CN112419733B - Detection method, device, equipment and storage medium for user nonstandard parking - Google Patents
Detection method, device, equipment and storage medium for user nonstandard parking Download PDFInfo
- Publication number
- CN112419733B CN112419733B CN202011108270.6A CN202011108270A CN112419733B CN 112419733 B CN112419733 B CN 112419733B CN 202011108270 A CN202011108270 A CN 202011108270A CN 112419733 B CN112419733 B CN 112419733B
- Authority
- CN
- China
- Prior art keywords
- parking
- detection frame
- vehicle
- detection
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Abstract
The invention discloses a detection method, a device, equipment and a storage medium for nonstandard parking of a user, wherein the method comprises the following steps: acquiring an image about a parking area shot by a camera; identifying a first detection frame in which an unnormalized parked target vehicle is located and a second detection frame containing personnel from the image; and determining the person responsible for the nonstandard parking based on the position relation of the first detection frame and the second detection frame in the image. According to the method, the first detection frame and the second detection frame containing the personnel of the target vehicle which are not normally parked are identified from the image, the relationship between the non-normal parking and the user is carried out by utilizing the position relationship between the two detection frames, and the person responsible for the non-normal parking is determined, so that data support is provided for the treatment of the non-normal parking, the operation and maintenance are convenient to carry out punishment or limitation measures of the related person responsible, and the aim of improving the management effect of the non-normal parking is achieved.
Description
Technical Field
The present invention relates to the field of machine vision, and in particular, to a method, an apparatus, a device, and a storage medium for detecting a user's nonstandard parking.
Background
The shared vehicles include shared automobiles, shared bicycles, shared mopeds and the like. Currently, shared vehicles are generally classified into two parking modes, one is to stop immediately after use and the other is to stop with a fixed parking pile. The mode of stopping immediately after being used can be greatly convenient for a user to go out, and the vehicle can be parked at any time and any place, but the mode often causes the phenomenon that the vehicle is stopped at random, and the urban capacity is seriously affected. At present, for shared vehicles, particularly shared bicycles, a public parking area is usually defined, and a user can park at any time in a relatively flexible parking mode without fixing a vehicle pile. However, irregular parking phenomena such as disordered parking, askew parking, irregular parking or vehicle falling are avoided, and the phenomena not only affect urban volumes, but also prevent other people from orderly parking. The inventor believes that although there are many schemes for detecting the non-standard parking behavior of the shared vehicle, if the non-standard parking behavior cannot be associated with a corresponding responsible person, and some credit penalties are carried out on the responsible person through a corresponding shared bicycle credit score mechanism, and the vehicle of the user is limited after the user does not park normally for many times, the consciousness of the user for parking the shared vehicle is difficult to be fundamentally improved, and the occurrence of the phenomenon is fundamentally reduced. At present, no technical scheme is disclosed to determine the responsible person who does not stop normally.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the present invention aims to solve the technical problem that the existing detection of the nonstandard parking behavior does not determine the responsible person of the nonstandard parking, so that the treatment effect of the nonstandard parking is not obvious.
In order to achieve the above object, the present invention provides a method for detecting a user's nonstandard parking, which is applied to a server and is divided into parking areas for parking vehicles in advance, wherein the parking areas are correspondingly provided with at least one camera for acquiring images in the parking areas, and the method comprises: acquiring an image of a parking area shot by the camera; identifying a first detection frame in which an unnormalized parked target vehicle is located and a second detection frame containing personnel from the image; and determining a person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
In a preferred embodiment of the present invention, the identifying, from the image, a first detection frame in which the target vehicle that is not parked normally is located includes: detecting a detection frame for parking the vehicle from the image; judging whether the coordinates of the detection frame for parking the vehicle are in a preset range or not, wherein the preset range is a region of a parking point position defined in the parking region; and when the coordinates are not in the preset range, determining that the parked vehicle is a target vehicle which is not parked normally, wherein a detection frame of the parked vehicle is used as the first detection frame.
In a preferred embodiment of the present invention, further comprising: judging whether the headstock of the parked vehicle accords with the headstock orientation preset in the parking area or not; and when the head orientation preset in the parking area is not met, determining that the parked vehicle is a target vehicle which is not parked normally, wherein the detection frame of the parked vehicle is used as the first detection frame.
In a preferred embodiment of the present invention, the detecting frame for detecting a parked vehicle from the image includes: inputting the image into a target detection model obtained by pre-training to obtain an initial detection frame output by the target monitoring model; and filtering the initial detection frames by utilizing the intersection ratio and the distance between the initial detection frames, and removing repeated initial detection frames to obtain the detection frames for parking the vehicle.
In a preferred embodiment of the present invention, the determining the person responsible for the non-standard parking based on the positional relationship between the first detection frame and the second detection frame in the image includes: calculating the distances between the center points of all the second detection frames and the center point of the first detection frame; judging whether the distance is smaller than or equal to a preset distance threshold value; when a second detection frame with the distance smaller than or equal to the preset distance threshold exists, judging the person in the second detection frame with the smallest distance from the center point of the first detection frame as a suspected responsible person without stopping normally.
