Driving assistance system and method

Abstract

TECHNICAL FIELD The present invention belongs to the field of intelligentization of vehicles, and more specifically, relates to a manner of recognizing whether there is interaction between an occupant and a driver in a vehicle, and correspondingly enabling or disabling a driving assistance function of the vehicle, thereby improving driving safety.SOLUTION: The present invention provides a driving assistance method and apparatus. The method includes: identifying an environmental state in a vehicle; and activating a voice assistant to perform a human-computer interaction with a driver by voice when the environmental state is a silent state within a preset time period.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to the field of smart vehicles, and more particularly to an aspect of improving driving safety by recognizing whether there is interaction between the vehicle occupants and the driver, and turning on or off the vehicle's driving assistance functions accordingly. [Background technology]

[0002] Maintaining a driver's alertness while driving a vehicle is extremely important for driving safety, and experience has shown that the atmosphere inside the vehicle has a significant impact on the driver. For example, if someone in the vehicle falls asleep while the driver is driving, especially if a passenger in the front seat (i.e., a passenger seat occupant) falls asleep, the following potential dangers and effects may occur to the driver's driving:

[0003] 1. Distraction: When a passenger falls asleep, the driver may feel lonely because they have no one to communicate with or share the driving task with. This loneliness can lead to driver distraction and a decrease in alertness.

[0004] 2. Lack of assistance: The passenger usually plays a support role, helping the driver observe road conditions (e.g., assisting with route decisions on unfamiliar roads), navigate, and adjust vehicle controls. If the passenger falls asleep, the driver may miss important information and assistance functions, making driving more difficult.

[0005] Therefore, from the driving experience, if someone in the car falls asleep, especially the front passenger, and there is no one to interact with the driver, this may increase driving risks as the driver has to deal with more challenges and stress. Summary of the Invention [Means for solving the problem]

[0006] The present invention focuses on the driver's driving atmosphere, such as whether the environmental condition inside the vehicle is silent. Here, a silent state may mean, for example, that there is no interaction inside the vehicle or that multimedia such as music playback is not in use. When the atmosphere may affect driving safety (for example, when the front passenger or all passengers are asleep), a smart voice assistant or driving assistance system is activated to provide a driving assistance aspect that improves driving safety. For example, activating the smart voice assistant to maintain voice-based man-machine interaction with the driver keeps the driver active, thereby achieving the goal of helping the driver maintain an excited state while driving. Furthermore, activating an advanced driver assistance system (ADAS) equipped in the vehicle can assist the driver in driving control, thereby avoiding driving safety issues caused by a silent state such as a drowsy atmosphere affecting the driver and reducing their vehicle control ability, such as slowed reaction. Therefore, aspects of the present invention can solve problems such as driver distraction and lack of assistance when there is no one to interact with the driver in silent situations where someone in the car falls asleep, for example. [Brief explanation of the drawings]

[0007] [Figure 1] 1 is a block diagram of a driving assistance system according to an embodiment of the present invention. [Figure 2] 2 is a flowchart of a driving assistance method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, numerous specific details will be set forth to enable those skilled in the art to more comprehensively understand the present invention. However, it will be apparent to those skilled in the art that the present invention may be realized without some of these specific details. It should also be understood that the present invention is not limited to the specific embodiments described above. On the contrary, the present invention can be implemented by any combination of the features and elements described herein.

[0009] FIG. 1 shows a block diagram of a driving assistance device according to an example of the present invention. As shown in the figure, the driving assistance device 100 can be communicatively connected to one or more cameras CAM1 and CAM2 installed in the vehicle cabin, and these cameras CAM are used to acquire a target image including all passengers in the vehicle. When multiple cameras are installed, these cameras can be configured to intentionally capture target images of passengers in specific areas or seats in the vehicle. For example, a camera CAM1 installed on or near the rearview mirror can capture a target image IMG1 including front passengers (including the driver and passenger), and another camera CAM2 installed in a central position on the roof can capture a target image IMG2 including rear passengers. Of course, in another example, a single camera can be installed to capture a target image including all passengers in the vehicle. For example, a single in-vehicle camera can be used to simultaneously capture images of all seats in the vehicle from the front, thereby acquiring a target image including all passengers in the vehicle. The driving assistance device 100 can also be communicatively connected to a microphone (not shown) installed in the vehicle cabin, thereby picking up in-vehicle sounds, including voice calls or multimedia, such as music playback, of people inside the vehicle through the microphone.

