System and method for detecting parking violations

Abstract

A system for detecting one or more parking violations includes a streaming gateway configured to receive a video stream from one or more mobile cameras; an image processing module configured to process the received video stream, identify a parked vehicle, identify traffic data associated with the parked vehicle, and identify a timestamp of the identified traffic data; a violation evaluation module configured to determine a parking violation based on the traffic data determined at the two or more timestamps and one or more parking violation rules; and a violation confirmation module configured to confirm the detected parking violation as an actual parking violation based on checking the one or more violation exceptions.

Description

Technical Field

[0001] The present disclosure relates to a system and method for detecting one or more parking violations. Background Art

[0002] In many densely populated cities, parking spaces are very limited. In highly densely populated urban centers, parking spaces on the street are particularly tight. Parking violations caused by illegal parking on the street or in other spaces, such as public parking lots or parking lot buildings or loading and unloading areas, are a problem faced in many cities. Parking violations can include several instances of illegal parking, including but not limited to, parking in non-parking spaces or non-parking areas on the street, parking near intersections or high traffic areas, parking near high pedestrian traffic areas, staying in a parking space for an extended period of time, parking at bus stops or loading and unloading areas.

[0003] Detecting and enforcing parking violations is challenging for relevant departments. Detecting and enforcing parking violations requires many staff members to manually patrol parking spaces and non-parking areas and regularly check these areas to identify parking violations. This manual way of detecting and enforcing parking violations puts certain financial and manpower pressure on government departments. Summary of the invention

[0004] According to a first aspect of the present disclosure, a system for detecting one or more parking violations is provided. The system includes: a streaming gateway configured to receive a video stream from one or more mobile cameras; an image processing module configured to process the received video stream, identify a parked vehicle, identify traffic data associated with the parked vehicle, and identify a timestamp of the identified traffic data; a violation assessment module configured to determine a parking violation based on traffic data determined at two or more timestamps and one or more parking violation rules; and a violation confirmation module configured to confirm a detected parking violation as an actual parking violation based on checking one or more violation exceptions.

[0005] According to a first aspect of the present disclosure, a system for detecting one or more parking violations is provided. The system includes: a streaming gateway configured to receive video streams from a plurality of mobile cameras; an image processing module configured to process the received video streams, identify parked vehicles, identify traffic data associated with the parked vehicles, and identify timestamps of the identified traffic data; a violation assessment module configured to determine a parking violation based on traffic data determined at two or more timestamps and one or more parking violation rules; and a violation confirmation module configured to confirm a detected parking violation as an actual parking violation based on checking one or more violation exceptions.

[0006] In one example, the traffic data includes one or more of the following: vehicle registration information; vehicle brand; vehicle model; vehicle location; and vehicle geographic location.

[0007] In one example, the image processing module is configured to: extract multiple frames from each video stream; identify one or more vehicles in each frame, one or more traffic data associated with the identified vehicles, and a timestamp corresponding to each frame.

[0008] In one example, the image processing module is configured to execute a machine learning network to perform object recognition on each frame, wherein the machine learning network is trained to recognize a vehicle and one or more traffic data associated with the recognized vehicle.

[0009] In one example, the image processing module is configured to execute a trained machine learning network, and the image processing module is configured to: extract multiple frames from each video stream; identify a vehicle in each frame; identify vehicle registration information on a license plate of the vehicle; identify a geographic location based on information from a satellite navigation module; and identify the location of the vehicle.

[0010] In one example, the system includes one or more mobile cameras, each mobile camera being movable or adapted to be mounted on a vehicle or adapted to be mounted on a person, each mobile camera including a network interface for transmitting a recorded video stream.

[0011] In one example, the video stream is a live video stream, and the streaming gateway is configured to receive the live video stream from the one or more mobile cameras and collate the video stream.

[0012] In one example, the violation assessment module is configured to: identify a vehicle at a first timestamp - identify a registration of the vehicle; identify the vehicle at a second timestamp - identify the vehicle registration information; in response to the registrations being the same, verify that the identified vehicles in the first and second timestamps are the same vehicle; determine the geographic location of the vehicle at the first and second timestamps; determine a parking violation based on whether the geographic location of the vehicle between the first and second timestamps has not changed and based on one or more violation rules.

[0013] In one example, the violation rules are one or more customizable rules that define parking violations.

[0014] In one example, the system includes: a violation rule system configured to receive rule input, define one or more rules, and store the one or more rules; and the violation evaluation module configured to access one or more violation rules from the violation rule system to determine a parking violation.

[0015] In one example, the violation rule system includes: a violation rule database; a violation rule interface; a violation rule engine;

[0016] wherein the violation rules interface is configured to present an interface for receiving one or more violation rules from a user;

[0017] The violation rule engine is configured to create one or more violation rules based on input in the violation rule interface; and the violation rule database is configured to store the one or more violation rules.

[0018] In one example, a vehicle is identified at a first time stamp from a first video stream, and a vehicle is identified at a second time stamp from a second video stream.

[0019] In one example, the system includes a violation exception module configured to store one or more predefined violation exceptions.

[0020] In one example, the violation confirmation module is configured to: receive a detected parking violation from the violation assessment module; check one or more violation exceptions from the violation exception module; and generate a parking violation confirmation if no violation exception is applicable, or deny the parking violation if an applicable violation exception is present.

[0021] In one example, the violation exception module is configured to identify one or more violation exceptions within a video stream or within a frame of a video stream based on one or more predefined violation exceptions.

[0022] In one example, the system includes: a parking violation reporting interface configured to: receive parking violation confirmation information from the violation confirmation module; and present the confirmed parking violation on the parking violation reporting interface.

[0023] In one example, the system includes: a reporting module or a prosecution module, which is configured to: receive a reporting input or a prosecution input from a user; automatically generate a violation notice including an appropriate fine; identify the owner of the violating vehicle based on the registration information of the vehicle; and transmit the violation notice to the owner.

[0024] In one example, the system is configured to utilize multiple video streams from multiple mobile cameras to identify a vehicle and a violation by the vehicle based on one or more parking violation rules and one or more violation exceptions.

[0025] In one example, each mobile camera is configured to transmit a video stream recorded for a parked vehicle to the streaming gateway in real time.

[0026] According to a second aspect of the present disclosure, a system for detecting one or more parking violations is provided. The system includes: a parking violation system configured to: receive multiple video streams of an environment; process each of the multiple video streams to identify one or more vehicles in each video stream; and determine whether one or more vehicles have constituted a parking violation based on one or more violation rules.

[0027] In one example, the parking violation system is further configured to: identify traffic data at two or more timestamps associated with each identified vehicle; and determine whether the one or more vehicles have constituted a parking violation based on the traffic data detected at the two or more timestamps and the one or more violation rules.

[0028] In one example, a parking violation system includes a violation assessment module configured to identify a vehicle based on a registration number and determine whether the identified vehicle has violated one or more violation rules.

