System and method for detecting freight vehicle measurement evasion

Abstract

A system for detecting freight vehicle measurement evasion according to an embodiment of the present invention comprises: a multi-lane driving vehicle object detection module for receiving a road image from a multi-lane surveillance camera and detecting a vehicle; a multi-lane driving vehicle license plate recognition module for receiving a road image from the multi-lane surveillance camera and recognizing a license plate of a vehicle; a single-lane driving vehicle object detection module for receiving a road image from a single-lane surveillance camera and detecting a vehicle; a single-lane driving vehicle license plate recognition module for receiving a road image from the single-lane surveillance camera and recognizing a license plate of a vehicle; and an operation-restriction-suspected vehicle measurement evasion determination module for determining whether a vehicle detected in the road image has evaded measurement.

Description

Truck Measurement Refusal Detection System and Method

[0001] The present invention relates to a truck measurement non-resistance detection system and a method thereof.

[0002] Overloaded vehicles are a major cause of road damage, as they cause roads to become potholes and uneven, increasing the risk of accidents for small vehicles such as passenger cars. Such road damage caused by overloaded vehicles requires frequent road repairs, leading to the problem of wasting tax money.

[0003] Therefore, there is a persistent need to crack down on overloaded vehicles traveling on the road. Overloaded vehicle enforcement equipment is implemented by photographing license plates using cameras, etc., by utilizing a vehicle detection trigger signal generated from automatic weighing equipment embedded in the road or from a vehicle detection sensor installed in supplement to the automatic weighing equipment.

[0004] The technical problem that the present invention aims to solve is to provide a cargo truck measurement non-refusal detection system and method characterized by recognizing the driving location and license plate of a specific vehicle suspected of being subject to driving restrictions, and determining whether to perform normal measurement or avoid measurement through an overload checkpoint.

[0005] To solve the above technical problem, the cargo truck measurement non-compliance detection system according to the present embodiment includes: a multi-lane driving vehicle object detection module that receives road images from a multi-lane monitoring camera and detects a vehicle; a multi-lane driving vehicle license plate recognition module that receives road images from the multi-lane monitoring camera and recognizes the license plate of a vehicle; a single-lane driving vehicle object detection module that receives road images from a single-lane monitoring camera and detects a vehicle; a single-lane driving vehicle license plate recognition module that receives road images from the single-lane monitoring camera and recognizes the license plate of a vehicle; and a suspected driving restriction vehicle measurement non-compliance determination module that determines whether the vehicle detected in the road images is non-compliance with measurement.

[0006] The above multi-lane driving vehicle detection module and the above single-lane driving vehicle detection module can detect the vehicle as a bounding box through a trained deep learning model.

[0007] The bounding box above can be represented by four x and y coordinates.

[0008] The above multi-lane driving vehicle license plate recognition module and the above single-lane driving vehicle license plate recognition module can generate a trigger line on the road image and recognize the license plate of the vehicle through the trigger line.

[0009] The above-mentioned measurement non-response determination module for vehicles suspected of restricted driving may determine that measurement is non-response when the number of the vehicle suspected of restricted driving is recognized by the multi-lane driving vehicle number plate recognition module, and determine that measurement is normal when it is recognized by the single-lane driving vehicle number plate recognition module.

[0010] To solve the above technical problem, the method for detecting non-compliance with measurement of a cargo truck according to the present embodiment comprises: a step of receiving a road image from a multi-lane monitoring camera and detecting a vehicle; a step of receiving a road image from the multi-lane monitoring camera and recognizing a license plate of a vehicle; a step of receiving a road image from a single-lane monitoring camera and detecting a vehicle; a step of receiving a road image from the single-lane monitoring camera and recognizing a license plate of a vehicle; and a step of determining whether the vehicle detected in the road image is non-compliance with measurement.

[0011] The step of detecting a vehicle by receiving a road image from the multi-lane monitoring camera and the step of detecting a vehicle by receiving a road image from the single-lane monitoring camera can detect the vehicle as a bounding box through a trained deep learning model.

[0012] The bounding box above can be represented by four x and y coordinates.

[0013] The step of receiving a road image from the multi-lane monitoring camera and recognizing a vehicle's license plate, and the step of receiving a road image from the single-lane monitoring camera and recognizing a vehicle's license plate, can generate a trigger line on the road image and recognize the vehicle's license plate through the trigger line.

