Method for preventing vehicle theft of one or more vehicles in monitoring area

Abstract

A method for controlling an outdoor lighting system to prevent vehicle theft of one or more vehicles in a monitored area is disclosed. The monitoring area includes a first zone containing one or more vehicles and a second zone adjacent to the first zone. The method includes controlling an outdoor lighting system to illuminate a lighting area according to a first light setting, the lighting area including at least a portion of a first partition and at least a portion of a second partition. The method further includes monitoring the monitored area via the one or more sensors, determining a presence feature of a person in the first zone based on input from the one or more sensors, the presence feature indicating a vehicle theft reconnaissance, and responsive to detecting the presence feature, determining a presence feature of the person in the first zone, the presence feature indicating a vehicle theft reconnaissance. The outdoor lighting system is controlled to illuminate the lighting area in accordance with a second light setting different from the first light setting, where the visibility of at least a portion of the one or more vehicles is reduced in the second light setting.

Description

Technical Field

[0001] The present invention relates to a method for controlling an outdoor lighting system to prevent vehicle theft of one or more vehicles in a monitored area. The present invention also relates to a controller and a computer program product for controlling an outdoor lighting system to prevent vehicle theft of one or more vehicles in a monitored area. The method also relates to a system for preventing vehicle theft of one or more vehicles in a monitored area. Background Art

[0002] Many cities are plagued by a surge in "smash and grab" incidents involving parked vehicles. Smash and grab incidents are often "crimes of opportunity." Offenders see an unattended vehicle in an area where no one is around, break the windows, and take any valuables left in the vehicle.

[0003] Lighting plays a huge role in creating a desired atmosphere and sense of security in a space. Therefore, outdoor lighting (such as lighting systems used for roads, streets, parking facilities, parks, landscaping, sidewalks and bike paths) can affect the rate of automobile crime in residential areas and can be used for the detection and prevention of such automobile break-ins. Summary of the invention

[0004] Interestingly, and contrary to popular belief, increased street lighting may help, rather than hinder, smash and grab incidents. With increased light levels on the street, criminals may be able to more easily see inside a vehicle or assess the security of a vehicle. Criminals may be more easily able to see if a parked car contains valuables when there is a lot of light entering the interior of a vehicle from a street lamp than when there is no light. Additionally, it may also be more difficult to remove objects such as hubcaps from a vehicle without adequate lighting.

[0005] However, reducing street lighting levels at night may raise many safety concerns as citizens worry about increased levels of violence, robbery or residential burglary. Increased lighting deters potential offenders by increasing the risk that they will be seen or recognized when committing a crime. According to the Association of Lighting Professionals, the lack of lighting during the night caused by partial night lighting in the UK has been shown to have an impact on the elderly and more vulnerable groups in society, especially their fear of crime leading to a reluctance to leave their homes at night. Therefore, it may not be desirable to simply reduce street lighting levels to discourage vehicle crime.

[0006] Therefore, an object is to provide a method for controlling an outdoor lighting system to prevent vehicle theft of one or more (motor) vehicles in a monitored area, such as theft from motor vehicles, upon detection of suspicious activity that typically precedes vehicle theft of one or more vehicles, such as car break-ins, for example when a person is scouting for vehicles in an area.

[0007] According to a first aspect, the purpose is achieved by a method for controlling an outdoor lighting system to prevent vehicle theft of one or more vehicles in a monitoring area. The monitoring area includes a first partition containing one or more vehicles and a second partition adjacent to the first partition. As an example, the first partition may include one or more parking spaces in a parking garage, and the second partition may include a sidewalk adjacent to the first partition, through which a person can walk away from his / her parked vehicle or walk to his / her parked vehicle. As another example, the first partition may include a parking lane along a street, and the second partition may include a sidewalk adjacent to the parking lane. Generally speaking, the first partition is where one or more vehicles can be parked, and the second partition is adjacent to the first partition and provides access to and from the vehicle for the legitimate occupants of the parked (multiple) vehicles and the perpetrators of vehicle theft. The second partition may also provide access to pedestrians who are not associated with any vehicle. Vehicle theft includes stealing vehicles and / or stealing internal objects (e.g., wallets, portable screens left in the vehicle) or parts of the vehicle (e.g., hubcaps, catalytic converters, expensive car radios, speakers, LCD dashboard screens). The method comprises controlling an outdoor lighting system to illuminate a lighting area according to a first light setting, the lighting area comprising at least a portion of a first subarea and at least a portion of a second subarea. The first light setting may be a general light setting that illuminates one or more vehicles in the first subarea and, for example, a pedestrian path or sidewalk in the second subarea in a similar manner, for example to meet general lighting regulations.

