Multi-degree-of-freedom part image automatic acquisition method and acquisition device
Through the multi-degree-of-freedom movement fixture and camera, combined with attitude data storage and automatic adjustment technology, the problem of automatic image acquisition of parts with multiple surfaces and angles is solved, and high-quality and consistent image acquisition and online detection are achieved.
Patent Information
- Application Number
- CN202411820405.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-11
- Publication Date
- 2025-03-18
AI Technical Summary
The prior art is difficult to realize automatic image acquisition of parts with multiple surfaces and angles, especially in the detection of metal parts with complex structures. The camera and light source configuration of fixed camera positions is difficult to meet the automatic acquisition needs.
The clamp and camera with multiple degrees of freedom are used to adjust the position of the clamp and camera through the driving mechanism, so that the various acquisition surfaces of the parts can achieve the desired effect in the camera's field of view, and the posture data of each node is stored through the control system to achieve the consistency of automatic acquisition and repeated acquisition of part images.
It realizes stable and high-quality automatic image acquisition of each surface of the part, improves the acquisition quality and consistency, meets the multi-angle and multi-surface image acquisition requirements of complex structural parts, and realizes efficient online detection.
Smart Images

Figure CN119653241A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to image acquisition equipment, and in particular to a multi-degree-of-freedom part image automatic acquisition method and acquisition device, which can automatically and accurately acquire multi-surface images of parts, and is particularly suitable for metal parts with complex structures, and belongs to the field of intelligent manufacturing and machine vision technology. Background Art
[0002] In the manufacturing industry, surface defect detection of parts (metal or non-metal) is an important part of ensuring product quality. Traditional manual detection methods are inefficient and rely on experience, which can easily lead to missed detection or false detection. In order to achieve high-quality surface detection, modern manufacturing has gradually applied machine vision for automated detection. However, the multi-surface structure, complex shape characteristics and multi-angle shooting requirements of parts increase the difficulty of image acquisition. In the prior art, the fixed-position camera and light source configuration method is difficult to adapt to the automatic acquisition requirements of multiple angles and multiple surfaces. Therefore, how to provide a collection device with multiple degrees of freedom to accurately adjust the relative position between the part and the camera and improve the image acquisition quality of different surfaces is an urgent problem to be solved and has a strong practical demand. Summary of the invention
[0003] In view of the above shortcomings of the prior art, the purpose of the present invention is to provide a multi-degree-of-freedom part image automatic acquisition method and acquisition device. The present invention can realize stable and high-quality automatic image acquisition of each surface of the part, ensure the consistency of multiple acquisitions, and realize efficient online detection.
[0004] The technical solution of the present invention is achieved in this way: The automatic acquisition method of multi-degree-of-freedom part images is carried out in the following steps: 1) Adjust the camera and the fixture to their initial positions through the driving mechanism respectively; 2) When capturing the image of the first part, the fixture is driven to pick up the first part, and the fixture and the camera are continuously adjusted through the driving mechanism to make the first acquisition surface of the part within the camera field of view to achieve the desired effect; then, the camera takes a picture to complete the image acquisition of the first acquisition surface of the part; 3) Based on the last position of the fixture and the camera, the fixture and the camera are adjusted again through the driving mechanism so that the next acquisition surface of the first part can achieve the desired effect within the camera field of view; the camera takes a photo to complete the image acquisition of the next acquisition surface of the first part; 4) Repeat step 3) until all the image acquisition of the first part's intended acquisition surface is completed; 5) Drive the fixture to put the parts that have completed image acquisition back to the preset position; 6) The control system stores the posture data corresponding to each node of the camera and the fixture when the first part is imaged; the camera node posture data includes the posture data of the movement into place when the image is collected on each acquisition surface; the fixture node posture data includes the posture data of the movement into place when the first part is picked up, the posture data of the movement into place when the image is collected on each acquisition surface, and step 5) the posture data of the movement into place of the fixture when the first part that has completed the image collection is placed back to the preset position; 7) When capturing images of other parts that are the same as the first part, the other parts are pre-positioned at the same position as the position where the first part was picked up; the control system automatically controls the fixture and the camera through the drive mechanism from the position after the end of step 5) according to the posture data stored in step 6), and executes the actions of steps 2)-5) in sequence; when executing the actions of steps 2)-5), the path of the fixture and the camera from the previous node position to the next node position is executed according to the preferred path obtained by the control system through the algorithm, thereby automatically completing the image capture of each acquisition surface of other parts.