[0010] As shown in the figure, the driving assistance device 100 includes an image receiving module 101, a state analysis module 102, and a function activation module 103. The image receiving module 101 receives one or more image frames IMG of passengers in the vehicle transmitted from a camera CAM in the vehicle, and as described above, these image frames include captured images of all passengers, including the driver and passengers in other seats.

[0011] The state analysis module 102 is configured to recognize the environmental state inside the vehicle. In the present invention, the environmental state is defined as whether there is sound inside the vehicle. If there is sound, the environmental state is a sound state, and if there is no sound, the environmental state is a silent state. Therefore, here, recognizing the environmental state inside the vehicle includes determining whether there is an interaction, such as a chat or conversation, between passengers inside the vehicle, between the passengers inside the vehicle and the driver, or whether there is an interaction between the driver and the outside, such as whether the driver is making a phone call, or whether multimedia playback is taking place inside the vehicle. If it is determined that there is no interaction or no multimedia playback, the state analysis module 102 determines that the current environmental state inside the vehicle is a silent state.

[0012] Since each person's sleep state clearly indicates whether that person is in a silent state, according to one embodiment of the present invention, the state analysis module 102 can recognize the sleep state of passengers in the vehicle to determine whether there is interaction between passengers or between a passenger and a driver. Therefore, the state analysis module 102 uses the camera CAM to acquire image frames IMG of each passenger and driver, and performs image analysis and recognition processing on the received image frames IMG to confirm the state of the passengers in the vehicle, including whether the driver's state and passengers in seats other than the driver in the vehicle are sleeping, thereby determining whether there is interaction between these passengers and the driver of the vehicle. According to an embodiment of the present invention, the state analysis module 102 needs to distinguish between the driver and other passengers in the vehicle and the specific positions where the passengers are located, i.e., whether they are in the passenger seat or the rear seat, based on the received images, and recognize whether these passengers are sleeping. Therefore, as shown in FIG. 1 , the state analysis module 102 includes a target detection unit 1021 and a state recognition unit 1022. The target detection unit 1021 is configured to distinguish between passengers in the vehicle, including the driver, a front passenger seat passenger, and a rear passenger seat passenger. The target detection unit 1021 can distinguish between passengers using a known target detection algorithm in the prior art. For example, the target detection unit 1021 can use an advanced target detection algorithm, such as YOLO, SSD, or Faster R-CNN, to recognize the position and posture of each person in the vehicle. For example, the target detection unit 1021 detects the driver, front passenger seat passenger, and rear passenger seat passenger by processing a target image IMG including all passengers captured by a single camera and recognizing the position and posture of each person in the vehicle. It is easy to understand that when multiple cameras, such as CAM1, capture only target images IMG1 of front passenger seat passengers and another camera, such as CAM2, capture only target images IMG2 of rear passenger seat passengers, the target detection unit 1021 only needs to distinguish between the driver and the front passenger seat passenger, thereby reducing the computational load of the target detection unit 1021.

[0013] The state recognition unit 1022 recognizes whether the driver, front passenger, and rear passenger are asleep based on the detection results provided by the target detection unit 1021. The state recognition unit 1022 uses a behavior recognition algorithm known in the prior art to analyze the passengers' facial expressions, eye closure, head posture, and the like to determine whether they are asleep and generates a sleep state indication signal SIG accordingly. The state analysis module 102 continuously analyzes continuously received images IMG to monitor and update the passengers' sleep states in real time, and determines whether there is an interaction between the front passenger or rear passenger and the vehicle driver based on the sleep state indication signal SIG of the passenger or rear passenger in the vehicle generated by the state recognition unit 1022. Figure 2 shows a control flow for determining an interaction according to an example of the present invention.

[0014] As shown in FIG. 2, in step 201, a sleep state indication signal SIG of an in-vehicle occupant is received, where the signal SIG includes occupant state information such as (1) a driver, (2) a front passenger, and (3) one or more rear passengers.

