CN111797796A

CN111797796A - Construction method of parking specification detection model, parking specification detection method, system, terminal and medium

Info

Publication number: CN111797796A
Application number: CN202010668124.2A
Authority: CN
Inventors: 曹学军
Original assignee: Hangzhou Zhixing Technology Co ltd
Current assignee: Hangzhou Zhixing Technology Co ltd
Priority date: 2020-07-13
Filing date: 2020-07-13
Publication date: 2020-10-20

Abstract

The invention provides a construction method of a parking specification detection model, a parking specification detection method, a system, a terminal and a medium, and relates to a parking specification detection scheme based on deep learning.

Description

Construction method of parking specification detection model, parking specification detection method, system, terminal and medium

Technical Field

The invention relates to the technical field of vehicle management and machine vision, in particular to a construction method of a parking specification detection model, a parking specification detection method, a parking specification detection system, a terminal and a medium.

Background

At present, the phenomena of disordered parking, inclined parking, irregular parking or vehicle falling down frequently exist in a public parking area of a non-motor vehicle, the phenomena not only affect the appearance of a city, but also prevent other people from orderly parking, and aiming at the problem, the main solution at present is to patrol the parking area in a manual mode and manually adjust the parking area when the phenomena are found, but the mode is time-consuming and labor-consuming and consumes great labor cost.

Disclosure of Invention

In view of the above defects in the prior art, the invention provides a method, a system, a terminal and a medium for constructing and predicting a parking specification detection model, which are used for solving the problems that in the prior art, parking of vehicles with non-specification specifications is manually adjusted, and further time and labor are wasted, and great labor cost is consumed.

In order to achieve the above object, a first aspect of the present invention provides a method for constructing a parking specification detection model, including: collecting images of a parking area; marking the vehicles in the image; dividing the marked samples into a training set and a testing set, inputting the training set into a deep learning network for training, and testing the trained deep learning network by using the testing set; and determining a parking specification detection model according to the test result of the deep learning network so as to perform parking specification detection on the image to be detected.

In some embodiments of the first aspect of the present invention, the method further comprises labeling the vehicle in the image by means of a bezel labeling.

In some embodiments of the first aspect of the present invention, the labeling information labeling the vehicle includes: one or more of coordinate information of the vehicle center point in the image, size information of the vehicle, angle information of vehicle parking, vehicle category information, and vehicle parking posture information; wherein the vehicle stopping posture information comprises posture information of vehicle standing or vehicle falling.

In order to achieve the above object, a second aspect of the present invention provides a parking regulation detecting method, including: acquiring an image to be detected; inputting the image to be detected into a parking standard detection model; obtaining vehicle parking information output by the parking specification detection model; and judging whether the vehicle in the image to be detected is subjected to nonstandard parking according to the vehicle parking information, and sending out prompt information when the nonstandard parking is detected.

In some embodiments of the second aspect of the present invention, the vehicle parking information includes one or more of parking position coordinate information, parking angle information, and information on whether to fall down the vehicle in the image to be detected.

In some embodiments of the second aspect of the present invention, the irregular parking comprises a vehicle parking super-zone; wherein, judge whether the vehicle in the image of waiting to detect parks the mode of super region and include: presetting a corresponding parking area coordinate for a parking area; judging whether the coordinates of the vehicle exceed the coordinates of the parking area in the image according to the parking position coordinate information of the vehicle output by the parking specification detection model; and if so, determining that the vehicle is parked in the super region.

In some embodiments of the second aspect of the present disclosure, the irregular stop comprises a vehicle tilting; wherein, judge whether the vehicle in the image of waiting to detect askew mode of stopping include: obtaining the parking angle information of the vehicle output by the parking specification detection model; judging whether the parking angle of the vehicle exceeds a preset angle range or not in the image according to the parking angle information of the vehicle; if so, determining that the vehicle is askew to stop.

In some embodiments of the second aspect of the present invention, the irregular parking comprises a vehicle falling over; wherein, judge whether the mode that the vehicle in the image of waiting to detect falls over includes: acquiring parking posture information of the vehicle output by the parking specification detection model; and judging whether the vehicle falls down according to the parking position information.