[0005] Furthermore, the camera is connected to a display screen, which can display the status of the part in the camera field of view in real time; when executing steps 2)-5) to capture the image of the first part, the manually operated driving mechanism is used to adjust the position of the fixture and the camera and the display screen is used to determine whether the position of the fixture and the camera meets the image capture requirements.
[0006] Furthermore, the fixture is connected to a flipping mechanism, a rotating mechanism and a translation mechanism. The flipping mechanism is used to drive the part to flip around a horizontal axis, the rotating mechanism is used to drive the part to rotate around a vertical axis, and the translation mechanism is used to drive the part to translate back and forth and left and right on a horizontal plane. By operating the flipping mechanism, the rotating mechanism and the translation mechanism, the three can work together to adjust the clamped part to a required position. The camera has vertical sliding and pitch adjustment functions. By continuously adjusting the fixture and the camera, the part collection surface can achieve the required effect within the camera's field of view.
[0007] Furthermore, the preset position described in step 5) is the position where the part is picked up in step 2), and the fixture places the part that has completed image acquisition back to the preset position and then moves away an appropriate distance to facilitate the placement of the next part to be imaged; in step 6), the fixture node position includes the position where the part is placed back to the preset position and then moved away an appropriate distance.
[0008] Furthermore, the fixture and the camera are arranged on the side of the conveyor belt. The conveyor belt stops running after conveying the part to the set position. The fixture picks up the part from the stopped conveyor belt, and puts the part back on the conveyor belt after completing the image acquisition. The conveyor belt runs to convey the part to the next position.
[0009] The present invention also provides a multi-degree-of-freedom part image automatic acquisition device, comprising: A fixture for clamping a part; the fixture is connected to a fixture driving mechanism, and the fixture driving mechanism can drive the fixture to perform multi-degree-of-freedom motion to transfer the part to a desired position; The camera is used to take pictures of the parts to realize image acquisition; the camera is connected to a camera driving mechanism, which can drive the camera to perform multi-dimensional movement and cooperate with the fixture to place the parts to be imaged in the camera field of view at a position that meets the image acquisition requirements; Display screen, which is connected to the camera and is used to display the status of the parts in the camera field of view in real time; The control system is connected with the fixture driving mechanism and the camera driving mechanism. The control system can store the posture data corresponding to each node of the camera and the fixture when the first part is imaged, and when other parts that are the same as the first part are imaged, the control system can automatically drive the fixture and the camera through the driving mechanism to sequentially execute the actions required for image acquisition based on the stored posture data; the control system uses an algorithm to derive the preferred path for the fixture and the camera to move from the previous node position to the next node position and drives the fixture and the camera to move along the preferred path.
[0010] Furthermore, the clamp includes a clamp seat, and two symmetrical electromagnet mounting plates in an eight-shaped shape are provided at the front end of the clamp seat. Electromagnets are installed on the inner sides of the two electromagnet mounting plates. When the electromagnets are energized, electromagnetic force is generated to adsorb metal parts to achieve clamping of the parts.
[0011] Furthermore, it also includes a light source, which includes a ring light source and a point light source. The ring light source and the point light source are coordinated to meet the lighting requirements of different surfaces of the parts during image acquisition.
[0012] Furthermore, it also includes a main frame, which includes a bottom plate and a top plate arranged in parallel up and down, four supporting columns are provided at the four corners of the bottom plate, and the top plate is horizontally installed on the upper ends of the four supporting columns; the clamp is installed on the upper surface of the bottom plate; the camera is installed on the lower surface of the top plate; the annular light source and the point light source are respectively installed on two of the supporting columns and can rotate in a horizontal plane around a vertical rotation axis under the drive of a driving mechanism.