[0015] In step 203, it is determined whether all passengers other than the driver, including the passenger in the front seat and passengers in the rear seats, are asleep. If the determination result is YES, the process proceeds to step 205. In step 205, since all passengers are asleep, it can be determined that there is no interaction between the passengers in the vehicle and the driver, and an indication signal SilenceSig indicating that the vehicle environment is silent is output. Otherwise, if it is determined in step 203 that all passengers are not asleep, the process proceeds to step 207.

[0016] In step 207, it is determined whether the front passenger is asleep. If the front passenger is not asleep, for example, if he is speaking, the process returns to step 201 and continues to monitor the state of the person in the vehicle. Otherwise, if it is determined that the front passenger is asleep, the process proceeds to step 209, where it is further determined whether the rear passenger is asleep. If the rear passenger is not asleep, the process proceeds to step 211.

[0017] In step 211, the in-vehicle microphone is turned on to pick up audio information that may be coming from people in the vehicle. For example, the state analysis module 102 may include a language recognition unit (not shown) that recognizes possible chat content between the driver and other passengers. In step 213, it is determined whether there is any verbal communication between the driver and the rear seat passenger based on the picked-up audio information. For example, the language recognition unit may call a natural language processing (NLP) algorithm to make a verbal communication judgment and further determine whether the verbal communication relates to substantial content. For example, based on a method such as semantic analysis, it may determine whether there is a normal chat between the driver and the rear seat passenger. Alternatively, the state analysis module 102 may further determine whether the content of the chat keeps the driver excited or interested. In step 213, if it is determined that normal chat is taking place between the two, the process returns to step 201; otherwise, if there is abnormal chat, for example, only short exclamations such as "yeah" or "ah," such content may cause negative factors such as boredom, so the process proceeds to step 205 and outputs a silent indication signal SilenceSig indicating that there is no interaction between the passenger in the car and the driver.

[0018] Furthermore, if step 209 determines that the rear seat passenger is asleep, the process proceeds to step 215. In step 215, the process determines whether there is interaction between the driver and the outside world or whether multimedia is being played in the vehicle, for example, whether music is being played. For example, the in-vehicle microphone is turned on to pick up voice calls or in-vehicle sounds that may be coming from the driver. Here, the state analysis module 102 may still use the speech recognition unit to analyze the sounds collected by the microphone to determine whether there is a call from the driver, or may combine this with image recognition technology to determine whether the driver is talking by, for example, assisting in the judgment of mouth shape. The state analysis module 102 may also determine whether the in-vehicle sound is multimedia playback sound. If step 215 determines that the driver is currently talking to the outside world and / or multimedia is being played in the vehicle, the process returns to step 201 to determine that the current in-vehicle environment is in a noisy state. Otherwise, if it is determined that there is no interaction between the driver and the outside or that multimedia is not being played inside the vehicle, it determines that the current vehicle interior is in a silent state, so it outputs a silent instruction signal SilenceSig and proceeds to step 205.

[0019] It should also be noted that, in the above example, the target detection unit 1021 and the state recognition unit 1022 are implemented as separate units, but the present invention is not limited thereto and may be implemented as an integrated module or algorithm, including performing preprocessing and state recognition on images using various image processing algorithms known in the prior art to recognize sleep states. For example, by recognizing images using a classifier or convolutional neural network that has been learned and trained on a large number of images, it is possible to first distinguish between occupants located at different positions in the vehicle, and then analyze and recognize the sleep states of the occupants in each vehicle. For example, as described above, if the images IMG are from a specific camera, for image IMG1 from camera CAM1, it is necessary to not only distinguish between the driver and the passenger seat occupant, but also to distinguish whether the passenger seat occupant is sleeping, while for image IMG2 from camera CAM2, it is only necessary to determine the sleep states of the occupants located in the rear seats.

[0020] As described above, if the state analysis module 102 determines that the vehicle environment is in a silent state, it sends a silent indication signal SilenceSig to the feature activation module 103, indicating that there is no interaction or multimedia playback.

[0021] In the present invention, the function activation module 103 is configured to activate a voice assistant to realize voice-based man-machine interaction with the driver, or further activate an advanced driver assistance system (ADAS), when the environment state in the vehicle is silent within a predetermined time, such as when there is no interaction between passengers in the vehicle, between passengers in the vehicle and the driver, or between the driver and the outside, or when there is no multimedia playback in the vehicle. For example, SilenceSig may be a low-level “0” signal indicating that the environment state in the vehicle is silent. If the level value of the signal SilenceSig is not inverted, for example, not changed to a high level “1” within a predetermined time, for example, three minutes, the function activation module 103 activates a voice assistant to realize voice-based man-machine interaction with the driver.