To achieve the above object, a third aspect of the present invention provides a parking regulation detecting system comprising: the image acquisition device is used for acquiring a real-time image corresponding to the parking area; the detection device is in communication connection with the image acquisition device; wherein the detection device receives real-time images from the image acquisition device; inputting the real-time image into a parking specification detection model obtained by training a deep learning network so as to obtain vehicle parking information output by the parking specification detection model; judging whether the vehicle in the image to be detected is subjected to nonstandard parking according to the vehicle parking information; and the prompting device is in communication connection with the detection device and is used for sending out prompting information when the detection device detects that the parking is not standardized.

In order to achieve the above object, a fourth aspect of the present invention provides a terminal for constructing a parking specification detection model, including: a first storage unit for storing at least one computer program; the first processing unit is used for running the at least one computer program to execute the construction method of the parking specification detection model.

To achieve the above object, a fifth aspect of the present invention provides a parking regulation detecting terminal, including: a second storage unit for storing at least one computer program; a second processing unit for running the at least one computer program to perform the parking specification detection method.

To achieve the above object, a sixth aspect of the present invention provides a computer-readable storage medium storing at least one computer program which, when executed, performs a method of constructing the parking specification detection model; or to perform the parking specification detection method.

The construction and prediction method, the system, the terminal and the medium of the parking specification detection model have the following technical effects: the invention provides a parking specification detection scheme based on deep learning, which is characterized in that a shot public parking area photo is directly detected through a deep neural network from the perspective of computer vision, whether the phenomenon that a vehicle is parked disorderly, askew, untidy in parking or falls over the ground and the like occurs in the photo is identified, if any one or more of the above conditions are detected, corresponding responsible personnel are informed to correct the situation, and the parking area does not need to be patrolled in a manual mode, so that the automatic maintenance of the public parking area is realized, the manpower and material resources of manual patrolling are saved, and the market appearance is improved.

The conception, the specific structure and the technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, the features and the effects of the present invention.

Drawings

Fig. 1 is a schematic view of a parking irregularity scene in the prior art.

Fig. 2 is a schematic flow chart of a method for constructing a parking specification detection model according to an embodiment of the present invention.

Fig. 3 is a flowchart illustrating a parking specification detection method according to an embodiment of the invention.

Fig. 4 is a schematic structural diagram of a parking specification detection system according to an embodiment of the present invention.

Fig. 5 is a schematic structural diagram of a terminal for building a parking specification detection model according to an embodiment of the present invention.

Fig. 6 is a schematic structural diagram of a parking specification detection terminal in an embodiment of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.

It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.

Also, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises," "comprising," and/or "comprising," when used in this specification, specify the presence of stated features, operations, elements, components, items, species, and/or groups, but do not preclude the presence, or addition of one or more other features, operations, elements, components, items, species, and/or groups thereof. The terms "or" and/or "as used herein are to be construed as inclusive or meaning any one or any combination. Thus, "A, B or C" or "A, B and/or C" means "any of the following: a; b; c; a and B; a and C; b and C; A. b and C ". An exception to this definition will occur only when a combination of elements, functions or operations are inherently mutually exclusive in some way.

Aiming at the phenomena of disordered parking, askew parking, irregular parking or vehicle falling down and the like in a non-motor vehicle public parking area, for example, in a vehicle returning scene shown in fig. 1, the vehicle 11 askew parking causes the vehicle to exceed the vehicle returning area, and the vehicle 12 not only exceeds the vehicle returning area but also falls down on the ground. For such a disorder, the current technical means is to patrol the parking area in a manual mode, and manually adjust the parking area when the phenomenon is found, but the mode is time-consuming and labor-consuming, and consumes great labor cost.

In view of the above, the invention provides a parking specification detection scheme based on deep learning, which directly detects a shot public parking area photo through a deep neural network from the perspective of computer vision, identifies whether the phenomena of vehicle disorderly parking, oblique parking, irregular parking or vehicle falling over occur in the photo, and if any one or more of the above phenomena are detected, informs corresponding responsible personnel to correct the situation, and does not need to patrol the parking area in a manual mode, so that automatic maintenance of the public parking area is realized, manpower and material resources for manual patrol are saved, and the market appearance is improved.