[0029] In one example, the parking violation system includes: an image processing module;

[0030] Wherein, the image processing module is configured to execute a trained machine learning network, and the image processing module is configured to: extract multiple frames from each video stream; identify the vehicle in each frame; identify the vehicle registration information on the license plate of the vehicle; identify the geographic location based on information from the satellite navigation module; and identify the location of the vehicle.

[0031] In one example, the violation assessment module is configured as follows:

[0032] identifying a vehicle at a first timestamp—identifying a registration of the vehicle;

[0033] identifying the vehicle at a second timestamp—identifying the vehicle registration information;

[0034] responsive to the registrations being identical, verifying that the identified vehicles in the first and second timestamps are the same vehicle;

[0035] determining a geographic location of the vehicle at the first and second timestamps;

[0036] A parking violation is determined based on whether the geographic location of the vehicle has not changed between the first and second timestamps and based on one or more violation rules.

[0037] In one example, the parking violation system includes:

[0038] A violation rule system configured to receive a rule input, define one or more rules, and store the one or more rules; wherein the violation rule system comprises:

[0039] Violation rule database, violation rule interface, violation rule engine;

[0040] wherein the violation rules interface is configured to present an interface for receiving one or more violation rules from a user;

[0041] Wherein, the violation rule engine is configured to create one or more violation rules based on input in the violation rule interface; the violation rule database is configured to store the one or more violation rules;

[0042] Wherein, the violation assessment module is configured to access one or more violation rules from a violation rule system to determine a parking violation.

[0043] In one example, the parking violation system includes: a violation confirmation module; wherein the violation confirmation module is configured to:

[0044] Receiving a detected parking violation from the violation assessment module; checking one or more violation exceptions from the violation exception module; generating a parking violation confirmation if no violation exceptions are applicable, or denying a parking violation if an applicable violation exception is present.

[0045] In one example, the violation exception module is configured to identify one or more violation exceptions within a video stream or within a frame of a video stream based on one or more predefined violation exceptions.

[0046] As used herein, the term "comprising" (and its grammatical variations) is used in the inclusive sense of "having" or "including" and not in the sense of "consisting only of. It should be understood that if any prior art information is cited herein, this citation does not constitute an admission that this information forms part of the common general knowledge in the art in any country. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Hereinafter, embodiments of the present disclosure will be described by way of examples with reference to the accompanying drawings.

[0048] Figure 1 An exemplary embodiment of a system for detecting parking violations is shown.

[0049] Figure 2 An exemplary embodiment of a parking violation calculation system and components thereof is shown, wherein the parking violation calculation system is part of a system for detecting parking violations.

[0050] Figure 3 A schematic diagram showing the hardware elements of a parking violation calculation system.

[0051] Figure 4 A map of a route captured by a vehicle equipped with a camera is shown presented on a parking report interface.

[0052] Figure 5 A first parking report interface is shown showing a detected vehicle.

[0053] Figure 6 A second parking report interface showing a detected vehicle is shown.

[0054] Figure 7 A parking report interface is shown showing a vehicle that has incurred a violation.

[0055] Figure 8 A parking report interface showing exceptions to parking violations is shown.

[0056] Fig. 9 Another example of a parking report interface is shown.

[0057] Fig.10 An exemplary method for detecting parking violations is shown. DETAILED DESCRIPTION

[0058] The present disclosure relates to a system and method for detecting parking violations. In one example, the described system and method are used to detect parking violations by drivers on the street. The described system and method for detecting parking violations can also be used to detect other parking violations, such as parking violations in parking lots, parking buildings, or any other parking instances.

[0059] The system and method can automatically detect parking violations based on processing video streams received from one or more mobile cameras. The system is advantageous because it uses video streams from multiple mobile cameras, rather than fixed cameras, such as CCTV cameras. This provides a significant advantage over fixed cameras, where the coverage area is limited due to the limited number and fixed nature of the cameras. To increase the coverage area, more fixed cameras can only be added. This can result in a significant increase in installation and operating costs, so that it may cost a significant amount of money to cover large areas of a city. In addition, due to the limitations of fixed cameras, the storage and processing of video as evidence is essentially sporadic, and therefore has a very limited role in reporting and enforcement.

[0060] In this exemplary embodiment, the mobile camera can be any camera mounted on a vehicle or a person. The camera can be a smart phone camera, a tablet camera, a camera on an Internet of Things (IoT) or smart device, or a web-enabled camera (such as a dashboard, or a specially designed tactical or law enforcement camera). The camera can also be mounted on or in any vehicle, such as a motorcycle, a tricycle, a unicycle, a scooter, a golf cart, a car, a truck, a bicycle, a cart, a tractor, a radio-controlled vehicle, an aerial vehicle, an unmanned aerial vehicle (UAV); or the camera can be mounted on a person (such as a body camera) or an animal (such as a mounted police or canine). The mobile camera is preferably mounted on a government-authorized vehicle, such as a police car, a fire truck, a government taxi or vehicle fleet, or on a private vehicle, such as a transit vehicle, a private taxi, a private security vehicle, etc. Additionally or alternatively, the mobile camera can be mounted on or used by a selected group of personnel, such as government personnel such as police officers, parking attendants, etc. The mobile camera can also be mounted on or used by other private citizens (such as security guards). Mobile cameras can be mounted like body cameras on government or other personnel.

[0061] refer to Figure 1 , showing an embodiment of the present disclosure. The present embodiment provides a system for detecting one or more parking violations. The system includes: a streaming gateway configured to receive a video stream from one or more mobile cameras (preferably, multiple mobile cameras); an image processing module configured to process the received video stream, identify a parked vehicle, identify traffic data associated with the parked vehicle, and identify a timestamp of the identified traffic data; a violation assessment module configured to determine a parking violation based on traffic data determined at two or more timestamps and one or more parking violation rules; a violation confirmation module configured to confirm a detected parking violation as an actual parking violation based on checking one or more violation exceptions.

[0062] Parking violations include any unauthorized or illegal occurrence of a vehicle remaining stationary or substantially stationary for a period of time. Parking may include occupied or unoccupied stationary vehicles. The definition of parking violations may be customized by an authorized party, such as the police or transportation authorities. Violation rules may be defined and customized by an authorized party, such as the police, and violation rules may define parking violations.

[0063] Figure 1An exemplary system for detecting parking violations is shown. System 100 includes a plurality of mobile cameras 102, 104, 106. The mobile cameras are configured to be mounted on one or more vehicles or persons. Mobile cameras 102-106 may be cameras capable of recording video. Alternatively, cameras 102-106 may be capable of capturing a plurality of still images.

[0064] Each mobile camera 102-106 may be mobile or adapted to be mounted on a vehicle or on a person. Each mobile camera 102-106 includes a network interface for transmitting a recorded video stream over a network 120 (e.g., a cellular network). Optionally, each camera 102-106 may include a mount that enables the camera to be mounted on a vehicle or a person. The mount may also include a stabilizing structure, such as a gimbal or a support frame.

