Description:Complete Specification:
The following specification describes and ascertains the nature of this invention and the manner in which it is to be performed.

Field of the invention
[0001] The present disclosure relates to a system and a method for detecting potential threat of vandalism to a vehicle in a parking place. More specifically, it also relates to providing real time threat assessment to a vehicle in a parking place using Artificial Intelligence (AI) models.
Background of the invention
[0002] Most of the parking lots and garages provide criminals with many opportunities such as theft, vandalism, etc. Most of the multi-storied and underground car parking lots are unmanned, where parked vehicles are monitored remotely. There are several instances, where vehicles are left unattended for long time which lead to the instances of vandalism and theft. Vandalism incudes car scratch, theft, smashed wind mirrors, windows, lights, snapped windscreen wipes, etc. At present, there is no mean by which the presence of an unusual event near the parked vehicle can be detected, so that the events related to the vandalism and theft can be avoided.
[0003] US2016096506AA provides a method and system to detect a predefined activity, e.g., vandalism, loitering, trespass, etc., associated with a parked vehicle using video.
Brief Disclosure of the Invention
[0004] The invention discloses a system for detecting potential threat of vandalism to a vehicle in a parking place. The system comprises one or more cameras and microphones installed in the parking place to capture image and sound in the parking place. A controller analyzes the image and the sound for judging probability of vandalism to the vehicle in the parking place based on one or more specific parameters.
[0005] Examples of the specific parameter include but not limited to, a location of the parking place, a type of the parking place, a parking duration, a time and/or season of the parking, a vehicle model, regularity of parking at the same parking place, and a duration of a presence of a human/object in a vicinity of the parking place, a sound pattern, etc.
[0006] The controller further performs real time threat assessment to a vehicle based on one or more of the parameters but not limited to, detection of a movable object in a vicinity of the vehicle, a visit to the vehicle by a human/object, a difference in a color pattern of a human/object in a vicinity of the vehicle, number of human/objects detected in the vicinity as compared to that at the time of parking, a duration of a movement around the vehicle, a sound detected during a movement around the vehicle, etc.
[0007] In an embodiment of the invention, the controller is an edge controller installed in the parking place.
[0008] In another embodiment of the invention, the controller is a remote controller connected to the camera and the microphone. The connection is a wired or wireless.
[0009] The controller uses at least one artificial intelligence (AI) model to judge the probability of vandalism based on a sound pattern and an image patten. The controller raises an alert to one or more of a user of the vehicle, a parking management personnel, a security personnel, a law enforcement personnel, etc. based on the probability of vandalism and/or the real time threat assessment.
[0010] The controller also triangulates a source of noise and takes an appropriate action.
[0011] The invention also discloses a method for detecting potential threat of vandalism to a vehicle. The method involves receiving visual and audio inputs from one or more cameras and microphone installed in a parking place. The image and the sound are analyzed for judging a probability of vandalism to the vehicle in the parking place based on at least one specific parameters.
[0012] The specific parameter includes, but not limited to, locations of the parking place, a type of the parking place, a parking duration, and a regularity of parking, a time and/or season of the parking, a vehicle model, regularity of parking at the same parking place, a duration of a presence of a human/object in a vicinity of the parking place, a sound pattern, etc.
[0013] The method further involves performing real time threat assessment to a vehicle based on one or more of the detections of a movable object in a vicinity of the vehicle, a visit to the vehicle by a human/object, a difference in a color pattern of a human/object in a vicinity of the vehicle, a duration of a movement around the vehicle, a sound detected during a movement around the vehicle, etc.
[0014] One or more artificial intelligence (AI) models are used to judge the probability of vandalism based on a sound pattern and an image patten.
Brief description of the drawings
[0015] An embodiment of the disclosure is described with reference to the following accompanying drawing.
[0016] Fig. 1 shows a schematic block diagram of a system for detecting potential threat of vandalism to a vehicle in a parking place, according to an embodiment of the present invention.
[0017] Fig. 2 shows a schematic block diagram of a system for detecting potential threat of vandalism to a vehicle in a parking place according to an embodiment of the present invention, where controller is a remote controller.
[0018] Fig. 3 shows a flow chart of a method for detecting potential threat of vandalism to a vehicle in a parking place, according to an embodiment of the present invention.
Detailed description of the drawings
[0019] Fig. 1 shows a schematic block diagram of a system for detecting potential threat of vandalism to a vehicle in a parking place, according to an embodiment of the present invention. In Fig. 1, a parking place 100 with a system 101 for detecting potential threat of vandalism to a vehicle is depicted. It should be understood at the outset that, although exemplary embodiments are illustrated in the figures and describer below, the present disclosure should in no way be limited to the exemplary implementation and techniques illustrated in the drawings and described below.