[0014] The step of determining whether a vehicle detected in the above road image is refusing to measure may determine that the vehicle is refusing to measure if the number of the vehicle suspected of being restricted from driving is recognized by the multi-lane driving vehicle number plate recognition module, and determine that it is normal measurement if it is recognized by the single-lane driving vehicle number plate recognition module.

[0015] According to embodiments of the present invention, the driving location and license plate of a specific vehicle suspected of being restricted from driving can be recognized to determine whether normal measurement or measurement avoidance is performed through an overload checkpoint.

[0016] FIG. 1 illustrates a truck measurement non-response detection system according to the present embodiment installed on a road.

[0017] FIG. 2 is a block diagram of a truck measurement non-response detection system according to the present embodiment.

[0018] FIG. 3 is a diagram illustrating a multi-lane driving vehicle object detection module of a cargo truck measurement non-response detection system according to the present embodiment.

[0019] FIG. 4 is a diagram illustrating a multi-lane driving vehicle license plate recognition module of a cargo truck measurement non-response detection system according to the present embodiment.

[0020] FIG. 5 is a diagram illustrating a single-lane driving vehicle object detection module of a cargo truck measurement non-response detection system according to the present embodiment.

[0021] FIG. 6 is a diagram illustrating a license plate recognition module for a single-lane driving vehicle of a cargo truck measurement non-response detection system according to the present embodiment.

[0022] FIGS. 7 and 8 are drawings for explaining the bounding boxes of the multi-lane driving vehicle object detection module and the single-lane driving vehicle object detection module of the truck measurement non-response detection system according to the present embodiment.

[0023] FIGS. 9 to 12 are drawings for explaining the multi-lane driving vehicle license plate recognition module and the single-lane driving vehicle license plate recognition module of a cargo truck measurement non-response detection system according to the present embodiment.

[0024] FIG. 13 illustrates a screen of a truck measurement non-response detection system according to the present embodiment.

[0025] FIG. 14 is a flowchart of a method for detecting non-response to truck measurement according to the present embodiment.

[0026] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.

[0027] However, the technical concept of the present invention is not limited to some of the described embodiments but can be implemented in various different forms, and within the scope of the technical concept of the present invention, one or more of the components among the embodiments may be selectively combined or substituted.

[0028] In addition, terms used in this embodiment (including technical and scientific terms) may be interpreted in a sense that is generally understood by those skilled in the art to which this embodiment belongs, unless explicitly and specifically defined otherwise. Terms that are commonly used, such as terms defined in advance, may be interpreted in consideration of their meaning in the context of the relevant technology.

[0029] Furthermore, the terms used in this embodiment are for the purpose of describing the embodiment and are not intended to limit the invention.

[0030] In this specification, the singular form may include the plural form unless specifically stated otherwise in the text, and when described as "at least one of A and B and C (or more than one)," it may include one or more of all combinations that can be formed from A, B, and C.

[0031] In addition, terms such as first, second, A, B, (a), (b), etc., may be used when describing the components of the present embodiment. These terms are used merely to distinguish the components from other components and are not intended to limit the essence, order, or sequence of the components.

[0032] And, where it is stated that a component is 'connected', 'combined', or 'connected' to another component, this may include not only cases where the component is directly 'connected', 'combined', or 'connected' to the other component, but also cases where it is 'connected', 'combined', or 'connected' due to another component located between the component and the other component.

[0033] Furthermore, when described as being formed or placed "above" or "below" each component, "above" or "below" includes not only cases where two components are in direct contact with each other, but also cases where one or more other components are formed or placed between the two components. Additionally, when expressed as "above" or "below," it may include the meaning of a downward direction as well as an upward direction relative to a single component.

[0034]

[0035] FIG. 1 illustrates a truck measurement non-response detection system according to the present embodiment installed on a road, FIG. 2 is a block diagram of a truck measurement non-response detection system according to the present embodiment, FIG. 3 is a diagram for explaining a multi-lane driving vehicle object detection module of a truck measurement non-response detection system according to the present embodiment, FIG. 4 is a diagram for explaining a multi-lane driving vehicle license plate recognition module of a truck measurement non-response detection system according to the present embodiment, FIG. 5 is a diagram for explaining a single-lane driving vehicle object detection module of a truck measurement non-response detection system according to the present embodiment, FIG. 6 is a diagram for explaining a single-lane driving vehicle license plate recognition module of a truck measurement non-response detection system according to the present embodiment, FIG. 7 and FIG. 8 are diagrams for explaining the bounding boxes of the multi-lane driving vehicle object detection module and the single-lane driving vehicle object detection module of a truck measurement non-response detection system according to the present embodiment, FIG. 9 to FIG. 12 are diagrams for explaining the multi-lane driving vehicle license plate recognition module and the single-lane driving vehicle license plate recognition module of a truck measurement non-response detection system according to the present embodiment, FIG. Figure 13 illustrates a screen of a truck measurement non-response detection system according to the present embodiment, and Figure 14 is a flowchart of a truck measurement non-response detection method according to the present embodiment.