[0022] In various embodiments of the present invention, the voice assistant may be an artificial intelligence chat tool, Talker. Currently, artificial intelligence technology has made remarkable progress, and application technologies such as generative artificial intelligence (AIGC) models have emerged, which are widely used in smart chat, video generation, and the like. Their excellent chatting and information understanding capabilities enable good interaction with humans. Furthermore, with the development of autonomous driving control technology, an increasing number of vehicles are equipped with advanced driver assistance systems (ADAS) that assist or replace the driver's driving control. Generally, ADAS systems can be selectively activated or deactivated by the vehicle driver. According to the present invention, when a vehicle is equipped with the smart chat tool Talker, activating the smart chat function allows the driver to maintain a high level of alertness and vitality without being affected by the sleep of other passengers. Furthermore, the ADAS system can assist the driver in controlling driving, thereby reducing the risk of driving safety caused by the driver's distraction due to the influence of other passengers' sleep.

[0023] According to one embodiment of the present invention, the function activation module 103 determines whether to activate the voice assistant and / or the driving assistance system based on whether there is an interaction with the driver in the vehicle. In another example, the function activation module 103 may be configured to activate the voice assistant based on whether there is an interaction between the passengers in the vehicle; obviously, if there is a voice interaction between the rear seat passengers in the vehicle, the driver may hear it, which reduces the driver's sense of isolation and contributes to improving driving safety.

[0024] In addition, in the above-described embodiment of the present invention, the sleeping state of the passengers is determined by recognizing images of the passengers inside the vehicle, and further, whether there is an interaction between the passengers and the vehicle driver is determined. However, the present invention is not limited to this form. For example, the state analysis module can simply monitor the voice interaction of the passengers inside the vehicle through a microphone to determine whether the vehicle driver is participating in the interaction or whether there is an interaction between the passengers, and thereby provide an indication signal to the function activation module regarding whether there is an interaction between the passengers and the vehicle driver.

[0025] Therefore, according to an embodiment of the present invention, the voice assistant Talker is activated by detecting the environmental conditions inside the vehicle, especially when all passengers, including the front passenger, are asleep and not interacting with the vehicle driver, thereby improving the safety and comfort of the driving experience. In this invention, the smart assistant Talker is not just a conversation partner, but can effectively relieve the driver's stress and fatigue through friendly conversation and interaction, creating a more relaxed and enjoyable driving environment. It should be noted that Talker can be realized by any method known in the prior art and can be used for the purposes of the present invention, for example, by AIGC.

[0026] In the above-described example of the present invention, the smart chat tool Talker and the ADAS system are activated synchronously or selectively depending on the system design task. However, in other embodiments, the chat tool Talker may be designed to learn the driver's emotional state during the chat process, for example, whether the driver is disinterested or drowsy. If it is determined that the driver's emotional state is not suitable for driving a car, the chat tool Talker is used to control the activation of the ADAS system, thereby intervening early and avoiding discomfort to the driver. In the present invention, the ADAS system is realized using a system similarly known in the prior art.

[0027] Although a general embodiment of the present invention has been described with reference to the drawings, the driving assistance device 100 of the present invention is not limited to the above configuration and can be modified. For example, in the above example, the state analysis module 102 of the driving assistance device 100 needs to execute an image recognition algorithm to analyze images received from a camera and determine the sleep state of each vehicle occupant. However, in another example of the present invention, taking into account current developments in smart cameras, images can be processed and recognized on the camera side. For example, with the development of camera systems for smart cockpits, it is now possible to use commercially available cameras to monitor the driver's fatigue level, recognize facial expressions (e.g., eyes closed / sleeping, frown, eye movement, etc.), and recognize language based on lip movement. Meanwhile, in another example of the present invention, the state analysis module 102 can directly receive feature recognition data for each vehicle occupant's facial expression features, such as eyes closed / sleeping, provided by the camera. Based on this, it can determine whether there is an interaction between the vehicle occupant and the driver based on the control policy shown in FIG. 2. This reduces the image reception and processing tasks of the driving assistance device 100, including state recognition.