[0013] Furthermore, an X-axis ball screw slide is installed on the upper surface of the bottom plate, and a Y-axis ball screw slide is installed on the sliding component of the X-axis ball screw slide; an electric rotating table whose rotating component can rotate around a vertical axis is installed on the sliding component of the Y-axis ball screw slide, and a steering gear is installed on the rotating component of the electric rotating table, and the steering gear output shaft is arranged horizontally, and the clamp is fixedly installed on the steering gear output shaft through a clamp seat so that it can be turned over with the steering gear output shaft; A downward vertical guide rail is fixedly installed on the lower surface of the top plate; the camera slider is installed on the vertical guide rail and can slide up and down along the vertical guide rail under the drive of the driving mechanism; the camera is rotatably connected to the camera slider through a horizontal rotation axis, and the camera pitch angle can be adjusted when the camera can rotate around the horizontal rotation axis under the drive of the driving mechanism.
[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. Flexible multi-surface acquisition capability: The device adopts a multi-degree-of-freedom motion structure to meet the image acquisition requirements of different surfaces and multiple angles, achieving efficient and accurate image acquisition.
[0015] 2. Posture data storage function: The relative position of the part and the camera is recorded through posture data, and the relative position of the camera and the part is automatically adjusted based on the recorded posture during repeated acquisition, thereby improving the acquisition quality and consistency.
[0016] 3. Efficient online detection: The device is integrated with the conveyor belt to automatically collect parts from the conveyor belt and then accurately put them back on the conveyor belt to achieve uninterrupted online detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 - Schematic diagram of the overall structure of the image acquisition device of the present invention.
[0018] Figure 2 -Schematic diagram of the movement direction of each movement mechanism of the image acquisition device of the present invention.
[0019] Figure 3 -Side structural diagram of the image acquisition device of the present invention.
[0020] Figure 4 - Schematic diagram of the device posture for collecting gear tooth surface images of the present invention (the gear is in a vertical state after being flipped by the steering gear).
[0021] Figure 5 -Overall system block diagram of the image acquisition device of the present invention.
[0022] Figure 6 -Flowchart of image acquisition integrated with conveyor belt of the present invention.
[0023] Figure 7 -Flowchart of debugging of multi-surface image acquisition device of parts and storage of motor position data of the present invention.
[0024] Figure 8 - GUI operation debugging interface diagram of a certain embodiment of the present invention. DETAILED DESCRIPTION
[0025] The multi-degree-of-freedom part image automatic acquisition method of the present invention is carried out according to the following steps. Figure 5-Figure 7 : 1) Adjust the camera and the fixture to their initial positions through the driving mechanism respectively; 2) When capturing the image of the first part, the fixture is driven to pick up the first part, and the fixture and the camera are continuously adjusted through the driving mechanism to make the first acquisition surface of the part within the camera field of view to achieve the desired effect; then, the camera takes a picture to complete the image acquisition of the first acquisition surface of the part; 3) Based on the last position of the fixture and the camera, the fixture and the camera are adjusted again through the driving mechanism so that the next acquisition surface of the first part can achieve the desired effect within the camera field of view; the camera takes a photo to complete the image acquisition of the next acquisition surface of the first part; 4) Repeat step 3) until all the image acquisition of the first part's intended acquisition surface is completed; 5) Drive the fixture to put the parts that have completed image acquisition back to the preset position; 6) The control system stores the posture data corresponding to each node of the camera and the fixture when the first part is imaged; the camera node posture data includes the posture data of the movement into place when the image is collected on each acquisition surface; the fixture node posture data includes the posture data of the movement into place when the first part is picked up, the posture data of the movement into place when the image is collected on each acquisition surface, and step 5) the posture data of the movement into place of the fixture when the first part that has completed the image collection is placed back to the preset position; 7) When capturing images of other parts that are the same as the first part, the other parts are pre-positioned at the same position as the position where the first part was picked up; the control system automatically controls the fixture and the camera through the drive mechanism from the position after the end of step 5) according to the posture data stored in step 6), and executes the actions of steps 2)-5) in sequence; when executing the actions of steps 2)-5), the path of the fixture and the camera from the previous node position to the next node position is executed according to the preferred path obtained by the control system through the algorithm, thereby automatically completing the image capture of each acquisition surface of other parts.
[0026] When the present invention collects images, it is sufficient for the first part to be collected manually. The control system will automatically record and store the posture data corresponding to each node of the camera and the fixture when collecting the image of the first part, record the relative position of the part (fixture) and the camera through the posture data, and automatically adjust the relative position of the camera and the fixture based on the recorded posture data in the subsequent repeated collection of the same part, thereby realizing automatic collection of the part image next time and improving the collection quality and consistency.