[0036] WIM stands for "Weight-In-Motion" and is a technology that measures the weight of a vehicle while it is in motion on the road. Using sensors installed on the road, the WIM system measures the weight of a vehicle in real time as it passes and is used for traffic monitoring, road maintenance, and the collection of taxes and tolls. The WIM system enhances road safety and enables efficient traffic management.

[0037] The truck measurement non-response detection system (10) according to the present embodiment may include a multi-lane driving vehicle object detection module (12), a multi-lane driving vehicle license plate recognition module (13), a single-lane driving vehicle object detection module (15), a single-lane driving vehicle license plate recognition module (16), and a measurement non-response determination module (17) for vehicles suspected of being restricted from driving. Referring to FIG. 2, the truck measurement non-response detection system (10) according to the present embodiment may have two arms installed on a single pole near an overload checkpoint, and a multi-lane monitoring camera (11) and a single-lane monitoring camera (14) may be installed thereon, and a multi-lane driving vehicle object detection module (12), a multi-lane driving vehicle license plate recognition module (13), a single-lane driving vehicle object detection module (15), a single-lane driving vehicle license plate recognition module (16), and a measurement non-response determination module (17) for vehicles suspected of being restricted from driving may be installed inside the enclosure.

[0038] The multi-lane driving vehicle object detection module (12) can detect vehicles by receiving road images through the multi-lane monitoring camera (11). The single-lane driving vehicle object detection module (15) can detect vehicles by receiving road images through the single-lane monitoring camera (14). The multi-lane monitoring camera (11) may be a device that collects images in real-time in the direction of the multi-lane near the overload checkpoint. The single-lane monitoring camera (14) may be a device that collects images in real-time at the measurement location of the overload checkpoint. In the following, since the process of detecting vehicles in the multi-lane driving vehicle object detection module (12) and the single-lane driving vehicle object detection module (15) is the same, the explanation will be based on the multi-lane driving vehicle object detection module (12), and the explanation for the multi-lane driving vehicle object detection module (12) can be applied in the same way to the single-lane driving vehicle object detection module (15).

[0039] The multi-lane driving vehicle object detection module (12) can detect vehicles as bounding boxes through a trained deep learning model (DCNN, deep convolutional neural network). The road image may include a background other than the truck. In the road image, the truck may be placed in the center area or in the corner area.

[0040] The multi-lane driving vehicle object detection module (12) can detect a vehicle in a road image and generate bounding boxes (P1, P2, P3, P4) by combining edge information for the upper, lower, left, and right sides. The bounding boxes can be represented by four x and y coordinates. The four coordinates of the bounding boxes can be set as P1 (x1,y1), P2 (x2,y2), P3 (x3,y3), and P4 (x4,y4), respectively.

[0041] The multi-lane driving vehicle object detection module (12) can generate classification data of the detected vehicles. The multi-lane driving vehicle object detection module (12) can generate classification data that separates the color, size, and type of the detected vehicles. The multi-lane driving vehicle object detection module (12) can transmit the detected vehicle information and classification data to the multi-lane driving vehicle license plate recognition module (13). The multi-lane driving vehicle license plate recognition module (13) can receive road images from the multi-lane surveillance camera (11), and according to another embodiment, can receive the detected vehicle information and classification data from the multi-lane driving vehicle object detection module (12).

[0042] The multi-lane driving vehicle license plate recognition module (13) can recognize the license plate of a vehicle by receiving a road image through a multi-lane monitoring camera (11). The single-lane driving vehicle license plate recognition module (16) can recognize the license plate of a vehicle by receiving a road image through a single-lane monitoring camera (14). In the following, since the process of detecting a vehicle in the multi-lane driving vehicle license plate recognition module (13) and the single-lane driving vehicle license plate recognition module (16) is the same, the explanation will be based on the multi-lane driving vehicle license plate recognition module (13), and the explanation for the multi-lane driving vehicle license plate recognition module (13) can be applied equally to the single-lane driving vehicle license plate recognition module (16).