[0065] In one example, the mobile cameras 102-106 can each include a stereo camera (i.e., two cameras positioned side by side as shown). The stereo cameras 102-106 allow depth to be calculated. In one example, the mobile cameras 102-106 can each include a smart phone or a tablet computer, which includes an integrated camera, a network interface, and other components. Each mobile camera 102-106 can include a smart phone. This has the benefit that the smart phone can capture and stream video over a cellular network. Alternatively, the cameras 102-106 can stream video over other networks, such as a WAN network or a Wi-Fi network, etc.

[0066] like Figure 1 As shown, each camera vehicle 112, 114, 116 may have a camera mounted thereon or therein. Figure 1 As shown) or a person or a combination thereof, the cameras 102-106 are movable. The system 100 may include some cameras mounted on a vehicle and some cameras mounted on a person.

[0067] The system 100 defines a decentralized network. The parking violation calculation system 200 can be configured to communicate with any camera. Cameras can be added to the system 100 or removed from the system 100. The camera can include appropriate software and hardware components to enable the camera to communicate with the violation calculation system 200 over a network (e.g., a cellular network).

[0068] refer to Figure 1, the cameras 102-106 can record video streams of one or more cars. Each mobile camera 102-106 is configured to transmit a video stream to the parking violation calculation system 200 via a network 120, such as a cellular network 120. When the camera vehicles 112-116 are driving around the road, the video streams can record multiple cars. Figure 1 As shown, parked vehicles AD can be captured. Parked vehicles AD can be captured by multiple cameras 102-106. Camera vehicles 112-116 can capture parked vehicles AD at different moments, that is, at different time stamps.

[0069] like Figure 1 As shown, when vehicle 112 is traveling, camera 102 can capture video of parked cars A and B. When vehicle 114 is traveling, camera 104 can capture video of cars C, A, and B. When a vehicle with a camera is stationary, the camera's video stream can also be captured. For example, Figure 1 As shown, vehicle 116 may stop while camera 106 records at least car D. Camera 106 may also capture vehicle C.

[0070] The cameras 102-106 capture video streams and continuously transmit the captured video to the parking violation computing system. In one example, the video streams may be continuous and streamed in real time. Alternatively, the cameras 102-106 may include a cache that may temporarily cache some video prior to streaming. In one example, each camera may be a smartphone, and the captured video may be streamed directly to the parking violation computing system 200 via the smartphone's cellular interface module.

[0071] The parking violation calculation system 200 is configured to receive video streams from a plurality of cameras, process the video streams, and determine parking violations. The captured video is streamed from the cameras to the system 200. The parking violation calculation system 200 can process the received video streams, identify one or more vehicles, and determine parking violations based on the detected vehicles, traffic data associated with the vehicles, and one or more violation rules.

[0072] In some embodiments, when multiple cameras 102, 104, and 106 are moved around a particular area being patrolled, the system and its multiple cameras 102, 104, and 106 are able to capture various images of parked vehicles and other traffic-related objects at different times. In the case where multiple cameras 102, 104, and 106 are mounted to a traffic enforcement unit or vehicle (including but not limited to a car, bicycle, motorcycle, scooter, or hiker), each of the multiple cameras will capture images of vehicles parked at a particular location but at different times or various timestamps. Therefore, when these images are processed by an instance of the parking violation calculation system 200, a vehicle parked at a particular location may be captured by different cameras at different time slots. In turn, parking violations can be determined based on the time at which the parked vehicle is captured at a particular location. Similarly, although multiple cameras are preferred, a single camera on a traffic enforcement unit may be able to capture enough video streams to determine parking violations. For example, a single traffic enforcement unit may patrol an area continuously, thereby providing enough video streams of different timestamps for detecting parking or other traffic violations.

[0073] Parking violation rules may also be customizable. Violation rules may be associated with a geographic location (e.g., a specific street, a school zone, a hospital) or a specific landmark (e.g., an intersection or a traffic light) or any other suitable condition. There may be multiple violation rules associated with a single location. Violation rules may be dynamic, customizable, and changeable.

[0074] For example, Figure 1 As shown, the parking violation detection system 100 can be used to detect a parking violation committed by a vehicle D parked in a non-parking zone. In another example, a vehicle A can be detected by the cameras 102, 104, and the parking violation calculation system 200 can determine that the vehicle A has committed a parking violation because the vehicle has been parked longer than the allowed time limit of 120 minutes. In another example, when the vehicle C has just come to rest on the road, it can be detected that the vehicle C has committed a parking violation.

[0075] The computing system 200 may automatically issue and send violation notices to violators and / or may also send parking violation information to other law enforcement systems 130 or may record and store detected parking violations. The computing system 200 may be configured to report the stored violations to the law enforcement agency computing system 130 or to the council or local government system for further processing. The stored parking violations may also be used as evidence in court during reporting.

[0076] Figure 2 A schematic diagram of components that are part of a parking violation calculation system 200 is shown. The system 100 for detecting parking violations includes Figure 2 Components shown.

[0077] In one example, the computing system 200 can be implemented by any computing architecture, including a portable computer, a tablet computer, a stand-alone personal computer (PC), a smart device, an Internet of Things (IoT) device, an edge computing device, a client / server architecture, a "dumb" terminal / host architecture, a cloud computing-based architecture, or any other appropriate architecture. The computing device can be appropriately programmed to detect one or more parking violations from multiple video streams or multiple images received from multiple cameras.

[0078] The cameras 102-106 may stream the recorded video in real-time. Alternatively, if no network connection is available, the recorded video may be cached locally in the camera. Figure 2 , computing system 200 includes streaming gateway 202. Streaming gateway 202 is configured to receive multiple streams. Streaming gateway 202 can collate received streams for further processing. Each video stream can be tagged and collated based on the corresponding camera. Alternatively, the streams can be processed in streaming gateway 202, which can classify the streams based on the recording location and collate the streams based on the recording location. Any suitable collation rules can be applied.

[0079] The collated video streams are transmitted through the image processing module 204. The image processing module is configured to extract a plurality of frames from each video stream. The image processing module 204 may be further configured to identify one or more vehicles in each frame, one or more traffic data associated with the identified vehicles, and a timestamp corresponding to each frame. The image processing module 204 is configured to perform image recognition to identify one or more of a vehicle, vehicle registration information (i.e., vehicle license plate) recognition, vehicle model, and / or vehicle location.

[0080] In one example, the image processing module 204 is configured to execute a machine learning network to perform object recognition for each frame. The machine learning network is trained to identify a vehicle and one or more traffic data associated with the identified vehicle. The machine learning network can be an AI model. The machine learning network can be further trained to perform environmental recognition for each frame to identify one or more environmental objects, such as a stop sign, a traffic sign, a traffic light, etc. The AI ​​model can be trained using AI training data.

[0081] The image processing module 204 may be further configured to implement one or more image stabilization systems to stabilize the frames of the video stream and filter light, blur artifacts, and other noise from each frame to aid in object recognition. The applied image stabilization system may improve vehicle detection.

[0082] The system 100 may include an AI training data database 208. The AI ​​training data database 208 includes training data for training a machine learning network. The training data may include one or more of vehicle identification data, vehicle license plate data, image stabilization data, environmental object data, and other data that may be required for training.

[0083] The traffic data extracted from the image processing module 204 may include any one or more of the following: vehicle registration information, vehicle brand, vehicle model, vehicle location, or vehicle geographic location. The traffic data is used to identify vehicles and detect traffic violations.

[0084] The image processing module 204 is configured to execute a trained machine learning network, and the image processing module is configured to extract multiple frames from each video stream. These frames can be stored, or can be processed and then deleted. The image processing module 204 is configured to identify the vehicle in each frame. The image processing module 204 can be configured to identify the vehicle registration information on the license plate of the vehicle. Optionally, the image processing module 204 can be configured to identify the geographic location based on information from a navigation module, such as a satellite navigation module (Beidou, GLONASS, Galileo, GPS, etc.) or a ground navigation module. The navigation module can be part of a smart phone that includes a camera. The navigation information can be encoded into a video stream sent to a streaming gateway. The image processing module 204 is further configured to identify the location of the vehicle.

[0085] The parking violation calculation system 200 includes an image storage module 210. The image storage module 210 can be a database configured to store processed images from the image processing module 204. The processed images can include one or more identified vehicles. Optionally, the images stored in the image storage can also include the geographic location of the image, such as the coordinates of the location that can be determined by the image processing module. The stored images can include other tags. The stored images can be used in the violation exception module 216 to determine one or more exceptions to the parking violation.

[0086] The traffic data determined in the image processing module 204 may be stored in a traffic data database 206. The traffic data database 206 may be separate from the image processing module 204 or may be incorporated into the image processing module. The traffic data database 206 may be configured to store the traffic data identified by the image processing module 204. The traffic data may be any data related to the parked vehicle, but may also relate to other data, such as geographic location, etc. Some examples of traffic data may include one or more of the following: vehicle registration information, vehicle brand, vehicle model, vehicle location, or geographic location of the vehicle, or other information.

[0087] From each mobile camera, a large amount of traffic data for various detected vehicles can be stored. The data collected can be greater than that typically collected from fixed cameras such as CCTV because mobile cameras can move over a wider area and can access areas where fixed cameras cannot be installed. This data can be stored in the traffic data database 206. This data can be used for further analysis, such as identifying areas where there are repeated parking violations or identifying serial violators.

[0088] The parking violation calculation system 200 includes a violation assessment module 212. The violation assessment module 212 is configured to determine a parking violation based on traffic data determined at two or more time stamps and one or more parking violation rules.

[0089] More specifically, the violation assessment module 212 is configured to identify the vehicle and the registration number of the vehicle at a first timestamp (i.e., at a first time); and to identify the registration information of the vehicle at a second timestamp. The violation assessment module 212 is configured to verify that the vehicle identified in the first and second timestamps is the same vehicle. In one example, the violation assessment module 212 is configured to verify that the same vehicle has been identified by comparing the vehicle registration information. If the vehicle registration is the same, it is the same vehicle. The violation assessment module 212 is further configured to determine the geographic location of the vehicle at the first and second timestamps. The geographic location can be defined as coordinates. Additionally or alternatively, the geographic location can include one or more of the following: street name, suburb, city, and province.

[0090] The violation assessment module 212 may be further configured to determine whether a parking violation has occurred based on whether the geographic location of the vehicle has not changed between the first and second timestamps and based on one or more violation rules.

[0091] The system 100 further includes a violation rule system 220. The violation rule system 220 is configured to receive rule input, define one or more rules, and store the one or more rules. These violation rules can be dynamic and can be changed. In one example, the violation rules are one or more customizable rules that define parking violations. The violation rules can be customized by an authorized party (e.g., law enforcement personnel or government personnel, etc.).

[0092] The violation assessment module 212 is configured to access one or more violation rules from the violation rule system 220 to determine a parking violation. The violation assessment module 212 is configured to determine whether a parking violation has occurred based on checking whether a vehicle detected at different time stamps has violated one or more defined violation rules.

[0093] In one example, the violation rule may be that the parking space has a predetermined time limit. A vehicle may be detected in the parking location in a video stream at a first time, and then in the same parking location in another video stream at a second time. The second time is after the first time. The video streams may be from the same camera or may be from two different cameras. For example, one camera may capture a parked vehicle at a first time (i.e., a timestamp), and then a second camera may capture the parked vehicle at a second time. The image processing module 210 may be configured to identify a specific parking location, for example, based on detecting landmarks (e.g., street names) and / or using navigation information or a combination thereof. The violation assessment module 212 is configured to access the violation rule for the identified location (i.e., for the parking space) from the violation rule system 220. In this example, the parking limit for the parking location may be 30 minutes. The violation assessment module 212 is configured to determine that the time difference between the first time and the second time is 45 minutes, and the parked vehicle has been detected at the same location. The violation assessment module 212 will determine that the vehicle has constituted a parking violation because the length of time it has been parked in the parking space has exceeded the allowed time limit. There may be exceptions to the violation, which are described later.

[0094] The violation rules system 220 includes: a rules engine 222, a violation rules interface 224, and a violation rules database 226. The violation rules interface 224 is configured as an interface, such as a screen that receives one or more violation rules from a user. The violation rules interface 224 can be presented on a screen of the parking violation computing system 200. Alternatively, the violation rules interface can be presented on another user device that can be used to access the computing system 200. At least the parking violation computing system 200 is adapted to communicate with a plurality of other user devices, such as a device of a law enforcement officer or a device of a government officer. The computing system 200 can be accessed by a law enforcement system or a government system or a legal professional (prosecution and defense) or any other authorized entity that can be involved in enforcing parking violations.

[0095] The violation rules interface 224 allows authorized users to create new violation rules or change or customize violation rules. The interface 224 also allows authorized users to delete or suspend existing violation rules. Authorized users will be required to complete a registration process to access the computing system 200. The access request will include relevant information, and the computing system 200 can be configured to verify that the user requiring access is an authorized user.

[0096] The violation rules engine 222 is configured to create one or more violation rules based on input in the violation rules interface 224. The violation rules engine 222 may be further configured to create rules based on current parking laws and regulations. The violation rules 222 may be configured to update existing rules based on regulatory changes. The created violation rules are transmitted to the violation rules database 226. The violation rules database 226 is configured to store one or more violation rules. Some example violation rules that may be stored may include one or more parking time limits, restricted locations, no parking time zones. Violation rules may be associated with geographic locations or landmarks or other suitable features.

[0097] The violation evaluation module 212 is configured to look up and retrieve violation rules from the violation rules engine 226 during the determination of a violation. The violation rules system 220 can be implemented on the computing system 200. Alternatively, the violation rules system 220 can be implemented in a separate server or a separate computing device that can be arranged to communicate with the parking violation computing system 200.

[0098] The parking violation calculation system 200 further includes a violation exception module 216. The violation exception module 216 is configured to store one or more predefined violation exceptions. The violation exception module 216 is configured to identify one or more violation exceptions within the video stream or within a frame of the video stream based on the one or more predefined violation exceptions. The violation exception module 216 can be configured to process the stored images, i.e., frames. The violation exception module 216 is configured to access the stored frames from the image storage database 210. Alternatively, the violation exception module 216 can be configured to access the processed video frames from the image processing module 204.

[0099] In one example, the violation exception module 216 can store predefined violation exceptions. These violation exceptions can be defined by an authorized party (e.g., a police officer or parking enforcement agency) or another suitable authorized party. These violation exceptions are exceptions to parking violations, i.e., they are instances of people, vehicles, etc. that are exempt from parking violations, for example. Some examples include emergency vehicles, school buses, elderly care vehicles, food transportation vehicles, and vehicles exempt from inspection. Exceptions can also include the detection of dangerous situations, i.e., determining the difference in vehicle position where the vehicle has moved. The exception module can also define certain rules, such as the minimum number of violations before a penalty or report is imposed or a time limit buffer for parking violations or any other rules. The violation exception module 216 can also define other vehicles that need to be arrested, such as wanted vehicles or wanted persons.

[0100] In another example, the violation exception module 216 may be configured to receive and store predefined exceptions to block violations from authorized users, and the violation exception module 216 may process an image or video stream to identify and mark one or more violation exceptions. The module 216 may implement object recognition, such as an AI model, to perform object recognition on the image or video stream. The violation exception module 216 may be configured to identify one or more exceptions based on object recognition. For example, the module 216 may identify an emergency vehicle parked in a non-parking area. The module 216 may identify a wanted vehicle and may send a message with location information to law enforcement personnel.

[0101] In another example, the violation exception module 216 may be configured to determine any differences in the position of a vehicle detected in multiple frames captured at multiple times. If there is a position difference, such as the difference is above a threshold, the exception module may mark the vehicle as an exception because the change in position may indicate that the vehicle has moved. The violation exception module 216 may be configured to mark, for example, a violation exception within an image or simply as a command or data point. The marked violation exception may be transmitted or accessed by the violation confirmation module 214.

[0102] In one example, the violation exception module 216 may include one or more permanent violation exceptions. These violation exceptions may be permanent exceptions. Examples of permanent exceptions may be emergency vehicles or law enforcement vehicles. The violation exception module 216 may also include time-limited exceptions that may be automatically deleted after a predetermined time, such as streets where temporary parking may be allowed. The module 216 may also include customizable violation exceptions.

[0103] The system 100 for detecting parking violations may further include a violation confirmation module 214. The violation confirmation module 214 may be implemented in the parking violation calculation system 200. The violation confirmation module 214 is configured to confirm a detected parking violation as an actual parking violation based on checking one or more violation exceptions.

[0104] In one example, the violation confirmation module 214 is configured to receive the detected parking violation from the violation assessment module 212. The parking violation can be captured in a data structure indicating the detected violation. The violation confirmation module 214 is configured to check one or more violation exceptions from the violation exception module 216. The violation exceptions can be accessed from the violation exception module 216. The violation confirmation module is further configured to generate a parking violation confirmation if there are no applicable violation exceptions, or to reject the detected parking violation (i.e., determine that the parking is not a violation) if there are applicable violation exceptions.

[0105] In one example, the violation assessment module 212 can detect a vehicle parked in a no-parking area or parked too close to a traffic light. If the vehicle is a civilian vehicle, the violation exception module will not return an exception and the verification module 214 will confirm the parking violation. In another example, if an ambulance is parked in a non-parking area, the violation assessment module 212 detects it as a parking violation (i.e., the vehicle is parked in a non-parking area). However, in this example, the violation exception module 216 will return a violation exception because the detected vehicle is an emergency vehicle. The violation confirmation module 214 will not return a parking violation or may not return an output because emergency vehicles are exempt from parking violations.

[0106] In another example, if a vehicle is detected at times with a large gap (e.g., several hours), and the location between the two detected times is substantially different (e.g., the direction of the vehicle in the second timestamp is different from the first timestamp), the violation exception module 216 may return or flag a violation exception. The change in location indicates that the user drove away and re-parked. In this example, the violation confirmation module 214 will return no violation or no information.

[0107] The system for detecting parking violations includes a parking violation reporting interface 230. The parking violation reporting interface 230 can be part of the parking violation computing system 200. The parking violation reporting interface 230 is configured to receive parking violation confirmations from the violation confirmation module 214. The parking violation reporting interface 230 is further configured to present confirmed parking violations on an interface. The parking violation reporting interface 230 can be presented on a display of the computing system 200, or can be presented on a device or display of an authorized user using the computing system 200.

[0108] The parking violation reporting interface 230 can present information on a display of the parking violation computing system 200. Alternatively, the parking violation reporting interface can be presented on a display of a user device, such as a tablet or computer. The parking violation reporting interface presents information captured by the image processing module 204 as the image processing module 204 processes the video stream.

[0109] The parking violation reporting interface 230 may present the recorded video as a video stream, or may present discrete frames or images. Figure 5 , 6 7 show some relevant examples. Parking violation reporting interface 230 presents at least detected vehicles, vehicle registration information (i.e., license plate), location information, and a timestamp. Parking violation reporting interface 230 may also present registration information, i.e., vehicle license plates of multiple vehicles that have been detected. Parking violation reporting interface 230 presents parking violations when parking violations are detected. If no parking violations are detected, parking violation reporting interface 230 is configured to present information of the detected vehicles.

[0110] Fig. 9 Another example of a parking violation reporting interface 230 is shown. Fig. 9 As shown, the parking violation reporting interface 230 includes an image 902 of a vehicle that constitutes a parking violation. The vehicle can be identified in the image by a bounding box 904. The vehicle license plate 906 is identified on the parking violation reporting interface 230. The parking violation reporting interface 230 displays location information 908 of the violation. The location information can also be stored as traffic data in the database 206. Timestamp information 910 is presented on the parking violation reporting interface 230. A video recording of the violation 912 can be stored and presented. The violation presented on the parking violation reporting interface 230 can be captured and stored, for example, in the image database 210 or the violation evidence database 234. Optionally, the data presented on the parking violation reporting interface 230 can be transmitted to the violation evidence database 234.

[0111] The parking violation reporting interface 230 may also present a parking violation confirmation 910 and a violation type 242. Fig. 9 As shown, a parking violation confirmation 920 is shown. The violation type 922 in the example shown is "parking in a bus lane". A bus lane can be a no-parking zone. Therefore, the black car 904 has been parked in a bus lane (no-parking zone).

[0112] The parking violation reporting interface 230 may also present a report option 930 or a prosecution option 930. The report option (prosecution option) may include two buttons, "Yes" and "No". The authorized user may choose to automatically report or prosecute by selecting the "Yes" button 932, or may decide not to report or prosecute by pressing the "No" button 934. This option allows the authorized user to apply some user discretion before reporting or prosecuting.