[0027] In order to facilitate manual operation to capture the image of the first part, the present invention connects the camera to the display screen, so that the display screen can display the state of the part in the camera field of view in real time; when performing steps 2)-5) to capture the image of the first part, the manual operation driving mechanism is used to adjust the position of the fixture and the camera, and the display screen is used to determine whether the position of the fixture and the camera meets the image acquisition requirements. In this way, the operator only needs to use the display screen to adjust the camera position and the fixture position, without looking at the part itself, nor does it need to check whether the part is in place through the camera observation window. Figure 8 This is a GUI operation debugging interface diagram of an embodiment of the present invention. Figure 8 It can be seen that the display interface in the figure can intuitively show the state of the parts observed under the camera observation window.
[0028] Furthermore, the fixture of the present invention is connected with a flipping mechanism, a rotating mechanism and a translation mechanism. The flipping mechanism is used to drive the part to flip around the horizontal axis, the rotating mechanism is used to drive the part to rotate around the vertical axis, and the translation mechanism is used to drive the part to translate back and forth and left and right on the horizontal plane. By operating the flipping mechanism, the rotating mechanism and the translation mechanism, the three can work together to achieve multi-degree-of-freedom adjustment and adjust the clamped part to the required position. At the same time, the camera has vertical sliding and pitch adjustment functions. By continuously adjusting the fixture and the camera, the part collection surface can finally achieve the required effect within the camera field of view.
[0029] The clamp and camera of the present invention adopt a multi-degree-of-freedom motion structure to meet the image acquisition requirements of different surfaces and multiple angles, and realize flexible, efficient and accurate multi-surface image acquisition.
[0030] In the present invention, the preset position described in step 5) is the position where the part is picked up in step 2), that is, the part is picked up and placed at the first position. Therefore, the fixture needs to move away a proper distance after placing the part that has completed image acquisition back to the preset position to facilitate the placement of the next part to be imaged. Correspondingly, in step 6), the fixture node position includes the position where the part is moved away a proper distance after being placed back to the preset position.
[0031] In the embodiment, the clamp and the camera are arranged on the side of the conveyor belt, that is, the parts are transmitted and taken away by the conveyor belt, and the conveyor belt automatically stops running after transmitting the parts to the set position (which can be easily achieved through mature technologies such as sensors, infrared rays, and induction probes). The clamp picks up the parts from the stopped conveyor belt, and after completing the image acquisition, puts the parts back on the conveyor belt; after the sensor detects that the parts are put back, the conveyor belt starts running again to transmit the parts to the next position.
[0032] The device of the present invention is installed beside the conveyor belt of the production line. By cooperating with the conveyor belt, the parts on the conveyor belt are automatically picked up by the clamp for image acquisition. During the acquisition process, the clamp remains stable to prevent the parts from shifting. After the acquisition is completed, the parts are accurately placed back on the conveyor belt to achieve uninterrupted online detection, improve acquisition efficiency, and avoid affecting the efficiency of the production line.
[0033] Based on the above-mentioned automatic image acquisition method, the present invention also designs a multi-degree-of-freedom part image automatic acquisition device, which can also be understood as the acquisition method is preferably implemented based on this acquisition device. Figure 1-Figure 4 , the image acquisition device of the present invention comprises, A fixture 1 is used to clamp a part; the fixture is connected to a fixture driving mechanism, and the fixture driving mechanism can drive the fixture to perform multi-degree-of-freedom motion to transfer the part to a desired position; Camera 2 is used to take pictures of parts to achieve image acquisition; the camera is connected to a camera driving mechanism, which can drive the camera to perform multi-dimensional movement and cooperate with the fixture to place the part to be imaged in the camera field of view at a position that meets the image acquisition requirements; The display screen is connected to the camera and is used to display the status of the parts in the camera field of view in real time; the figure of the present invention only reflects the content of the mechanical structure part of the device, so the display screen and control system are not shown.
[0034] The control system is connected with the fixture driving mechanism and the camera driving mechanism. The control system can store the posture data corresponding to each node of the camera and the fixture when the first part is imaged, and when other parts that are the same as the first part are imaged, the control system can automatically drive the fixture and the camera through the driving mechanism to sequentially execute the actions required for image acquisition based on the stored posture data; the control system uses an algorithm to derive the preferred path for the fixture and the camera to move from the previous node position to the next node position and drives the fixture and the camera to move along the preferred path.