[0043] The multi-lane driving vehicle license plate recognition module (13) can capture vehicle detection image data through the lanes and trigger lines of a predefined license plate recognition zone. The multi-lane driving vehicle license plate recognition module (13) can recognize the license plate and driving lane of the detected vehicle. The multi-lane driving vehicle license plate recognition module (13) can transmit the recognized vehicle license plate and lane data to the vehicle measurement non-compliance judgment module (17) suspected of driving restriction. For example, the trigger line may have coordinate values ​​TrigLine(xp, yp) and TrigLine(xq, yq). The lane line may have coordinate values ​​Lane1(xa, ya), Lane2(xb, yb), Lane3(xc, yc), and Lane4(xd, yd).

[0044] Vehicle objects detected by the multi-lane driving vehicle object detection module (12) and the single-lane driving vehicle object detection module (15) can be used to calculate Obj_CenterOfButtom(x,y). For example, Obj_CenterOfButtom(x,y) = ((x2-x1) / 2), y2. If Obj_CenterOfButtom(y) is greater than TrigLine(yp) and TrigLine(yq), then the detected object can be extracted. If Obj_CenterOfButtom(x) exists between Lane1(xa) and Lane2(xb), it can be determined as driving in Lane 1; if Obj_CenterOfButtom(x) exists between Lane2(xb) and Lane3(xc), it can be determined as driving in Lane 2; and if Obj_CenterOfButtom(x) exists between Lane3(xc) and Lane4(xd), it can be determined as driving in Lane 3.

[0045] License plates inside a vehicle can be extracted using Color Pattern Matching based on images within the bounding box of objects extracted via trigger lines. Color Pattern Matching is a technology that recognizes and compares specific color patterns or shapes in images or videos. Template images for various cargo trucks can be generated in advance to utilize the Color Pattern Matching function.

[0046] Color pattern matching can extract a license plate within an object similar to a template image based on a weight of 50 for the shape of the license plate and a weight of 50 for the color of the license plate. Based on the license plate extracted through color pattern matching, the license plate string can be recognized through OCR (Optical Character Recognition). To use OCR, the strings of license plates of multiple trucks can be trained one character at a time in advance. The recognized license plate string result value can be transmitted to the measurement refusal judgment module (17) for vehicles suspected of driving restrictions.

[0047] The measurement refusal determination module (17) for vehicles suspected of being restricted from driving can determine whether measurement is normal or avoidance by receiving the license plate of a vehicle suspected of being restricted from driving identified through an external overload enforcement system and comparing it with the license plate of a vehicle received from a multi-lane driving vehicle license plate recognition module (13) and a single-lane driving vehicle license plate recognition module (16). The measurement refusal determination module (17) for vehicles suspected of being restricted from driving can determine that measurement is refusal if the number of the vehicle suspected of being restricted from driving is recognized by the multi-lane driving vehicle license plate recognition module (13). The measurement refusal determination module (17) for vehicles suspected of being restricted from driving can determine that measurement is normal if the number of the vehicle suspected of being restricted from driving is recognized by the single-lane driving vehicle license plate recognition module (16). The measurement refusal determination module (17) for vehicles suspected of being restricted from driving can display all received data (vehicle detection and classification, vehicle license plate, driving lane) on a single screen.

[0048]

[0049] FIG. 14 is a flowchart of a method for detecting non-compliance with measurement of a cargo truck according to the present embodiment. Since the detailed description of each step in FIG. 14 corresponds to the detailed description of the cargo truck non-compliance detection system in FIG. 1 to FIG. 13, redundant descriptions will be omitted below.

[0050] In detecting non-compliance with measurement of a cargo truck, in step S1, a road image is received from a multi-lane monitoring camera to detect a vehicle, in step S2, a road image is received from a multi-lane monitoring camera to recognize the license plate of the vehicle, in step S3, a road image is received from a single-lane monitoring camera to detect a vehicle, in step S4, a road image is received from a single-lane monitoring camera to recognize the license plate of the vehicle, and in step S5, it is determined whether the vehicle detected in the road image is non-compliance with measurement.