[0008] The method may include receiving data indicating the first and second partitions. For example, the data may include a floor plan indicating the first and second partitions. The method also includes monitoring the surveillance area via one or more sensors. The method may include identifying and recognizing the first and second partitions based on the input from the one or more sensors, identifying objects such as vehicles and / or people in the first and / or second partitions, and determining a presence signature of a person in the first partition based on the input from the one or more sensors, the presence signature indicating a vehicle theft reconnaissance. A perpetrator of a vehicle theft will need to determine whether the vehicle is worth stealing. For example, whether there are valuable items (e.g., wallets) left inside the vehicle, whether there are people in the vehicle, whether there are valuable items in the vehicle (e.g., hubcaps, expensive radios, high-end LCD user interfaces integrated into the vehicle dashboard, etc.), or whether the vehicle is worth stealing. If there are multiple vehicles, the perpetrator may want to choose the vehicle with the most valuable items inside to steal. In addition, the perpetrator may want to look around to see what happens when he (she) approaches the vehicle, whether bystanders notice, whether alarms sound, etc. Therefore, the perpetrator will conduct a vehicle theft reconnaissance to determine whether to steal the vehicle. Such presence characteristics may include the duration of the person's presence in the first partition, the distance the person has traveled in the first partition, the number of times the person has entered the first partition within a given time period, the person's time-series gaze direction, the person's stress level, or other behaviors or a combination of such behaviors.

[0009] The method also includes, in response to detecting the presence characteristic, controlling the outdoor lighting system to illuminate the lighting area according to a second light setting different from the first light setting, wherein visibility of at least a portion of one or more vehicles is reduced in the second light setting.

[0010] The inventors have appreciated that reducing the visibility of at least a portion of one or more vehicles in a monitoring area can help prevent vehicle theft of one or more vehicles in a monitoring area. In this way, a potential criminal is not able to clearly assess objects within the vehicle and / or components of the vehicle (e.g., a hubcap, an expensive radio) that may be worth stealing. Thus, vehicle theft is prevented.

[0011] One or more sensors (such as cameras, Wi-Fi nodes, ultra-wideband (UWB) sensors, occupancy sensors such as PIR sensors, radar sensors, LiDAR sensors, time-of-flight sensors, audio sensors, vibration sensors, thermopile sensors, etc.) can be used to monitor the surveillance area. Input from one or more sensors can be used to identify and recognize the first and second partitions, and further identify the presence of objects and / or people in the partitions. Input from one or more sensors can also be used to determine the presence characteristics of a person indicating vehicle theft detection, for example, activities / actions that typically precede vehicle thefts of one or more vehicles. For example, time series sensor data from one or more sensors can be used as input to a machine learning model (ML) to determine the presence characteristics of a person indicating vehicle theft detection. Such an ML model may have been trained using labeled instances of presence characteristics indicating vehicle theft detection and corresponding sensor data streams. Additionally and or alternatively, time series or frequency analysis can be applied to the time series sensor data from one or more sensors to determine the presence characteristics of a person indicating vehicle theft detection.

[0012] The method comprises controlling an outdoor lighting system to reduce visibility of at least part of vehicles in a monitored area only when a presence signature of a person indicative of vehicle theft detection is determined. In this way, the outdoor lighting system typically illuminates an illuminated area, such as a road, street, parking facility, etc., according to a first light setting that is considered safe for citizens, and only when a presence signature indicative of vehicle theft detection is determined, the outdoor lighting system illuminates the illuminated area according to a second light setting, wherein visibility of at least part of one or more vehicles is reduced in the second light setting. This is beneficial because the outdoor lighting system is controlled in such a way that the illuminated area is both considered safe for citizens and vehicle theft of one or more vehicles in the area is prevented by conditionally reducing (at least part of) visibility of (one or more) vehicles.

[0013] In a second lighting setting, the outdoor lighting system is controlled so that the lighting level in a first partition in the second lighting setting can be lower than the corresponding lighting level in the first partition in the first lighting setting, so that the visibility of at least part of one or more vehicles is reduced. This is beneficial because by reducing the lighting level in the first partition including one or more vehicles, the visibility of at least part of the vehicles is reduced, and therefore, potential offenders may not be able to clearly assess the vehicle. For example, assess whether objects within the vehicle and / or parts of the vehicle are worth stealing. In the second lighting setting, the lighting level in the adjacent second partition can be higher than the first partition, so that pedestrians and citizens can feel safe. The well-lit adjacent second partition can also increase the recognition of people's faces, their head gestures, and emotions through computer vision, and help improve the effectiveness of video surveillance cameras (CCTV) (if used).

[0014] In the second light setting, the outdoor lighting system may be controlled such that the lighting level in the first zone may also depend on the time of day. For example, when detecting a potential offender, the outdoor lighting system may reduce the level of light emitted in the first zone to a greater extent during periods of low pedestrian traffic (so that the first zone has a lower ambient lighting level) than when detecting a potential offender during periods of high pedestrian traffic.