[0028] In the above embodiment, the state analysis module 102 primarily uses image recognition to determine whether a person in the vehicle is asleep and then determines whether there is interaction between passengers in the vehicle, between a passenger in the vehicle and the driver, or between the driver and the outside. However, the present invention is not limited to this. It can also analyze whether the vehicle is in a silent state based entirely on voice recognition technology to activate the voice assistant Talker. According to one embodiment of the present invention, the state analysis module 102 uses voice recognition to determine whether there is a voice interaction signal in the vehicle based on signals picked up by the in-vehicle microphone. If it determines that there is a voice interaction signal, whether it is a conversation between passengers in the vehicle, between a passenger in the vehicle and the driver, or a voice call between the driver and the outside, it determines that the in-vehicle environment is in a noisy state; otherwise, it determines that the in-vehicle environment is in a silent state. In a further embodiment, the state analysis module 102 can also determine whether multimedia audio is being played in the vehicle based on the audio signals picked up by the in-vehicle microphone. It determines that the vehicle is in a silent state only if it determines that there is no voice interaction signal or multimedia playback. As a result, the state analysis module 102 outputs a signal SilenceSig indicating a silent state so that the function activation module activates the voice assistant Talker and / or the ADAS system.

[0029] Although different embodiments of the present invention have been described above with reference to specific examples, those skilled in the art will understand that the various exemplary logic modules and method steps described with reference to the contents disclosed herein can be realized as electronic hardware, computer software, or a combination of both. For example, a control device according to the present invention may be realized as a processor or main controller and memory, with each module in the form of a computer program stored in the memory, and the processor can implement the method of the present invention by executing these modules. Another embodiment of the present invention provides a machine-readable medium having machine-readable instructions stored therein, which, when executed by a processor, cause the processor to perform any of the methods disclosed herein. These embodiments also fall within the scope of protection of the present invention. [Explanation of symbols]

[0030] 100 Driving assistance device 101 Image receiving module 102 Condition Analysis Module 103 Feature Activation Module 201 steps 203 steps 205 steps 207 steps 209 steps 211 steps 213 steps 215 steps 1021 Target Detection Unit 1022 State Recognition Unit

Claims

1. a state analysis module that recognizes the environmental state inside the vehicle; A function activation module configured to activate a voice assistant to realize voice-based human-machine interaction with a driver when the environmental state is silent within a predetermined time; A driving assistance device comprising:

2. 2. The device of claim 1, wherein the state analysis module is further configured to determine whether there is an interaction between passengers in the vehicle, between the passengers in the vehicle and the driver, or between the driver and the outside, or whether multimedia playback is occurring in the vehicle, and determine that the environmental state is a silent state if it determines that there is no interaction or that multimedia playback is not occurring.

3. 3. The device of claim 2, wherein the state analysis module recognizes a sleep state of the passenger or the driver based on state data indicating various states of the passenger and the driver provided from at least one camera, the sleep state indicating whether there is an interaction between the passengers, between the passenger and the driver, or between the driver and the outside, and the camera is capable of generating the state data based on captured images.

4. further comprising an image receiving module for receiving one or more image frames of the driver and the passenger; 4. The device of claim 3, wherein the state analysis module is further configured to perform image recognition on the image frames to recognize whether the passenger and the driver are asleep, thereby determining whether there is an interaction between the passenger and the driver of the vehicle.

5. The condition analysis module a target detection unit configured to distinguish between a front passenger seat occupant positioned in a front passenger seat and a rear passenger seat occupant positioned in a rear seat among the passengers; The device according to any one of claims 1 to 4, further comprising: a state recognition unit for recognizing whether each of the passengers is in a sleeping state or not.

6. The state analysis module is further configured to determine that there is no interaction between the passengers and the driver of the vehicle when it determines that all passengers other than the driver are asleep; 6. The device of claim 5, wherein the function activation module activates the voice assistant to realize voice-based human-machine interaction with the driver and / or activates an advanced driver assistance system (ADAS) to assist the driver in controlling the vehicle.

7. If the state recognition unit recognizes that the front passenger is in a sleeping state, further determining whether the rear passenger is in a sleeping state; 6. The device of claim 5, wherein if the rear seat passenger is not asleep, the state analysis module is further configured to turn on an in-vehicle microphone to monitor whether the driver and the rear seat passenger are having a conversation, and determine whether the interaction is occurring based on the conversation.