[0020] The system 101 comprises one or more cameras 103 and microphones 104 installed in different locations of the parking place 100. It is possible to install the cameras 103 and microphones 104 separately. Alternatively, several modules comprising integrated cameras 103 and the microphones 104 together can also be installed. The captured images and sound are analyzed by a controller 102 for judging the probability of vandalism based on one or more specific parameters.
[0021] In an embodiment of the invention, the controller 102 is an edge controller installed in the parking place 100. The edge controller 102 is a part of a module comprising camera 103 and microphone 104.
[0022] In another embodiment of the invention, the edge controller 102 is an independent controller 102 connected to one or more cameras 103 and microphones 104.
[0023] In yet another embodiment of the invention, the controller 102 is a remote controller connected to the camera 103 and the microphone 104 through wired or wireless connections such as Bluetooth, wifi, etc. Fig. 2 shows a schematic block diagram of a system for detecting potential threat of vandalism to a vehicle in a parking place according to this embodiment of the present invention, where controller is a remote controller 102.
[0024] Examples of the specific parameters include, but not limited to, a location of parking place of vehicle, type of the parking place, parking duration, regularity of parking in that place, time of the parking, season of the parking, vehicle model, frequency of the parking, duration of human/object presence in the vicinity of the parking place, sound pattern, etc.
[0025] For instance, if the image and sound captured indicate that the location of the parking place is at a remote place, type of the parking place is a public parking place (not a secured private parking place), and there is an indication from the sound pattern that the vehicle door is being opened, then it indicates that the possibility of vandalism is high. Similarly, different combinations of such specific parameters indicate the possibility of the vandalism to the vehicle.
[0026] The controller 102 uses one or more artificial intelligence (AI) models to judge the probability of vandalism based on a sound pattern and an image patten. Based on the captured image, sound and one or more specific parameters, the controller 102 detects one or more unusual events. Based on a pre-programmed learned sound pattern and a reference image patten, the AI module inside the controller 102 judges the possibility of vandalism to the vehicle. The controller 102 triangulates the source of noise and takes appropriate action.
[0027] For instance, when the controller 102 identifies that there is an intruder in the parking place 100 and he is having some weapon from the image and sound, the controller 102 raises an alert to different people like the user, parking management personnel, police, or necessary public authorities.
[0028] In other embodiment, as described in figure 2, the images and sounds captured by camera 103 and microphone 104 are sent to the remote controller 102 having an artificial intelligence module through cloud for processing to judge the probability of vandalism based on a sound and an image pattern.
[0029] The system 101 also performs the real time threat assessment of threat to the vehicle based on one or more following:
- Movable objects in the vicinity of the parked vehicle.
- Multiple visits to the car by human/object within short time.
- Differing color pattern of the human/object in the vicinity of the vehicle.
- Duration of movement around the vehicle.
- Number of human/objects detected in the vicinity as compared to that at the time of parking.
- Sound detected when there is movement of the car and movement of the vehicle, etc.
[0030] Based on the probability of vandalism and/or real time threat assessment, the alert is raised to a user of the vehicle, a parking management personnel, a security personnel, a law enforcement personnel, etc. However, the communication of the alert is not restricted to only those mentioned above.
[0031] The reference image and audio patterns for detecting the possibility of vandalism and unusual event are updated by controller 102 on regular intervals in the case of any change in the parking place.
[0032] Fig. 3 shows a flow chart of a method for detecting potential threat of
vandalism to a vehicle in a parking place, according to an embodiment of the present invention.
[0033] The method involves receiving real time visual and audio inputs from one or more cameras 103 and microphones 104 in a parking place 100 at step 302. The camera 103 and microphone 104 are not restricted to only one place The method step 304 involves analyzing the image and the sound for judging a probability of vandalism to the vehicle in the parking place 100 based on at least one specific parameter by a controller 102.
[0034] The specific parameter includes, but not limited to, locations of the parking place, a type of the parking place, a parking duration, and a regularity of parking, a time and/or season of the parking, a vehicle model, regularity of parking at the same parking place, and a duration of a presence of a human/object in a vicinity of the parking place, a sound pattern, etc.
[0035] The method further involves performing real time threat assessment to a vehicle based on one or more of the detections of a movable object in a vicinity of the vehicle, a visit to the vehicle by a human/object, a difference in a color pattern of a human/object in a vicinity of the vehicle, a duration of a movement around the vehicle, a sound detected during a movement around the vehicle, etc.
