Locomotive sand box opening position identification and positioning system and method

Abstract

The invention belongs to the field of image processing and automation, and particularly relates to a locomotive sand box opening position identification and positioning system and method, and the system comprises a visual host which is used for remotely browsing, querying and storing sand box opening image information and position data in a visual extension; meanwhile, according to the sand box opening image information, manual identification is conducted, a sand box opening is marked, sand box opening position data are sent to the sand feeding robot through the visual extension set, and automatic sand feeding of a locomotive sand box through the sand feeding robot is achieved; the visual extension set is used for intelligently matching the sand box opening image information collected by the visual camera to obtain sand box opening position coordinate data and sending the sand box opening position coordinate data to the visual host; meanwhile, the distance, sent by the distance measuring sensor, between the visual camera and the target sand box opening is corrected and compensated. The method has the functions of re-learning and training, and the original corresponding sand box identification model file is supplemented and updated. The identification efficiency and the identification reliability of the system can be improved.

Description

Technical Field

[0001] The invention belongs to the field of image processing and automation, and in particular is a system and method for identifying and locating the position of a locomotive sand box opening. Background Art

[0002] In the traditional locomotive industry, sanding of locomotive sand boxes is done manually, which is labor-intensive and has a poor working environment. Using intelligent sanding robots to automatically add sand to railway locomotive sand boxes can reduce manpower and increase efficiency, and improve the level of automation in locomotive maintenance. Only by accurately identifying and locating the exact position of the locomotive sand box opening can the intelligent sanding robot realize automatic sanding. Therefore, a system that can realize the automatic identification and positioning of the sand box openings of various models of locomotives in an all-weather environment is developed to realize the automation of locomotive sand box sanding. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent system that can automatically identify and locate the position of the locomotive sand box mouth during the automatic sand filling of the electric locomotive sand box. The system can be used in the electric locomotive industry. The system can automatically identify and locate the position of the sand box mouth, and send its position information to the sand loading robot to guide the sand loading robot to automatically add sand to the locomotive sand box. The present invention can realize the automatic identification and positioning of the locomotive sand box mouth of various models in an all-weather environment, thereby realizing the automation of locomotive sand box sand filling. The problem of automatic position positioning and identification of the sand box mouth of the sand loading robot is solved, thereby realizing the automation of locomotive sand box sand filling.

[0004] The technical solution adopted by the present invention to achieve the above-mentioned purpose is: a system for identifying and locating the position of a locomotive sand box mouth, comprising: a visual host, a visual extension, a visual camera, a distance sensor and a visual light source;

[0005] The visual host is used to remotely browse, query and store the sand box opening image information and position data in the visual extension. At the same time, the sand box opening is manually identified and marked according to the sand box opening image information, and the sand box opening position data is sent to the sand loading robot through the visual extension, so that the sand loading robot can automatically add sand to the sand box.

[0006] The visual extension is used to intelligently match the collected sand box opening image information, obtain the sand box opening position coordinate data and send it to the visual host; at the same time, correct and compensate the distance between the visual camera and the target sand box opening sent by the ranging sensor;

[0007] The distance sensor is used to measure the distance between the visual camera and the target sand box opening and send it to the visual extension;

[0008] Visual camera, used to collect images of the sand box opening and change the exposure time;

[0009] Vision light source, used to provide light source illumination for vision camera.

[0010] The visual camera, the distance measuring sensor and the visual light source are arranged on the side of the sand loading robot and are all connected to the visual extension;

[0011] The visual extension is arranged in the control cabinet of the sand loading robot body. The input port is connected to the visual camera for communication via Ethernet, and the output port is connected to the ranging sensor, the visual light source, and the visual host arranged in the central control room through the bus, so as to receive the instructions sent by the visual host and control the visual camera, the ranging sensor, and the visual light source to start. At the same time, the obtained sand box mouth position coordinate data is sent to the visual host.

[0012] There are multiple visual light sources, all arranged above the visual camera;

[0013] The central axes of the illumination directions of the multiple visual light sources are focused on the center of the visual camera's field of view, and the illumination range is covered by the visual camera's field of view.