In a preferred embodiment of the present invention, further comprising: judging whether a face detection result exists in a second detection frame corresponding to the suspected responsible person; and when a face detection result exists in the second detection frame corresponding to the suspected responsible person, extracting a face picture from the second detection frame corresponding to the suspected responsible person.
In a preferred embodiment of the present invention, when there is no face detection result in the second detection frame corresponding to the suspected responsible person, an image including the second detection frame corresponding to the suspected responsible person is stored; and when a face detection result exists in the second detection frame corresponding to the suspected responsible person, storing the first detection frame, the second detection frame corresponding to the suspected responsible person and the extracted face picture.
In a preferred embodiment of the present invention, the image is M images, and the determining the person responsible for the nonstandard parking based on the positional relationship between the first detection frame and the second detection frame in the image includes: judging whether N images in M images detect an irregular parking vehicle at the same position of the target vehicle, and judging whether P images in the M images detect suspected liability persons corresponding to the irregular parking vehicle, wherein M, N and P are natural numbers larger than 1, M is larger than N, and M is larger than P; and when N images in the M images detect the non-standard parking vehicles at the same position of the target vehicle and P images in the M images detect the suspected responsible persons corresponding to the non-standard parking vehicles, determining the suspected responsible persons as the non-standard parking responsible persons.
In order to achieve the above object, the present invention provides a detection device for a user to stop without specification, which is applied to a server and is divided into a parking area for parking a vehicle in advance, wherein the parking area is correspondingly provided with at least one camera for acquiring an image in the parking area, and the device comprises:
the acquisition module is used for acquiring an image about a parking area shot by the camera;
the detection module is used for identifying a first detection frame where the target vehicle which is not parked normally is located and a second detection frame containing personnel from the image;
and the determining module is used for determining the person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
To achieve the above object, the present invention provides a computer apparatus comprising: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to perform the method for detecting the user's non-canonical parking.
To achieve the above object, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the method for detecting non-stop of a user.
The device or the method provided by the invention has the following technical effects:
1. the first detection frame and the second detection frame containing personnel of the target vehicle which is not normally parked are identified from the image by acquiring the image of the parking area shot by the camera, the relationship between the non-normal parking and the user is carried out by utilizing the position relationship between the two detection frames, and the responsible person of the non-normal parking is determined, so that data support is provided for the treatment of the non-normal parking, the operation and maintenance are convenient for punishment or limiting measures of the relevant responsible person, and the aim of improving the management effect of the non-normal parking is fulfilled.
2. The parking position is divided in the parking area, whether the parked vehicle is in a preset range of the parking position is judged, and when the parked vehicle is not in the range, the parked vehicle is determined to be in nonstandard parking. Whether the vehicle is parked normally or not is judged by the area of the parking point, the difficulty of identifying the image information by utilizing machine vision can be reduced, and meanwhile, the accuracy of non-normal parking identification is improved.
3. The vehicle head is used as a judging basis for standard parking, so that the characteristics of the vehicle head in parking are fully utilized, and the accuracy of non-standard vehicle detection is improved.
4. And filtering the repeated detection frames by using the cross-over ratio and the center point distance between the detection frames in a DIoU-NMS mode, and obtaining a final detection result. In conventional NMS, ioU index is commonly used to suppress redundant test boxes (boxes), where overlapping areas are the only factor, often resulting in error suppression for occlusion situations. The DIoU-NMS takes DIoU as a criterion of the NMS because not only the overlapping area but also the center point distance between the two boxes should be considered in the suppression criterion, and DIoU considers both the overlapping area and the center distance of the two boxes at the same time, thereby improving the accuracy of filtering.
5. The distance between the first detection frame and the second detection frame is utilized to determine the position relation between the first detection frame and the second detection frame and determine the responsible person or the suspected responsible person, so that the operation speed can be improved, the response time can be reduced, and the load of an operation chip can be reduced under the condition that the accuracy is not reduced.
The conception, specific structure, and technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, features, and effects of the present invention.
Drawings
FIG. 1 is a flow chart of a preferred embodiment of a method for detecting user-out-of-specification parking in accordance with the present invention;
FIG. 2 is a schematic diagram of a preferred embodiment of the detection result of the present invention;
FIG. 3 is a schematic diagram of a preferred embodiment of a user out-of-specification parking detection device of the present invention;
FIG. 4 is a schematic diagram of the architecture of one embodiment of a computer device of the present invention.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the illustrations, not according to the number, shape and size of the components in actual implementation, and the form, number and proportion of each component in actual implementation may be arbitrarily changed, and the layout of the components may be more complex.
Some exemplary embodiments of the invention have been described for illustrative purposes, it being understood that the invention may be practiced otherwise than as specifically shown in the accompanying drawings.