In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention are further described in detail by the following embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The first embodiment is as follows:

fig. 2 is a schematic flow chart illustrating a method for constructing a parking specification detection model according to an embodiment of the present invention.

It should be noted that the method for constructing the parking specification detection model in this embodiment can be applied to various types of hardware devices. The hardware device may be a controller, such as an arm (advanced RISC machines) controller, an fpga (field Programmable Gate array) controller, a soc (system on chip) controller, a dsp (digital signal processing) controller, or an mcu (micro controller unit) controller; the hardware device may also be a Personal computer, such as a desktop computer, a notebook computer, a tablet computer, a smart phone, a smart television, a Personal Digital Assistant (PDA for short), and the like; the hardware device may also be a server, and the server may be arranged on one or more entity servers according to various factors such as functions, loads, and the like, or may be formed by a distributed or centralized server cluster, which is not limited in this embodiment.

In this embodiment, the method for constructing the parking specification detection model mainly includes the following steps:

step S201: an image of a parking area is acquired.

In some examples, images of the parking area may be captured by the camera module. The camera module comprises a camera device, a storage device and a processing device. The image capturing device includes but is not limited to: cameras, video cameras, camera modules integrated with optical systems or CCD chips, camera modules integrated with optical systems and CMOS chips, and the like.

Step S202: and marking the vehicles in the image, and dividing the samples obtained by marking into a training set and a testing set.

In some examples, after the vehicles in the image are labeled, each vehicle is labeled with a labeling frame, and the labeling information is obtained through the labeling frame in the image. The labeling information includes, but is not limited to, coordinate information of a vehicle center point in the image, size information of the vehicle, parking angle information of the vehicle, vehicle category information, and vehicle parking posture information.

The coordinate information of the vehicle center point in the image refers to the coordinate information of the center point of the marking frame of the vehicle in the image. The size information of the vehicle at least comprises length and width information, and the length and width information of the vehicle can be obtained by marking the length and width of the frame. The angle information of the vehicle parking can be obtained through the angle between the marking frame and the edge of the parking area; typically, the angle between a regularly parked vehicle and the curb of a parking area is about 90, while an unregulated vehicle deviates from the curb of a parking area by 90. The vehicle category information is labeled manually and can be divided into non-motor vehicle categories and other categories. The vehicle parking posture information refers to the state information of the vehicle falling or standing, and can be marked manually.

In some examples, the vehicles in the image are preferably cross-hatched. It should be noted that, the labeling methods commonly used at present are all rectangular labels perpendicular to the horizontal axis, the angle between the labeling method and the horizontal axis cannot be changed, and the inclined frame labels can form any angle with the horizontal axis. The invention aims to detect vehicles which are not parked normally, so that the vehicles are parked at any angle, and the vehicle in the image can be labeled more flexibly and accurately by virtue of the inclined frame labeling.

Step S203: inputting the training set into a deep learning network for training, and testing the trained deep learning network by using the test set.

In particular, the data set of the parking area may be divided into a training set and a test set after the labeling is completed. The ratio of the training set to the testing set can be adjusted according to the practical application, for example, 7:3 or 6: 4. And then, inputting the training set into a deep learning network for training, and verifying the training effect by adopting the test set after training for a certain number of rounds.

In some examples, the deep learning network may select an image target detection network model. The image target detection network model integrates the segmentation and the identification of the target based on the geometric and statistical characteristics of the target, positions the target and determines the position and the size of the target. The target detection network model related to the embodiment includes, but is not limited to, network models such as R3det, R-CNN, Fast R-CNN, FPN, SSD, or YOLO.

Step S204: and determining a parking specification detection model according to the test result of the deep learning network so as to perform parking specification detection on the image to be detected. Preferably, the deep learning network with the best test effect can be captured as the parking specification detection model.

Specifically, after the deep learning network completes training, the deep learning network with the best test effect on the test set is selected as the parking specification detection model. The parking specification detection model is used for detecting the parking specification of the current image to be detected, and particularly inputs the current image to be detected into the parking specification detection model, and the parking specification detection model outputs the results of the coordinates of the vehicle center point, the parking angle of the vehicle, whether the vehicle falls down or not and the like of the vehicle in the image.