[0015] The outdoor lighting system may be controlled such that the amount of light entering the interior of one or more vehicles is reduced, such that visibility of the interior of one or more vehicles is reduced. This ensures that a pedestrian's sense of safety can be maintained while the interior of the vehicle is less illuminated than before (and therefore the interior is less visible), and thus the ability of a potential offender to identify and assess objects within the vehicle can be reduced. For example, this ensures that there is sufficient bright light not only on the trail, but also on bushes adjacent to the trail (low light conditions in bushes are known to induce fear as an attacker may be able to hide in the bushes without being seen from a distance).

[0016] The outdoor lighting system may be controlled according to the second light setting such that the vertical illuminance in the first sub-zone is lower than the corresponding vertical illuminance in the first sub-zone at the first light setting, such that visibility of at least part of the interior of one or more vehicles is reduced. Since most vehicles, such as cars, trucks, etc., have opaque metal roofs, the illumination provided by the outdoor lighting system enters the interior of the vehicle through the windows. Therefore, by adaptively controlling the outdoor lighting system to reduce the vertical illuminance in the first sub-zone, this ensures that the amount of light entering the interior of the vehicle is reduced (and therefore the interior is less visible). Preferably, the amount of light falling on the surface of the vehicle (horizontal illuminance) is increased so that the area around the parked vehicle can be adequately illuminated to provide a sense of security for passers-by.

[0017] When the outdoor lighting system is controlled according to the second light setting, the light beam direction of the outdoor lighting system can be directed towards the second sub-area so that the amount of light directed towards the first sub-area is lower than the amount of light directed towards the second sub-area. This ensures that the lighting level in the first sub-area including one or more vehicles is reduced, while the adjacent second sub-area remains sufficiently illuminated.

[0018] The method may also include determining a location of a person in the first zone, and wherein the second lighting setting includes illuminating a coverage area around the location of the person with a first lighting level and illuminating the environment outside the coverage area according to a second lighting level, the second lighting level being higher than the first lighting level. By selectively controlling the outdoor lighting system to reduce the lighting level of the coverage area around the potential offender, visibility of at least a portion of one or more vehicles in the vicinity of the location of the person is reduced. By managing a higher lighting level outside the coverage area, the person can still maintain a sense of security despite the dimming of the lighting level in his / her immediate vicinity.

[0019] The method may further comprise determining, based on input from one or more sensors, that a duration of a person's presence in the first partition may be above a time threshold, and the presence characteristic comprises the duration of the presence. If the person spends a significant amount of time in the first partition, the duration being above a (predetermined) time threshold, this may indicate that the person is scouting / checking inside one or more vehicles (typically activities preceding a vehicle theft of one or more vehicles). Thus, the duration of presence in the first partition may be considered a presence characteristic indicative of vehicle theft detection in the surveillance area.

[0020] The method may also include determining based on input from one or more sensors that a distance traveled by a person in the first partition is above a distance threshold, and the presence characteristic includes the traveled distance. A person scouting (checking inside) more than one vehicle in the first partition may travel a considerable distance (above a predetermined distance threshold) within the first partition. For example, a person moves from one vehicle to another. Therefore, the distance traveled by a person in the first partition may be considered a presence characteristic indicating vehicle theft detection of one or more vehicles in the surveillance area.

[0021] The method may also include determining, based on input from one or more sensors, that the number of times a person has entered the first partition is above a threshold, and the presence characteristic includes the number of times the person has entered the first partition within a predetermined time period. If the person has entered (and correspondingly left) the first partition a considerable number of times (above a predetermined threshold), this may further indicate that the person is scouting (checking) one or more vehicles in the first partition. For example, a person moving from one vehicle (and checking inside the vehicle) to another vehicle may pass between the first and second partitions multiple times. Therefore, the number of times the person enters the first partition can be considered as a presence characteristic indicating vehicle theft detection. The duration of presence can be determined based on the person's behavior over the past few minutes. Alternatively, the monitoring system can identify that the person is returning to the partition, for example after 10 minutes, an hour, or a day; therefore, the duration of presence can also take into account past visits to the partition.

[0022] The method may also include determining, based on input from one or more sensors, that a duration of a gaze directed toward each of the one or more vehicles by a person in the first partition is above a threshold, and the presence characteristic may include the duration of the gaze. If a person in the first partition has looked at a parked vehicle for more than a predetermined time threshold, such as more than 20 seconds, or has looked at multiple parked vehicles, each for more than a predetermined time threshold, this may be an indication that the person is scouting (checking) one or more vehicles in the first partition. Thus, the duration of a person's gaze directed toward each of the one or more vehicles may be considered a presence characteristic indicative of vehicle theft detection.