The embodiment of the invention provides a detection method for nonstandard parking of a user, which is applied to a server, is executed by the server, and is used for carrying out subsequent processing based on image data acquired by a camera at the front end. In the embodiment of the invention, a parking area for parking a vehicle is divided in advance, and at least one camera is correspondingly arranged in the parking area and used for collecting images in the parking area. In the embodiment of the invention, the parking areas are defined vehicle parking places, a plurality of parking areas with different positions can be arranged in the same city, and a plurality of parking areas can be arranged in a region range at a place where vehicles are dense. The user can rent the vehicle in the parking area, and then the rented vehicle used elsewhere can be returned in the parking area and parked. The parking area may also be provided with parking locations for individual vehicles, for example, a plurality of parking spots, each for parking a vehicle, being divided in the parking area. It should be noted that, the vehicle related to the embodiment of the present invention may be a shared automobile, a shared bicycle, or a shared booster vehicle. For convenience of description, embodiments of the present invention will be collectively referred to as a vehicle.
In order to realize detection of vehicle parking, at least one camera is erected in each parking area and used for shooting and collecting images of the corresponding parking area. Each camera is in communication connection with the server and is used for transmitting the acquired image data or video data back to the server, and the cameras can be network cameras. According to the embodiment of the invention, the data obtained by shooting the camera can be subjected to the processes of data acquisition, data cleaning, data labeling and model training to obtain a target detection model, such as a yolov4 target detection network. That is, in the embodiment of the present invention, a large amount of image data captured by a camera may be taken as a sample in advance, and a corresponding model may be constructed by using a deep learning technique and a machine vision technique, and training and learning may be performed to obtain a target detection model that may be used for target detection. The target detection model can be used for detecting targets such as various brands of bicycles, moped vehicles, pedestrians, riders, faces and the like in a parking area, and can detect the directions (up, down, left and right) of the heads of various brands of bicycles and moped vehicles in images, and on the basis, the recognition of vehicles and personnel is carried out.
The detection method for the nonstandard parking of the user is mainly used for detecting the nonstandard parking vehicle and the corresponding responsible person and correlating the nonstandard parking vehicle with the corresponding responsible person, so that the nonstandard parking responsible person is determined, and corresponding punishment or limitation measures are adopted for the responsible person, so that the aim of reminding the user of the standard parking is fulfilled.
As shown in fig. 1, the method for detecting the user's non-standard parking in the embodiment of the invention includes:
step S101, acquiring an image about a parking area, which is shot by the camera.
The camera can shoot the image of the parking area at regular time according to a time period, and because the video data also comprises a frame of image, in the embodiment of the invention, the camera can also be the shot video data, that is, the image of the parking area can be a multi-frame image set. For example, the camera transmits the acquired images to the server every 1s, and the server receives the images so as to facilitate subsequent processing and judgment.
Step S102, a first detection frame where the target vehicle which is not parked normally is located and a second detection frame containing personnel are identified from the image.
In the embodiment of the invention, the target vehicle detection can be directly carried out from the image, namely the vehicle which is not stopped in a standard way, and the first detection frame where the target vehicle is positioned is determined. Meanwhile, image recognition is directly carried out to identify a second detection frame containing personnel. That is, after an image on a parking area is acquired, a vehicle (target vehicle) that is not normally parked and a person in the image are identified from the image, the first detection frame indicates the position of the target vehicle in the image, and the second detection frame indicates the position of the person in the image.
Optionally, in the embodiment of the present invention, the server may perform target detection by using the above-mentioned middle target detection model (such as yolov4 target detection model), take the acquired image as an input of the target detection model, output all detection frames of each brand of bicycle, each brand of moped, pedestrian, and rider after passing through the target detection model, where a specific format of each detection result is [ a horizontal coordinate of a target center point in the image, a vertical coordinate of a target center point in the image, a width of the target, a height of the target, a confidence of the target, and a specific category of the target ], and the finally obtained visual effect is shown in fig. 2.
The detection frame of each vehicle is detected by using the target detection model, and then the detection frame is used as a first detection frame after judging whether the vehicle is normally parked. The detection frame for the person is then directly used as a second detection frame.
Step S103, determining a person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
In the embodiment of the invention, after the first detection frame and the second detection frame are detected, the position relationship between the first detection frame and the second detection frame can be utilized to determine the person responsible for the nonstandard parking. The inventors have found that when capturing an image of a parking area, a user using a vehicle is typically able to capture an image of the association between the user and the vehicle during parking. A main feature of the associated image is that the positional relationship between the user and the vehicle is relatively tight. Based on the characteristics, in the embodiment of the invention, when the first detection frame of the target vehicle quantity which is not parked normally and the second detection frame containing the personnel are detected, whether the personnel is a user which is not parked normally can be determined through the position relation between the first detection frame and the second detection frame.
Specifically, the determination may be made using the IoU index between the first and second detection frames. Among them, ioU is called the intersection ratio (Intersection over Union), and IoU calculates the ratio of the intersection and union of the "predicted border" and the "real border". The embodiment of the invention utilizes the calculation mode and carries out corresponding adjustment to calculate the ratio of the intersection between the first detection frame and the second detection frame to the union. Wherein, the value range of IoU is 0-1, and when the value of IoU is larger, the overlapping area of the first detection frame and the second detection frame is larger, and the positions are closer; otherwise, the farther the position is. Optionally, in the embodiment of the present invention, the distance between the center points of the first detection frame and the second detection frame may be directly used to determine the positional relationship between the two, so as to determine the responsible person who does not normally park.
