light source 3 at the left-view camera 2; the light source controller 9 controls the four groups of strip light sources 7 to expose and image separately, and the visual processor 10 controls the overhead camera 6 to obtain four strip frontal light images in four directions of the standard aircraft connector; the light source controller 9 controls the high-angle ring light source 3 to expose and image, and the visual processor 10 controls the left-view camera 2 to obtain the frontal light image of the left side of the aircraft connector.

[0069] Step 240: After performing highlight area elimination and image enhancement processing on the four acquired strip frontal light images of the standard aircraft connector, the processed four frontal light images are synthesized to obtain the strip frontal light image of the standard aircraft connector.

[0070] After removing the highlight area and performing image enhancement processing on the acquired left side front light image of the aircraft connector, a standard left side front light image of the aircraft connector is obtained; the standard aircraft connector front light image and the left side front light image are used as image templates and stored in a specific folder.

[0071] At the same time, the address of the image template is stored in the aircraft connector feature database to obtain the final aircraft connector feature set L j ′=(L j1 ,L j2 ,L j3 ,…,L j6 ,δ j ), j represents the backlit image of the aircraft connector at the jth position, L jk(k=1,2,3,…,6) represents the shape factor of the backlit image of the aircraft connector at the jth position, δ j Represents the j-th orientation image template of the aircraft connector.

[0072] Preferably, a number import module is provided on the aircraft connection part feature database, and before visual import of each standard aircraft connection part, the number information of each standard aircraft connection part is pre-imported into the number import module.

[0073] After extracting the shape factor or image template of the backlight image, the numbering information of the corresponding standard aircraft connector is attached to the shape factor data or image template data, and then the shape factor data or image template data is transmitted to the aircraft connector feature database. The aircraft connector feature database determines whether the numbering information attached to the shape factor data or image template data corresponds to the pre-imported numbering information. If so, it is directly stored; if not, the database is updated according to the numbering information attached to the shape factor data or image template data of the standard aircraft connector and the corresponding data is stored.

[0074] Step 250: Place various standard aircraft connectors on the placement table 5 in sequence, obtain the shape factors and image templates of various standard aircraft connectors, store them in the aircraft connector feature database, and establish the entire aircraft connector feature database. i represents the i-th standard aircraft connector, j represents the j-th backlit image of the standard aircraft connector, L jk(k=1,2,3,…,6) represents the shape factor of the backlit image of the jth orientation of the standard aircraft connector, δ j It represents the j-th orientation image template of the standard aircraft connector, so as to realize the rapid matching of the image of the aircraft connector 4 to be identified later.

[0075] Step S300, the aircraft connector 4 to be identified is placed on the placement table 5, and the bottom parallel surface light source 1 is first controlled by the light source controller 9 to perform exposure imaging to obtain a minimum external matrix, and then the XYR motion platform 13 is adjusted through the minimum external matrix to adjust the backlight image of the aircraft connector 4 to be identified to the center of the field of view.

[0076] Here are the steps:

[0077] Step 310: placing the aircraft connector 4 to be identified on the placement table 5, and then the control center controller 11 sends corresponding control instructions to the light source controller 9, the visual processor 10 and the motion controller 12 respectively;

[0078] Step 320: The light source controller 9 controls the bottom parallel surface light source 1 to expose and image, and the visual processor 10 controls the overhead camera 6 to obtain the bottom backlight image of the aircraft connector 4 to be identified. The minimum circumscribed matrix of the outer contour of the image is extracted through the backlight image of the aircraft connector 4 to be identified obtained by the overhead camera 6, and the center coordinates (x i ′,y i ′) and attitude angle value θ i ′;

[0079] Step 330: Calculate the deviation (Δx i ′,Δy i′) and Δθ i ', control the XYR motion platform 13 to move along the X axis 131 to Δx i ', movement along the Y axis 132 is Δy i ', the movement along the R axis 133 is Δθ i ', adjust the backlight image of the aircraft connector 4 to be identified to be located in the center of the field of view, and its outer contour is a regular rectangular contour;

