Cheating prevention method, apparatus and system for unattended weight inspection, and readable medium

Abstract

The present invention relates to the field of automatic weighing, and specifically relates to a cheating prevention method, apparatus and system for unattended weight inspection, and a readable medium. The cheating prevention method for unattended weight inspection comprises: acquiring a first image, wherein the first image includes a water discharge pipe image of a water tank of a truck to be weighed; performing recognition on the first image by using a first recognition model, so as to obtain corresponding discrimination data of the water discharge pipe image; and when the discrimination data indicates that a water pipe is open, executing a weighing operation on said truck. A water discharge pipe image of a water tank of a truck to be weighed is acquired, and recognition is then performed on the water discharge pipe image to determine whether a water discharge pipe is in an open state; if the water discharge pipe is in an open state, it is determined that water has been discharged, and said truck is then weighed, that is, it is ensured that a driver cannot interfere with accurate weighing by means of discharging water from the water tank, and the detection of whether water in the water tank has been completely discharged is completed fully automatically without requiring human involvement, thereby effectively reducing the labor cost; and whether the water tank has been emptied is detected in real time, thereby avoiding the need for follow-up investigation.

Description

Unattended weighing and anti-cheating method, device, system and readable medium Technical Field

[0001] The present invention relates to the field of automatic weighing, and in particular to an anti-cheating method, device, system and readable medium for unattended weighing. Background Art

[0002] Unmanned weighing systems have been widely used in industries such as coal mining, coking, steelmaking, mining and metallurgy, petrochemicals, cement, building materials, grain, food, feed, agricultural product procurement, animal husbandry, and ports. They have played a significant role in improving enterprise management, plugging management loopholes, reducing consumption, controlling costs, standardizing production procedures, improving work efficiency, and increasing economic benefits. Traditional unmanned intelligent weighing systems replace manual anti-cheating monitoring with various methods, including monitoring for incomplete vehicle weighing, license plate replacement, and remote control interference with the meter. However, water tank cheating can occur when drivers fill the tank with water before gross weighing and then drain the water after unloading, affecting the net weight and causing losses to the receiving company.

[0003] The defects and shortcomings of the current automatic weighing system's weight monitoring system include: 1. Manual inspection at the weighing site to check whether the vehicle's water tank is empty requires inspectors to be on duty 24 hours a day, which is a great waste of human resources; 2. Video surveillance and image capture are used to check whether the water tank switch is open. This method can only be verified after the fact and cannot achieve real-time monitoring; 3. The vehicle water tank inspection process is time-consuming, and vehicles need to wait in line to be weighed.

[0004] Summary of the Invention

[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide an unattended weighing anti-cheating method, device, system and readable medium, which can effectively prevent the driver from filling the water tank with water before the gross weight and draining the water after unloading the goods, thereby affecting the net weight value and causing losses to the receiving unit.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In one aspect, the present invention provides an unattended weighing and anti-cheating method, comprising:

[0008] Acquire a first image; the first image includes an image of a water outlet pipe of a water tank of a truck to be weighed;

[0009] Using a first recognition model to recognize the first image, obtaining corresponding determination data of the water outlet pipe image; the determination data includes whether the water pipe is open or closed; the first recognition model is constructed based on a neural network model;

[0010] When the determination data indicates that the water pipe is open, a weighing operation is performed on the truck to be weighed.

[0011] Furthermore, before acquiring the first image, the method further includes:

[0012] Obtain an image of the truck to be weighed;

[0013] Using a vehicle type recognition model to identify the image of the truck to be weighed to obtain vehicle type data; the vehicle type recognition model is constructed based on a neural network model;

[0014] determining a water tank position based on the vehicle model data;

[0015] A first image corresponding to the water tank position is acquired.

[0016] Furthermore, after obtaining the vehicle model data, the following steps are executed:

[0017] Acquire multiple risk locations of the truck to be weighed based on the vehicle model data;

[0018] Acquire a second image corresponding to each risk location of the truck to be weighed;

[0019] Recognize the second image based on a second recognition model to obtain risk data; the second recognition model is constructed based on a neural network model;

[0020] An alerting operation is performed based on the risk data.

[0021] Furthermore, the risk data includes abnormal information and non-abnormal information;

[0022] Performing warning operations based on the risk data specifically includes:

[0023] When the risk data is abnormal information, a warning message is issued to the outside.

[0024] Furthermore, the second image is identified based on a second recognition model to obtain risk data, specifically including:

[0025] If the second image is recognized by the second recognition model and it is determined that the second image contains a water tank image, the risk data is abnormal information; otherwise, the risk data is non-abnormal information.