In the embodiment of the invention, the person in the second detection frame with the closest position can be used as the user for using and parking the target vehicle in the first detection frame, namely, the person responsible for non-standard parking; also, a responsible person may be present. In the embodiment of the invention, the person in the second detection frame with the closest position can be used as a suspected responsible person, then the suspected responsible person is detected for multiple times, and if the multiple times meet the same condition, the person is determined to be the responsible person; or, in combination with some behavior detection of the user, auxiliary judgment is performed, for example, the position relationship between the hand of the user and the vehicle is detected, or the operation action between the hand and the vehicle is determined as a responsible person, so that the accuracy of the detection judgment is improved.
According to the embodiment of the invention, the first detection frame and the second detection frame containing the personnel of the target vehicle which is not normally parked are identified from the image by acquiring the image of the parking area shot by the camera, the relationship between the non-normally parked and the user is carried out by utilizing the position relationship between the two detection frames, and the responsible person of the non-normally parked is determined, so that the data support is provided for the treatment of the non-normally parked, the operation and maintenance are convenient for punishment or limitation measures of the relevant responsible person, and the aim of improving the management effect of the non-normally parked is fulfilled.
It should be noted that although there may be a technical means for performing the target detection in the prior art, a main invention point of the present invention is that the first detection frame and the second detection frame are used to associate the non-standard parking with the user, so as to provide data support for the vehicle parking management.
As an optional implementation manner, in the step S102, the identifying, from the image, the first detection frame in which the target vehicle that is not parked in a standard manner is located includes: detecting a detection frame for parking the vehicle from the image; judging whether the coordinates of the detection frame for parking the vehicle are in a preset range or not, wherein the preset range is a region of a parking point position defined in the parking region; and when the coordinates are not in the preset range, determining that the parked vehicle is a target vehicle which is not parked normally, wherein a detection frame of the parked vehicle is used as the first detection frame.
In the embodiment of the invention, a parking point can be defined in each parking area, and each parking point is used for parking a vehicle. In the embodiment of the invention, the camera shoots images aiming at the parking area, so that the visual angle is fixed, and the shot image range is fixed. In this way, the position of each parking spot in the photographed image is also a relatively fixed position, and the range of each parking spot can be determined from the pixel coordinates in the image, with the pixel range of each parking spot as a preset range. After detecting the detection frame of the parked vehicle, it may be determined whether the detection frame is within a preset range. Because the preset range is a plurality of, the detection frame and the preset range of each parking point position need to be matched and judged, and the range of the parking point position closest to the position can be preferentially selected for judgment, so that the data volume of matching processing is reduced.
When the coordinates of the detection frame are not within the preset range of any parking position, it can be determined that the individual vehicle is not parked according to the fixed parking position, and the phenomenon of non-standard parking is caused, so that the corresponding detection frame is used as the first detection frame, and the vehicle is used as the target vehicle for non-standard parking. Otherwise, the vehicle is considered to be parked as normal, or other determinations are made later.
According to the embodiment of the invention, the parking point is divided in the parking area, then whether the parked vehicle is in the preset range of the parking point is judged, and when the parked vehicle is not in the range, the parked vehicle is determined to be in nonstandard parking. Whether the vehicle is parked normally or not is judged by the area of the parking point, the difficulty of identifying the image information by utilizing machine vision can be reduced, and meanwhile, the accuracy of non-normal parking identification is improved.
As an alternative implementation manner, in an embodiment of the present invention, after obtaining a detection frame of a vehicle, the method further includes: judging whether the headstock of the parked vehicle accords with the headstock orientation preset in the parking area or not; and when the head orientation preset in the parking area is not met, determining that the parked vehicle is a target vehicle which is not parked normally, wherein the detection frame of the parked vehicle is used as the first detection frame.
The inventor finds that the headstock is often a critical judging part for nonstandard parking, and the headstock has the characteristic of obvious difference when the headstock is in nonstandard parking and standard parking, can be identified in an image through the headstock in a detection frame of the vehicle, and then judges whether the headstock accords with the headstock orientation preset in a parking area or not; when the vehicle is out of compliance, determining that the vehicle is out of specification for parking; otherwise, it may be considered canonical or may be judged later.
In the embodiment of the invention, the headstock is used as the judging basis for standard parking, so that the characteristics of the headstock during parking are fully utilized, and the accuracy of non-standard vehicle detection is improved.
It should be noted that in the above embodiment of the present invention, no technical conflict exists between the preset range of the parking spot and the vehicle head judgment mode, and in the process of applying the scheme, the two schemes may be used together, or one scheme may be selected to perform the judgment of the nonstandard vehicle, which is not limited by the present invention.
Optionally, in an embodiment of the present invention, the detecting frame for detecting a parked vehicle from the image includes: inputting the image into a target detection model obtained by pre-training to obtain an initial detection frame output by the target monitoring model; and filtering the initial detection frames by utilizing the intersection ratio and the distance between the initial detection frames, and removing repeated initial detection frames to obtain the detection frames for parking the vehicle.