In some examples, an evaluation index of target detection, mAP, may be employed to evaluate whether the deep learning network is trained completely. The higher the mAP value of the deep learning network on the test set is, the better the test effect is, and the highest mAP value is selected as the optimal neural network model. Since the calculation of the mAP value is already the prior art, it is not described in detail.

According to the method for constructing the parking specification detection model, the neural network model constructed according to the method can be used for intelligently detecting whether abnormal phenomena such as vehicle disorderly parking, oblique parking, irregular parking or vehicle falling down and the like occur in the picture through the deep neural network, manual detection is not needed, automatic maintenance of a public parking area is achieved, manpower and material resources for manual inspection are saved, and the market appearance is improved.

Example two:

fig. 3 is a schematic flow chart illustrating a parking specification detection method according to an embodiment of the invention.

It should be noted that the parking specification detection method in the present embodiment can be applied to various types of hardware devices. The hardware device may be a controller, such as an arm (advanced RISC machines) controller, an fpga (field programmable Gate array) controller, a soc (system on chip) controller, a dsp (digital signal processing) controller, or an mcu (micro controller unit) controller; the hardware device may also be a Personal computer, such as a desktop computer, a notebook computer, a tablet computer, a smart phone, a smart television, a Personal Digital Assistant (PDA for short), and the like; the hardware device may also be a server, and the server may be arranged on one or more entity servers according to various factors such as functions, loads, and the like, or may be formed by a distributed or centralized server cluster, which is not limited in this embodiment.

In this embodiment, the parking specification detection method includes the following steps:

step S301: and acquiring an image to be detected.

In some examples, the images to be detected are taken periodically from cameras located in the parking area, for example, the cameras take one picture per second and upload.

Step S302: and inputting the image to be detected into a parking standard detection model.

In some examples, the parking specification detection model may be trained by a construction method of the parking specification detection model, such as that illustrated in fig. 2.

Step S303: and obtaining the vehicle parking information output by the parking specification detection model. The vehicle parking information comprises one or more of parking position coordinate information, parking angle information, whether the vehicle falls down or not and the like of the vehicle in the image to be detected.

Step S304: and judging whether the vehicle in the image to be detected is subjected to nonstandard parking according to the vehicle parking information.

In some examples, the irregular parking includes a vehicle parking super zone; wherein, judge whether the vehicle in the image of waiting to detect parks the mode of super region and include: presetting a corresponding parking area coordinate for a parking area; judging whether the coordinates of the vehicle exceed the coordinates of the parking area in the image according to the parking position coordinate information of the vehicle output by the parking specification detection model; and if the coordinates of the vehicle exceed the coordinates of the parking area, determining that the vehicle is parked in the super area.

Specifically, for each parking area, a corresponding parking area coordinate may be preset, and if the coordinate of a certain vehicle parking position output by the parking specification detection model of this embodiment exceeds the preset parking area coordinate, the vehicle parking super area may be considered as belonging to an irregular parking behavior.

In some examples, the irregular stop includes a vehicle rolling; wherein, judge whether the vehicle in the image of waiting to detect askew mode of stopping include: obtaining the parking angle information of the vehicle output by the parking specification detection model; judging whether the parking angle of the vehicle exceeds a preset angle range or not in the image according to the parking angle information of the vehicle; and if the parking angle of the vehicle exceeds the preset angle range, determining that the vehicle is askew to park.

Specifically, an acceptable parking angle range may be preset for each parking area, and if a certain vehicle parking angle output by the parking specification detection model of the present embodiment exceeds the acceptable parking angle range, the vehicle may be considered to be parked askew and belong to a parking irregularity behavior. It should be noted that the case where the parking angle of the vehicle exceeds the range in the present embodiment does not necessarily result in the vehicle being parked beyond the area, and the present embodiment is not limited to this.

In some examples, the irregular stop includes a vehicle falling over; wherein, judge whether the mode that the vehicle in the image of waiting to detect falls over includes: acquiring parking posture information of the vehicle output by the parking specification detection model; and judging whether the vehicle falls down according to the parking position information.