[0014] The distance measuring sensor is any one of an ultrasonic distance measuring sensor, a laser distance measuring sensor, an infrared distance measuring sensor or an optical distance sensor.

[0015] The sand box opening image information and position data include: the original sand box image, the intelligent matching image, and the sand box opening position coordinate data.

[0016] A method for a locomotive sand box mouth position identification and positioning system comprises the following steps:

[0017] 1) The visual host sends an opening command to the visual extension. After receiving the command, the visual extension controls the visual camera, ranging sensor, and visual light source to turn on respectively;

[0018] 2) The sand loading robot moves to the sand box opening of the locomotive, and the visual extension controls the visual camera to collect images of the sand box opening in various environments and send them to the visual extension;

[0019] 3) The visual extension uses the AI ​​algorithm to train the model of the sand box opening, and generates the corresponding sand box recognition model file after the training is completed; the visual extension matches the sand box opening through the intelligent matching algorithm according to the sand box recognition model file, obtains the sand box opening position coordinate data, and executes step 5), otherwise it cannot match and executes step 4);

[0020] 4) If the corresponding sand box opening cannot be matched, the loop algorithm is used until the corresponding sand box opening is matched; if the corresponding sand box opening still cannot be matched after the set number of loops, the visual host automatically alarms;

[0021] 5) The distance sensor is used to measure the distance between the visual camera and the target sand box opening. The visual extension reads the distance data between the visual camera and the target sand box opening from the distance sensor, and corrects and compensates the sand box opening position data identified by intelligent matching to improve the positioning accuracy of the sand box opening;

[0022] 6) The visual host remotely browses and queries the original image, intelligent matching image, and position data information of the sand box stored in the visual extension, manually identifies the image, marks the sand box opening, and then sends the sand box opening position data to the sand loading robot through the visual extension to guide the sand loading robot to automatically add sand to the sand box.

[0023] In step 3), the visual submachine uses the AI ​​algorithm to perform model training on the sand box opening, and generates a corresponding sand box recognition model file after the training is completed, specifically:

[0024] 2-1) The visual camera collects images of the sand box mouth at various positions and exposure times;

[0025] 2-2) manually marking the sand box opening target in the sand box opening image;

[0026] 2-3) Import the sandbox opening image with the marked target into the AI ​​cloud platform for model training. After the training is completed, a sandbox recognition model file is generated;

[0027] 2-4) The visual extension classifies, stores and records the sand box mouth images collected during work; after the collected sand box mouth image data has accumulated for a set time, the stored sand box mouth image data is retrieved, recognized and annotated by the AI ​​algorithm again for model training, and is supplemented and updated to the original corresponding sand box recognition model file to improve the system's recognition efficiency, recognition reliability and environmental adaptability.

[0028] In step 3), the visual submachine matches the sand box opening through the intelligent matching algorithm according to the sand box recognition model file to obtain the sand box opening position coordinate data, specifically:

[0029] 3-1) Collect the sand box mouth image, perform DL image matching search on the sand box mouth image according to the corresponding sand box model, and obtain the sand box mouth position coordinate data in the image coordinate system of the optimal target parameter;

[0030] 3-2) Read the actual distance between the visual camera and the target sand box opening collected by the ranging sensor, and convert the image coordinates of the sand box opening position at different distances to the image coordinates of the sand box opening at the camera calibration distance according to the visual image focal length formula;

[0031] 3-3) Perform image coordinate-physical coordinate calibration conversion on the sand box mouth position coordinate data under the image coordinate system to obtain the sand box mouth position coordinate data under the camera physical coordinate system.

[0032] In step 4), the loop algorithm is used until the corresponding sand box opening is matched, specifically:

[0033] 4-1) The visual extension controls the visual camera to change the exposure time, reads the sand box opening image data again, and then matches the sand box opening until the corresponding sand box opening is matched, completes the sand box opening identification and positioning, and calculates the sand box opening position coordinate data;

[0034] 4-2) If the image of the sand box mouth still cannot be matched after changing the camera exposure time, the visual extension controls the visual light source to change the light source brightness and read the sand box mouth image data again;

[0035] 4-3) Match the sand box opening to improve the environmental adaptability of the sand box opening intelligent matching, identification and positioning.