[0023] The method may also include determining, based on input from one or more sensors, that a stress level of a person in the first partition is above a stress threshold, and the presence signature may include the stress level. If it is inferred that a person in the first partition is experiencing a high stress level, such as by higher breathing and / or heart rate, etc., this may be an indication that the person is performing vehicle theft detection. Thus, the person's stress level may be considered a presence signature indicative of vehicle theft detection.

[0024] The method may also include determining two or more presence features, such as the duration of a person's presence in the first partition, the distance traveled by the person in the first partition, the number of times the person enters the first partition, the person's stress level, etc., analyzing the presence features and determining the vehicle theft detection based on the analysis. For example, a rule-based algorithm may be applied to the presence features to determine actions (activities) prior to vehicle theft of one or more vehicles (vehicle theft detection). Additionally and or alternatively, a machine learning model (ML) may be applied to the presence features to determine actions (activities) prior to vehicle theft of one or more vehicles. Such ML models may have been trained using labeled instances of vehicle theft detection and corresponding presence features. Additionally and or alternatively, a probabilistic machine learning model may be applied to the presence features to determine a probability score for vehicle theft detection (e.g., there is an 80% probability that a person is detecting a car break-in). Additionally and or alternatively, a further probabilistic ML model may be trained to output a probability score for a further action indicating further activity in the monitored area (e.g., there is a 20% probability that the person is waiting for a bus). The output probability score may affect the second lighting setting, for example, the degree of illumination reduction in the first zone may depend on (eg, similar to) the probability score of a person detecting a car break-in.

[0025] According to a second aspect, the object is achieved by a controller for controlling an outdoor lighting system to prevent vehicle theft of one or more vehicles in a monitored area, the monitored area comprising a first partition containing the one or more vehicles and a second partition adjacent to the first partition, the controller being configured to:

[0026] - controlling the outdoor lighting system to illuminate a lighting area according to the first light setting, the lighting area comprising at least part of the first subarea and at least part of the second subarea;

[0027] - receiving input from one or more sensors monitoring a surveillance area;

[0028] - determining a presence characteristic of a person in the first zone based on input from the one or more sensors, the presence characteristic being indicative of vehicle theft detection;

[0029] - in response to determining the presence characteristic, controlling the outdoor lighting system to illuminate the lighting area according to a second light setting different from the first light setting, wherein in the second light setting visibility of at least part of the one or more vehicles is reduced.

[0030] According to a third aspect, the object is achieved by a system for preventing vehicle theft of one or more vehicles in a monitored area, the monitored area comprising a first subarea containing the one or more vehicles and a second subarea adjacent to the first subarea, the system comprising:

[0031] - an outdoor lighting system as described herein;

[0032] - one or more sensors that monitor the surveillance area;

[0033] -A controller as described herein.

[0034] According to a fourth aspect, the object is achieved by a computer program product for a computing device, the computer program product comprising a computer program code for performing the method when the computer program product is run on a processing unit of the computing device.

[0035] It should be understood that the system, method, and computer program product may have similar and / or the same embodiments and advantages as the lighting device described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The above and additional objects, features and advantages of the disclosed systems, apparatus and methods will be better understood through the following illustrative and non-limiting detailed description of embodiments of the apparatus and methods with reference to the accompanying drawings, in which:

[0037] Figure 1 An example of a system for preventing vehicle theft of one or more vehicles is schematically illustrated; and

[0038] Figure 2 A method of controlling an outdoor lighting system to prevent vehicle theft of one or more vehicles is schematically illustrated.

[0039] All the figures are schematic, not necessarily to scale, and generally show only parts which are necessary in order to elucidate the invention, wherein other parts may be omitted or merely suggested. DETAILED DESCRIPTION

[0040] Figure 1 An example of a system 100 for preventing vehicle theft of one or more vehicles 110-116 (such as cars, trucks, etc.) in a monitoring area 120 is shown. The monitoring area 120 can be a road, a street, an indoor / outdoor parking facility, a park, etc. The monitoring area 120 includes a first partition 122 containing one or more vehicles 110-116, such as a dedicated parking partition on one side of the street, a dedicated parking partition in an indoor parking space, and a second partition 124 adjacent to the first partition, such as a street, a passage in a garage. The system 100 includes an outdoor lighting system 140, which includes one or more outdoor lighting devices (such as light poles 141-143) to illuminate a lighting area, which includes at least a portion of the first partition 122 and at least a portion of the second partition 124. The lighting devices 141-143 can be configured to illuminate the lighting area using light technologies such as high intensity discharge (HID), LED, OLED, bioluminescent light source, laser lighting, fluorescent lighting, etc.