In the embodiment of the invention, the repeated detection frames can be filtered by using the cross-over ratio and the center point distance between the detection frames in a DIoU-NMS mode, and a final detection result is obtained. In conventional NMS, ioU index is commonly used to suppress redundant test boxes (boxes), where overlapping areas are the only factor, often resulting in error suppression for occlusion situations. The DIoU-NMS takes DIoU as a criterion of the NMS because not only the overlapping area but also the center point distance between the two boxes should be considered in the suppression criterion, and DIoU considers both the overlapping area and the center distance of the two boxes at the same time, thereby improving the accuracy of filtering.
The inventor finds that when the responsible person judges, the calculation process with IoU index needs large calculation amount, so that the calculation pressure is brought to the calculation chip in the server, and the response speed is relatively slow. Therefore, as an invention point of the embodiment of the present invention, the distance between the first detection frame and the second detection frame is used to determine the positional relationship between the two and determine the responsible person or the suspected responsible person, so that the operation speed can be increased, the response time can be reduced, and the load of the operation chip can be reduced without reducing the accuracy.
Specifically, in the embodiment of the present invention, the determining, based on the positional relationship between the first detection frame and the second detection frame in the image, a person responsible for non-standard parking includes: calculating the distances between the center points of all the second detection frames and the center point of the first detection frame; judging whether the distance is smaller than or equal to a preset distance threshold value; when a second detection frame with the distance smaller than or equal to the preset distance threshold exists, judging the person in the second detection frame with the smallest distance from the center point of the first detection frame as a suspected responsible person without stopping normally.
If the second detection frames of the personnel exist, for each nonstandard parking vehicle, firstly calculating the distances between all the second detection frames and the central points of the first detection frames of the nonstandard parking vehicle, if the calculated distances between all the central points are larger than a preset distance threshold value, continuing to process the next picture, and if the calculated distances between the central points are smaller than the preset distance threshold value, taking the first detection frame with the smallest distance between the first detection frame and the central point of the first detection frame as a suspected responsible person of the nonstandard parking vehicle.
In the embodiment of the invention, the complexity of calculating the center point distance is lower than that of calculating IoU, so that the response time can be improved and the load of an operation chip can be reduced without reducing the accuracy.
Further, after determining the suspected responsible person, the embodiment of the invention further comprises: judging whether a face detection result exists in a second detection frame corresponding to the suspected responsible person; and when a face detection result exists in the second detection frame corresponding to the suspected responsible person, extracting a face picture from the second detection frame corresponding to the suspected responsible person. When the second detection frame corresponding to the suspected responsible person does not have a face detection result, storing an image containing the second detection frame corresponding to the suspected responsible person; and when a face detection result exists in the second detection frame corresponding to the suspected responsible person, storing the first detection frame, the second detection frame corresponding to the suspected responsible person and the extracted face picture.
In the embodiment of the invention, whether the face detection results exist in all the detection frames can be judged first, if yes, whether the face detection results exist in a suspected responsible person detection frame (namely, a second detection frame which is judged to be a suspected responsible person) of the irregular parking or not can also be judged, if not, the current image is directly saved, and if yes, the face picture is extracted, and the detection frame results and the current image of all the irregular parking vehicles and the irregular parking responsible persons detected in the current frame are saved.
According to the embodiment of the invention, after the suspected responsible person is judged, the face picture is detected and stored, so that the identity of the specific responsible person can be conveniently and subsequently identified.
In the embodiment of the invention, the images are M, that is, when the images are identified, the responsible person who does not stop normally can be determined after the images are identified.
Specifically, the determining, based on the positional relationship between the first detection frame and the second detection frame in the image, a responsible person who does not normally park includes: judging whether N images in M images detect an irregular parking vehicle at the same position of the target vehicle, and judging whether P images in the M images detect suspected liability persons corresponding to the irregular parking vehicle, wherein M, N and P are natural numbers larger than 1, M is larger than N, and M is larger than P; and when N images in the M images detect the non-standard parking vehicles at the same position of the target vehicle and P images in the M images detect the suspected responsible persons corresponding to the non-standard parking vehicles, determining the suspected responsible persons as the non-standard parking responsible persons.
The M images may be a plurality of continuous images, and when the irregular vehicle is judged, the M images may be judged one by one, and after the first image is judged, the second image is judged. The specific judging process can be adopted in the embodiment of the invention. Similarly, the embodiments described in the above examples may be used for determination of a person who is suspected to be responsible. When N images in the M images are judged to detect the non-standard parking vehicles at the same position of the target vehicle and P images in the M images detect the suspected responsible persons corresponding to the non-standard parking vehicles, the suspected responsible persons can be completely determined to be the non-standard parking responsible persons in terms of frequency and statistical probability.
For example, considering that there is a situation that a pedestrian may just pass through an irregular parking vehicle, the detection results in five pictures corresponding to the next five seconds are taken to make the following judgment: (1) Next, at least three of the five pictures detect an out-of-specification vehicle at the same location as the first out-of-specification vehicle. When the two conditions are met, the suspected responsible person is determined to be the responsible person of the nonstandard parking, and the corresponding face information of the pictures after the pictures form the video is stored in a related database.