[0051] Steps S1 and S3 detect vehicles as bounding boxes using a trained deep learning model. The bounding box can be represented by four x and y coordinates. Steps S2 and S4 generate trigger lines on the road image and can recognize the vehicle's license plate through the trigger lines. In step S5, if the license plate of a vehicle suspected of being restricted from driving is recognized by a multi-lane driving vehicle license plate recognition module, it is determined to be non-compliance with measurement, and if it is recognized by a single-lane driving vehicle license plate recognition module, it is determined to be normal measurement.

[0052]

[0053] Meanwhile, embodiments of the present invention can be implemented as computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored.

[0054] Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. Additionally, computer-readable recording media may be distributed across networked computer systems, allowing computer-readable code to be stored and executed in a distributed manner. Furthermore, functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention belongs.

[0055] Those skilled in the art related to the embodiments described above will understand that they may be implemented in modified forms without departing from the essential characteristics of the description. Therefore, the disclosed methods should be considered in an illustrative rather than a restrictive sense. The scope of the invention is defined by the claims, not by the foregoing description, and all variations within the scope of equivalence should be interpreted as being included in the invention.

Claims

1. A multi-lane driving vehicle object detection module that receives road images from a multi-lane surveillance camera and detects vehicles; A multi-lane driving vehicle license plate recognition module that receives road images from the above-mentioned multi-lane monitoring camera and recognizes the license plate of a vehicle; A single-lane driving vehicle object detection module that detects vehicles by receiving road images from a single-lane monitoring camera; A single-lane driving vehicle license plate recognition module that receives road images from the single-lane monitoring camera and recognizes the license plate of a vehicle; and A freight vehicle measurement non-compliance detection system comprising a measurement non-compliance determination module for a suspected vehicle with driving restrictions that determines whether a vehicle detected in the above road image is refusing measurement.

2. In Paragraph 1, The above multi-lane driving vehicle detection module and the above single-lane driving vehicle detection module are a cargo truck measurement non-response detection system that detects the vehicle as a bounding box through a trained deep learning model.

3. In Paragraph 1, The above bounding box is a truck measurement non-response detection system represented by four x and y coordinates.

4. In Paragraph 1, A truck measurement non-response detection system in which the above multi-lane driving vehicle license plate recognition module and the above single-lane driving vehicle license plate recognition module generate a trigger line in the road image and recognize the license plate of the vehicle through the trigger line.

5. In Paragraph 1, A cargo truck measurement non-resistance detection system in which the above-mentioned measurement non-resistance determination module for vehicles suspected of driving restrictions determines that measurement is non-resistance when the number of the vehicle suspected of driving restriction is recognized by the multi-lane driving vehicle number plate recognition module, and determines that measurement is normal when recognized by the single-lane driving vehicle number plate recognition module.

6. A step of receiving road images from a multi-lane surveillance camera and detecting vehicles; A step of receiving road images from the above-mentioned multi-lane surveillance camera and recognizing the license plate of a vehicle; A step of receiving road images from a single-lane monitoring camera and detecting vehicles; A step of receiving a road image from the above-mentioned single-lane monitoring camera and recognizing the license plate of a vehicle; and A method for detecting a cargo truck's refusal to measure, comprising the step of determining whether the vehicle detected in the above road image is refusing to measure.

7. In Paragraph 6, A method for detecting a truck that does not measure, comprising the steps of receiving a road image from a multi-lane monitoring camera to detect a vehicle and receiving a road image from a single-lane monitoring camera to detect a vehicle, wherein the vehicle is detected as a bounding box through a trained deep learning model.

8. In Paragraph 7, The above bounding box is a method for detecting non-compliance with truck measurement, indicated by four x and y coordinates.

9. In Paragraph 6, A method for detecting non-compliance with measurement of a cargo truck, comprising the steps of receiving a road image from a multi-lane monitoring camera and recognizing a vehicle's license plate, and receiving a road image from a single-lane monitoring camera and recognizing a vehicle's license plate, wherein a trigger line is generated in the road image and the vehicle's license plate is recognized through the trigger line.

10. In Paragraph 6, A method for detecting non-compliance with measurement of a cargo truck, wherein the step of determining whether a vehicle detected in the above road image is non-compliance with measurement is determined as non-compliance with measurement if the number of the vehicle suspected of being restricted from driving is recognized by the multi-lane driving vehicle number plate recognition module, and determined as normal measurement if it is recognized by the single-lane driving vehicle number plate recognition module.

Images