[0035] In order to clamp metal parts more conveniently and friendly, the present invention designs an automatic adsorption type electromagnet clamp. Figure 3The clamp 1 includes a clamp seat 11, and two electromagnet mounting plates 12 symmetrically arranged in an eight-shaped shape are provided at the front end of the clamp seat 11. Electromagnets 13 are installed on the inner side surfaces of the two electromagnet mounting plates 12. When the electromagnet 13 is energized, it generates electromagnetic force to adsorb metal parts to achieve clamping of the parts, and automatically releases when the power is off. The present invention adopts an electromagnet structure symmetrically arranged in an eight-shaped shape to clamp parts, which has the following advantages: 1. The open structure of the eight-shaped shape has better adaptability to the size of parts, that is, it can adapt to the clamping of parts within a certain size range; 2. Because the clamping position of the clamp is fixed each time under the control of the control system, and the electromagnet has a self-adsorption function, even if the placement position of the metal part deviates from the standard position, the clamp can adsorb the part to achieve clamping, and there is a certain fault tolerance space for the placement position of the metal part. Based on this clamp structure, the present invention shows very satisfactory reliability in the clamping of metal disc gears.
[0036] Considering that image acquisition requires suitable lighting, the present invention also includes a light source, which includes a ring light source 4 and a point light source 5. The cooperation of the ring light source 4 and the point light source 5 can meet the lighting requirements of different surfaces of the part 8 during image acquisition.
[0037] In order to make the components of the present invention form an organic whole and have a stronger product quality, the present invention also includes a main frame 3, which includes a bottom plate 31 and a top plate 32 arranged in parallel up and down, four support columns 33 are arranged at the four corners of the bottom plate 31, and the top plate 32 is horizontally installed at the upper ends of the four support columns 33. The camera 2, fixture 1, light source, etc. involved in the present invention are arranged on the main frame. Specifically, the fixture 1 is installed on the upper surface of the bottom plate 31; the camera 2 is installed on the lower surface of the top plate 32; the annular light source 4 and the point light source 5 are respectively installed on two of the support columns 33 through the bracket 6 and can rotate in the horizontal plane around the vertical rotation axis under the drive of the driving mechanism. In the embodiment, the rotation of the annular light source 4 and the point light source 5 is realized by the steering gear 7. When a light source is needed, the corresponding light source is rotated to the inside of the frame and located below the camera 2 by the steering gear 7, and when it is not needed, it can be moved away. The main frame of the present invention is made of aluminum alloy material, which can effectively ensure the structural stability of the device.
[0038] In order to realize the multi-dimensional movement of the fixture 1, the present invention installs an X-axis ball screw slide 14 on the upper surface of the base plate 31, and a Y-axis ball screw slide 15 is installed on the sliding part of the X-axis ball screw slide 14; an electric rotary table 16 whose rotating part can rotate around the vertical axis is installed on the sliding part of the Y-axis ball screw slide 15, and a steering gear 7 is installed on the rotating part of the electric rotary table 16, and the steering gear output shaft is arranged horizontally, and the fixture is fixedly installed on the steering gear output shaft through the fixture seat so that it can be flipped with the steering gear output shaft. The ball screw slide, the electric rotary table and the steering gear are all existing mature products, which are not repeated here. The forward and backward, left and right movement and rotation of the fixture are respectively realized by three stepping motors. The present invention realizes the flipping movement of the fixture through the steering gear, allowing the parts to flip within a range of 270 degrees, which is convenient for image acquisition of different parts surfaces. Through the above design, the forward and backward, left and right translation, rotation and flipping of the fixture are realized, and the parts can be moved to any position within the travel range.
[0039] Furthermore, a telescopic cylinder 17 is provided on the steering gear output shaft, and the clamp is fixedly mounted on the piston rod of the telescopic cylinder 17 through the clamp seat 11, which further increases the moving range of the clamp in the front, back, left and right directions.