In the embodiment of the invention, when the nonstandard parking behavior of the shared bicycle of the user is detected, video storage is carried out on the user behavior of the nonstandard parking, the face photo of the responsible person is extracted, and the face photo is corresponding to the responsible person of the nonstandard parking through a face recognition method, so that the vehicle of the user for limiting the nonstandard parking for a plurality of times is laid.
The embodiment of the invention also provides a detection device for the nonstandard parking of the user, which is applied to the server, and is divided into parking areas for parking the vehicles in advance, wherein the parking areas are correspondingly provided with at least one camera and are used for collecting images in the parking areas. The apparatus may be used to perform the method described in the above embodiments, as shown in fig. 3, where the apparatus includes:
an acquiring module 301, configured to acquire an image about a parking area captured by the camera.
And the detection module 302 is used for identifying a first detection frame where the target vehicle which is not parked normally is located and a second detection frame containing personnel from the image.
And the determining module 303 is used for determining the person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
According to the embodiment of the invention, the first detection frame and the second detection frame containing the personnel of the target vehicle which is not normally parked are identified from the image by acquiring the image of the parking area shot by the camera, the relationship between the non-normally parked and the user is carried out by utilizing the position relationship between the two detection frames, and the responsible person of the non-normally parked is determined, so that the data support is provided for the treatment of the non-normally parked, the operation and maintenance are convenient for punishment or limitation measures of the relevant responsible person, and the aim of improving the management effect of the non-normally parked is fulfilled.
The specific description of the above device may be referred to the above method embodiments, and will not be repeated here.
In one embodiment of the present invention, there is further provided a computer device, which may be a server in the above embodiment, or may refer to an intelligent device installed or disposed on a vehicle, and an internal structure thereof may be as shown in fig. 4. The computer device includes a processor, memory, a network interface, and may also include a display screen and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external computer device through a network connection. The computer program when executed by the processor is used for realizing a vehicle control method, the computer equipment also comprises a display screen and an input device, wherein the display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen or a key, a track ball or a touch pad arranged on the shell of the computer equipment.
On the other hand, the computer device may not include a display screen and an input device, and those skilled in the art will understand that the structure shown in fig. 4 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation of the computer device to which the present application is applied, and a specific computer device may include more or less components than those shown in the drawings, or may combine some components, or have different component arrangements.
In one embodiment, a computer device is provided that includes at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to perform the steps of:
acquiring an image of a parking area shot by the camera;
identifying a first detection frame in which an unnormalized parked target vehicle is located and a second detection frame containing personnel from the image;
and determining a person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
In one embodiment, a readable storage medium is provided, the computer readable storage medium storing computer instructions for causing the computer to perform:
acquiring an image of a parking area shot by the camera;
identifying a first detection frame in which an unnormalized parked target vehicle is located and a second detection frame containing personnel from the image;
and determining a person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
The foregoing describes in detail preferred embodiments of the present invention. It should be understood that numerous modifications and variations can be made in accordance with the concepts of the invention without requiring creative effort by one of ordinary skill in the art. Therefore, all technical solutions which can be obtained by logic analysis, reasoning or limited experiments based on the prior art by the person skilled in the art according to the inventive concept shall be within the scope of protection defined by the claims.
Claims (11)
1. The utility model provides a detection method that user does not standardize parks, its characterized in that is applied to the server, has divided in advance and is used for parking the parking area of vehicle, the vehicle is shared vehicle, parking area corresponds to and sets up at least one camera for gather the image in this parking area, the method includes:
acquiring an image of a parking area shot by the camera;
identifying a first detection frame in which an unnormalized parked target vehicle is located and a second detection frame containing personnel from the image;
and determining a person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
2. The method for detecting irregular parking of a user according to claim 1, wherein the first detection frame in which the target vehicle which is irregular parked is identified from the image includes:
detecting a detection frame for parking the vehicle from the image;
judging whether the coordinates of the detection frame for parking the vehicle are in a preset range or not, wherein the preset range is a region of a parking point position defined in the parking region;
and when the coordinates are not in the preset range, determining that the parked vehicle is a target vehicle which is not parked normally, wherein a detection frame of the parked vehicle is used as the first detection frame.
3. The method for detecting non-standard parking of a user according to claim 2, further comprising:
judging whether the headstock of the parked vehicle accords with the headstock orientation preset in the parking area or not;
and when the head orientation preset in the parking area is not met, determining that the parked vehicle is a target vehicle which is not parked normally, wherein the detection frame of the parked vehicle is used as the first detection frame.
4. A method for detecting a user's non-stop as claimed in claim 2 or 3, wherein the detecting a detection frame of a parked vehicle from the image comprises:
inputting the image into a target detection model obtained by pre-training to obtain an initial detection frame output by the target monitoring model;
and filtering the initial detection frames by utilizing the intersection ratio and the distance between the initial detection frames, and removing repeated initial detection frames to obtain the detection frames for parking the vehicle.