Specifically, the parking position information of the vehicle output by the parking specification detection model includes position information of the vehicle standing up or position information of the vehicle falling down, and if the position information of the vehicle falling down is output, it can be determined that the vehicle falling down belongs to a parking irregularity.

It should be noted that the above-mentioned irregular parking behaviors can be regarded as irregular parking behaviors only by the presence of any one of them.

Step S305: if the parking is not standardized, a prompt message is sent out.

Specifically, when the situation that the vehicle in the current image is parked in a super-region, askew parking or falling down is detected, prompt information is sent to the relevant responsible party, and the corresponding responsible party is informed to process in time, so that the city appearance is improved, and the purpose of standard parking is achieved.

Step S306: otherwise, ending.

According to the parking specification detection method provided by the embodiment, the parking specification detection model obtained through the training deep learning network can be used for automatically detecting whether the vehicle in the image has the phenomena of irregular parking, askew parking, irregular parking or irregular parking such as falling of the vehicle and the like, manual detection is not needed, automatic maintenance of a public parking area is realized, manpower and material resources for manual inspection are saved, and the appearance of the city is improved.

Example three:

fig. 4 is a schematic structural diagram of a parking specification detection system according to an embodiment of the present invention. The parking specification detection system of the present embodiment includes an image acquisition device 41, a detection device 42, and a prompt device 43.

Specifically, the image capturing device 41 is used to capture a real-time image corresponding to a parking area; the detection device 42 is in communication connection with the image acquisition device 41; the detection device 42 receives a real-time image from the image acquisition device 41; inputting the real-time image into a parking specification detection model obtained by training a deep learning network so as to obtain vehicle parking information output by the parking specification detection model; and judging whether the vehicle in the image to be detected is subjected to nonstandard parking according to the vehicle parking information.

It should be noted that, in the present embodiment, the image capturing device 41 may be a camera module, and the camera module includes a camera device, a storage device, and a processing device. The image capturing device includes but is not limited to: cameras, video cameras, camera modules integrated with optical systems or CCD chips, camera modules integrated with optical systems and CMOS chips, and the like. In this embodiment, the detection device 42 may be a server, where a parking specification detection model obtained by training a deep learning network is deployed, and the server may be arranged on one or more entity servers according to various factors such as functions and loads, or may be formed by a distributed or centralized server cluster, which is not limited in this embodiment.

In some examples, the parking specification detection system further includes a prompting device 43. The prompting device 43 is in communication connection with the detecting device 42, and is used for sending out a prompting message when the detecting device 42 detects that the parking is not standardized, and timely notifying a corresponding responsible party to process the prompting message, so that the city appearance is improved, and the purpose of standardized parking is achieved. The prompting device 43 can optionally send out prompting information in a voice prompt mode, a text prompt mode, an image prompt mode, a vibration mode, or an on/off mode of an indicator light, and the embodiment is not limited.

In some examples, the images to be live are taken periodically by the image capturing devices 41, for example, the image capturing devices take one picture per second and upload the pictures to the detecting device 42.

In some examples, the irregular parking includes a vehicle parking super zone; the manner for determining whether the vehicle in the image to be detected is parked in the super-region by the detection device 42 includes: presetting a corresponding parking area coordinate for a parking area; judging whether the coordinates of the vehicle exceed the coordinates of the parking area in the image according to the parking position coordinate information of the vehicle output by the parking specification detection model; and if the coordinates of the vehicle exceed the coordinates of the parking area, determining that the vehicle is parked in the super area.

In some examples, the irregular stop includes a vehicle rolling; the method for determining whether the vehicle in the image to be detected is inclined by the detection device 42 includes: obtaining the parking angle information of the vehicle output by the parking specification detection model; judging whether the parking angle of the vehicle exceeds a preset angle range or not in the image according to the parking angle information of the vehicle; and if the parking angle of the vehicle exceeds the preset angle range, determining that the vehicle is askew to park.

In some examples, the irregular stop includes a vehicle falling over; the manner for determining whether the vehicle in the image to be detected falls down by the detection device 42 includes: acquiring parking posture information of the vehicle output by the parking specification detection model; and judging whether the vehicle falls down according to the parking position information.