[0080] In step S400, the light source controller 9 controls the right parallel surface light source 8 to expose and image, and obtains the shape factor and image template of the backlit image of the aircraft connector 4 to be identified by shooting with the overhead camera 6 and the left camera 2, and then compares them with the shape factor and image template of the standard aircraft connector in the aircraft connector feature database. When the deviation of the shape factor and the image template are both less than the set value, it is judged that the identification is successful; otherwise, it is judged that the aircraft connector does not exist; by repeatedly obtaining the shape factors and image templates of the backlit images of different aircraft connectors 4 to be identified, and comparing and identifying, the identification of all aircraft connectors 4 to be identified is completed.

[0081] Here are the steps:

[0082] Step 410: the light source controller 9 controls the right parallel surface light source 8 to expose and image; the visual processor 10 controls the top view camera 6 and the left view camera 2 to respectively obtain the bottom backlight image of the aircraft connector 4 to be identified, as well as the front light image and the left front light image;

[0083] Step 420: extract the shape factor of the backlit image of the aircraft connector 4 to be identified acquired by the overhead camera 6 and compare it with the aircraft connector feature database If the deviation of each shape factor is less than 5% (which can be adjusted according to the actual situation), it is determined that the feature recognition is successful and the process goes to step 430. Otherwise, the system determines that the aircraft connector does not exist.

[0084] Step 430: extract the shape factor of the backlit image of the aircraft connector 4 to be identified acquired by the left-view camera 2 and compare it with the aircraft connector feature database established in S100 If the deviation of each shape factor is less than 5% (which can be adjusted according to the actual situation), it is determined that the feature recognition is successful and the process goes to step 440. Otherwise, the system determines that the aircraft connector does not exist.

[0085] Step 440: Extract the front light image of the aircraft connector 4 to be identified, obtain the image template, and compare it with the aircraft connector feature database established in S100 The image template is compared to calculate the similarity value and the local deviation. If the similarity value is less than 5% (which can be adjusted according to the actual situation) and the local deviation is less than 1% (which can be adjusted according to the actual situation), then the process goes to step 450. Otherwise, the system determines that the aircraft connector does not exist.

[0086] Step 450: Extract the left frontal light image of the aircraft connector 4 to be identified, obtain an image template, and compare it with the aircraft connector feature database established in S100 The image template is compared to calculate the similarity value and the local deviation. If the similarity value is less than 5% (which can be adjusted according to the actual situation) and the local deviation is less than 1% (which can be adjusted according to the actual situation), it is determined that the feature recognition is successful and the process goes to step 460. Otherwise, the system determines that the aircraft connector does not exist.

[0087] Step 460: the system displays a prompt indicating that the aircraft connector has been successfully identified, and outputs the aircraft connector drawing number information at the same time, and executes steps S300-S400 in a loop, placing various aircraft connectors 4 to be identified on the placement table 5 in turn and identifying the shape factor and image template, thereby achieving accurate identification of various aircraft connectors 4 to be identified.

[0088] The present application collects standard aircraft connector feature data and establishes an aircraft connector feature database by controlling the overhead camera 6 and the left-view camera 2; realizes the primary recognition of aircraft connectors by extracting the shape factor of the backlight image of the aircraft connector 4 to be identified and comparing it with the shape factor in the aircraft connector feature database; realizes the secondary recognition of aircraft connectors by extracting the front light image and the left front light image of the aircraft connector 4 to be identified, obtaining the image template, and calculating its similarity with the image template in the aircraft connector feature database; realizes the tertiary recognition of aircraft connectors by obtaining the front light image and the left front light image of the aircraft connector 4 to be identified when it is in the standard sampling position, obtaining the image template of the standard sampling position, and performing a local deviation comparison between it and the image template in the aircraft connector feature database. The above method solves the problem of efficient and accurate recognition of aircraft connectors.