[0026] Furthermore, when the risk data is abnormal information, it also includes:

[0027] The second image is input into the first recognition model for recognition to obtain the corresponding judgment data of the water outlet pipe image. When the judgment data shows that the water pipe is open, the truck to be weighed is weighed, otherwise a warning message is issued.

[0028] Furthermore, when the determination data indicates that the water pipe is closed, an alarm message is issued to the outside.

[0029] In another aspect, the present invention provides an unattended weighing and anti-cheating device, comprising:

[0030] An acquisition module, configured to acquire a first image; the first image includes an image of a water outlet pipe of a water tank of a truck to be weighed;

[0031] a recognition module, configured to recognize the first image using a first recognition model to obtain determination data corresponding to the water outlet pipe image; the determination data including whether the water pipe is open or closed; the first recognition model being constructed based on a neural network model;

[0032] The weighing module is used to perform a weighing operation on the truck to be weighed when the determination data indicates that the water pipe is open.

[0033] In another aspect, the present invention provides an unattended weighing and anti-cheating system, comprising:

[0034] weighbridge;

[0035] One or more cameras, configured to acquire detection images; the detection images comprising a first image including an image of a water outlet pipe of a water tank of a truck to be weighed, a second image corresponding to each risk location, and an image of the truck to be weighed;

[0036] The unattended weighing and anti-cheating device is connected to the floor scale and the camera respectively.

[0037] On the other hand, the present invention provides a computer-readable medium storing a computer program, which, when executed by a processor, implements the unattended weighing and anti-cheating method.

[0038] Compared with the prior art, the unattended weighing and anti-cheating method, device, system, and readable medium provided by the present invention have the following beneficial effects:

[0039] The unattended weighing and anti-cheating method provided by the present invention obtains an image of the water outlet pipe of the truck to be weighed, and then identifies the water outlet pipe image to determine whether it is in an open state. If so, it is determined that the water has been drained, and then the truck is weighed. In other words, it is ensured that the driver cannot interfere with accurate weighing by draining water from the water tank. The water tank is automatically detected to see if it is empty, without the need for human intervention, effectively saving labor costs, and real-time detection of whether the water tank is empty, without the need for subsequent investigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] FIG1 is a flow chart of the unattended weighing and anti-cheating method provided by the present invention.

[0041] FIG2 is a flow chart of an implementation method of the unattended weighing and anti-cheating method provided by the present invention.

[0042] FIG3 is a flow chart of a process of acquiring a first image based on a vehicle type provided by the present invention.

[0043] FIG4 is a flow chart of risk detection provided by the present invention.

[0044] FIG5 is a structural block diagram of the unattended weighing and anti-cheating device provided by the present invention.

[0045] FIG6 is a schematic diagram of the unattended weighing and anti-cheating system provided by the present invention. Implementation Method

[0046] In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] Those skilled in the art will understand that the foregoing general description and the following detailed description are exemplary and illustrative of specific embodiments of the present invention and are not intended to limit the present invention.

[0048] As used herein, the terms "comprises," "comprising," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps may include not only those steps but may also include other steps not expressly listed or inherent to such process or method. Similarly, a reference to one or more devices or subsystems, elements or structures or components preceded by "comprising ... a" does not, without further limitations, preclude the presence of other devices or other subsystems or other elements or other structures or other components or other devices or other subsystems or other elements or other structures or other components. Throughout this specification, appearances of the phrases "in one embodiment," "in another embodiment," and similar language may, but do not necessarily, all refer to the same embodiment.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0050] When a truck brakes for a long time downhill, the friction between the brake shoes and the drum generates a large amount of heat. The brake drum is very hot and needs to be cooled by water droplets sent from the water tank. Temporarily draining the water tank before the vehicle is weighed will not affect the truck's driving safety. The water tank can be refilled when the truck leaves the site. Therefore, in accordance with the requirements of the management system, the driver of the delivery vehicle first opens the vehicle's water tank and drains the water in the tank before weighing. Then, after the delivery vehicle empties the water tank, the water tank switch remains open while weighing.

[0051] Please refer to Figures 1-2 and 6. The present invention provides an unattended weight checking and anti-cheating method, which is applied to an automatic weighing system and is suitable for monitoring when the truck water tank is installed on both sides of the vehicle and the water tank drain switch is not blocked.

[0052] The automatic weighing system includes a control device, a sound column, an LED display, a camera, a floor scale, and a fill light. It primarily uses camera monitoring combined with intelligent AI algorithms to determine whether the water tank switch is open. If the switch is open, it determines that the water in the tank has been drained. If the switch is closed, the driver is prompted to empty the tank and keep the drain switch open before re-weighing via text prompts on the display and voice announcements. The control device includes intelligent devices such as computers and servers. Data exchange between the various components of the automatic weighing system is based on the Industrial Internet.

[0053] Unattended weighing and anti-cheating methods include:

[0054] A first image is acquired; the first image includes an image of the water outlet pipe of the water tank of the truck to be weighed. Specifically, in this embodiment, the first image is acquired in real time by a camera installed at a location corresponding to the water outlet pipe of the water tank. Furthermore, multiple cameras may be deployed around the weighbridge, and during use, data transmitted by only one of the cameras is acquired, or each of the multiple images captured is used as the first image for recognition.

[0055] The first image is recognized using a first recognition model to obtain corresponding judgment data of the water outlet pipe image; the judgment data includes whether the water pipe is open or closed; the first recognition model is constructed based on a neural network model; in specific operations, if images of a truck to be weighed are acquired at multiple angles and orientations, then recognition is performed on multiple first images. If the judgment data of the first image of the water outlet pipe are all that the water pipe is open, the truck is continued to be weighed.

[0056] When the determination data indicates that the water pipe is open, a weighing operation is performed on the truck to be weighed.

[0057] When the determination data indicates that the water pipe is closed, an alarm message is issued externally.

[0058] The unattended weighing and anti-cheating method provided by the present invention obtains an image of the water outlet pipe of the truck to be weighed, and then identifies the water outlet pipe image to determine whether it is in an open state. If so, it is determined that the water has been drained, and then the truck is weighed. In other words, it is ensured that the driver cannot interfere with accurate weighing by draining water from the water tank. The water tank is automatically detected to see if it is empty, without the need for human intervention, effectively saving labor costs, and real-time detection of whether the water tank is empty, without the need for subsequent investigation.

[0059] In some embodiments, the process of generating the first recognition model includes:

[0060] Obtaining a first training set, the first training set including a plurality of first training samples, the first training samples being images of a water outlet pipe of a water tank of a truck and corresponding working state labels; the working state labels including an open state and a closed state;

[0061] The first training set is used to train the initialized neural network model to obtain the first recognition model. Preferably, the first recognition model is built into the control device.

[0062] Furthermore, in some embodiments, the first recognition model is built into a camera to form a first smart camera, which outputs the determination data to a control device. After receiving the determination data, the control device drives the scale to perform a weighing operation on the truck to be weighed if the determination data indicates that the water pipe is open. If the determination data indicates that the water pipe is closed, the control device drives a sound bar or an LED display to issue a warning message.

[0063] In specific implementation, the delivery vehicle may enter the recognition area of ​​the smart camera after being weighed. The smart recognition camera will identify the switch status of the water tank and report the recognition result to the control device in the weighing room through the network.

[0064] The computer receives the recognition result returned by the smart camera. If the received result is that the water tank switch is in the off state, the computer program sends a text reminder "Please turn on the water tank switch, empty the water tank and weigh it again" to the LED screen through the network, and at the same time sends a text voice broadcast reminder to the sound column.

[0065] The LED display and sound column receive information sent by the computer and respond.

[0066] Further, referring to FIG3 , as a preferred solution, in this embodiment, before acquiring the first image, the process further includes:

[0067] Obtain an image of the truck to be weighed;

[0068] Using a vehicle type recognition model to identify the image of the truck to be weighed to obtain vehicle type data; the vehicle type recognition model is constructed based on a neural network model;

[0069] determining a water tank position based on the vehicle model data;

[0070] Acquire a first image of the corresponding water tank position. In this embodiment, an image of the truck to be weighed is first captured, and then the vehicle type of the truck to be weighed is determined, and then the location of the water outlet pipe of the water tank of the truck of this model is determined, and then a first image of the corresponding water outlet pipe is captured. This allows for more flexible photography of the truck's water tank.

[0071] In some embodiments, the process of generating the vehicle type identification model includes:

[0072] Obtain a vehicle model training set, where the vehicle model training set includes multiple vehicle model training samples, each of which is a truck image and a vehicle model label;

[0073] The vehicle type recognition model is obtained by training the initialized neural network model using the vehicle type training set. Preferably, the vehicle type recognition model is built into the control device.

[0074] In some embodiments, corresponding water tank location data is associated with different truck models. That is, as long as the truck model data is obtained, the water tank location data for the model can be quickly known, and then the corresponding water tank position can be quickly found, and further the water outlet pipe position can be found.

[0075] In some embodiments, the vehicle type recognition model is embedded in a camera to form a vehicle type recognition camera, which outputs vehicle type data to a control device. Upon receiving the vehicle type data, the control device retrieves water tank location data for the truck corresponding to the vehicle type data and drives the corresponding first intelligent camera to capture first image data, thereby obtaining judgment data.

[0076] Furthermore, in order to further prevent the driver from installing a water tank at other locations on the truck and being unable to observe it directly, please refer to FIG4 . As a preferred solution, in this embodiment, after obtaining the vehicle model data, the following steps are further executed:

[0077] Acquire multiple risk locations of the truck to be weighed based on the vehicle model data;

[0078] Acquire a second image corresponding to each risk location of the truck to be weighed;

[0079] Recognize the second image based on a second recognition model to obtain risk data; the second recognition model is constructed based on a neural network model;

[0080] A warning is executed based on the risk data. In this embodiment, after determining the type of truck equipped with a scale, several locations on the vehicle where water tanks are easily installed are identified as risk locations, such as the chassis, rear end, and other inconspicuous locations. Furthermore, the camera that captures the corresponding second image is also installed in a location capable of capturing the corresponding risk locations, such as on a weighbridge, facing upward to capture the underside of the vehicle. A second image is then captured for each risk location to determine whether any external hanging objects are attached, including whether they are water tanks, buckets, or other such objects. If so, a warning is issued, quickly identifying whether the truck has been modified. The warning can be provided through voice notification, such as "External water tank. Please open the water outlet and re-weigh." The LED display will display the text message "Please open the water tank, drain the water tank, and re-weigh." Simultaneously, the speaker will play a corresponding voice prompt.

[0081] In some embodiments, the generation process of the second recognition model includes:

[0082] Obtaining a second training set, the second training set including a plurality of second training samples, the second training samples being images of a water outlet pipe of a water tank of a truck and corresponding working state labels; the working state labels including an open state and a closed state;

[0083] The initialized neural network model is trained using the second training set to obtain the second recognition model. Preferably, the second recognition model is built into the control device.

[0084] In some embodiments, the first recognition model is built into a camera to form a first smart camera, which outputs the risk data to a control device.

[0085] Furthermore, as a preferred solution, in this embodiment, the risk data includes abnormal information and non-abnormal information;

[0086] Performing warning operations based on the risk data specifically includes:

[0087] When the risk data is abnormal, a warning message is issued to the outside;

[0088] When the risk data is non-abnormal information, the truck to be weighed is weighed.

[0089] Furthermore, as a preferred solution, in this embodiment, identifying the second image based on the second recognition model to obtain risk data specifically includes:

[0090] If the second recognition model identifies the second image as containing a water tank, the risk data is considered abnormal; otherwise, the risk data is considered normal. Specifically, if an external water tank or bucket is mounted on the truck, the risk data is considered abnormal; otherwise, the risk data is considered normal. In this embodiment, the presence of an external water tank triggers an alert, prompting on-duty personnel to intervene on-site, and the image information is archived.

[0091] Furthermore, as a preferred solution, in this embodiment, when the risk data is abnormal information, the following is further included:

[0092] The second image is input into the first recognition model for recognition, and corresponding water outlet pipe image determination data is obtained. If the determination data indicates that the water pipe is open, the truck to be weighed is weighed; otherwise, a warning message is issued. In this embodiment, if an external water tank is mounted and its outlet pipe is closed, the risk information is abnormal; otherwise, no abnormality is displayed. This ensures that even if an external water tank is mounted on a truck, the system can still detect whether the tank has been drained, further enhancing anti-cheating effectiveness.

[0093] The team took photos of various vehicle water tanks in advance and trained on pictures of the water tanks in open and closed states. Using AI technologies such as target detection and motion recognition, a model was developed through training that can identify the on / off status of vehicle water tanks.

[0094] Furthermore, in some embodiments, the first recognition model and the second recognition model can be integrated into the same camera to form an AI smart camera.

[0095] Correspondingly, referring to FIG5 , the present invention further provides an unattended weighing and anti-cheating device, comprising:

[0096] An acquisition module, configured to acquire a first image; the first image includes an image of a water outlet pipe of a water tank of a truck to be weighed;

[0097] a recognition module, configured to recognize the first image using a first recognition model to obtain determination data corresponding to the water outlet pipe image; the determination data including whether the water pipe is open or closed; the first recognition model being constructed based on a neural network model;

[0098] The weighing module is used to perform a weighing operation on the truck to be weighed when the determination data indicates that the water pipe is open.

[0099] Correspondingly, referring to FIG6 , the present invention further provides an unattended weighing and anti-cheating system, comprising:

[0100] weighbridge;

[0101] One or more cameras, configured to acquire detection images; the detection images comprising a first image including an image of a water outlet pipe of a water tank of a truck to be weighed, a second image corresponding to each risk location, and an image of the truck to be weighed;

[0102] The unattended weighing and anti-cheating device is connected to the floor scale and the camera respectively.

[0103] During specific implementation, a computer can be installed in the weighing room for the linkage control of unmanned weighing equipment. An AI intelligent recognition camera is installed in the center of the left and right sides of the truck scale. The camera's field of view covers the location where the water tank is installed on the side of the truck. The camera is connected to the computer in the weighing room through the network. A fill light is installed next to the camera to illuminate the drain switch of the truck water tank to improve the camera recognition rate. A sound column is installed on the wall of the weighing room and connected to the computer in the weighing room through the network. LED display screens are installed at the front and rear of the truck scale and connected to the computer in the weighing room through the network. A software program is installed on the computer in the weighing room to receive data from the AI ​​intelligent recognition camera and to linkage control the LED display screen and sound column to send prompt instructions.

[0104] Accordingly, the present invention also provides a computer-readable medium storing a computer program, which, when executed by a processor, implements the unattended weighing and anti-cheating method.

[0105] More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0106] In this application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0107] It is understandable that those skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of the present invention, and all these changes or substitutions should fall within the scope of protection of the claims attached to the present invention.

Claims

1. An unattended weighing and anti-cheating method, characterized in that: include: Acquire a first image; the first image includes an image of a water outlet pipe of a water tank of a truck to be weighed; The first image is recognized using a first recognition model to obtain corresponding determination data of the water outlet pipe image; the determination data includes whether the water pipe is open or closed; the first recognition model is constructed based on a neural network model; When the determination data indicates that the water pipe is open, a weighing operation is performed on the truck to be weighed.

2. The unattended weighing anti-cheating method according to claim 1 is characterized in that: Before acquiring the first image, the method further includes: Acquire the image of the truck to be weighed; Using a vehicle type recognition model to identify the image of the truck to be weighed to obtain vehicle type data; the vehicle type recognition model is constructed based on a neural network model; determining a water tank position based on the vehicle model data; A first image corresponding to the water tank position is acquired.

3. The unattended weighing anti-cheating method according to claim 2 is characterized in that: After obtaining the vehicle model data, the following is also executed: Acquire multiple risk locations of the truck to be weighed based on the vehicle model data; Acquire a second image of each risk location of the truck to be weighed; Recognize the second image based on a second recognition model to obtain risk data; The second recognition model is constructed based on a neural network model; An alerting operation is performed based on the risk data.

4. The unattended weighing anti-cheating method according to claim 3 is characterized in that: The risk data includes abnormal information and non-abnormal information; Performing warning operations based on the risk data specifically includes: When the risk data is abnormal information, a warning message is issued to the outside.

5. The unattended weighing anti-cheating method according to claim 4 is characterized in that: Recognizing the second image based on the second recognition model to obtain risk data specifically includes: If the second image is recognized by the second recognition model and it is determined that the second image contains a water tank image, the risk data is abnormal information; otherwise, the risk data is non-abnormal information.

6. The unattended weighing anti-cheating method according to claim 5, characterized in that: When the risk data is abnormal information, it also includes: The second image is input into the first recognition model for recognition to obtain the corresponding determination data of the water outlet pipe image. When the determination data indicates that the water pipe is open, the truck to be weighed is weighed, otherwise a warning message is issued.

7. The unattended weighing anti-cheating method according to claim 1, characterized in that: When the determination data indicates that the water pipe is closed, an alarm message is issued externally.

8. An unattended weighing and anti-cheating device, characterized in that: include: An acquisition module, configured to acquire a first image; the first image includes an image of a water outlet pipe of a water tank of a truck to be weighed; A recognition module, configured to recognize the first image using a first recognition model to obtain determination data of the corresponding water outlet pipe image; the determination data includes whether the water pipe is open or closed; the first recognition model is constructed based on a neural network model; The weighing module is used to perform a weighing operation on the truck to be weighed when the determination data indicates that the water pipe is open.

9. An unattended weighing and anti-cheating system, characterized in that: include: Weighbridge; One or more cameras, used to obtain detection images; the detection images include a first image including an image of a water outlet pipe of a water tank of a truck to be weighed, a second image corresponding to each risk location, and an image of the truck to be weighed; The unattended weighing and anti-cheating device described in claim 8 is connected to the floor scale and the camera respectively.

10. A computer-readable medium, characterized in that A computer program is stored, and when the computer program is executed by a processor, the unattended weighing and anti-cheating method according to any one of claims 1 to 7 is implemented.

Images