[0036] In step 5), the sand box opening position data identified by intelligent matching is corrected and compensated, specifically:

[0037] 5-1) The visual extension reads the distance data between the visual camera and the target sand box mouth collected by the ranging sensor, and converts the sand box mouth image matching coordinates at different distances to the image coordinates at the camera calibration distance according to the visual image focal length formula, that is:

[0038]

[0039] Among them, (PR x , P.R. y ) distance corrected sand box mouth position coordinates, (PC x , PC y ) The coordinates of the sand box mouth before distance correction, L R Actual value of the distance between the visual camera and the target sand box opening, L C Calibration value of the distance between the visual camera and the target sand box opening;

[0040] 5-2) The sand box mouth position data is corrected and compensated according to the standard calibration template to improve the positioning accuracy of the sand box mouth at different distances.

[0041] The present invention has the following beneficial effects and advantages:

[0042] 1. The intelligent system for identifying and locating the position of the locomotive sand box mouth of the present invention can complete the intelligent identification and positioning of the locomotive sand box mouth position in the railway canopy in an all-weather environment, and can work normally under high and low temperature conditions.

[0043] 2. The visual submachine of the present invention classifies, stores and records the sand box mouth images collected during work. After the sand box mouth image data to be collected is accumulated for a period of time, the stored sand box mouth image data is again retrieved, recognized, labeled and other model training is performed through a high-performance AI algorithm, and supplemented and updated to the original corresponding sand box recognition model file. The recognition efficiency, recognition reliability and environmental adaptability of the system can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of the system composition of a system for automatically identifying and locating the position of a locomotive sand box opening according to the present invention;

[0045] Figure 2 Flowchart of the method for automatically identifying and locating the position of a locomotive sand box opening of the present invention. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0047] like Figure 1 As shown, it is a schematic diagram of the system composition of a system for automatically identifying and locating the position of a locomotive sand box mouth of the present invention. The system for automatically identifying and locating the position of a locomotive sand box mouth of the present invention comprises: a visual host, a visual extension, a visual camera, a distance sensor and a visual light source;

[0048] The visual host is set in the central control room, and can remotely browse and query the original image of the sand box, intelligent matching image, position data and other information stored in the visual extension. When the visual extension cannot automatically and intelligently match the corresponding sand box opening, an automatic alarm prompt will be given, and the sand box opening will be manually identified and marked on the visual host, and then the sand box opening position data will be transmitted to the visual extension, and then the visual extension will send the sand box opening position information to the sand loading robot to guide the sand loading robot to automatically add sand to the sand box.

[0049] The visual extension is used to intelligently match the collected sand box opening image information, obtain the sand box opening position coordinate data and send it to the visual host; at the same time, the distance between the visual camera and the target sand box opening sent by the ranging sensor is corrected and compensated; the visual extension is set in the control cabinet of the sand loading robot body, and the input port is connected to the visual camera through Ethernet for communication, and the output port is connected to the ranging sensor, the visual light source, and the visual host set in the central control room through the bus, so as to receive the instructions sent by the visual host and control the visual camera, the ranging sensor, and the visual light source to start, and at the same time, send the obtained sand box opening position coordinate data to the visual host.

[0050] The distance measuring sensor is used to measure the distance between the visual camera and the target sand box opening. The visual extension reads the distance data between the visual camera and the target sand box opening from the distance measuring sensor, and corrects and compensates the sand box opening position data identified by intelligent matching to improve the positioning accuracy of the sand box opening.

[0051] The visual camera is mainly used to collect the image of the sand box mouth and change the exposure time. The visual light source provides light source illumination for the visual camera.

[0052] The visual camera, distance sensor and visual light source are arranged on the side of the sand loading robot and are all connected to the visual extension unit;

[0053] There are multiple visual light sources, all arranged above the visual camera;

[0054] The central axes of the irradiation directions of the multiple visual light sources are focused on the center of the visual camera's field of view, and the irradiation range covers the visual camera's field of view.

[0055] The distance measuring sensor is any one of an ultrasonic distance measuring sensor, a laser distance measuring sensor, an infrared distance measuring sensor or an optical distance sensor.

[0056] like Figure 2 As shown, it is a flow chart of the method of the present invention. The present invention provides a method for a locomotive sand box mouth position identification and positioning system, comprising the following steps:

[0057] 1) The visual host sends an opening command to the visual extension. After receiving the command, the visual extension controls the visual camera, ranging sensor, and visual light source to turn on respectively;

[0058] 2) The sand loading robot moves to the sand box opening of the locomotive, and the visual extension controls the visual camera to collect images of the sand box opening in various environments and send them to the visual extension;

[0059] 3-1) The visual extension uses the AI ​​algorithm to train the model of the sand box opening, and generates the corresponding sand box recognition model file after the training is completed; specifically:

[0060] 3-1-1) The visual camera collects images of the sand box opening at various positions and exposure times;

[0061] 3-1-2) manually marking the sand box opening target in the sand box opening image;

[0062] 3-1-3) Import the sandbox opening image with the marked target into the AI ​​cloud platform for model training. After the training is completed, a sandbox recognition model file is generated;

[0063] 3-1-4) The visual extension classifies, stores and records the sand box mouth images collected during work; after the collected sand box mouth image data has accumulated for a set time, the stored sand box mouth image data is retrieved, recognized and annotated by the AI ​​algorithm again for model training, and supplemented and updated to the original corresponding sand box recognition model file to improve the system's recognition efficiency, recognition reliability and environmental adaptability.

[0064] 3-2) The visual submachine matches the sand box opening through the intelligent matching algorithm according to the sand box recognition model file, obtains the sand box opening position coordinate data, and executes step 5). Otherwise, if the match fails, execute step 4); specifically:

[0065] 3-2-1) Collect the sand box mouth image, perform DL image matching search on the sand box mouth image according to the corresponding sand box model, and obtain the sand box mouth position coordinate data in the image coordinate system of the optimal target parameter;

[0066] 3-2-2) Read the actual distance between the visual camera and the target sand box opening collected by the ranging sensor, and convert the image coordinates of the sand box opening position at different distances to the image coordinates of the sand box opening at the camera calibration distance according to the visual image focal length formula;

[0067] 3-2-3) Perform image coordinate-to-physical coordinate calibration conversion on the sand box mouth position coordinate data under the image coordinate system to obtain the sand box mouth position coordinate data under the camera physical coordinate system.

[0068] 4) If the corresponding sand box opening cannot be matched, the loop algorithm is used until the corresponding sand box opening is matched; if the corresponding sand box opening still cannot be matched after the set number of loops, the visual host automatically alarms;

[0069] Through the loop algorithm, until the corresponding sand box opening is matched, specifically:

[0070] 4-1) The visual extension controls the visual camera to change the exposure time, reads the sand box opening image data again, and then matches the sand box opening until the corresponding sand box opening is matched, completes the sand box opening identification and positioning, and calculates the sand box opening position coordinate data;

[0071] 4-2) If the image of the sand box mouth still cannot be matched after changing the camera exposure time, the visual extension controls the visual light source to change the light source brightness and read the sand box mouth image data again;

[0072] 4-3) Match the sand box opening to improve the environmental adaptability of the sand box opening intelligent matching, identification and positioning.

[0073] 5) The distance sensor is used to measure the distance between the visual camera and the target sand box opening. The visual extension reads the distance data between the visual camera and the target sand box opening from the distance sensor, and corrects and compensates the sand box opening position data identified by intelligent matching to improve the positioning accuracy of the sand box opening;

[0074] 5-1) The visual extension reads the distance data between the visual camera and the target sand box mouth collected by the ranging sensor, and converts the sand box mouth image matching coordinates at different distances to the image coordinates at the camera calibration distance according to the visual image focal length formula, that is:

[0075]

[0076] Among them, (PR x , P.R. y ) distance corrected sand box mouth position coordinates, (PC x , PC y ) The coordinates of the sand box mouth before distance correction, L R Actual value of the distance between the visual camera and the target sand box opening, L C Calibration value of the distance between the visual camera and the target sand box opening;

[0077] 5-2) The sand box mouth position data is corrected and compensated according to the standard calibration template to improve the positioning accuracy of the sand box mouth at different distances.

[0078] 6) The visual host remotely browses and queries the original image, intelligent matching image, and position data information of the sand box stored in the visual extension, manually identifies the image, marks the sand box opening, and then sends the sand box opening position data to the sand loading robot through the visual extension to guide the sand loading robot to automatically add sand to the sand box.

[0079] In summary, the intelligent system for identifying and locating the locomotive sand box opening of the present invention has the functions of re-learning and training, and is supplemented and updated to the original corresponding sand box identification model file, thereby improving the recognition efficiency and reliability of the system.

[0080] Those skilled in the art will appreciate that the above are only preferred embodiments of the present invention, and the various embodiments of the present disclosure and / or the features described in the claims may be combined or combined in various ways, even if such combinations or combinations are not explicitly described in the present disclosure. It is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art may still modify the technical solutions described in the aforementioned embodiments, or perform equivalent substitutions on some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

[0081] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A system for identifying and locating the position of a locomotive sand box opening, characterized in that: include: Vision host, vision extension, vision camera, distance sensor and vision light source; The visual host is used to remotely browse, query and store the sand box opening image information and position data in the visual extension. At the same time, the sand box opening is manually identified and marked according to the sand box opening image information, and the sand box opening position data is sent to the sand loading robot through the visual extension, so that the sand loading robot can automatically add sand to the sand box. The visual extension is used to intelligently match the collected sand box opening image information, obtain the sand box opening position coordinate data and send it to the visual host; at the same time, correct and compensate the distance between the visual camera and the target sand box opening sent by the ranging sensor; The distance sensor is used to measure the distance between the visual camera and the target sand box opening and send it to the visual extension; Visual camera, used to collect images of the sand box opening and change the exposure time; Vision light source, used to provide light source illumination for vision camera.

2. A locomotive sand box opening position identification and positioning system according to claim 1, characterized in that: The visual camera, the distance measuring sensor and the visual light source are arranged on the side of the sand loading robot and are all connected to the visual extension; The visual extension is arranged in the control cabinet of the sand loading robot body. The input port is connected to the visual camera for communication via Ethernet, and the output port is connected to the ranging sensor, the visual light source, and the visual host arranged in the central control room through the bus, so as to receive the instructions sent by the visual host and control the visual camera, the ranging sensor, and the visual light source to start. At the same time, the obtained sand box mouth position coordinate data is sent to the visual host.

3. The intelligent system for identifying and locating the position of the locomotive sand box mouth according to claim 1, characterized in that: There are multiple visual light sources, all arranged above the visual camera; The central axes of the illumination directions of the multiple visual light sources are focused on the center of the visual camera's field of view, and the illumination range is covered by the visual camera's field of view.

4. The intelligent system for identifying and locating the position of the locomotive sand box mouth according to claim 1, characterized in that: The distance measuring sensor is any one of an ultrasonic distance measuring sensor, a laser distance measuring sensor, an infrared distance measuring sensor or an optical distance sensor.

5. The intelligent system for identifying and locating the position of the locomotive sand box mouth according to claim 1, characterized in that: The sand box opening image information and position data include: the original sand box image, the intelligent matching image, and the sand box opening position coordinate data.

6. The method for identifying and locating the position of a locomotive sand box opening according to claim 1, characterized in that: The following steps are involved: 1) The visual host sends an opening command to the visual extension. After receiving the command, the visual extension controls the visual camera, ranging sensor, and visual light source to turn on respectively; 2) The sand loading robot moves to the sand box opening of the locomotive, and the visual extension controls the visual camera to collect images of the sand box opening in various environments and send them to the visual extension; 3) The visual extension uses the AI ​​algorithm to train the model of the sand box opening, and generates the corresponding sand box recognition model file after the training is completed; the visual extension matches the sand box opening through the intelligent matching algorithm according to the sand box recognition model file, obtains the sand box opening position coordinate data, and executes step 5), otherwise it cannot match and executes step 4); 4) If the corresponding sand box opening cannot be matched, the loop algorithm is used until the corresponding sand box opening is matched; if the corresponding sand box opening still cannot be matched after the set number of loops, the visual host automatically alarms; 5) The distance sensor is used to measure the distance between the visual camera and the target sand box opening. The visual extension reads the distance data between the visual camera and the target sand box opening from the distance sensor, and corrects and compensates the sand box opening position data identified by intelligent matching to improve the positioning accuracy of the sand box opening; 6) The visual host remotely browses and queries the original image, intelligent matching image, and position data information of the sand box stored in the visual extension, manually identifies the image, marks the sand box opening, and then sends the sand box opening position data to the sand loading robot through the visual extension to guide the sand loading robot to automatically add sand to the sand box.

7. A method for identifying and locating a locomotive sand box opening according to claim 6, characterized in that: In step 3), the visual submachine uses the AI ​​algorithm to perform model training on the sand box opening, and generates a corresponding sand box recognition model file after the training is completed, specifically: 2-1) The visual camera collects images of the sand box mouth at various positions and exposure times; 2-2) manually marking the sand box opening target in the sand box opening image; 2-3) Import the sandbox opening image with the marked target into the AI ​​cloud platform for model training. After the training is completed, a sandbox recognition model file is generated; 2-4) The visual extension classifies, stores and records the sand box mouth images collected during work; after the collected sand box mouth image data has accumulated for a set time, the stored sand box mouth image data is retrieved, recognized and annotated by the AI ​​algorithm again for model training, and is supplemented and updated to the original corresponding sand box recognition model file to improve the system's recognition efficiency, recognition reliability and environmental adaptability.

8. A method for identifying and locating a locomotive sand box opening position according to claim 6, characterized in that: In step 3), the visual submachine matches the sand box opening through the intelligent matching algorithm according to the sand box recognition model file to obtain the sand box opening position coordinate data, specifically: 3-1) Collect the sand box mouth image, perform DL image matching search on the sand box mouth image according to the corresponding sand box model, and obtain the sand box mouth position coordinate data in the image coordinate system of the optimal target parameter; 3-2) Read the actual distance between the visual camera and the target sand box opening collected by the ranging sensor, and convert the image coordinates of the sand box opening position at different distances to the image coordinates of the sand box opening at the camera calibration distance according to the visual image focal length formula; 3-3) Perform image coordinate-physical coordinate calibration conversion on the sand box mouth position coordinate data under the image coordinate system to obtain the sand box mouth position coordinate data under the camera physical coordinate system.

9. A method for a locomotive sand box opening position identification and positioning system according to claim 6, characterized in that: In step 4), the loop algorithm is used until the corresponding sand box opening is matched, specifically: 4-1) the visual extension controls the visual camera to change the exposure time, reads the sand box opening image data again, and then matches the sand box opening until the corresponding sand box opening is matched, completing the sand box opening identification and positioning, and calculating the sand box opening position coordinate data; 4-2) If the image of the sand box mouth still cannot be matched after changing the camera exposure time, the visual extension controls the visual light source to change the light source brightness and read the sand box mouth image data again; 4-3) Match the sand box opening to improve the environmental adaptability of the sand box opening intelligent matching, identification and positioning.

10. The method for identifying and locating the position of a locomotive sand box opening according to claim 6, characterized in that: In step 5), the sand box opening position data identified by intelligent matching is corrected and compensated, specifically: 5-1) The visual extension reads the distance data between the visual camera and the target sand box mouth collected by the ranging sensor, and converts the sand box mouth image matching coordinates at different distances to the image coordinates at the camera calibration distance according to the visual image focal length formula, that is: Among them, (PR x , P.R. y ) distance corrected sand box mouth position coordinates, (PC x , PC y ) The coordinates of the sand box mouth before distance correction, L R Actual value of the distance between the visual camera and the target sand box opening, L C Calibration value of the distance between the visual camera and the target sand box opening; 5-2) The sand box mouth position data is corrected and compensated according to the standard calibration template to improve the positioning accuracy of the sand box mouth at different distances.

Images