[0036] One or more artificial intelligence (AI) models are used to judge the probability of vandalism based on a sound pattern and an image patten. The AI models are trained continuously based on different sound and image data collected at different times from same parking place or similar parking place.
[0037] The image and sound pattern for detecting probability of vandalism are collected specifically for each parking lot/places. Using these collected data and incident history, a heat map or a threat map indicating different levels threat probability is created for each parking place. The heat map or the threat map comprises different ways of indicating the threat level such as color, symbol, etc.
[0038] The invention enables the user to understand the safety level of different parking areas at different times based on combination of past analyzed data and real time threat assessment. The system users update AI models to precisely determine the possibility of vandalism to a vehicle. The invention can be implemented in the form of a parking app or a web version, whereby user can see different threat levels associated with different parking lots and different. Thereby, the system can be used to adopt selective parking fee charging system based on threat level.
[0039] The invention enables providing real time alert to user and different stake holders to improve the safety of the vehicles. When a user is driving to a new place and looking for a reliable private or public parking place, the invention can be very much useful. The invention can be used by the parking management companies, public authorities, city corporations, event management companies, etc.
[0040] The invention is very useful during major fests, or sports events, political events, gatherings, etc., where large number of parking of vehicles is expected. The system can provide a user-friendly color coded or symbolic threat level assessment to user before he/she makes a parking decision. By implementing this invention into existing parking places/ lots, garages, current incident of vandalism, theft can be avoided.
[0041] It must be understood that the embodiments explained in the above detailed description are only illustrative and do not limit the scope of this invention. Any modification to the vandalism detection system using microphone and camera having artificial intelligence model and the method thereof are envisaged and form a part of this invention. The scope of this invention is limited only by the claims.
, Claims:We claim:
1. A system (101) for detecting potential threat of vandalism to a vehicle in a parking place, said system comprises,
- at least one camera (103) and a microphone (104) installed in said parking place (100) to capture at least one image and sound in said parking place (100), characterized in that system:
- a controller (102) to analyze said image and said sound for judging probability of vandalism to said vehicle in said parking place (100) based on at least one specific parameter.
2. The system (101) as claimed in claim 1, wherein said at least one specific parameter includes at least one of a location of said parking place, a type of said parking place, a parking duration, a time and/or season of the parking, a vehicle model, regularity of parking at the same parking place, and a duration of a presence of a human/object in a vicinity of the parking place, and a sound pattern.
3. The system (101) as claimed in claim 1, wherein said controller (102) further performs real time threat assessment to a vehicle based on at least one of a detection of a movable object in a vicinity of said vehicle, a visit to said vehicle by a human/object, a difference in a color pattern of a human/object in a vicinity of said vehicle, a duration of a movement around said vehicle, a sound detected during a movement around said vehicle.
4. The system (101) as claimed in claim 1, wherein said controller (102) is an edge controller installed in the parking place (100).
5. The system (101) as claimed in claim 1, wherein said controller (102) is a remote controller connected to said at least one camera (103) and at least one microphone (104).
6. The system (101) as claimed in claim 1, wherein said controller (102) uses at least one artificial intelligence (AI) model to judge the probability of vandalism based on a sound pattern and an image patten.
7. The system (101) as claimed in claim 3, wherein said controller (102) raises an alert to at least one user of said vehicle, a parking management personnel, a security personnel, a law enforcement personnel based on said probability of vandalism and/or said real time threat assessment.
8. The system (101) as claimed in claim 1, wherein said controller (102) triangulates a source of noise and takes an appropriate action.
9. A method for detecting potential threat of vandalism to a vehicle, the method comprises:
- receiving visual and audio inputs from at least one of a camera (103) and a microphone (104) installed at a several locations in a parking place (100), characterized in that method:
- analyzing said image and said sound for judging a probability of vandalism to said vehicle in said parking place (100) based on at least one specific parameters.
10. The method as claimed in claim 9, wherein said at least one specific parameter includes at least one of the locations of said parking place (100), a type of said parking place, a parking duration, and a regularity of parking, a time and/or season of the parking, a vehicle model, regularity of parking at the same parking place, and a duration of a presence of a human/object in a vicinity of the parking place, and a sound pattern.
11. The method as claimed in claim 9, further comprises performing real time threat assessment to a vehicle based on at least one of a detection of a movable object in a vicinity of said vehicle, a visit to said vehicle by a human/object, a difference in a color pattern of a human/object in a vicinity of said vehicle, a duration of a movement around said vehicle, a sound detected during a movement around said vehicle.
12. The method as claimed in claim 9, wherein at least one artificial intelligence (AI) model is used to judge the probability of vandalism based on a sound pattern and an image pattern