[0041] The system 100 also includes one or more sensors 130 that monitor the surveillance area 120. The one or more sensors 130 can be occupancy or motion sensors, such as passive infrared, ultrasonic, time-of-flight sensors, radar sensors, thermopile sensors, Wi-Fi sensor nodes, or even more advanced sensors, such as cameras. The one or more sensors 130 can be stand-alone sensors, or sensors co-located with the outdoor lighting system 140. For example, the outdoor one or more lighting devices 141-143 can include a camera sensor 130 for capturing video and audio information (e.g., pictures, video streams, video recordings).

[0042] The system 100 also includes at least one data processor or controller 106. The controller 106 can be connected and communicate with one or more sensors 130 via a wireless connection, such as via a radio frequency or optical communication link. For example, Wi-Fi, ZigBee, BLE, Lo-Ra, UWB, VLC, IR, Li-Fi, cellular communication, low-orbit satellite communication, etc. The connection can alternatively be wired and power line communication.

[0043] The one or more sensors 130 may include a transmitter (not shown) for transmitting the corresponding signal 41 to the controller 106 via a wired or wireless connection. The controller 106 may include a receiver (not shown) for receiving each corresponding signal 41 from the corresponding one or more sensors 130.

[0044] The controller 106 is configured to control the outdoor lighting system 140 to illuminate a lighting area according to a first light setting, the lighting area including at least a portion of the first zone 122 and at least a portion of the second zone 124. The first light setting may be predefined based on general lighting regulations and guidelines for street lighting applications, parking lot lighting applications, etc.

[0045] The controller 106 is configured to determine a presence feature of the person 104 in the first partition 122 based on input from one or more sensors 130, the presence feature indicating vehicle theft detection. For example, the controller 106 can be configured to first identify and recognize the first and second partitions based on input from the one or more sensors 130, and identify objects and / or people in the first and second partitions. There are several methods and techniques in the prior art that can be used for the purpose of identifying partitions in an area (e.g., the monitoring area 120) and detecting objects in the partitions. For example, image segmentation techniques can be used to identify partitions in an area by grouping pixels with similar features together. In another example, color or texture-based segmentation techniques can be used to distinguish different partitions in the monitoring area 120. Object detection algorithms, such as Haar cascades, deep learning-based methods, such as You Only Look Once (YOLO) or Single Shot MultiBox Detector (SSD) methods, can be used to detect vehicles and / or people in images or video frames. These algorithms can be trained on large data sets, such as annotated parking lot images, to learn features of vehicles, etc. The controller 106 may be configured to determine the presence of features by directly applying a machine learning model, such as a support vector machine model (SVM), a neural network model, a logistic regression model, a graph neural network, an autoencoder, etc., to the time series signals 41 received from one or more sensors 130. A trained machine learning model may make such a determination because the machine learning model may have been trained with inputs that may include instances or segments (time series data) of signals 41 received from one or more sensors 130 monitoring the surveillance area 120, and output corresponding labeled instances of such presence features. In another example, the controller 106 may be configured to determine the presence of features by applying a time series or frequency domain based data analysis to the received time series signals 41 from one or more sensors 130.

[0046] Alternatively, the one or more sensors 130 may have processing capabilities, and the presence characteristic determination may be performed (at least in part) at the one or more sensors 130. For example, a camera sensor with processing capabilities may process video stream data from the surveillance area 120 to determine the presence of the person 104 in the first zone 122, and optionally determine one or more presence characteristics of the person 104 that are indicative of vehicle theft detection. In this case, the controller 106 receives the one or more presence characteristics as input from the one or more sensors 130.

[0047] The presence feature may include a duration of the presence of the person 104 in the first partition 122. The controller 106 may be configured to determine whether the duration of the presence of the person 104 in the first partition 122 is above a time threshold. The time threshold may be predetermined based on expert knowledge, may be adaptively adjusted based on, for example, time of day, day of the week, etc., or may be adaptively adjusted based on adaptive feedback, etc. For example, if the presence of a person entering his / her car or starting one of the (rear) doors of the car in the first partition takes no more than 30 seconds, then a time threshold close to or much higher than 30 seconds may be set. The time threshold may also be adjusted based on the location type. For example, near a parking meter (where people are waiting to pay for parking tickets), a different time threshold may be applied than near an ATM machine, because the typical dwell time and human behavior associated with each of these locations is different. A self-learning sensor system may be used to automatically identify (e.g., during a learning-in period) location types, such as residential streets, near a parking meter, near an ATM, etc. Subsequently, the time threshold may be adjusted based on the identified location type. For example, for a parking meter, the time threshold may be increased.