[0113] The violation evidence database 234 can be configured to store all confirmed violations from the violation confirmation module. The violation evidence database 234 can store the information presented on the parking violation reporting interface 230 for any court case or law enforcement action or reporting action. Examples of stored information are vehicle registration information, images of parking violations, timestamps, locations of violations, video recordings or still images of violations, and other relevant information needed as evidence.

[0114] The system 100 may also include a reporting (prosecution) module 232. The reporting (prosecution) module 232 may be part of the parking violation calculation system 200. The reporting module may be configured to receive a reporting input from a user, such as through the parking violation reporting interface 230. If an authorized user chooses to report a violation, the reporting (prosecution) module 232 may be configured to automatically generate a violation notice including an appropriate fine. For example, if the authorized user selects the reporting option 932, the reporting (prosecution) module 232 may generate a violation notice and send the violation notice. The reporting (prosecution) module 232 may be configured to identify the owner of the violating vehicle based on the vehicle's registration information and send the violation notice to the owner.

[0115] In some embodiments of the system 100 for detecting parking violations, the system may be arranged or adapted to detect specific traffic or parking violations targeted by laws or regulations of a particular region, city, state, or jurisdiction. Such traffic or parking violations may include, but are not limited to:

[0116] Vehicles that are stopped or parked in areas that obstruct the road or create a hazard to traffic or road users, such as school areas, hospital entrances, etc.;

[0117] Vehicles stopped at zebra crossings or zebra-controlled areas;

[0118] Vehicles stopped near fire hydrants;

[0119] Vehicles stopped in non-standing or non-parking areas;

[0120] · Vehicles parked in areas other than authorized parking spaces;

[0121] Parking on sidewalks, crosswalks, medians, curbs, emergency parking areas, and in areas that are

[0122] vehicles of other users, traffic islands or other areas or spaces reserved or appropriate for traffic or activities;

[0123] · Vehicles parked in a manner that obstructs entry into or exit from the property or driveway;

[0124] · Vehicles parked in parking spaces that violate traffic directions, signs or markings;

[0125] Vehicles parked across multiple parking spaces or across parking space lines;

[0126] Vehicles parked in parking spaces where parking is suspended at government offices;

[0127] · Vehicles parked in authorized parking spaces beyond the permitted time limit;

[0128] · a vehicle parked in an authorized parking space without paying; or

[0129] · Vehicles parked in a specific parking space without the necessary authorization, credentials or vehicle type.

[0130] The system may also be used to assist relevant government authorities, including transportation departments, traffic wardens, police, or other government agencies, in reporting or issuing violation notices to violators who have been detected to have committed traffic or parking violations. In some exemplary embodiments, the system may provide an interface to allow reporting of detected violations by allowing authorized users to review and publish such violations. The system may also be used to semi-automatically or fully automatically issue violation notices to reported users.

[0131] The system may also provide the functionality of electronic or mail service for violation notifications as required by the laws of the enforcement jurisdiction. In addition, an online or computer-based user interface, such as a web portal or an App portal, may also be provided to the user who issued the violation notice so that the user is aware of their violation, and to the user so that the user can view the violation involved in the report. Such an online or computer-based system may also provide video or photographic evidence of the violation to the reported user, thus providing an easier method. Users can be summoned to court, pay their fines, and identify the parties involved or prepare a defense.

[0132] like Figure 3, a schematic diagram of a parking violation computing system 200 is shown. The computing system 200 is used to implement an example embodiment of a system for detecting parking violations. In this embodiment, a parking violation computing system (i.e., a server) 200 is included, which includes appropriate components necessary to receive, store, and execute appropriate computer instructions. These components may include: a processing unit 302, which includes a central processing unit (CPU), a math co-processing unit (Math Processor), a graphics processing unit (GPU), or a tensor processing unit (TPU) for tensor or multi-dimensional array calculation or manipulation operations; a read-only memory (ROM) 304; a random-access memory (RAM) 306; an input / output (I / O) device (e.g., a disk drive 308); an input device 310 (e.g., an Ethernet port, a USB port, etc.); a display 312, such as a liquid crystal display, a light-emitting display, or any other suitable display; and a communication link 314. The computing device 200 may include instructions that may be stored in the ROM 304, the RAM 306, or the disk drive 308 and executed by the processing unit 302.

[0133] A plurality of communication links 314 may be provided that may be variously connected to one or more computing devices, such as servers, personal computers, terminals, wireless or handheld computing devices, IoT devices, smart devices, and edge computing devices. At least one of the plurality of communication links may be connected to an external computing network via a telephone line or other type of communication link. The communication links 314 allow the parking violation computing system 200 to communicate with a plurality of cameras 102-106 at the site and receive a plurality of video streams.

[0134] The parking violation computing system 200 may include a storage device such as a disk drive 308, which may include a solid-state drive, a hard disk drive, an optical drive, a tape drive, or a remote or cloud-based storage device. The parking violation computing system 200 may use a single disk drive or multiple disk drives, or a remote storage service. The server 200 may also have a suitable operating system resident on the disk drive or in ROM of the server 200.

[0135] The system 100 for detecting parking violations may utilize one or more machine learning networks. The parking violation calculation system 200 may also provide the necessary computing power to operate or interface with a machine learning network, such as a neural network, to provide various functions and outputs. The neural network may be implemented locally, or may be accessed or partially accessed through a server or cloud-based service. The machine learning network may also be untrained, partially trained, or fully trained, and / or may also be retrained, modified, or updated over time.

[0136] Figure 2 The components of the parking violation calculation system 200 shown in and described herein may be software modules or firmware modules or a combination thereof, which may be implemented by the parking violation calculation system 200 and its hardware components described herein. Alternatively, Figure 2 One or more of the components described in may be Figure 3 The hardware elements of the parking violation calculation system 200 described in the accompanying drawings may be used in conjunction with hardware modules or hardware elements.

[0137] Figure 4 A map 400 of a route 402 captured by a vehicle equipped with a camera is shown. The route taken by the vehicle equipped with the cameras 102-106 is tracked and the route can be presented on the display 312. The route 402 can be presented on the display of an authorized party that can access the parking violation computing system 200. The map 400 can include a starting point 404 and an ending point. In an example, the route 402 can be a predefined route, such as a predefined patrol route that a police car or parking enforcement vehicle can follow. The route 402 can indicate the current location 406 of the vehicle and the camera 102-106. The location of the camera 102-106 can be tracked by satellite navigation. Each camera can enable satellite navigation. In an example, each camera 102-106 is part of a smart phone, and the satellite navigation function of the smart phone can be used to track the location of the camera.

[0138] One or more detected parking violations may be displayed on the map 400. Routes with detected violations may be stored in a memory or appropriate database. The map 400 presented to the user may also present a license plate search function 408. The license plate search function allows the user to search for captured license plates and / or license plates of parking violators. The search results may be displayed on the map screen 400 along with the violation location. Alternatively, the search results may be displayed on a separate screen.

[0139] Figures 5 to 8 Shown are screen shots presented when the system 100 for detecting parking violations is running. Figures 5 to 8 The screen shots shown in illustrate the operation of the parking violation calculation system 200.