[0040] Similarly, in order to realize the multi-dimensional movement of the camera 2, the present invention has a downward vertical guide rail 21 fixedly installed on the lower surface of the top plate 32; the camera slider 22 is installed on the vertical guide rail 21 and can slide up and down along the vertical guide rail under the drive of the driving mechanism. The structure for realizing vertical sliding of the camera can also directly adopt a mature lifting ball screw slide module. At the same time, the camera 2 is rotatably connected to the camera slider 22 (or the sliding component on the lifting ball screw slide module) through a horizontal rotating shaft, and the camera can rotate around the horizontal rotating shaft under the drive of the driving mechanism, thereby realizing the adjustment of the camera pitch angle. In the embodiment, the pitch adjustment of the camera 2 is also achieved by the servo 7. In the present invention, the camera slider 22 is driven up and down by a stepping motor to control the fine adjustment of the camera height.
[0041] The camera of the present invention adopts a high-resolution industrial camera, and cooperates with a ring light source and a point light source to adapt to the lighting requirements of different surface structures. According to the needs of image acquisition, the position and type of the light source can be changed to achieve a more flexible lighting configuration.
[0042] Figure 2The schematic diagram of the motion direction structure of each motion module of the present invention is shown, showing the distribution of stepper motors and servos and the various motion directions. Among them, each arrow indicates the motion direction of the relevant components, ① indicates the up and down movement of the camera; ② indicates the pitch movement of the camera driven by the servo; ③ indicates the left and right rotation of the annular light source driven by the servo; ④ indicates the left and right rotation of the point light source driven by the servo; ⑤ indicates the X-direction (left and right) movement of the fixture; ⑥ indicates the Y-direction (front and back) movement of the fixture; ⑦ indicates the rotation movement of the fixture in the horizontal plane; ⑧ indicates the flipping movement of the fixture driven by the servo; ⑨ indicates the telescopic movement of the fixture driven by the telescopic cylinder. The embodiment of the present invention adopts a total of 4 servos, which are respectively used for camera pitch adjustment, rotation of two light sources and flipping of the fixture. When collecting images, the device can change the relative positions of the light source, camera and parts by adjusting the motor and servo to meet the requirements of different types of parts for bright field shooting and dark field shooting.
[0043] All driving mechanisms of the present invention, including various driving motors, such as the driving motor of the X-axis ball screw slide, the driving motor of the Y-axis ball screw slide, the driving motor of the electric rotating platform, the driving motor of the steering gear, the driving motor of the camera, the driving mechanism of the ring light source and the point light source, and the electromagnet are connected to the control system. The driving board of the control system of the present invention is composed of a steering gear driving board, a stepping motor driving board, and a master control board, which can communicate with the host computer through WiFi, TTL and 485 serial ports, receive signals and control the movement of the stepping motor and the steering gear, and realize multi-degree-of-freedom movement and precise control of the device.
[0044] The operation process of the automatic image acquisition device of the present invention is as follows: Step 1: After power on, the system first initializes the status of each stepper motor and servo to ensure that the device position is zeroed, providing a reference for subsequent actions. The device identifies the position of the part on the conveyor belt 9, controls the fixture 1 to clamp the target part 8 from the conveyor belt 9 and move it to the collection position.
[0045] Step 2: The device selects the corresponding light source and camera height and angle according to the surface structure of the part, and makes preliminary posture adjustments. If necessary, the fixture position can be adjusted again until the part position meets the image acquisition requirements.
[0046] Step 3: During the image acquisition process, the control system records the posture data of each key position in real time, including the position of fixture 1, the angle of camera 2 and the position of the light source, to ensure that the posture data is accurately stored in the database.
[0047] Step 4: When performing repeated acquisition tasks, the control system directly calls the recorded posture data and adjusts the camera 2 and part 8 to the optimal relative position of the original record to ensure the consistency of image quality in the same part acquisition tasks, which is suitable for multiple inspection tasks of the same part.
[0048] Step 5: After the acquisition task is completed, the control system automatically calibrates the current posture data to ensure the accuracy of the next call.
[0049] The image acquisition process after the present invention is integrated with the conveyor belt 9 is shown in Figure 6 .
[0050] When the device of the present invention first collects the image of the part, the system records the key node posture data such as the current position of each motion mechanism, the angle of the camera, the position of the light source, etc. The posture data is stored in the control system and used as a posture reference for subsequent identical or similar parts. When performing multiple collections of the same part or similar collection tasks of similar parts in the future, the system can directly call the stored posture data and automatically adjust the part and the camera to the best relative position recorded, ensuring the consistency of image collection and improving the collection efficiency and quality.