[0089] Finally, it should be noted that: the drawings of the embodiments disclosed in the present invention only involve structures related to the embodiments disclosed in the present invention, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0090] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A high-similarity aircraft connector visual recognition method, characterized in that: include: Setting standard sampling positions of different standard aircraft connectors, first placing any standard aircraft connector on a placement table (5), obtaining the minimum circumscribed matrix of the standard aircraft connector, and then adjusting the position of the corresponding standard aircraft connector according to the deviation between the position information of the corresponding minimum circumscribed matrix and the standard sampling position, until the standard aircraft connector is adjusted to the standard sampling position; Establishing an aircraft connector feature database, respectively controlling the bottom parallel surface light source (1), the right parallel surface light source (8), the strip light source (7) and the high-angle ring light source (3) for exposure imaging, obtaining the shape factor and image template of the standard aircraft connector, and storing them in the aircraft connector feature database; Then, other standard aircraft connectors are placed on the placement table (5) respectively, and the shape factors and image templates of different standard aircraft connectors are obtained and stored in the aircraft connector feature database; The aircraft connector (4) to be identified is placed on a placement table (5), and the bottom parallel surface light source (1) is first controlled by a light source controller (9) to perform exposure imaging to obtain a minimum external matrix, and then the XYR motion platform (13) is adjusted by the minimum external matrix to adjust the backlight image of the aircraft connector (4) to be identified to the center of the field of view; The light source controller (9) then controls the right parallel surface light source (8) to expose and image, and obtains the shape factor and image template of the backlight image of the aircraft connector (4) to be identified by shooting with the overhead camera (6) and the left camera (2), and then compares the shape factor and image template with the shape factor and image template of the standard aircraft connector in the aircraft connector feature database. When the deviation of the shape factor and the image template is less than the set value, it is judged that the identification is successful; otherwise, it is judged that the aircraft connector does not exist; by repeatedly obtaining the shape factor and image template of the backlight image of different aircraft connectors (4) to be identified, and performing comparison and identification, the identification of all aircraft connectors (4) to be identified is completed.

2. The high-similarity aircraft connector visual recognition method according to claim 1, characterized in that: The specific method for adjusting the standard aircraft connector to the standard sampling position is: The standard aircraft connector is placed in a rectangular frame marked on the placement table (5), and the control center controller (11) issues instructions to the visual processor (10), thereby controlling the overhead camera (6) to capture a backlit image of the standard aircraft connector, and obtaining the minimum circumscribed matrix of the standard aircraft connector by extracting data of the outer contour of the backlit image, and calculating the position information of the minimum circumscribed matrix corresponding to the standard aircraft connector, including the center coordinates (x i ,y i ) and attitude angle value θ i ; The position coordinates of the corresponding standard aircraft connector when placed at the exact center of the XYR motion platform (13) are set as the standard sampling position; the control center controller (11) sends instructions to the motion controller (12), thereby controlling the operation of the XYR motion platform (13), and calculating the deviation (Δx i ,Δy i ) and Δθ i , according to the deviation (Δx i ,Δy i ) and Δθ i The backlit image of the standard aircraft connector is adjusted to the exact center of the field of view, and its outer contour is a regular rectangular contour.

3. The high-similarity aircraft connector visual recognition method according to claim 2, characterized in that: The specific method of adjusting the backlight image of the standard aircraft connector to the center of the field of view is: controlling the XYR motion platform (13) to move along the X axis (131) by a distance Δx i , the movement distance along the Y axis (132) is Δy i , the movement distance along the R axis (133) is Δθ i , and obtain the standard sampling position (x i ′,y i ′) and θ i ′, where the formula is as follows:

4. The high-similarity aircraft connector visual recognition method according to claim 2, characterized in that: Each standard aircraft connector is numbered, and after the minimum circumscribed matrix of the aircraft connector is obtained, the minimum circumscribed matrix of each identification standard aircraft connector is stored separately according to the numbering information.

5. The high-similarity aircraft connector visual recognition method according to claim 4, characterized in that: The specific method for storing the shape factor or image template of the standard aircraft connector is as follows: The control center controller (11) issues instructions to the light source controller (9) and the visual processor (10) respectively, wherein the light source controller (9) controls the bottom parallel surface light source (1) and the right parallel surface light source (8) to expose and image respectively; the visual processor (10) controls the top view camera (6) to obtain the bottom backlight image of the standard aircraft connecting part, and controls the left view camera (2) to obtain the left backlight image of the standard aircraft connecting part, thereby obtaining the backlight images of the standard aircraft connecting part in two directions; The shape factors of the backlit images of the standard aircraft connectors in two directions are extracted respectively and stored in the aircraft connector feature database; the shape factors are the aircraft connector features, including length, width, perimeter, area, aspect ratio and circularity, which constitute the aircraft connector feature set L j =(L j1 ,L j2 ,L j3 ,…,L j6 ), j represents the backlight image of the aircraft connector at the jth (j=1,2) orientation, j=1 represents the backlight image of the aircraft connector acquired by the top-view camera (6), and j=2 represents the standard backlight image of the aircraft connector acquired by the left-view camera (2); The directions of the four groups of strip light sources (7) are calibrated, and the high-angle ring light source (3) at the left-view camera (2) is calibrated; the light source controller (9) controls the four groups of strip light sources (7) to expose and image respectively, and the visual processor (10) controls the overhead camera (6) to obtain four strip frontal light images in four directions of the standard aircraft connector; the light source controller (9) controls the high-angle ring light source (3) to expose and image, and the visual processor (10) controls the left-view camera (2) to obtain the frontal light image on the left side of the aircraft connector; After performing highlight area elimination and image enhancement processing on the four strip frontal light images of the standard aircraft connector, the processed four frontal light images are synthesized to obtain a strip frontal light image of the standard aircraft connector; after performing highlight area elimination and image enhancement processing on the left side frontal light image of the aircraft connector, the left side frontal light image of the standard aircraft connector is obtained; the standard aircraft connector frontal light image and the left side frontal light image are used as image templates and stored in a specific folder; Various standard aircraft connectors are placed on a placement table (5) in sequence, and shape factors and image templates of various standard aircraft connectors are obtained and stored in an aircraft connector feature database to establish a database of all aircraft connector features. i represents the i-th standard aircraft connector, j represents the j-th backlit image of the standard aircraft connector, L jk(k=1,2,3,…,6) represents the shape factor of the backlit image of the jth orientation of the standard aircraft connector, δ j Represents the j-th orientation image template of a standard aircraft connector.

6. The high-similarity aircraft connector visual recognition method according to claim 5, characterized in that: The method of adjusting the backlit image of the aircraft connector (4) to be identified to the center of the field of view is as follows: The aircraft connector (4) to be identified is placed on a placement table (5), and then the control center controller (11) issues corresponding control instructions to the light source controller (9), the visual processor (10) and the motion controller (12) respectively; The light source controller (9) controls the bottom parallel surface light source (1) to expose and image, and the visual processor (10) controls the overhead camera (6) to obtain the bottom backlight image of the aircraft connector (4) to be identified. The minimum circumscribed matrix of the outer contour of the image is extracted from the backlight image of the aircraft connector (4) to be identified obtained by the overhead camera (6), and the center coordinate (x i ′,y i ′) and attitude angle value θ i ′; Calculate the deviation (Δx) between the minimum circumscribed matrix center coordinates and attitude angle value of the aircraft connector (4) to be identified and the standard sampling position i ′,Δy i ′) and Δθ i ', control the XYR motion platform (13) to move along the X axis (131) to Δx i ', the movement along the Y axis (132) is Δy i ', and the movement along the R axis (133) is Δθ i ', adjust the backlight image of the aircraft connector (4) to be identified to be located at the center of the field of view, and its outer contour is a regular rectangular contour.