[0140] Figure 5 A captured image 500 of a detected vehicle 501 and a license plate 502 is shown. The captured image also shows street parking where parking is allowed. The image shows a stop sign 504 that can be detected. It shows the location 506, timestamp 508 and coordinates 510 of the detected vehicle. The image can be viewed at Figure 4 The image 500 may be a captured frame of a video stream. The system 100 has determined that the vehicle 501 has not committed a parking violation.

[0141] Figure 6 Another image 600 of another vehicle being detected is shown. Vehicle 601 is detected and license plate 602 is displayed. Figure 5 The information shown can be a location, coordinates, and a timestamp. The detected vehicles 501, 601 do not constitute a parking violation. Screens 500 and 600 are examples of information presented through the parking violation reporting interface 230. Figure 5 and Figure 6 As shown, the detected vehicle registration information is highlighted. Other detected vehicle registration information is also presented. The registration information detected in the video stream is presented sequentially from left to right.

[0142] Figure 7 An image 700 of a vehicle that has committed a parking violation is shown. Image 700 is an example of a parking violation reporting interface 230. Figure 7 As shown, a detection vehicle 701 is detected and its license plate (i.e., registration number) 702 is highlighted and displayed. Other detected vehicles and their license plates are also shown. The license plate of the currently detected vehicle is highlighted. The bus lane 704 is detected by the system 200. The location 706, coordinates 710, and timestamp 708 are also presented. Optionally, satellite navigation coordinates may also be presented. Surrounding the vehicle 701, a bounding box 712 is presented. Surrounding the violating vehicle, a bounding vehicle 712 is identified. The vehicle 701 is parked in the bus lane, which is considered a parking violation. Once the violation is confirmed after exception detection, a screen (e.g., Fig. 9 ). Figure 7 As shown, the vehicle in front of vehicle 701 is also detected as constituting a parking violation.

[0143] Figure 8An exemplary screen 800 presented by the parking violation reporting interface 230 is shown. A taxi 801 is detected as constituting a parking violation, so a bounding box 810 is presented around it. Since the system initially detected a parking violation, a bounding box 810 is presented around the taxi 801. The detected location may be a taxi zone. A registration number 802, location 806, and timestamp 808 are given. Optionally, satellite navigation coordinates 804 may be presented. The detected vehicle 801 is a taxi, so parking is allowed, i.e., the violation rule is not satisfied. This is because taxi parking is an exception. The exception module will generate a flag. The parking violation reporting interface 230 then removes the bounding box, and the system 200 will ignore the violation.

[0144] Optionally, parking violation reporting interface 230 may allow an authorized user to make a decision regarding ignoring a violation based on detected exceptions and the user's discretion.

[0145] Fig.10 An exemplary method of detecting parking violations 1000 is shown. The method 1000 is a method implemented by the parking violation computing system 200 and its components. The method 1000 begins at step 1002. Step 1002 includes: receiving a plurality of video streams from a plurality of cameras. The cameras are configured to record vehicles to detect parking violations. Step 1002 includes: receiving a plurality of video streams from various mobile cameras moving around a city or town or region. Optionally, the video streams may be collated.

[0146] Step 1004 includes identifying one or more vehicles in the received video stream. The identified vehicles may be parked vehicles or may be any stationary vehicles. Moving vehicles may be filtered. Step 1006 includes identifying traffic data associated with each identified parked vehicle and identifying a timestamp for each detected vehicle.

[0147] Step 1008 includes receiving or accessing one or more parking violation rules. Step 1010 includes comparing the traffic data and vehicles at two or more time stamps to the one or more parking violation rules. The violation rules may be associated with a specific location and may be time-limited.

[0148] Step 1012 includes determining a parking violation if the traffic data at a plurality (ie, two or more) of time stamps satisfies one or more parking violation rules. The violation assessment is based on information collated from a plurality of video streams from a plurality of cameras.

[0149] Step 1014 includes presenting a positive violation indicator on a parking report interface. A user may confirm the violation by applying user judgment. Alternatively, the method may optionally include automatically issuing a parking violation notification or other report, such as issuing a fine. Step 1016 includes storing the detected violation in a database. The stored violation may be used as evidence. The stored detected violation includes traffic data for the violating vehicle and the specific violation constituted.

[0150] Optionally, method 1000 may include step 1018. Step 1018 may be performed before step 1014. Step 1018 includes: checking one or more violation exceptions, i.e., checking whether one or more exemptions apply. If a violation exception is detected, such as the violating vehicle is an emergency vehicle, the detected violation may be invalidated or ignored or filtered out. The parking violation interface may present the detected vehicle, for example, Figure 5 and 6 If this happens, for example Figure 7 , the interface may further indicate a violation. Method 1000 may be repeated continuously.

[0151] The system for detecting parking violations is advantageous because it uses video streams from multiple mobile cameras, which provide wider geographic coverage than fixed cameras. The system can operate using video streams received over a network (e.g., via a cellular network or WAN or other network). It is also easier and cheaper to set up multiple mobile cameras than CCTV cameras. Mobile cameras are mobile, which is advantageous because the system can be used to monitor multiple locations due to the use of multiple mobile cameras. Mobile cameras are also mobile, which provides increased coverage compared to CCTV cameras. As described, at least one camera can be mounted on a vehicle or on a person, which allows parking violations to be captured more clearly because there are multiple angles and closer distances that can be captured. Using mobile cameras in a system for detecting street parking violations is superior to fixed cameras. This is because mobile cameras can move to avoid obstacles. Because mobile cameras capture video streams while moving, they do not really have blind spots. Moving devices, obstacles, etc. are not limitations.

[0152] The system for detecting street parking violations is advantageous. This is because it utilizes video streams from multiple different cameras. This is advantageous in that different cameras can again follow multiple different paths, thereby achieving wider coverage. The video streams ensure that images are captured substantially in real time, making the assessment of parking violations more accurate. Therefore, when combined with multiple vehicles and their cameras, the video streams provide substantially thorough coverage of the enforcement area due to the large number of mobile cameras within the enforcement area. The cameras can also be used interchangeably between vehicles. Thus, even if some vehicles (e.g., police vehicles) may not be available for assessing parking violations, the system is still more modular and available. Mobile cameras are advantageous. This is because they can be used with any vehicle or mounted on a person, rather than being fixed.

[0153] In one example, a system for detecting parking violations is also advantageous. This is because the system is configured to utilize multiple video streams from multiple mobile cameras to identify vehicles and violations by the vehicles based on one or more parking violation rules and one or more violation exceptions. Multiple video streams from multiple cameras allow for independent verification of violations and wider geographic coverage. This means that the system 100 can be used to capture more parking violations.

[0154] The system 100 is also advantageous. This is because, in one example, each mobile camera is configured to send a recorded video stream of a parked vehicle to a streaming gateway in real time. Real-time capture and storage provides stronger evidence of the violation and more convincing evidence in a report. In addition, the real-time transmission of the video stream is advantageous. This is because the system 100 can detect and enforce parking violations faster than current methods that require more manual processing.