8. The condition analysis module further configured to determine the absence of interaction if there is no conversation or if the conversation does not include substantive content or conversational content sufficient to interest or excite the driver; 8. The device of claim 7, wherein, in the absence of the interaction, the feature activation module activates the voice assistant to facilitate voice interaction with the driver, and the voice assistant selectively activates an advanced driver assistance system to assist the driver in controlling the vehicle.

9. The condition analysis module The apparatus of claim 8 , further comprising a language analysis unit configured to invoke a natural language processing algorithm to recognize the conversation content and determine whether it contains the substantive content.

10. Recognizing an environmental condition inside the vehicle; If the environmental state is silent within a predetermined time, activating a voice assistant to realize voice-based man-machine interaction with the driver; A driving assistance method comprising:

11. The step of recognizing an environmental condition inside the vehicle includes: determining whether there is interaction between passengers in the vehicle, between the passengers in the vehicle and the driver, or between the driver and the outside, or whether multimedia playback is occurring in the vehicle; The driving assistance method of claim 10, comprising determining that the environmental state is silent if it is determined that there is no interaction or no multimedia playback.

12. 12. The driving assistance method according to claim 11, further comprising: a step of recognizing a sleeping state of the passenger or the driver based on status data indicating various states of the passenger and the driver provided from at least one camera, the sleeping state indicating whether there is an interaction between the passengers, between the passenger and the driver, or between the driver and the outside, and the camera being capable of generating the status data based on captured images.

13. receiving one or more image frames of occupants, the occupants including the driver and the passenger; 13. The driving assistance method according to claim 12, further comprising: performing image recognition on the image frames to recognize whether the passenger is asleep, thereby determining whether there is an interaction between the passenger and a driver of the vehicle.

14. The step of recognizing the state of the passenger in the vehicle includes: A step of distinguishing between a front passenger seat occupant positioned in a front passenger seat and a rear passenger seat occupant positioned in a rear seat among the passengers; The driving assistance method according to any one of claims 10 to 13, further comprising the step of recognizing whether each of the passengers is asleep or not.

15. 15. The driving assistance method of claim 14, further comprising the step of: if it is determined that all passengers other than the driver are asleep, determining that there is no interaction between the passengers and the driver of the vehicle, activating the voice assistant to realize voice-based man-machine interaction with the driver, and / or activating an advanced driver assistance system (ADAS) to assist the driver in controlling the vehicle.

16. If the front passenger seat occupant is determined to be asleep, further determining whether the rear passenger seat occupant is asleep; 15. The driving assistance method of claim 14, further comprising: if the rear seat passenger is not asleep, turning on a microphone in the vehicle to monitor whether the driver and the rear seat passenger are having a conversation, and determining whether to activate a voice assistant and / or an advanced driver assistance system based on the conversation.

17. 17. The driving assistance method of claim 16, wherein if the conversation is not occurring or if the conversation does not contain substantial content or sufficient conversational content to interest or excite the driver, the voice assistant is activated to realize voice-based man-machine interaction with the driver, and the voice assistant selectively activates an advanced driver assistance system to assist the driver in controlling the vehicle.

18. The condition analysis module further Turn on the microphone in the car to monitor whether there is conversation or multimedia playback in the car, The driving assistance method of claim 11 , configured to determine that the environmental state inside the vehicle is silent if the conversation or multimedia playback is not monitored.

19. determining a speech position of the voice based on the conversation; determining whether the driver has participated in the conversation based on the speech location; 20. The driving assistance method of claim 18, further comprising the step of: determining that there is no interaction in the vehicle if it is determined that the driver is not participating in a conversation.

20. at least one camera that captures one or more image frames of occupants in a vehicle, the occupants including a driver and at least one passenger; A driving assistance system comprising: the driving assistance device according to any one of claims 1 to 9.

21. a machine-readable medium having machine-readable instructions stored thereon; one or more processors, wherein the instructions, when executed by the one or more processors, cause the processors to perform the method of any one of claims 10 to 19.

22. A computer program product comprising machine-readable instructions which, when executed by one or more processors, cause the processors to perform the method of any one of claims 10 to 19.

Images