[0051] The present invention realizes stable and high-quality image acquisition of each surface of the part through multi-angle and precise structural design, and ensures consistency in multiple acquisitions through posture data storage function. After the acquisition device is integrated with the conveyor belt of the production line, it can automatically clamp the parts when the conveyor belt is running, complete multi-angle, high-precision and efficient acquisition of each surface through multi-angle adjustment, and put the parts back on the conveyor belt after detection, realizing efficient online detection and automated production integration.
[0052] Finally, it should be pointed out that the specific examples mentioned above are only used to explain the present invention and do not constitute a limitation on the embodiments of the present invention. Although the present invention selects preferred examples and makes relevant descriptions, for those skilled in the art, other forms of adjustments and improvements can be made based on the above explanations. It is impossible to list all possible implementation schemes here. Any direct or indirect adjustments and improvements within the scope of the technical solution of the present invention can be regarded as the protection scope of the present invention.
Claims
1. A multi-degree-of-freedom part image automatic acquisition method, characterized in that: Proceed as follows: 1) Adjust the camera and the fixture to their initial positions through the driving mechanism respectively; 2) When capturing the image of the first part, the fixture is driven to pick up the first part, and the fixture and the camera are continuously adjusted through the driving mechanism to make the first acquisition surface of the part within the camera field of view to achieve the desired effect; then, the camera takes a picture to complete the image acquisition of the first acquisition surface of the part; 3) Based on the last position of the fixture and the camera, the fixture and the camera are adjusted again through the driving mechanism so that the next acquisition surface of the first part can achieve the desired effect within the camera field of view; the camera takes a photo to complete the image acquisition of the next acquisition surface of the first part; 4) Repeat step 3) until all the image acquisition of the first part's intended acquisition surface is completed; 5) Drive the fixture to put the parts that have completed image acquisition back to the preset position; 6) The control system stores the posture data corresponding to each node of the camera and the fixture when the first part is imaged; the camera node posture data includes the posture data of the movement into place when the image is collected on each acquisition surface; the fixture node posture data includes the posture data of the movement into place when the first part is picked up, the posture data of the movement into place when the image is collected on each acquisition surface, and step 5) the posture data of the movement into place of the fixture when the first part that has completed the image collection is placed back to the preset position; 7) When capturing images of other parts that are the same as the first part, the other parts are pre-positioned at the same position as the position where the first part was picked up; the control system automatically controls the fixture and the camera through the drive mechanism from the position after the end of step 5) according to the posture data stored in step 6), and executes the actions of steps 2)-5) in sequence; when executing the actions of steps 2)-5), the path of the fixture and the camera from the previous node position to the next node position is executed according to the preferred path obtained by the control system through the algorithm, thereby automatically completing the image capture of each acquisition surface of other parts.
2. The multi-degree-of-freedom part image automatic acquisition method according to claim 1 is characterized in that: The camera is connected to a display screen, which can display the status of the part in the camera field of view in real time; when executing steps 2)-5) to capture the image of the first part, the manually operated driving mechanism is used to adjust the position of the fixture and the camera and the display screen is used to determine whether the position of the fixture and the camera meets the image capture requirements.
3. The multi-degree-of-freedom part image automatic acquisition method according to claim 1 is characterized in that: The fixture is connected with a flipping mechanism, a rotating mechanism and a translation mechanism. The flipping mechanism is used to drive the part to flip around a horizontal axis, the rotating mechanism is used to drive the part to rotate around a vertical axis, and the translation mechanism is used to drive the part to translate back and forth and left and right on a horizontal plane. By operating the flipping mechanism, the rotating mechanism and the translation mechanism, the three can work together to adjust the clamped part to a required position. The camera has vertical sliding and pitch adjustment functions. By continuously adjusting the fixture and the camera, the part collection surface can achieve the required effect within the camera field of view.
4. The multi-degree-of-freedom part image automatic acquisition method according to claim 1 is characterized in that: The preset position described in step 5) is the position where the part is picked up in step 2). The fixture puts the part that has completed image acquisition back to the preset position and then moves it away an appropriate distance to facilitate the placement of the next part to be imaged. In step 6), the fixture node position includes the position where the part is placed back to the preset position and then moved away an appropriate distance.