7. The high-similarity aircraft connector visual recognition method according to claim 5, characterized in that: The method for comparing the shape factor and image template of the backlit image of the aircraft connection part (4) to be identified with the shape factor and image template of the standard aircraft connection part is as follows: The light source controller (9) controls the right parallel surface light source (8) to expose and image; the visual processor (10) controls the downward-looking camera (6) and the left-looking camera (2) to respectively acquire a backlight image of the bottom of the aircraft connector (4) to be identified, as well as a front light image and a left front light image; Extract the shape factor of the backlit image of the aircraft connector (4) to be identified obtained by the overhead camera (6) and compare it with the aircraft connector feature database If the deviation of each shape factor is less than the first threshold, the feature recognition is judged to be successful and the next step is entered; otherwise, the system judges that the aircraft connector does not exist; Extract the shape factor of the backlit image of the aircraft connection part (4) to be identified obtained by the left-view camera (2), and the aircraft connection part feature database If the deviation of each shape factor is less than the first threshold, the feature recognition is judged to be successful and the next step is entered; otherwise, the system judges that the aircraft connector does not exist; Extract the left frontal light image of the aircraft connector (4) to be identified, obtain the image template, and compare it with the aircraft connector feature database The image template is compared to calculate the similarity value and the local deviation. If the similarity value is less than the second threshold value and the local deviation is less than the third threshold value, the feature recognition is judged to be successful and the next step is entered. Otherwise, the system judges that the aircraft connector does not exist. The system displays a prompt indicating successful aircraft connector identification and outputs aircraft connector drawing number information at the same time, and then various aircraft connectors (4) to be identified are sequentially placed on a placement table (5) and the shape factors and image templates are identified until accurate identification of various aircraft connectors (4) to be identified is achieved.

8. The high-similarity aircraft connector visual recognition method according to claim 7, characterized in that: The first threshold is 5%, the second threshold is 5%, and the third threshold is 1%.

9. A high-similarity aircraft connector visual recognition device, using the method according to any one of claims 1 to 8, characterized in that: It comprises an identification unit, a light source unit and a control unit; the identification unit comprises a left-view camera (2) and a top-view camera (6); the control unit comprises a light source controller (9), a visual controller, a center controller (11), a motion controller (12) and an XYR motion platform (13); the light source unit comprises a bottom parallel surface light source (1), a high-angle annular light source (3), a strip light source (7) and a right parallel surface light source (8); The top-view camera (6) is vertically arranged at the center of the XYR motion platform (13); four groups of strip light sources (7) are arranged and distributed on the end surface of the XYR motion platform (13) at 90° around the top-view camera (6) as the center; the left-view camera (2) is arranged on one side of the XYR motion platform (13); the optical axis of the left-view camera (2) is parallel to and higher than the XYR motion platform (13); the high-angle annular light source (3) is installed on the working port of the left-view camera (2); the bottom parallel surface light source (1) is arranged on the moving seat of the XYR motion platform (13); a placement table (5) is arranged on the bottom parallel surface light source (1); a rectangular frame is opened on the placement table (5) for placing the aircraft connector (4) to be identified and the standard aircraft connector; The central controller (11) is electrically connected to the light source controller (9), the visual processor (10) and the motion controller (12); the visual processor (10) is electrically connected to the left-view camera (2) and the top-view camera (6); the right-side parallel surface light source (8) is located on the right side of the XYR motion platform (13) and is parallel to the high-angle annular light source (3); the light source controller (9) is connected to the high-angle annular light source (3), the bottom parallel surface light source (1), the strip light source (7) and the right-side parallel surface light source (8); and the motion controller (12) is electrically connected to the XYR motion platform (13).

10. The high-similarity aircraft connector visual recognition device according to claim 9, characterized in that: The placement table (5) is made of an acrylic plate, which is a rectangular frame with a size of 120 mm×120 mm. The size of the acrylic plate is the same as the size of the bottom parallel surface light source (1); the XYR motion platform (13) is composed of an X-axis (131), a Y-axis (132) and an R-axis (133).

Citation Information

Patent Citations

  • Safety monitoring analysis system for contact net running state

    CN105652154A

  • Cross-granularity sheet metal part identification system and method based on machine vision technology

    CN110866894A

  • Systems and methods for displaying a pilot display for an aircraft

    US20230348093A1