[0048] The presence feature may include a distance traveled by the person 104 in the first partition 122. The controller 106 may be configured to determine whether the distance traveled by the person 104 in the first partition 122 is above a distance threshold. The distance threshold may be predetermined based on expert knowledge, may be adaptively adjusted based on, for example, the time of day, the day of the week, etc., or may be adaptively adjusted based on adaptive feedback (e.g., adaptive learning of system parameters), etc. For example, if the maximum distance required for a person from the moment he (she) enters the first partition 122 to the moment he (she) enters his (her) car is approximately 100m, then a distance threshold close to or much higher than 100m may be set. The distance threshold may be adjusted for a specific partition, for example, by averaging the measured distance that people need to travel from the moment they enter the first partition 122 to the moment they enter their vehicle (which may depend on the size of the first partition 122).

[0049] The presence feature may include the number of times the person 104 enters (and respectively leaves) the first partition 122. The controller 106 may be configured to determine whether the number of times the person 104 enters the first partition 122 is above a (predetermined) threshold. The threshold may be predetermined based on expert knowledge, may be adaptively adjusted based on, for example, the time of day, the day of the week, etc., or may be adaptively adjusted based on adaptive feedback (e.g., adaptive learning of system parameters), etc. For example, a person who wants to enter his / her vehicle typically walks in an adjusted area 124 (e.g., a pedestrian area) and enters the first partition 122 (once) only to move toward his / her vehicle. A person follows the opposite path to leave his / her vehicle. Therefore, a person entering / leaving the first partition 122 multiple times may be an indication that the person is scouting / checking multiple vehicles in the first partition 122. Note that a person may enter the first partition 122 more than once without intending to perform a vehicle theft, for example because he / she has forgotten the exact location of his / her vehicle, he / she may return to a parked vehicle to retrieve an object, etc. The threshold may be adjusted depending on the time of day or the day of the week. For example, during busy hours / days when there are many vehicles in the first partition 122, a person may need to put in more effort to find his / her car, and thus, the threshold may be adjusted accordingly.

[0050] The presence feature may include a duration that the person 104 in the first partition 122 is directed toward looking at each of the one or more vehicles 110-116. Several methods and techniques for sensor-based gaze direction determination are known in the art. For example, by analyzing the head posture of the person 104 (by monitoring the person's neck, etc.). The controller 106 may be configured to determine whether the duration that the person 104 looks toward each of the one or more vehicles 110-116 is above a threshold. The threshold may be predetermined based on expert knowledge, may be adaptively adjusted based on, for example, the time of day, the day of the week, etc., or may be adaptively adjusted based on adaptive feedback (e.g., adaptive learning of system parameters), etc. For example, it may be typical for the person 104 to look at his / her car before entering the car, but this may not last more than, say, 20 seconds, so a duration threshold close to or much higher than 20 seconds may be set.

[0051] The controller 106 may also be configured to determine a stress level of the person 104 based on input from one or more sensors 130. Several methods and techniques for remotely detecting stress levels based on sensors are known in the art and may be used to determine the stress level of the person 104 moving in the surveillance area 120. For example, a frequency modulated continuous wave (FMCW) radar sensor or Wi-Fi sensing may be used to remotely determine a person's breathing rate and / or heart rate, from which the person's current stress level may be inferred. Video-based stress detection via deep learning is also well known in the art. Potential thieves experience high stress levels when investigating a theft. Thus, the stress level of the person 104 may be a presence signature indicative of vehicle theft detection.

[0052] The controller 106 may also be configured to determine one or more presence features, analyze the presence features, and determine actions / activities prior to vehicle theft, such as vehicle theft reconnaissance, for one or more vehicles 110-116 based on the analysis of the one or more presence features. For example, the controller 106 may be configured to determine vehicle theft reconnaissance by applying a rule-based algorithm or machine learning model (such as an SVM, a neural network model, a logistic regression model, etc.) to one or more presence features. A trained machine learning model may make such a determination because the machine learning model may have been trained with inputs that may include instances of the presence features and output a corresponding action labeled as vehicle theft reconnaissance. Additionally and or alternatively, the controller 106 may be configured to determine the probability of vehicle theft reconnaissance by applying a probabilistic machine learning model (such as a probabilistic neural network model, a Bayesian model, etc.) to one or more presence features.

[0053] The controller 106 may be configured to, in response to detecting a presence signature indicative of vehicle theft detection of one or more vehicles 110 - 116 , control the outdoor lighting system 140 to illuminate the lighting area according to a second light setting different from the first light setting in which visibility of at least a portion of the one or more vehicles 110 - 116 is reduced.

[0054] For example, the controller 106 may be configured to set the lighting level in the first partition 122 including one or more vehicles 110-116 to a lower level than the first lighting setting in the second lighting setting. In this way, the visibility of at least part of the one or more vehicles 110-116 is reduced, while the visibility of the adjacent second partition 124 (e.g., including a corridor in a parking garage, a sidewalk in a street, etc.) is maintained at a light level that is comfortable enough for citizens. The level of the lighting setting in the first partition 122 may depend on the time of day. For example, when detecting potential offenders, the light level in the first partition may be reduced to a greater extent (at a lower lighting level) during periods of low pedestrian traffic than when detecting potential offenders during periods of heavy pedestrian traffic. The light level may also depend on the age of the person. If the system 100 infers that the person is older, the elderly require higher light levels due to poorer eyesight and a higher risk of falling and a higher risk of being robbed, so the light level is increased.

[0055] The controller 106 may be configured to set the lighting level in the first zone 122 at the second light setting such that the vertical illuminance of the first zone 122 is lower than the corresponding vertical illuminance in the first zone 122 at the first light setting. Techniques and design tools for independently controlling the vertical illuminance (the amount of light falling on vertical surfaces, such as the amount of light entering the interior of a car through a window) and the horizontal illuminance (the amount of light falling on horizontal surfaces, such as the roof of a car) of a space are well known in the retail lighting and hospitality industries and will not be discussed further in the context of this application.

[0056] The controller 106 can be configured to direct the light beam direction of one or more lighting devices 141-143 of the outdoor lighting system 140 toward the second subarea 124, so that the amount of light directed toward the first subarea 122 is lower than the amount of light directed toward the second subarea 124. For example, by using a turning film installed inside or above one or more lighting devices 141-143, which allows redirecting and reducing or increasing the beam angle of the light source.

[0057] The controller 106 may also be configured to determine the location of the person 104 based on input from one or more sensors 130. Several methods and techniques for sensor-based detection, location, and tracking of people in dynamic environments are known in the art and may be used to determine and track the location of the person 104 moving in the surveillance area 120. The controller 106 may also be configured to illuminate a coverage area 150 around the location of the person with a first illumination level and illuminate the environment outside the coverage area 150 according to a second illumination level. The second illumination level may be higher than the first illumination level.

[0058] The system 100 may also include at least one data repository or storage or memory for storing computer program code instructions. Alternatively, however, the system 100 may include a server. One or more sensors 130 may transmit their corresponding signals 41 to the server so that the server can obtain each corresponding signal 41. The controller 106 may then be configured to retrieve (receive) the corresponding signal 41 from the server. The controller 206 may be communicatively coupled to the cloud.

[0059] Figure 2 A method 200 is shown of controlling the outdoor lighting system 140 to prevent vehicle theft of one or more vehicles 110-116 in the monitoring area 120. The method 200 may include the following steps:

[0060] - controlling 202 the outdoor lighting system 140 by the controller 106 according to the first light setting to illuminate a lighting area, the lighting area comprising at least a portion of the first subarea 122 and at least a portion of the second subarea 124;

[0061] - monitoring 204 a surveillance area via one or more sensors 130 ;

[0062] - determining 206 , by the controller 106 , a presence characteristic of a person 104 in the first zone 122 based on input from the one or more sensors 130 , the presence characteristic being indicative of the vehicle theft detection;

[0063] In response to detecting the presence feature, controlling 208 the outdoor lighting system 140 by the controller 106 to illuminate the lighting area according to a second light setting different from the first light setting, wherein in the second light setting visibility inside one or more vehicles 110 - 116 is reduced.

[0064] The method 200 may be performed by computer program code of a computer program product when the computer program product is executed on a processing unit of a computing device, such as the controller 106 .

[0065] It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims.

[0066] In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The use of the verb "comprise" and its conjugations does not exclude the presence of elements or steps other than those stated in a claim. The article "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer or processing unit. In a device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0067] Aspects of the present invention can be implemented in a computer program product, which can be a set of computer program instructions stored on a computer-readable storage device that can be executed by a computer. Instructions of the present invention can be any interpretable or executable code mechanism, including but not limited to scripts, interpretable programs, dynamic link libraries (DLLs), or Java classes. Instructions can be provided as complete executable programs, partial executable programs, as modifications (e.g., updates) of existing programs, or as extensions (e.g., plug-ins) of existing programs. In addition, partial processing of the present invention can be distributed over multiple computers or processors or even "clouds".

[0068] Storage media suitable for storing computer program instructions include all forms of non-volatile memory, including but not limited to EPROM, EEPROM and flash memory devices, magnetic disks such as internal and external hard drives, removable disks and CD-ROM disks. The computer program product may be distributed on such storage media or may be provided for download via HTTP, FTP, email or through a server connected to a network such as the Internet.