5. The multi-degree-of-freedom part image automatic acquisition method according to claim 1 is characterized in that: The fixture and camera are set on the side of the conveyor belt. The conveyor belt stops running after transporting the part to the set position. The fixture picks up the part from the stopped conveyor belt, and puts the part back on the conveyor belt after completing the image acquisition. The conveyor belt runs to transport the part to the next position.
6. Multi-degree-of-freedom part image automatic acquisition device, characterized by: include A fixture for clamping a part; the fixture is connected to a fixture driving mechanism, and the fixture driving mechanism can drive the fixture to perform multi-degree-of-freedom motion to transfer the part to a desired position; The camera is used to take pictures of the parts to realize image acquisition; the camera is connected to a camera driving mechanism, which can drive the camera to perform multi-dimensional movement and cooperate with the fixture to place the parts to be imaged in the camera field of view at a position that meets the image acquisition requirements; Display screen, which is connected to the camera and is used to display the status of the parts in the camera field of view in real time; The control system is connected with the fixture driving mechanism and the camera driving mechanism. The control system can store the posture data corresponding to each node of the camera and the fixture when the first part is imaged, and when other parts that are the same as the first part are imaged, the control system can automatically drive the fixture and the camera through the driving mechanism to sequentially execute the actions required for image acquisition based on the stored posture data; the control system uses an algorithm to derive the preferred path for the fixture and the camera to move from the previous node position to the next node position and drives the fixture and the camera to move along the preferred path.
7. The multi-degree-of-freedom part image automatic acquisition device according to claim 6 is characterized in that: The clamp comprises a clamp seat, at the front end of which are two symmetrical electromagnet mounting plates in an eight-shaped shape. Electromagnets are mounted on the inner sides of the two electromagnet mounting plates. When the electromagnets are energized, electromagnetic force is generated to adsorb metal parts to achieve clamping of the parts.
8. The multi-degree-of-freedom part image automatic acquisition device according to claim 6, characterized in that: It also includes a light source, which includes a ring light source and a point light source. The ring light source and the point light source are coordinated to meet the lighting requirements of different surfaces of the parts during image acquisition.
9. The multi-degree-of-freedom part image automatic acquisition device according to claim 8, characterized in that: It also includes a main frame, which includes a bottom plate and a top plate arranged in parallel up and down, four supporting columns are arranged at the four corners of the bottom plate, and the top plate is horizontally installed on the upper ends of the four supporting columns; the clamp is installed on the upper surface of the bottom plate; the camera is installed on the lower surface of the top plate; the annular light source and the point light source are respectively installed on two of the supporting columns and can rotate in a horizontal plane around a vertical rotation axis under the drive of a driving mechanism.
10. The multi-degree-of-freedom part image automatic acquisition device according to claim 9, characterized in that: An X-axis ball screw slide is installed on the upper surface of the bottom plate, and a Y-axis ball screw slide is installed on the sliding part of the X-axis ball screw slide; an electric rotating table whose rotating part can rotate around a vertical axis is installed on the sliding part of the Y-axis ball screw slide, and a steering gear is installed on the rotating part of the electric rotating table, and the steering gear output shaft is arranged horizontally, and the clamp is fixedly installed on the steering gear output shaft through a clamp seat so that it can be turned over with the steering gear output shaft; A downward vertical guide rail is fixedly installed on the lower surface of the top plate; The camera slider is installed on the vertical guide rail and can slide up and down along the vertical guide rail under the drive of the driving mechanism; The camera is rotatably connected to the camera slider via a horizontal rotation axis, and the camera pitch angle can be adjusted when the camera is rotated around the horizontal rotation axis under the drive of the driving mechanism.
Citation Information
Patent Citations
Adaptive calibration vision online detection device and method
CN106441094A
Robot rapid teaching system based on stereo vision
CN110170996A
Product defect detection system and method
CN110609037A
Line laser welding seam automatic tracking system and method based on visual demonstration
CN113102880A
Data acquisition and labeling device for line-scan digital camera and labeling method of data acquisition and labeling device
CN113569841A
Cited By
Recognition system and method based on photometric stereo
CN120495774A
A photometric stereo based identification system and method
CN120495774B