5. The method for detecting irregular parking of a user according to claim 1, wherein the determining of the person responsible for irregular parking based on the positional relationship of the first detection frame and the second detection frame in the image includes:
calculating the distances between the center points of all the second detection frames and the center point of the first detection frame;
judging whether the distance is smaller than or equal to a preset distance threshold value;
when a second detection frame with the distance smaller than or equal to the preset distance threshold exists, judging the person in the second detection frame with the smallest distance from the center point of the first detection frame as a suspected responsible person without stopping normally.
6. The method for detecting a user-out-of-specification parking of claim 5, further comprising:
judging whether a face detection result exists in a second detection frame corresponding to the suspected responsible person;
and when a face detection result exists in the second detection frame corresponding to the suspected responsible person, extracting a face picture from the second detection frame corresponding to the suspected responsible person.
7. The method for detecting irregular parking of a user according to claim 6, wherein,
when the second detection frame corresponding to the suspected responsible person does not have a face detection result, storing an image containing the second detection frame corresponding to the suspected responsible person;
and when a face detection result exists in the second detection frame corresponding to the suspected responsible person, storing the first detection frame, the second detection frame corresponding to the suspected responsible person and the extracted face picture.
8. The method for detecting irregular parking of a user according to claim 7, wherein the image is M sheets, and the determining the responsible person for irregular parking based on the positional relationship of the first detection frame and the second detection frame in the image includes:
judging whether N images in M images detect an irregular parking vehicle at the same position of the target vehicle, and judging whether P images in the M images detect suspected liability persons corresponding to the irregular parking vehicle, wherein M, N and P are natural numbers larger than 1, M is larger than N, and M is larger than P;
and when N images in the M images detect the non-standard parking vehicles at the same position of the target vehicle and P images in the M images detect the suspected responsible persons corresponding to the non-standard parking vehicles, determining the suspected responsible persons as the non-standard parking responsible persons.
9. The utility model provides a detection device that user's is not standard parks, its characterized in that is applied to the server, has divided in advance and is used for parking the parking area of vehicle, the vehicle is the sharing vehicle, the parking area corresponds and sets up at least one camera for gather the image in this parking area, the device includes:
the acquisition module is used for acquiring an image about a parking area shot by the camera;
the detection module is used for identifying a first detection frame where the target vehicle which is not parked normally is located and a second detection frame containing personnel from the image;
and the determining module is used for determining the person responsible for non-standard parking based on the position relation of the first detection frame and the second detection frame in the image.
10. A computer device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to perform the method of detecting user-non-canonical stops of any of claims 1-8.
11. A computer-readable storage medium storing computer instructions for causing a computer to perform the method of detecting a user-non-canonical stop of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011108270.6A CN112419733B (en) | 2020-10-16 | 2020-10-16 | Detection method, device, equipment and storage medium for user nonstandard parking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011108270.6A CN112419733B (en) | 2020-10-16 | 2020-10-16 | Detection method, device, equipment and storage medium for user nonstandard parking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112419733A CN112419733A (en) | 2021-02-26 |
| CN112419733B true CN112419733B (en) | 2023-05-05 |
Family
ID=74853999
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011108270.6A Active CN112419733B (en) | 2020-10-16 | 2020-10-16 | Detection method, device, equipment and storage medium for user nonstandard parking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112419733B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113055823B (en) * | 2021-04-01 | 2023-07-11 | 超级视线科技有限公司 | Method and device for managing shared bicycle based on road side parking |
| CN113920741A (en) * | 2021-07-01 | 2022-01-11 | 上海市城市建设设计研究总院（集团）有限公司 | Linkage device for standard parking of non-motor vehicles on sidewalk |
| CN113470080A (en) * | 2021-07-20 | 2021-10-01 | 浙江大华技术股份有限公司 | Illegal behavior identification method |
| CN113762272A (en) * | 2021-09-10 | 2021-12-07 | 北京精英路通科技有限公司 | Road information determination method and device and electronic equipment |
| CN114255584B (en) * | 2021-12-20 | 2023-04-07 | 济南博观智能科技有限公司 | Positioning method and system for parking vehicle, storage medium and electronic equipment |
| CN115641727B (en) * | 2022-10-18 | 2023-06-16 | 重庆交通大学 | Parking space management method based on video identification |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003340A (en) * | 2017-05-31 | 2018-12-14 | 李刚 | Bicycle parking management method and its system |
| WO2019148683A1 (en) * | 2018-01-30 | 2019-08-08 | 深圳市元征软件开发有限公司 | Vehicle sharing method and server |