Claims

1. A method for controlling an outdoor lighting system (140) to prevent vehicle theft of one or more vehicles (110-116) in a monitored area, the monitored area comprising a first zone (122) containing the one or more vehicles and a second zone (124) adjacent to the first zone, the method comprising the following steps: - controlling (202) the outdoor lighting system to illuminate a lighting area according to the first light setting, the lighting area comprising at least part of the first subarea and at least part of the second subarea; - monitoring (204) a surveillance area via one or more sensors (130); - based on input from the one or more sensors, determining (206) a presence characteristic of a person (104) in the first zone, the presence characteristic being indicative of vehicle theft detection; - in response to detecting the presence characteristic, controlling (208) the outdoor lighting system to illuminate the lighting area according to a second light setting different from the first light setting, wherein in the second light setting: - reduced visibility of at least part of the one or more vehicles, and - The lighting level in the second subarea is higher than the lighting level in the first subarea.

2. The method according to claim 1, wherein: In the second lighting setting, the outdoor lighting system (140) is controlled so that the amount of light entering the interior of the one or more vehicles (110-116) is reduced, so that visibility of the interior of the one or more vehicles is reduced.

3. A method according to any of the preceding claims, wherein the outdoor lighting system (140) is controlled so that the lighting level in the first partition (122) in the second lighting setting is lower than the lighting level in the first partition in the first lighting setting, so that the visibility of at least part of the one or more vehicles (110-116) is reduced.

4. The method according to any of the preceding claims, wherein in the second light setting the outdoor lighting system (140) is controlled such that the lighting level in the first subarea (122) depends on the time of day.

5. The method of claim 2, wherein the outdoor lighting system (140) is controlled so that in the second light setting, the vertical illuminance in the first partition (122) is lower than in the first lighting setting, thereby reducing visibility of at least part of the interior of the one or more vehicles (110-116).

6. A method according to any of the preceding claims, wherein the method further comprises determining a position of a person (104) in the first partition (122), and wherein the second lighting setting comprises illuminating a coverage area (150) around the position of the person with a first lighting level and illuminating the environment outside the coverage area according to a second lighting level, the second lighting level being higher than the first lighting level.

7. A method according to any of the preceding claims, wherein the method includes determining, based on input from the one or more sensors (130), that a duration of presence of a person (104) in the first partition (122) is above a time threshold, and wherein the presence characteristic includes the duration of the presence.

8. A method according to any of the preceding claims, wherein the method includes determining that a distance traveled by a person (104) in the first partition (122) is above a distance threshold based on input from the one or more sensors (130), and wherein the presence feature includes the travel distance.

9. A method according to any of the preceding claims, wherein the method includes determining, based on input from the one or more sensors (130), that the number of times a person (104) enters the first partition is above a threshold, and wherein the presence characteristic includes the number of times the person enters the first partition within a predetermined time period.

10. A method according to any of the preceding claims, wherein the method includes determining, based on input from the one or more sensors (130), that a duration of a gaze of a person (104) directed toward each of the one or more vehicles (110-116) in the first partition (122) is above a threshold, and wherein the presence characteristic includes the duration of the gaze.

11. A method according to any of the preceding claims, wherein the method comprises determining, based on input from the one or more sensors (130), that a stress level of a person (104) in the first partition (122) is above a stress threshold, and wherein the presence characteristic comprises the stress level.

12. The method according to any of the preceding claims, wherein the method further comprises determining two or more presence signatures according to claims 7-11, analyzing the presence signatures, and determining the vehicle theft detection based on the analysis.

13. A controller (106) for controlling an outdoor lighting system (140) to prevent vehicle theft of one or more vehicles (110-116) in a monitored area, the monitored area comprising a first zone (122) containing the one or more vehicles and a second zone (124) adjacent to the first zone, the controller being configured to: - controlling the outdoor lighting system to illuminate a lighting area according to the first light setting, the lighting area comprising at least part of the first subarea and at least part of the second subarea; - receiving input from one or more sensors (130) monitoring a surveillance area; - determining a presence characteristic of a person in the first zone based on input from the one or more sensors, the presence characteristic being indicative of vehicle theft detection; - in response to determining the presence characteristic, controlling the outdoor lighting system to illuminate the lighting area according to a second light setting different from the first light setting, wherein in the second light setting: - reduced visibility of at least part of the one or more vehicles, and - The lighting level in the second subarea is higher than the lighting level in the first subarea.

14. A system for preventing vehicle theft of one or more vehicles in a monitored area, the monitored area comprising a first subarea containing the one or more vehicles and a second subarea adjacent to the first subarea, the system comprising: - The outdoor lighting system according to claim 1; - one or more sensors that monitor the surveillance area; - A controller according to claim 13.

15. A computer program product for a computing device, the computer program product comprising computer program code for performing the method of any one of claims 1 to 11 when the computer program product is run on a processing unit of the computing device.

Images