Example four:

fig. 5 is a schematic structural diagram illustrating a construction terminal of a parking specification detection model according to an embodiment of the present invention.

The terminal 500 for constructing the parking specification detection model in this embodiment includes:

the first storage unit 501 is used for storing at least one computer program. Illustratively, the first storage unit 501 may include one or more memories. The memory may include high speed random access memory and may also include non-volatile memory, such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid state storage devices. In certain embodiments, the memory may also include memory that is remote from the one or more processors, such as network attached memory that is accessed via RF circuitry or external ports and a communications network, which may be the internet, one or more intranets, local area networks, wide area networks, storage area networks, and the like, or suitable combinations thereof. The memory controller may control access to the memory by other components of the device, such as the CPU and peripheral interfaces.

A first processing unit 502 for executing the at least one computer program for performing, for example, the method of fig. 1. Illustratively, the first processing unit 502 may include one or more processors, which may be one or more general purpose microprocessors, one or more special purpose processors, one or more field programmable logic arrays, or any combination thereof.

For example, the terminal 500 for constructing the parking specification detection model may be implemented in various processing terminals, such as a server, a desktop computer, a notebook computer, a smart phone, a tablet computer, smart glasses, a smart band, a smart watch, and the like, which is not limited in this embodiment.

Example five:

fig. 6 is a schematic structural diagram of a parking specification detection terminal according to an embodiment of the present invention.

The parking specification detecting terminal 600 in this embodiment includes:

the second storage unit 601 is used for storing at least one computer program. Illustratively, the second storage unit 601 may include one or more memories. The memory may include high speed random access memory and may also include non-volatile memory, such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid state storage devices. In certain embodiments, the memory may also include memory that is remote from the one or more processors, such as network attached memory that is accessed via RF circuitry or external ports and a communications network, which may be the internet, one or more intranets, local area networks, wide area networks, storage area networks, and the like, or suitable combinations thereof. The memory controller may control access to the memory by other components of the device, such as the CPU and peripheral interfaces.

A second processing unit 602 for executing the at least one computer program for performing, for example, the method of fig. 1. Illustratively, the second processing unit 602 may include one or more processors, which may be one or more general purpose microprocessors, one or more special purpose processors, one or more field programmable logic arrays, or any combination thereof.

For example, the parking specification detecting terminal 600 may be implemented in various processing terminals, such as a server, a desktop computer, a notebook computer, a smart phone, a tablet computer, smart glasses, a smart band, a smart watch, and the like, which is not limited in this embodiment.

Example six:

the present embodiment provides a computer-readable storage medium storing at least one computer program, the at least one computer program being executed to execute the method for constructing the parking specification detection model; alternatively, the parking specification detection method is executed.

It will be appreciated that the various functions performed in the foregoing embodiments relate to computer software products; the computer software product is stored in a storage medium, and is used for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention, such as the steps of the flowcharts in the embodiments of the methods in fig. 2 and 3, when the computer software product is executed.

In embodiments provided herein, the computer-readable and writable storage medium may comprise read-only memory, random-access memory, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory, a USB flash drive, a removable hard disk, or any other medium which can be used to store desired program code in the form of instructions or data structures and which can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the instructions are transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. It should be understood, however, that computer-readable-writable storage media and data storage media do not include connections, carrier waves, signals, or other transitory media, but are intended to be non-transitory, tangible storage media. Disk and disc, as used in this application, includes Compact Disc (CD), laser disc, optical disc, Digital Versatile Disc (DVD), floppy disk and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers.

In one or more exemplary aspects, the functions described by the computer program referred to in the method flow of the invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. The steps of a disclosed method or algorithm may be embodied in processor-executable software modules, which may be located on a tangible, non-transitory computer-readable and/or writable storage medium. Tangible, non-transitory computer readable and writable storage media may be any available media that can be accessed by a computer.

The flowcharts and block diagrams in the above-described figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In summary, the invention provides a method, a system, a terminal and a medium for constructing and predicting a parking specification detection model, and provides a parking specification detection scheme based on deep learning, which directly detects a shot public parking area photo through a deep neural network from the perspective of computer vision, identifies whether the phenomena of vehicle disordering, vehicle askew parking, irregular parking or vehicle falling over occur in the photo, and if any one or more of the above phenomena are detected, informs corresponding responsible personnel to correct the situation without manually inspecting the parking area, so that the automatic maintenance of the public parking area is realized, the manpower and material resources for manual inspection are saved, and the market appearance is improved. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.

The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims

1. A method for constructing a parking specification detection model is characterized by comprising the following steps:

collecting images of a parking area;

marking the vehicles in the image;

dividing the marked samples into a training set and a testing set, inputting the training set into a deep learning network for training, and testing the trained deep learning network by using the testing set;

and determining a parking specification detection model according to the test result of the deep learning network so as to perform parking specification detection on the image to be detected.

2. The method of constructing a parking specification detection model according to claim 1, further comprising: and marking the vehicle in the image by adopting an inclined frame marking mode.

3. The method of constructing a parking specification detection model according to claim 1 or 2, wherein the labeling information for labeling the vehicle includes: one or more of coordinate information of the vehicle center point in the image, size information of the vehicle, angle information of vehicle parking, vehicle category information, and vehicle parking posture information; wherein the vehicle stopping posture information comprises posture information of vehicle standing or vehicle falling.

4. A parking specification detection method, comprising:

acquiring an image to be detected;

inputting the image to be detected into a parking standard detection model;

obtaining vehicle parking information output by the parking specification detection model;

and judging whether the vehicle in the image to be detected is subjected to nonstandard parking according to the vehicle parking information.

5. The parking specification detecting method according to claim 4, wherein the vehicle parking information includes one or more of parking position coordinate information, parking angle information, and information on whether to fall down of the vehicle in the image to be detected.

6. The parking specification detection method as claimed in claim 4, wherein said irregular parking includes a vehicle parking super zone; wherein, judge whether the vehicle in the image of waiting to detect parks the mode of super region and include:

presetting a corresponding parking area coordinate for a parking area;

judging whether the coordinates of the vehicle exceed the coordinates of the parking area in the image according to the parking position coordinate information of the vehicle output by the parking specification detection model;

and if so, determining the vehicle parking super-region.

7. The parking specification detection method according to claim 4, wherein the irregular parking includes a vehicle being askew parked; wherein, judge whether the vehicle in the image of waiting to detect askew mode of stopping include:

obtaining the parking angle information of the vehicle output by the parking specification detection model;

judging whether the parking angle of the vehicle exceeds a preset angle range or not in the image according to the parking angle information of the vehicle;

and if so, determining that the vehicle is askew to stop.

8. The parking specification detection method according to claim 4, wherein the irregular parking includes a vehicle falling over; wherein, judge whether the mode that the vehicle in the image of waiting to detect falls over includes:

acquiring parking posture information of the vehicle output by the parking specification detection model;

and judging whether the vehicle falls down according to the parking position information.

9. A parking code detection system, comprising:

the image acquisition device is used for acquiring a real-time image corresponding to the parking area;

the detection device is in communication connection with the image acquisition device; wherein the detection device receives real-time images from the image acquisition device; inputting the real-time image into a parking specification detection model obtained by training a deep learning network so as to obtain vehicle parking information output by the parking specification detection model; and judging whether the vehicle in the image to be detected is subjected to nonstandard parking according to the vehicle parking information.

10. A construction terminal of a parking specification detection model is characterized by comprising:

a first storage unit for storing at least one computer program;

a first processing unit for executing the at least one computer program to perform the method of constructing a parking specification detection model according to any one of claims 1 to 3.

11. A parking regulation detection terminal, comprising:

a second storage unit for storing at least one computer program;

a second processing unit for executing the at least one computer program to perform the parking specification detection method according to any one of claims 4 to 8.

12. A computer-readable storage medium, in which at least one computer program is stored, the at least one computer program being executed to perform a method of constructing a parking specification detection model according to any one of claims 1 to 3; alternatively, the parking specification detection method according to any one of claims 4 to 8 is performed.