[0155] Although not required, the embodiments described with reference to the accompanying drawings may be implemented as an application programming interface (API) or a series of libraries used by developers, or may be included in another software application, such as a terminal or personal computer operating system or a portable computing device operating system. Generally, since program modules include routines, programs, objects, components, and data files that help perform specific functions, those skilled in the art will understand that the functions of software applications may be distributed between multiple routines, objects, or components to achieve the same functions required herein. It should also be understood that in the case where the method and system of the present disclosure are all implemented by a computing system or partially implemented by a computing system, any appropriate computing system architecture may be used. This will include stand-alone computers, network computers, and dedicated hardware devices. Where the terms "computing system" and "computing device" are used, these terms are intended to include any appropriate configuration of computer hardware that can implement the described functions. Those skilled in the art will understand that, without departing from the spirit or scope of the present disclosure as broadly described, various changes and / or modifications may be made to the present disclosure as shown in the specific embodiments. Therefore, the present embodiment is considered to be illustrative and not restrictive in all respects.

[0156] In addition, it should be noted that the embodiments may be described as processes, which are depicted as flow charts, flow charts, structure diagrams, or block diagrams. Although flow charts may describe operations as sequential processes, many operations may be performed in parallel or simultaneously. In addition, the order of operations may be rearranged. The process terminates when its operation is completed. The process may correspond to a method, function, process, subroutine, subprogram, etc. in a computer program. When a process corresponds to a function, its termination corresponds to the function returning to the calling function or the main function. The methods or algorithms described in conjunction with the examples disclosed herein may be implemented directly in the form of hardware, software modules executable by a processor, or a combination of the two, in the form of processing units, programming instructions, or other directions, and may be included in a single device or distributed on multiple devices. Without departing from the scope of the present disclosure, one or more components and functions shown in the drawings may be rearranged and / or combined into a single component or embodied as several components. Without departing from the scope of the present disclosure, additional elements or components may also be added.

Claims

1. A system for detecting one or more parking violations, characterized in that include: a streaming gateway configured to receive video streams from one or more mobile cameras; an image processing module configured to process the received video stream, identify a parked vehicle, identify traffic data associated with the parked vehicle, and identify a timestamp for the identified traffic data; a violation assessment module configured to determine a parking violation based on the traffic data determined at two or more time stamps and one or more parking violation rules; A violation confirmation module is configured to confirm the detected parking violation as an actual parking violation based on checking the one or more violation exceptions.

2. The system according to claim 1, characterized in that in, The traffic data includes one or more of the following: Vehicle registration information; Vehicle brand; Vehicle model; Vehicle location; The geographic location of the vehicle.

3. The system according to claim 2, characterized in that in, The image processing module is configured as follows: Extracting a plurality of frames from each video stream; One or more vehicles in each frame, one or more traffic data associated with the identified vehicles, and a time stamp corresponding to each frame are identified.

4. The system according to claim 3, characterized in that in, The image processing module is configured to execute a machine learning network to perform object recognition on each frame; wherein the machine learning network is trained to recognize a vehicle and one or more traffic data associated with the recognized vehicle.

5. The system according to claim 3, characterized in that in, The image processing module is configured to execute a trained machine learning network, and the image processing module is configured to: Extracting a plurality of frames from each video stream; Identify the vehicle in each frame; identifying vehicle registration information on the license plate of the vehicle; identifying a geographic location based on information from a satellite navigation module; A location of the vehicle is identified.

6. The system according to claim 1, characterized in that in, The system comprises one or more mobile cameras, each mobile camera being movable or adapted to be mounted on a vehicle or adapted to be mounted on a person, each mobile camera comprising a network interface for transmitting a recorded video stream.

7. The system according to claim 6, characterized in that in, The video stream is a live video stream, and the streaming gateway is configured to receive the live video stream from the one or more mobile cameras and collate the video stream.

8. The system according to claim 1, characterized in that in, The violation assessment module is configured as follows: identifying a vehicle at a first timestamp—identifying a registration number of the vehicle; identifying the vehicle at a second timestamp—identifying the vehicle registration information; responsive to the registrations being identical, verifying that the identified vehicles in the first and second timestamps are the same vehicle; determining a geographic location of the vehicle at the first and second timestamps; A parking violation is determined based on whether the geographic location of the vehicle has not changed between the first and second timestamps and based on one or more violation rules.

9. The system according to claim 8, characterized in that in, The violation rules are one or more customizable rules that define parking violations.

10. The system according to claim 9, characterized in that Further including: a violation rule system configured to receive a rule input, define one or more rules, and store the one or more rules; Wherein, the violation assessment module is configured to access one or more violation rules from a violation rule system to determine a parking violation.

11. The system according to claim 9, characterized in that in, The violation rule system includes: Violation rules database; Violation rules interface; Violation rules engine; wherein the violation rules interface is configured to present an interface for receiving one or more violation rules from a user; wherein the violation rule engine is configured to create one or more violation rules based on input in the violation rule interface; The violation rules database is configured to store the one or more violation rules.

12. The system according to claim 8, characterized in that in, The violation rules include one or more of the following: Parking time limit; Restricted location; No parking period.

13. The system according to claim 8, characterized in that in, A vehicle is identified at a first timestamp from the first video stream and a vehicle is identified at a second timestamp from the second video stream.

14. The system according to claim 1, characterized in that Further including: A violation exception module is configured to store one or more predefined violation exceptions.

15. The system according to claim 14, characterized in that in, The violation confirmation module is configured as follows: receiving a detected parking violation from the violation assessment module; checking one or more violation exceptions from the violation exception module; If no violation exception applies, a parking violation confirmation is generated, or if an applicable violation exception applies, a parking violation is denied.

16. The system according to claim 15, characterized in that in, The violation exception module is configured to identify one or more violation exceptions within a video stream or within a frame of a video stream based on one or more predefined violation exceptions.

17. The system according to claim 15, characterized in that in, The violation exceptions may include one or more of the following: Emergency vehicles detected; Food transport vehicles detected; An exempt vehicle is detected; A wanted vehicle was detected; Image difference detection is performed to determine one or more differences in the vehicle's position to determine that the vehicle has moved.

18. The system according to claim 1, characterized in that Further including: Parking violation reporting interface, which is configured as follows: receiving parking violation confirmation information from the violation confirmation module; Confirmed parking violations are presented on the parking violation reporting interface.

19. The system according to claim 18, characterized in that Further comprising a reporting module, which is configured as follows: receiving reporting input from a user; Automatically generate violation notices including appropriate fines; Based on the registration information of the vehicle, identifying the owner of the illegal vehicle; Transmitting the violation notification to the owner.

20. The system according to claim 1, characterized in that in, The system is configured to utilize multiple video streams from one or more mobile cameras; identify a vehicle and a violation by the vehicle based on one or more parking violation rules and one or more violation exceptions.

21. The system according to claim 3, characterized in that in, Each mobile camera is configured to transmit a video stream recorded for a parked vehicle to the streaming gateway in real time.

Images