| CN110337064A (en) * | 2019-07-04 | 2019-10-15 | 南京大学 | A kind of shared bicycle parking management system based on bluetooth and machine vision |
| CN110491135A (en) * | 2019-08-20 | 2019-11-22 | 深圳市商汤科技有限公司 | Detect the method and relevant apparatus of parking offense |
| CN111339994A (en) * | 2019-12-31 | 2020-06-26 | 智慧互通科技有限公司 | Method and device for judging temporary illegal parking |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7355527B2 (en) * | 2005-01-10 | 2008-04-08 | William Franklin | System and method for parking infraction detection |
| US11244171B2 (en) * | 2014-01-22 | 2022-02-08 | Conduent Business Services Llc | Video-based system for automated detection of double parking violations |
| US9997071B2 (en) * | 2014-12-02 | 2018-06-12 | Operr Technologies, Inc. | Method and system for avoidance of parking violations |
| US11132538B2 (en) * | 2016-06-22 | 2021-09-28 | Sony Corporation | Image processing apparatus, image processing system, and image processing method |
| CN107230346B (en) * | 2017-08-01 | 2019-05-31 | 何永安 | Confirmation method, device, server and the storage medium of shared bicycle parking specification |
| CN109547917A (en) * | 2017-08-19 | 2019-03-29 | 包小波 | The handover of shared bicycle and parking method and handover identifying system |
| CN115830837A (en) * | 2017-12-17 | 2023-03-21 | 杭州青奇科技有限公司 | Method and system for guiding user to park |
| CN109389829A (en) * | 2018-11-05 | 2019-02-26 | 湖北民族学院 | A kind of illegal parking recognition methods and system |
| CN111223289B (en) * | 2018-11-27 | 2021-04-27 | 杭州海康威视数字技术股份有限公司 | Method and system for snapshot of illegal parking event of shared vehicle and storage medium |
| CN109800696A (en) * | 2019-01-09 | 2019-05-24 | 深圳中兴网信科技有限公司 | Monitoring method, system and the computer readable storage medium of target vehicle |
| CN209486891U (en) * | 2019-03-28 | 2019-10-11 | 武汉大学 | Shared bicycle, which is disobeyed, stops detection system |
| CN110135356A (en) * | 2019-05-17 | 2019-08-16 | 北京百度网讯科技有限公司 | The detection method and device of parking offense, electronic equipment, computer-readable medium |
| CN111563469A (en) * | 2020-05-13 | 2020-08-21 | 南京师范大学 | Method and device for identifying irregular parking behaviors |
| CN111723718A (en) * | 2020-06-11 | 2020-09-29 | 杭州智行星科技有限公司 | Face recognition-based car returning confirmation method, system, terminal and storage medium |
-
2020
- 2020-10-16 CN CN202011108270.6A patent/CN112419733B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003340A (en) * | 2017-05-31 | 2018-12-14 | 李刚 | Bicycle parking management method and its system |
| WO2019148683A1 (en) * | 2018-01-30 | 2019-08-08 | 深圳市元征软件开发有限公司 | Vehicle sharing method and server |
| CN110337064A (en) * | 2019-07-04 | 2019-10-15 | 南京大学 | A kind of shared bicycle parking management system based on bluetooth and machine vision |
| CN110491135A (en) * | 2019-08-20 | 2019-11-22 | 深圳市商汤科技有限公司 | Detect the method and relevant apparatus of parking offense |
| CN111339994A (en) * | 2019-12-31 | 2020-06-26 | 智慧互通科技有限公司 | Method and device for judging temporary illegal parking |
Non-Patent Citations (1)
| Title |
|---|
| 各地如何解决共享单车停放问题.汽车与安全.2017,(第07期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112419733A (en) | 2021-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112419733B (en) | Detection method, device, equipment and storage medium for user nonstandard parking | |
| US10762338B2 (en) | Method and apparatus for detecting fake license plates of vehicles, readable storage medium, and electronic device | |
| JP7218535B2 (en) | Traffic violation vehicle identification system and server | |
| CN105702048B (en) | Highway front truck illegal road occupation identifying system based on automobile data recorder and method | |
| TWI497422B (en) | A system and method for recognizing license plate image | |
| CN112349144B (en) | Monocular vision-based vehicle collision early warning method and system | |
| CN107688764B (en) | Method and device for detecting vehicle violation | |
| CN111815959B (en) | Vehicle violation detection method and device and computer readable storage medium | |
| CN110298300B (en) | Method for detecting vehicle illegal line pressing | |
| CN111891061B (en) | Vehicle collision detection method and device and computer equipment | |
| CN112487908B (en) | Front vehicle line pressing behavior detection and dynamic tracking method based on vehicle-mounted video | |
| CN104574993B (en) | A kind of method of road monitoring and device | |
| CN113055823B (en) | Method and device for managing shared bicycle based on road side parking | |
| CN111311766A (en) | Roadside parking intelligent charging system and method based on license plate recognition and tracking technology | |
| CN106650730A (en) | Turn signal lamp detection method and system in car lane change process | |
| CN112818834A (en) | Method, device and medium for judging avoidance of emergency vehicle at intersection | |
| JP3541774B2 (en) | Vehicle type identification system | |
| CN114882393B (en) | Road reverse running and traffic accident event detection method based on target detection | |
| CN110796099A (en) | Vehicle overrun detection method and device | |
| CN111582171B (en) | Pedestrian red light running monitoring method, device and system and readable storage medium | |
| KR101738025B1 (en) | System for Deciding Accident using Black Box and Driving Method Thereof | |
| CN115880632A (en) | Timeout stay detection method, monitoring device, computer-readable storage medium, and chip | |
| CN107993446A (en) | A kind of traffic prohibition parking area domain parking offense monitoring device | |
| CN113516852A (en) | Multi-parking-space multi-target vehicle parking identification method and system | |
| CN114141022A (en) | Emergency lane occupation behavior detection method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |