System and method for dynamic surveillance management

Abstract

A system (100) and a method (600) for dynamic surveillance management in an environment are disclosed. The system (100) comprises cameras (102A, 102B-102N) and a processing unit (104) communicably coupled to the cameras (102A, 102B-102N). The processing unit (104) is configured to dynamically obtain one or more data streams from one or more data sources, analyse each obtained data stream to detect one or more events, dynamically determine a criticality level of at least one camera based on the detection of the one or more events in the corresponding data stream, and dynamically manage at least one configuration setting associated with the at least one camera based on the determined criticality level. The configuration of the system (100) ensures optimal system performance and data integrity while minimizing potential disruptions or loss of data.

Description

[0001] 202318179

[0002] 1

[0003] Description

[0004] SYSTEM AND METHOD FOR DYNAMIC SURVEILLANCE MANAGEMENT

[0005] The present invention generally relates to surveillance management. More specifically, the present invention relates to a system and a method for dynamic surveillance management in an environment, involving dynamic management of configuration settings associated with cameras based on their criticality levels.

[0006] Surveillance systems have become an indispensable component of modern security infrastructure, deployed across a wide range of environments, from residential homes to industrial establishments. These systems typically rely on a network of sensors and cameras, often interconnected through the Internet-of-Things (loT), to collect and process vast amounts of data. This data provides valuable insights for monitoring activities, detecting anomalies, and responding to incidents within the environment.

[0007] However, conventional surveillance systems present several challenges in the realm of data management. One of the most critical challenges is the potential for data disruption or data loss, particularly for data that is considered highly sensitive or critical.

[0008] The sensitivity or criticality of a camera is influenced by several factors, including location, temporal considerations, and certain events. Cameras that monitor areas with high security or privacy implications, such as executive offices, data centers, sensitive laboratories, or critical infrastructure, are generally considered to be highly critical. For example, a camera that monitors the entrance to a highly secure data center or a research laboratory may be considered highly sensitive. This is because the loss of data from such a camera would compromise the security of sensitive information or disrupt critical operations. In some cases, the loss of sensitive data can have far-reaching implications, such as legal liability and financial losses. On the other hand, cameras that monitor certain public areas such as parking lots or lift lobbies may be considered less sensitive. While the loss of data from such a camera could still have negative consequences, the impact is likely to be less severe.

[0009] It has been observed that data sensitivity may not always be static but may be influenced by events that are dynamic in nature. Correspondingly, criticality of a camera may depend on real- world factors and dynamic considerations, such as but not limited to, change in location / field of view of the camera, and specific events or activities occurring in the field of view of the camera. 202318179

[0010] 2

[0011] As an example, if a known suspect is identified within the camera’s field of view, the sensitivity of the data would be significantly increased due to the potential security implications.

[0012] A significant challenge in surveillance systems is the potential for data loss or data disruption during camera maintenance activities. Operators or maintenance personnel may not be aware of the sensitivity of the data or the criticality of the camera, which may lead to inadvertent disruption or data loss due to maintenance activity. This can have serious consequences. If critical data is lost or corrupted due to maintenance activity, it may compromise the security of the environment. Further, data loss or disruption can prevent the detection of anomalies or incidents, potentially leading to missed opportunities for intervention. In some cases, data loss or disruption can have legal or financial consequences. For example, if the surveillance system fails to capture a crime due to a maintenance-related issue, the organization may be held liable.

[0013] Another challenge exists in conventional surveillance systems in that, a camera may not sufficiently capture all the data necessary for intervention during sensitive events. This can be due to factors such as fixed and limited resolution and / or frame rate, poor lighting conditions, and the like. The consequences of insufficient data can be severe, including compromised security, missed opportunities for intervention, and legal or financial implications.

[0014] Another challenge exists in conventional surveillance systems in that, network parameters associated with a camera may not ensure optimal network performance and reliability in all situations. For example, consider a camera adapted to monitor an area where highly sensitive events may occur, albeit rarely. Despite the potential for such events, data from the camera may be assigned a low network priority due to the infrequency of these occurrences. If a sensitive event does happen, delays or interruptions in data transmission could occur, which could lead to missed detections or compromised security. As another example, consider another camera which monitors a low-priority area where no significant events that compromise security can occur. If this camera is assigned a higher network priority, it may consume excessive bandwidth and resources, thereby adversely affecting the performance of other cameras or network devices. This can lead to network congestion and potential data loss for other cameras, which is undesirable.

[0015] Another challenge has been identified in conventional surveillance systems in that, data compression parameters associated with data streams captured by a camera may not always be properly adjusted to account for dynamic real-world events. An improper adjustment of compression settings may lead to loss of detail, increased processing time, and reduced data 202318179

[0016] 3 quality. This can compromise the effectiveness of the surveillance systems by hindering analysis, diagnostics, and decision-making. For example, consider a camera in a power plant which monitors a critical operation (such as testing of a safety valve) only for a few minutes of the day. If the data compression ratio for the data captured by the camera is too high, it may be difficult to identify subtle changes or anomalies in the captured footage of the critical operation. This could potentially lead to missed detections of threats or equipment failures, resulting in significant financial losses or safety hazards.

[0017] Another challenge exists in conventional surveillance systems in that, security configuration parameters associated with a camera, such as access control and encryption settings, may not always be properly adjusted to account for real-world dynamic events. For example, consider a camera adapted to monitor an area where sensitive events may occur, albeit rarely. Despite the potential for such events, the camera may be assigned a low level of access control due to the infrequency of these occurrences. If a critical event does happen, unauthorized individuals may be able to view or download the footage of the critical event captured by the camera. Likewise, if the encryption settings associated with the data captured by the camera are weak, the data may be vulnerable to hacking or other cyberattacks. The consequences of compromised data confidentiality can be severe, including financial and legal repercussions. For instance, if sensitive surveillance data is leaked to a competitor or a malicious actor, it could lead to reputational damage, financial loss, or legal liability.

[0018] Another challenge exists in conventional surveillance systems in that, alignment and / or orientation of a camera may not be adjusted to account for real-world dynamic events. For instance, a limited field of view of a camera may hinder its ability to capture certain sensitive information. As an example, if the camera is monitoring a hallway and the camera detects a suspect passing across the hallway, the limited field of the view of the camera may not be able to capture certain data pertaining to the movement of the suspect.

[0019] In light of the above, there exists a need for a system and a method for dynamic surveillance management that addresses one or more of the aforementioned challenges in conventional systems.

[0020] Further limitations and disadvantages of conventional surveillance systems will become apparent to one of skill in the art, through comparison of described systems with some aspects of the present disclosure, as set forth in the remainder of the present specification and with reference to the drawings. 202318179

[0021] 4

[0022] It is an object of the present disclosure to provide a system and a method for dynamic surveillance management in an environment, that involves dynamic management of at least one configuration setting associated with at least one camera based on the criticality level of the at least one camera.

[0023] Throughout the disclosure, the phrases “one or more” and “at least one” may encompass both singular and plural forms. Even if these specific terms are not used, each element can be interpreted as either singular or plural, which may be self-evident. Additionally, depending on the embodiment and based on feasibility as may be appreciable by a person skilled in the art, each component may be provided in either singular or plural form.

[0024] Throughout the present disclosure, the term “unit” may include a unit realized by hardware, a unit realized by software, or a unit realized using both hardware and software. Furthermore, one unit may be realized using two or more hardware, or two or more units may be realized using one hardware.

[0025] One aspect of the present disclosure relates to a system for dynamic surveillance management in an environment. Across embodiments, the environment may be a building, such as but not limited to, a residential building, an office, a research facility, and the like. The environment may also include open spaces, or a combination of buildings and open spaces. Therefore, the system can be deployed in a variety of settings, from indoor spaces to outdoor spaces.

[0026] The system comprises one or more cameras. Each of the one or more cameras is capable of capturing data within a corresponding field of view in the environment. The one or more cameras may be configured to capture data in one or more forms, including but not limited to visual data, infrared radiation data, or other forms of data associated with electromagnetic radiation.

[0027] In an exemplary embodiment, the one or more cameras are capable of capturing visual data, either digitally using an electronic image sensor, or chemically via a light-sensitive material such as a photographic film. In a preferred embodiment, the one or more cameras are configured to capture visual data by using digital means such as electronic image sensors. In a preferred embodiment, the one or more cameras are closed-circuit television (CCTV) cameras. Each CCTV camera may be equipped with multiple features, such as night vision, motion detection, and remote access capabilities. 202318179

[0028] 5

[0029] In another exemplary embodiment, the one or more cameras may be thermal imaging cameras specifically configured to capture infrared radiation, which can be utilized to detect and visualize temperature differences in the respective fields of view. This can be beneficial in applications where visual data may be obscured or insufficient, such as in low-light conditions or through smoke or fog.

[0030] In a preferred embodiment, the one or more cameras are Internet-of-Things (loT) devices. This enables integration of the one or more cameras with other loT devices and systems within the environment, allowing for enhanced data sharing, automation, and remote management. The one or more cameras may be equipped with various sensors, such as environmental sensors, and, actuators that manage their respective positions, fields of view and allow for efficient monitoring of the environment.

[0031] The system comprises a processing unit. Even if the specific term “processing unit” is used, it is understood that it may encompass one or more processing units. The processing unit is communicably coupled to the one or more cameras through a network connection. Across embodiments, the network connection may be established using various communication means such as data transmission cables, Ethernet, WiFi, Bluetooth, cellular networks and the like. The processing unit may be embodied in a computer or a server. This could be a computer located off-site or an on-site computer located within the environment. The choice of location of the computer may depend on factors such as the size and complexity of the system, the need for local processing capabilities and network connectivity.

[0032] The processing unit is configured to dynamically obtain one or more data streams from one or more data sources. Across embodiments, the processing unit may be configured to dynamically obtain the one or more data streams associated with the one or more cameras and / or the environment. In an embodiment, the one or more data streams comprises at least one of: data captured by the one or more cameras, data pertaining to location and / or field of view of the one or more cameras within the environment, and ambient data associated with the surroundings of the one or more cameras. The ambient data may include, but not be limited to, information pertaining to temperature, humidity, light levels, noise levels, air quality, and the like.

[0033] In an embodiment, the one or more data sources are the one or more cameras. In an embodiment, the processing unit may be configured to directly receive data streams from the cameras, either through a wired or a wireless network connection. In an embodiment, the one or more data sources may comprise of one or more intermediary servers. The data streams may be routed to 202318179

[0034] 6 the processing unit through the one or more intermediary servers, which may be adapted for providing additional processing and / or storage capabilities. In another embodiment, the one or more data sources may comprise a remote server. The data streams may be routed to the processing unit from a remote server hosted by a remote vendor or operator, such as a cloudbased surveillance service provider. In yet another embodiment, the one or more data sources can be a combination of the aforementioned data sources. The configuration of the processing unit to obtain data streams from multiple sources may account for enhanced flexibility and scalability of the system. Additionally, integration of data from various sources can provide comprehensive surveillance of the environment.

[0035] The processing unit is configured to analyse each obtained data stream to detect one or more events. Analysing each obtained data stream involves processing each data stream, which involves detection of significant occurrences or changes within the system and / or the environment. The one or more events may be manually configured by a user, automatically considered by the system based on abnormal, unusual, or anomalous behaviour, or may be a combination of manual configuration and automatic considerations. The one or more events may comprise at least one of: a change in the location and / or the field of view of the one or more cameras, a change in one or more ambient parameters in the surroundings of the one or more cameras, an occurrence of a predetermined time-period, and detection of at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the field of view of the one or more cameras, and detection of an abnormal pattern in the fields of view of the one or more cameras.

[0036] Changes in the location and / or the field of view of the one or more cameras may occur when the one or more cameras are moved or repositioned. For example, if a camera is moved from one location to a different location, it may capture new areas of the environment, potentially revealing new information or detecting previously unseen events, and the corresponding change in the camera’s location can be an event. The one or more cameras may also be configured to change their respective field of view in response to another event, such as but not limited to, a movement of one or more subjects or objects away from their initial field of view.

[0037] Changes in one or more ambient parameters in the surroundings of the one or more cameras may also indicate significant events. The one or more ambient parameters may include temperature, humidity, light levels, noise levels, air quality, and the like. For example, a sudden increase in temperature might indicate a fire or other hazardous situation, which could warrant suitable intervention activities in the environment. Likewise, other events associated with the 202318179

[0038] 7 change in the one or more ambient parameters could be linked with varied criticality levels of the one or more cameras.

[0039] Occurrence of a predetermined time-period, such as specific hours of a day, may also be a trigger for other events of varied significance. For example, occurrence of a scheduled time of the day for performing a critical forensic experiment or for an important meeting can be an event detected by the processing unit. The processing unit may be further configured to detect and alert on events that occur during the predetermined time-period.

[0040] The detection of at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the fields of view of the one or more cameras may be another feature of event detection. This may involve using computer vision algorithms to identify and track specific objects, individuals, or patterns within each obtained data stream.

[0041] The predefined identifiers may refer to unique codes or labels assigned to individuals, objects, or other entities. The predefined identifiers can be used to track and identify specific items or persons in the environment. Examples of predetermined identifiers include barcodes and QR codes that can be scanned by the cameras and relayed to the processing unit for identifying products, individuals or other objects, and ID cards worn by individuals that may contain unique identification information (such as name, ID number, or codes).

[0042] The predefined patterns may refer to specific events or behaviours that indicate suspicious activity. The predefined patterns may be identified using computer vision algorithms and machine learning techniques. Examples of predefined patterns include unusual movement patterns, such as a person loitering in a restricted area or a vehicle driving erratically, and unusual object interactions, such as a person tampering with a piece of equipment or a vehicle approaching a restricted area.

[0043] The predetermined individuals may refer to individuals identified as persons of interest or potential threats. These individuals may be known suspects, or individuals with a history of harmful or criminal activity, or individuals who have been flagged for other reasons. Examples of predetermined individuals include individuals who have been identified as suspects in a crime or other incident, and individuals who have a previous criminal record or who are known to pose a security threat. 202318179

[0044] 8

[0045] The predetermined objects may refer to objects that are considered to be potential security risks or threats. These objects may be unusual or out of place for the specific environment. Examples of predetermined objects may include suspicious packages and prohibited items (i.e., items that are prohibited in the particular environment, such as weapons and hazardous materials).

[0046] The detection of an abnormal pattern in the fields of view of the one or more cameras may be another feature of event detection. In this context, the abnormal pattern may refer to a deviation from a statistically expected or an established norm or trend within the environment. This deviation may be indicative of an anomaly, an irregularity, or an event that is outside a normal range of parameters or behaviours. In an industrial environment, a normal pattern might be the consistent movement of machinery, workers, and materials in a predictable sequence or manner. However, an abnormal pattern could arise from a disruption to this routine. For example, detection of a person lying motionless on the factory floor, fall or collapse of an individual or equipment, and accidents within the environment could be considered as abnormal patterns. The system may employ a vision based artificial intelligence model for the detection of such abnormal patterns. Such abnormal patterns raise a red flag, prompting further investigation and / or immediate action. The occurrence of such abnormal events directly influences the criticality level or sensitivity level of the camera(s) monitoring such events.

[0047] Advantageously, detection of the one or more events by the processing unit allows for proactive monitoring, enhanced situational awareness, and improved security of the environment. Furthermore, detection of the one or more events provides valuable insights, enabling data-driven decision-making for optimization of surveillance strategies.

[0048] In an embodiment, the processing unit is configured to analyse each dynamically obtained data stream by execution of a machine learning model for detecting the one or more events. Execution of a machine learning model is particularly advantageous for event detection, owing to learning from large datasets and adaptability to changing conditions. The machine learning model can be trained on extensive labelled data to recognize the one or more events. Once trained, the model can be applied to new data streams to detect similar events in real-time. Advantageously, the machine learning model can achieve higher accuracy rates than conventional rule-based methods for event detection. The machine learning model may also be configured to handle large volumes of data and can advantageously analyse data streams from a scalable number of cameras. 202318179

[0049] 9

[0050] In a further embodiment, the processing unit is configured to analyse each dynamically obtained data stream by execution of a multimodal machine learning model for the detection of the one or more events. The multimodal machine learning model may be well-suited for dynamic surveillance management, as the model can process multiple formats of data such as image feed data from the one or more cameras, location data, sensory data, etc. and provide more accurate and reliable event detection, especially in a noisy environment. Additionally, the multimodal machine learning model may advantageously detect patterns and relationships that may be difficult to identify using a single-modal machine learning model.

[0051] In a further embodiment, the machine learning model is a multimodal Retrieval-Augmented Generation (RAG) based Generative Artificial Intelligence (GenAI) model (i.e., RAG system). The RAG system is trained using a comprehensive dataset which comprises a plurality of annotated images, a plurality of video feeds, and a plurality of sensor data from a plurality of three- dimensional environments. The training dataset comprises labelled examples of various objects, a plurality of object classifications, one or more spatial positions, and dynamic changes over time. The RAG system is trained by application of supervised learning techniques on a plurality of neurons of the RAG system. The RAG system may be configured to ingest data streams obtained from the one or more data sources (such as the one or more cameras), detect the one or more events in each data stream, generate structured queries, retrieve relevant information, and generate responses.

[0052] Advantageously, the RAG system can leverage comprehensive training data to provide accurate and relevant responses based on the specific context of the analysed data streams. Furthermore, the RAG system can generate human-readable summaries in natural language, making it convenient for users to understand and appropriately respond to the detected events.

[0053] In an embodiment, the processing unit is configured to implement a Convolutional Neural Network (CNN) model for detecting the one or more events in the data streams captured by the one or more cameras. CNNs are well-suited for image processing and can be used to detect objects, faces, and visual patterns in the footages captured by the one or more cameras. In an embodiment, the processing unit is configured to implement a Recurrent Neural Networks (RNN) model for detecting the one or more events associated with time-series data, such as traffic data within the environment. RNNs are well-suited for processing sequential or time-series data, making them suitable for detecting anomalies or patterns over time, such as unusual behaviour or changes in activity levels in the respective fields of view being monitored. 202318179

[0054] 10

[0055] The processing unit is configured to dynamically determine a criticality level of at least one camera of the one or more cameras based on the detection of the one or more events in the corresponding data stream that is analysed. The processing unit may implement a rule-based logic, a machine learning model, or employ a hybrid approach of utilizing a combination of rule-based logic and machine learning model to dynamically determine the criticality level of the at least one camera.

[0056] Dynamically determining the criticality level of the at least one camera refers to the process of continuously assessing and updating the criticality or the importance of the at least one camera based on the analysis of the one or more data streams. In an embodiment, the criticality level of the at least one camera can be reassessed at regular intervals, for example, every two or three seconds. This approach accounts for periodic updates in the criticality level based on the detection of the one or more events during the specified interval. In an embodiment, the criticality level of the at least one camera can be updated in real-time as the data streams are analysed. This approach allows for rapid adjustments to the criticality level of the at least one camera based on current conditions. In an embodiment, in addition to periodic or real-time analysis, the processing unit may be configured to analyse multiple data streams in parallel. Advantageously, such parallel processing enables the system to process data more efficiently when dealing with large volumes of data.

[0057] The criticality level of the at least one camera can be determined using various metrics or classifications. For example, a numerical rating system may be employed, where the at least one camera is assigned a criticality level on a scale of 1 to 10, with 1 representing the lowest level of criticality and 10 representing the highest level of criticality. Alternatively, a categorial system can be used for classifying the at least one camera into categories such as “low”, “medium”, “high” or “severe” criticality. Alternatively, a percentage-based rating system can be used to quantify the criticality level of the at least one camera, with higher percentages indicating higher criticality.

[0058] Advantageously, determination of the criticality level provides valuable data that can be used for informed decision-making to optimise surveillance strategies. By identifying criticality trends for the cameras, an enterprise can allocate resources and attention to most important areas in the environment, ensuring that critical data is protected and monitored effectively.

[0059] In an embodiment, if the one or more events are deterministic in nature and can be classified into predefined categories, the processing unit may utilize a lookup table to map the detected one or more events to corresponding criticality levels. Advantageously, this approach can simplify the determination of the criticality level and enable consistent application of rule-based logic. 202318179

[0060] 11

[0061] However, it is noteworthy that real-world events may not always fit neatly into predefined categories, and the lookup table may need to be updated periodically to account for such real- world events.

[0062] In an embodiment, the processing unit is configured to dynamically determine the criticality level of the at least one camera by: determining, by execution of a semantic contextualization module on the corresponding data stream, at least one label associable with the at least one camera based on the detection of the one or more events in the corresponding data stream; and tagging the at least one label to the at least one camera, whereby the at least one label is associated with the criticality level of the at least one camera.

[0063] As used herein, the “semantic contextualization module” corresponds to a machine-readable set of instructions accessible by the processing unit. The semantic contextualization module may leverage various techniques, including rule-based methods, statistical models, and machinelearning algorithms to extract meaningful and contextual information from the data streams processed by execution of the machine-learning model. The semantic contextualization module may also incorporate knowledge graphs or ontologies for provide additional context and understanding of the data streams.

[0064] Once the semantic contextualization module has analysed the corresponding data stream and the detected one or more events, the semantic contextualization module determines at least one label associable with the at least one camera. The at least one label may be a descriptive term or a category that represents the camera’s importance or sensitivity. For example, the at least one label might include “high-priority”, “medium-priority”, or “low-priority”.

[0065] The processing unit is configured to tag the determined at least one label to the at least one camera. In an exemplary embodiment, a dedicated database can be created to store information about each camera, including its unique identifier, location, and associated labels. The database can be queried to retrieve the criticality level of the corresponding camera. In an embodiment, a custom data structure can be defined to store information about each camera, including its unique identifier, location, and associated labels. This data structure can be integrated into the system architecture and accessed by the processing unit to determine the criticality of each camera.

[0066] Advantageously, the system can effectively tag cameras with appropriate labels and associate the cameras with the corresponding criticality levels. This information can then be used for 202318179

[0067] 12 prioritizing monitoring and maintenance activities, allocating resources in an efficient manner, and improve overall system performance.

[0068] The processing unit is configured to dynamically manage at least one configuration setting associated with the at least one camera based on the determined criticality level. The at least one configuration setting is indicative of the criticality level of the at least one camera in relation to the detection of the one or more events. Advantageously, by tailoring configuration settings associated with the cameras to the respective criticality levels, the system can ensure that the critical data is captured. Further, the system is capable of adjusting the configuration settings associated with the cameras in real-time to optimize their performance and security based on the nature of the one or more events and the perceived criticality level. Furthermore, the system’s ability to dynamically adjust configuration settings provides greater flexibility and adaptability in response to evolving threats and changing environment conditions.

[0069] Furthermore, the system is capable of detecting one or more events that may be manually configured by the user and / or automatically configured by the processing unit based on abnormal, unusual, or anomalous behaviour. Advantageously, the system allows for managing the at least one configuration setting in relation to the criticality level of the at least one camera based on the occurrence of the one or more events. For example, in a scenario where the at least one camera captures an event involving a trusted individual (trust being a predefined criteria set for the individual) loitering in a corridor area, the criticality level of the at least one camera may be set as low or normal. Correspondingly, the configuration setting associated with the at least one camera may be managed accordingly, such as maintaining the same network priority and the frame rate of the at least one camera. In a scenario where the at least one camera captures an event involving a suspect individual loitering in the corridor area, the criticality level of the at least one camera can be set as high or severe. Correspondingly, the configuration setting associated with the at least one camera may be managed accordingly, such as increasing the network priority and the frame rate of the at least one camera to capture essential surveillance data.

[0070] Furthermore, the system of the present disclosure employs criticality-based configuration management that can significantly reduce the occurrence of false alarms. That is, the system can dynamically adjust camera configuration settings based on the respective criticality levels, reducing the likelihood of false alarms caused by inappropriate configurations. Further, the system can dynamically adjust configuration settings associated with cameras to ensure that resources are used efficiently and effectively. By minimising false alarms and optimising resource 202318179

[0071] 13 allocation, the system of the present disclosure also helps reduce operational costs associated with surveillance management.

[0072] In one embodiment, the system comprises one or more indication devices associated with the one or more cameras. The one or more indication devices are communicatively coupled to the processing unit. The one or more indication devices may be communicatively coupled to the processing unit through a variety of wired or wireless communication techniques. Wired communication techniques may include the use of Ethernet cables, fiber optic cables, or other data transmission cables. Wireless communication techniques may include the use of Wi-Fi, Bluetooth, cellular networks, or radio frequency (RF) technologies. The specific method used may depend on factors such as the distance between the one or more indication devices and the one or more cameras, the required data transmission rate, and the security requirements of the system.

[0073] The one or more indication devices may include one or more of illumination devices, sound devices, visual displays, handheld devices, and the like. In an embodiment, the one or more indication devices are illumination devices such as light lamps, light emitting diodes (LEDs), or projection systems. In an embodiment, the one or more indication devices are audio devices, such as buzzers or speakers. In an embodiment, the one or more indication devices are visual displays, such as LCD screens, LED displays, or heads-up displays. In an embodiment, the one or more indication devices can be handheld devices or terminal devices, such as a mobile phone or a tablet that provides a user interface. In one embodiment, there could be a single indication device associated with multiple cameras. For example, the single indication device could be a central control panel or a dedicated monitoring station. In another embodiment, each camera could have its own dedicated indication device. This could provide more localized and specific information about the status of individual cameras. The choice of the indication device depends on factors such as but not limited to size and complexity of the system or the environment, the desired level of detail in indication, and the preferences of the users.

[0074] As already mentioned, the processing unit is configured to dynamically manage the at least one configuration setting associated with the at least one camera based on the determined criticality level. In an embodiment, the at least one configuration setting comprises an indication of the criticality level of the at least one camera by at least one indication device of the one or more indication devices. The at least one indication device is associated with the at least one camera. 202318179

[0075] 14

[0076] In a preferred embodiment, the at least one indication device is an illumination device, such as an LED, configured to emit light at a predefined wavelength (or colour) corresponding to the determined criticality level of the at least one camera. The illumination device may be positioned on, adjacent to, or proximal to the at least one camera. This provides a clear and visually intuitive indication of the criticality status of the at least one camera.

[0077] In an embodiment, a colour coding scheme may be employed to indicate the criticality level / status of the cameras. For example, the indication devices could be configured to emit red light for cameras with severe criticality level, orange light for cameras with a high criticality level, yellow light for cameras with nominal or normal criticality level, and green light for cameras with a low criticality level. Such a colour coding scheme allows operators to quickly and easily identify the criticality levels of individual cameras based on the colour of the emitted light. In another embodiment, blinking rate of the illumination devices may be dynamically adjusted by the processing unit to visually indicate the criticality level of the associated cameras. For example, a rapid, flashing pattern of illumination by the illumination devices may correspond to severe criticality of the associated cameras, which may indicate a severe-priority threat, such as a breach or intrusion. As another example, a slow and steady blinking pattern may correspond to low criticality of the associated cameras, which may indicate a less urgent event, such as a minor traffic incident or a routine activity in the environment.

[0078] Advantageously, by visually indicating the criticality levels of the cameras, operators can prioritize maintenance activities and avoid disrupting critical data streams. For example, cameras with a high or severe criticality level can be scheduled for maintenance during off-peak hours or when they are not capturing sensitive data, thereby reducing the risk of data loss or corruption.

[0079] In an embodiment, some indication devices could be configured to indicate a traffic level of the data to the users. This would provide the users with information about the volume and intensity of the data being processed by the system. Advantageously, this information can then be used to make informed decision about resource allocation, such as adjusting the number of cameras being monitored or allocating additional processing power. By understanding the volume of data being processed, the users can identify potential bottlenecks and allocate resources accordingly. This can help to ensure that the system is running efficiently, and that critical data is not being lost or delayed.

[0080] In one embodiment, the one or more indication devices comprise a mobile phone or a terminal device that supports an application. The application includes a map of the environment displaying 202318179

[0081] 15 locations of the one or more cameras and corresponding icons (for example, LED icons) that indicate the criticality levels of the one or more cameras. Additionally, the application may support pop-up notifications that provide real-time information about the criticality level of each camera. This allows users to simultaneously monitor the status of multiple cameras and identify any areas of concern.

[0082] Advantageously, managing the indication of the criticality levels of the cameras provides users / operators with real-time information about the criticality status of the cameras. By indicating the criticality levels of the cameras, users can make informed decisions pertaining to incident response. Further, managing the indication of the criticality levels of the cameras by the indication devices may help in reducing response times by alerting users of critical situations as soon as they occur. Furthermore, a record of the indications being managed by the system can be used for accountability and compliance purpose. Moreover, cameras with a high or severe criticality level can be scheduled for maintenance during off-peak hours or when they are not capturing sensitive data, thereby reducing the risk of data loss or corruption.

[0083] In an embodiment, the one or more cameras comprises a plurality of cameras with overlapping fields of view adapted to monitor a common region or area in the environment. In the embodiment, the processing unit is configured to determine a relative priority of each camera based on one or more criticality levels associated with each camera determined over a predetermined period. For example, the predetermined period could be 30 minutes. The processing unit is further configured to activate the one or more indication devices associated with the plurality of cameras to indicate the determined relative priority of each camera to the user. Thereby, the processing unit provides a recommendation of order of maintenance activity for the plurality of cameras that minimizes the potential for data loss due to interruption in the corresponding data streams.

[0084] For example, consider an exemplary system for dynamic surveillance management of a ballroom. The system comprises a first camera adapted for monitoring an entrance of the ballroom and a second camera adapted for monitoring an exit of the ballroom. The first camera and the second camera have overlapping fields of view, meaning they can capture some of the same areas within the ballroom. As an example, the overlapping field of view might be around 20%, indicating that the first camera and the second camera share a significant portion of their coverage.

[0085] Over a predetermined period, the system would analyse the data streams captured by both the first camera and the second camera to determine their respective criticality levels. Based on factors such as but not limited to the frequency of the detected events, the sensitivity of the 202318179

[0086] 16 monitored area, and the potential consequences of data loss, the first camera may be assigned a higher criticality level than the second camera. The first camera might be considered more critical due to its potential for capturing important security footage, such as individuals entering the ballroom. The second camera might be considered less critical, as it primarily captures footage of people leaving the ballroom.

[0087] The processing unit of the system would then activate the indication devices associated with both cameras to indicate their relative priorities. For example, the indication device associated with the first camera might emit a red light for indicating severe criticality of the first camera, and the indication device associated with the second camera might emit a green light for indicating low criticality of the second camera. If a situation requiring maintenance of both cameras should arise, which would not affect immediate functioning of both the cameras, the second camera could be maintained first as it streams less critical data in comparison to the first camera. Once the second camera is maintained, the first camera could be maintained. Advantageously, the system of the present disclosure is adapted to prioritize maintenance of the cameras based on their relative criticality so as to minimize the potential for data loss.

[0088] As already mentioned, the processing unit is configured to dynamically manage the at least one configuration setting associated with the at least one camera based on the determined criticality level. In an embodiment, the at least one configuration setting comprises at least one of: one or more imaging parameters of the at least one camera, one or more network parameters associated with the at least one camera, one or more data compression parameters associated with the one or more data streams captured by the at least one camera, one or more security parameters associated with the at least one camera, and an alignment and / or an orientation of the at least one camera.

[0089] In an embodiment, the at least one configuration setting comprises one or more imaging parameters of the at least one camera. The one or more imaging parameters are dynamically configured based on the determined criticality level of the at least one camera, ensuring that the captured data is of sufficient quality and relevance for specific surveillance scenario. The one or more imaging parameters may include, but not be limited to, frame rate, resolution, focus setting, zoom level, noise level, colour correction, exposure compensation, and the like.

[0090] For example, with an increase in criticality level of the at least one camera, the frame rate configuration of the at least one camera may be increased. Advantageously, higher frame rates can capture more details and facilitate the analysis of fast-moving objects or events, which may 202318179

[0091] 17 be necessary when the criticality level of the at least one camera is high. As another example, with an increase in criticality level of the at least one camera, resolution of the at least one camera may be increased. Accordingly, the at least one camera can monitor the respective field of view with a higher level of detail with an increase in the corresponding criticality level. Furthermore, based on the criticality level of the at least one camera, the focus setting of the at least one camera may be suitably adjusted to ensure that objects or subjects within the field of view are in sharp focus, providing clearer and more accurate information. With an increase in criticality level of the camera, the precision and / or the resolution associated with the zoom level of the at least one camera may be increased, allowing for better control by a remote operator. The extent of noise reduction in a footage captured by the at least one camera can be increased with an increase in the criticality level of the at least one camera, thereby improving footage quality. Additionally, based on the criticality level of the at least one camera, the extent of colour correction and exposure compensation on the image / video feed captured by the at least one camera may be suitably adjusted to give a clearer image / video feed for further analysis and intervention as may be necessary to ensure security.

[0092] Advantageously, by dynamically adjusting the imaging parameters based on the criticality of the camera, the system can optimize the captured data for specific use cases, ensuring that the data is of sufficient quality and relevance for analysis and decision making.

[0093] In an embodiment, the at least one configuration setting comprises one or more network parameters associated with the at least one camera. The one or more network parameters associated with the at least one camera may include, but not be limited to, data transmission rate of the data streams captured by the at least one camera and network priority associated with the at least one camera. The one or more network parameters may be dynamically adjusted based on the determined criticality level of the camera, ensuring that the camera has the necessary resources and the bandwidth to transmit data effectively.

[0094] When the criticality level of the at least one camera is high, the processing unit can dynamically adjust the network parameters to increase the rate of data transmission from the at least one camera to the processing unit. This can involve allocating more bandwidth to the camera, assigning it a higher network priority, and optimizing network routing to minimize latency.

[0095] When the criticality level of the at least one camera is low, the processing unit can dynamically adjust the network parameters to reduce resource consumption and prioritise other cameras with higher criticality levels. This could involve decreasing the data transmission rate from the at least 202318179

[0096] 18 one camera to the processing unit to conserve bandwidth and network resources, assigning a lower priority to the at least one camera’s network traffic, and optimizing network routing such that the data from critical cameras is prioritised.

[0097] Advantageously, by dynamically adjusting the network parameters based on the criticality of the camera, the system can ensure that critical data is transmitted efficiently and reliably, even during periods of high network load or congestion. Consequently, the system can also ensure that resources are allocated efficiently, and that critical or sensitive data is prioritised. This can help prevent data loss from the critical cameras and enhance the overall performance of the system.

[0098] In an embodiment, the at least one configuration setting comprises one or more data compression parameters associated with the one or more data streams captured by the at least one camera. The one or more data compression parameters may include compression ratio. When the criticality level of the at least one camera is high, the processing unit configures a lower compression ratio for the one or more data streams captured by the at least one camera, to preserve more details and ensure that the captured data is sufficient for analysis and decisionmaking. Conversely, when the criticality level of the at least one camera is low, the processing unit configures a higher compression ratio for the one or more data streams captured by the at least one camera, to reduce storage requirements and bandwidth consumption.

[0099] Advantageously, by dynamically adjusting data compression parameters based on the criticality level of the camera, the system can optimize the balance between data quality and resource utilization. Consequently, this helps to ensure that critical data is captured and transmitted effectively, while also minimizing the impact on system performance and costs.

[0100] In an embodiment, the at least one configuration setting comprises one or more security parameters associated with the at least one camera. The one or more security parameters may include, but not be limited to, access control settings and encryption settings. Access control settings regulate who can access and control the at least one camera and its captured data. If the criticality level of the at least one camera is high, the processing unit may configure a more stringent access control setting to prevent unauthorized access and data breaches. Correspondingly, the processing unit may configure stronger encryption levels for the network to help protect the data captured by the at least one camera from unauthorized access and intervention. Conversely, for cameras with lower criticality levels, less stringent security measures may be sufficient. This can help reduce the computational overhead and resource requirements associated with security measures, while still maintaining a reasonable level of protection. 202318179

[0101] 19

[0102] Advantageously, by dynamically adjusting security parameters based on camera criticality, the system can ensure that sensitive data is adequately protected while minimizing the impact on system performance and resource consumption.

[0103] In an embodiment, the processing unit may be configured to dynamically manage security parameters, such as access control, based on the criticality level of the one or more cameras monitoring specific spaces within the environment. The processing unit may implement various access control mechanisms corresponding to the criticality level of the one or more cameras.

[0104] In one implementation, the processing unit may implement an access control mechanism of restricting access to doors based on criticality levels. For example, doors of rooms having cameras of severe criticality level may be locked down owing to heightened security threats, while doors of rooms having cameras of low criticality level may have more relaxed access controls.

[0105] In another implementation, the processing unit may be configured to generate a traffic route for directing individuals or objects along specific exits or entrances corresponding to the security situation, which may be based on the determined criticality levels of the one or more cameras and their respective locations.

[0106] In yet another implementation, the processing unit may implement an access control mechanism involving authentication to access certain areas, based on the criticality levels of the cameras monitoring said areas. As an example, the access control mechanism may require individuals to present valid access cards or credentials before entering sensitive areas being monitored by cameras of high criticality level. Further, the access control mechanism may employ biometric authentication systems, such as fingerprint or facial recognition systems, to verify the identity of individuals before granting access to such individuals to sensitive areas monitored by cameras of high criticality level. Further, a multi-factor authentication protocol may be implemented for accessing regions monitored by cameras with high criticality levels. The number of authentication factors may be dynamically adjusted based on the real-time assessment of camera criticality.

[0107] Advantageously, by dynamically adjusting access control measures based on camera criticality, the system can optimise security protocols and minimise the risk of unauthorized access or breaches. For instance, if a sensitive event is detected in a specific area, the criticality level of the camera monitoring the area may be set as high, and the system may automatically lock down doors, restrict access, and reroute individuals to safer locations. Conversely, during low-criticality periods, access controls can be relaxed to facilitate normal operations. 202318179

[0108] 20

[0109] In an embodiment, the at least one configuration setting comprises an alignment and / or an orientation of the at least one camera. This configuration setting may be particularly useful in situations where a suspect or a person of interest moves beyond the initial field of view of the camera. For example, consider a case wherein the at least one camera is monitoring a hallway. If a suspect is captured by the at least one camera, the criticality level of the camera may be assigned as high. If the suspect who is initially captured by the at least one camera then moves out of its field of view, the processing unit may cause the alignment and / or the orientation of the at least one camera to be dynamically adjusted to capture the suspect’s continued movement.

[0110] Advantageously, by dynamically adjusting the camera’s alignment and / or orientation, a wider range of area may be captured when the one or more events are detected, ensuring that important events are captured to the extent possible. Furthermore, operators can gain a better understanding of the overall situation within the environment, thereby enabling them to address the situation.

[0111] In an embodiment, the system comprises the one or more cameras including a first camera and a second camera. In the embodiment, the second camera is located within a predetermined proximity (such as 10 m) to the first camera. The field of view of the first camera may at least partially overlap with the field of view of the second camera. The processing unit is configured to dynamically obtain one or more data streams associated with the first camera. The processing unit is configured to analyse each obtained data stream associated with the first camera to detect the one or more events. The processing unit is configured to dynamically determine the criticality level of the first camera based on the detection of the one or more events in the corresponding data stream. The processing unit is further configured to dynamically manage at least one configuration setting associated with at least the second camera based on the determined criticality level of the first camera. In a further embodiment, the processing unit may dynamically manage configuration setting(s) associated with the first camera as well as the second camera. The at least one configuration setting is indicative of the criticality level of the first camera in relation to the detection of the one or more events. The at least one configuration setting is indicative of the criticality level of the first camera in relation to the detection of the one or more events.

[0112] For example, consider a scenario wherein a vehicle entering a restricted area within the environment is captured by the first camera. Based on the nature of the event, the first camera whose footage initially captured the vehicle may be assigned a high criticality level. Additionally, since the second camera is located proximate to the first camera (and may have an overlapping 202318179

[0113] 21 field of view with that of the first camera), it may also be crucial to monitor the vehicle’s movements and activities within the second camera’s field of view. Therefore, the processing unit could assign high criticality level to the second camera and adjust its configuration settings accordingly. This could involve increasing the resolution, frame rate, or encryption parameters for the second camera to ensure that it can capture and transmit relevant data effectively. Advantageously, by dynamically managing the configuration settings as described herein, the system can ensure that critical events are captured and analysed effectively even when subjects or objects move between the fields of view of proximate cameras. This can contribute towards improving the overall security and effectiveness of the system.

[0114] The object of the present disclosure is also achieved by a method for dynamic surveillance management in an environment. The features, embodiments, and advantages of embodiments described hereinabove may be suitably applicable to the method steps and features described in the following.

[0115] The method comprises dynamically obtaining, by a processing unit, one or more data streams from one or more data sources. In an embodiment, the one or more data streams are associated with at least one of the one or more cameras and the environment. The one or more data streams may comprise at least one of: the data captured by the one or more cameras data pertaining to location or field of view of the one or more cameras, and ambient data associated with the surroundings of the one or more cameras.

[0116] The method comprises analysing, by the processing unit, each obtained data stream to detect one or more events. Detecting the one or more events may comprise detecting at least one of: change in the location and / or the field of view of the one or more cameras; a change in one or more ambient parameters in the surroundings of the one or more cameras; an occurrence of a predetermined time-period, at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the field of view of the one or more cameras; and an abnormal pattern in the fields of view of the one or more cameras.

[0117] In an embodiment, the method comprises analysing, by the processing unit, each dynamically obtained data stream by execution of a machine learning model for detecting the one or more events. The method further comprises dynamically determining the criticality level of the at least one camera, by the processing unit, by: determining, by execution of a semantic contextualization module on the corresponding data stream, at least one label associable with the at least one camera based on the detection of the one or more events in the corresponding data stream; and 202318179

[0118] 22 tagging the at least one label to the at least one camera. The at least one label is associated with the criticality level of the at least one camera.

[0119] The method comprises dynamically determining, by processing unit, a criticality level of at least one camera of one or more cameras, based on the detection of the one or more events in the corresponding data stream.

[0120] The method comprises dynamically managing by the processing unit, at least one configuration setting associated with the at least one camera based on the determined criticality level. The at least one configuration setting is indicative of the criticality level of the at least one camera in relation to the detection of the one or more events. In an embodiment, dynamically managing the at least one configuration setting associated with the at least one camera comprises managing at least one of: an indication of the criticality level of the at least one camera by at least one indication device associated with the at least one camera; one or more imaging parameters of the at least one camera; one or more network parameters associated with the at least one camera; one or more data compression parameters associated with the one or more data streams captured by the at least one camera; one or more security parameters associated with the at least one camera; an alignment and / or an orientation of the at least one camera.

[0121] In an embodiment wherein when the one or more cameras comprises a first camera and a second camera located within a predetermined proximity to the first camera, the method comprises dynamically obtaining, by the processing unit, one or more data streams associated the first camera. The method further comprises analysing, by the processing unit, each obtained data stream associated with the first camera to detect the one or more events. The method further comprises dynamically determining, by the processing unit, the criticality level of the first camera based on the detection of the one or more events in the corresponding data stream. The method further comprises dynamically managing, by the processing unit, at least one configuration setting associated with the at least the second camera based on the determined criticality level of the first camera. In a further embodiment, the method may involve dynamically manage configuration setting(s) associated with the first camera as well as the second camera by the processing unit. The at least one configuration setting is indicative of the criticality level of the first camera in relation to the detection of the one or more events.

[0122] The object of the present disclosure is also achieved by a computer-program product having machine-readable instructions stored therein, that when executed by the processing unit, cause the processing unit to perform the aforementioned method steps. 202318179

[0123] 23

[0124] The object of the present disclosure is also achieved by a computer-readable storage medium which comprises instructions which, when executed by a processing unit, cause the processing unit to perform the aforementioned method steps.

[0125] Features which are described in the context of separable aspects and embodiments of the disclosure may be used together and / or be interchangeable. Similarly, features described in the context of a single embodiment may also be provided separately or in any suitable subcombination.

[0126] The above-mentioned aspects, features, embodiments, and advantages will become more apparent and understandable with the following description of embodiments of the disclosure in conjunction with the corresponding drawings. The illustrated embodiments are intended to illustrate, but not limit the disclosure.

[0127] The present disclosure is further described hereinafter with reference to illustrated embodiments shown in the accompanying drawings, in which:

[0128] FIG 1 is a block diagram of a system for dynamic surveillance management in an environment, in accordance with an embodiment of the present disclosure;

[0129] FIG 2 illustrates exemplary events occurring in an exemplary environment, in accordance with an embodiment of the present disclosure;

[0130] FIG 3 illustrates a block diagram of an exemplary architecture, in which an embodiment of the present disclosure can be implemented;

[0131] FIG 4 is a block diagram of an exemplary embodiment of the system for dynamic surveillance management incorporating indication devices;

[0132] FIG 5 illustrates a scene within an exemplary environment, in accordance with an embodiment of the present disclosure; and

[0133] FIG 6 is a flow diagram of a method for dynamic surveillance management in an environment, in accordance with an embodiment of the present disclosure. 202318179

[0134] 24

[0135] The drawings are for illustrative purposes only and may not be drawn to scale. The specific proportions of the components may vary depending on the implementation. Further, like reference signs have been used to indicate corresponding or similar elements throughout the drawings. It should be understood that like reference signs may be used to refer to the same or similar elements in different embodiments or views illustrated in the drawings.

[0136] Hereinafter, embodiments for carrying out the present disclosure are described in detail. The various embodiments are described with reference to the accompanying drawings, wherein like reference signs are used to refer to like elements throughout. In the following description, for the purpose of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more embodiments. It may be evident that such embodiments may be practiced without these specific details.

[0137] FIG 1 is a block diagram of a system 100 for dynamic surveillance management in an environment, in accordance with an embodiment of the present disclosure. Across embodiments, the environment may be a building, such as but not limited to, a residential building, an office, a research facility, and the like. The environment may also include open spaces, or a combination of buildings and open spaces. Therefore, the system 100 may be deployed in a variety of settings, from indoor spaces to outdoor spaces.

[0138] The system 100 comprises one or more cameras 102A, 102B -102N. In the embodiment depicted, the system 100 comprises a plurality of cameras 102A, 102B-102N. The number of the one or more cameras 102A, 102B-102N may vary depending on factors such as, but not limited to, size of the environment to be monitored, range of coverage of the cameras, associated costs, and other surveillance requirements.

[0139] Each of the one or more cameras 102A, 102B-102N is capable of capturing data within a corresponding field of view within the environment. The one or more cameras 102A, 102B-102N may be configured to capture data in one or more forms, including but not limited to, visual data, infrared radiation data, or other forms of data associated with electromagnetic radiation.

[0140] In an exemplary embodiment, the one or more cameras 102A, 102B-102N are capable of capturing visual data, either digitally using an electronic image sensor, or chemically via a lightsensitive material such as a photographic film. In a preferred embodiment, the one or more cameras 102A, 102B-102N are configured to capture visual data by using digital means such as electronic image sensors. In a preferred embodiment, the one or more cameras 102A, 102B- 202318179

[0141] 25

[0142] 102N are closed-circuit television cameras (CCTV) cameras. Each CCTV camera may be equipped with multiple features, such as night vision, motion detection, and remote access capabilities.

[0143] In another exemplary embodiment, the one or more cameras 102A, 102B-102N may be thermal imaging cameras specifically configured to capture infrared radiation, which can be utilized to detect and visualise temperature differences in the respective fields of view. This can be beneficial in applications where visual data may be obscured or insufficient, such as in low-light conditions or through smoke or fog.

[0144] In one example, the one or more cameras 102A, 102B-102N may be mounted on ceilings or tall structures within the environment, to capture a wide-angle view of the three-dimensional environment. In another example, the one or more cameras 102A, 102B-102N may be mounted on an Unmanned Aerial Vehicle (UAV) flying overhead, offering a dynamic and flexible vantage point. The mounting location of the one or more cameras 102A, 102B-102N may be suitably selected to meet specific surveillance requirements in the environment.

[0145] In a preferred embodiment, the one or more cameras 102A, 102B-102N are Internet-of-Things (loT) devices. This enables integration of the one or more cameras 102A, 102B-102N with other loT devices and systems within the environment, allowing for enhanced data sharing, automation, and remote management. The one or more cameras 102A, 102B-102N may be equipped with various sensors, such as environmental sensors and actuators, to manage their respective positions, fields of view and to allow for efficient monitoring of the environment.

[0146] The system 100 comprises a processing unit 104. The processing unit 104 is communicably coupled to the one or more cameras 102A, 102B-102N through a network 106. In some embodiments, the network connection between the processing unit 104 and the one or more cameras 102A, 102B-102N may be established using various communication means such as data transmission cables, Ethernet, Wi-Fi, Bluetooth, cellular networks, and the like. In some embodiments, the network 106 may include, but is not limited to, one or more components such as the Internet, a Local Area Network (LAN), a Wide Area Network (WAN), a mobile telecommunications network like a cellular network, a Wi-Fi network or Wi-Max network, or any suitable combination thereof. The one or more components of the network 106 may communicate information via a transmission medium. As used herein, the term “transmission medium” may refer to any intangible medium that is capable of communicating or transmitting instructions for 202318179

[0147] 26 execution by the processing unit 104 and can include digital or analog communication signals or other intangible media to facilitate communication of such software.

[0148] The processing unit 104 is configured to dynamically obtain one or more data streams from one or more data sources. The processing unit 104 may be configured to dynamically obtain the one or more data streams associated with the one or more cameras 102A, 102B-102N and / or the environment. In an embodiment, the one or more data streams comprises at least one of: data captured by the one or more cameras 102A, 102B-102N, data pertaining to location and / or field of view of the one or more cameras 102A, 102B-102N within the environment, and ambient data associated with the surroundings of the one or more cameras 102A, 102B-102N. The ambient data may include, but not be limited to, information pertaining to temperature, humidity, light levels, noise levels, air quality, and the like.

[0149] In an embodiment, the one or more data sources are the one or more cameras 102A, 102B-102N. In the embodiment, the processing unit 104 may be configured to directly receive data streams from the one or more cameras 102A, 102B-102N, either through a wired or a wireless network 106. In an embodiment, the one or more data sources may comprise of one or more intermediary servers (not shown). The data streams may be routed to the processing unit 104 through the one or more intermediary servers, which may be adapted for providing additional processing or storage capabilities. In another embodiment, the one or more data sources may comprise a remote server (not shown). The data streams may be routed to the processing unit 104 from the remote server by a remote vendor or operator, such as a cloud-based surveillance service provider. In yet another embodiment, the one or more data sources can be a combination of the aforementioned data sources. The configuration of the processing unit 104 to obtain data streams from multiple data sources may account for enhanced flexibility and scalability of the system 100. Additionally, integration of data from various data sources can provide comprehensive surveillance of the environment.

[0150] The processing unit 104 is configured to analyse each obtained data stream to detect one or more events. The processing unit 104 is configured to dynamically determine a criticality level of at least one camera of the one or more cameras 102A, 102B-102N based on the detection of the one or more events in the corresponding data stream that is analysed. Hereinafter, reference to “at least one camera” may refer to any of the depicted cameras 102A, 102B-102N, considered either individually or in any combination. 202318179

[0151] 27

[0152] The processing unit 104 is further configured to dynamically manage at least one configuration setting associated with the at least one camera of the one or more cameras 102A, 102B-102N based on the determined criticality level. The at least one configuration setting is indicative of the criticality level of the at least one camera in relation to the detection of the one or more events. In this context, the term “at least one configuration setting” refers to a customisable parameter associated with the at least one camera that controls the behaviour, operation, or performance of the system in relation to the at least one camera.

[0153] Across embodiments, the at least one configuration setting may encompass an indication of the criticality level of the at least one camera (for example, by means of colour coded LEDs), imaging parameters (such as resolution, frame rate etc.) associated with the at least one camera, data compression parameters (such as data compression ratio, compression algorithm etc.) associated with the at least one data stream captured by the at least one camera, security parameters (such as access control, encryption settings, etc.) Across embodiments, the processing unit 104 may be configured to dynamically manage the at least one configuration setting associated with the at least one camera by employing artificial intelligence algorithms, rule-based logic, or a combination thereof.

[0154] In one implementation, the processing unit 104 may be configured to leverage machine learning models to analyse historical data and identify correlations between event occurrence, system performance, and configuration settings. These models can be trained on various datasets, including camera footage, sensor readings, system logs, and event metadata. The processing unit may employ an artificial intelligence algorithm for analysing and correlating the training data with the determined criticality level of the at least one camera. The artificial intelligence algorithm can learn to optimize the at least one configuration setting associated with the at least one camera based on factors such as the type of the one or more events being monitored, the location of the at least one camera, and the current system load.

[0155] In another implementation, the processing unit 104 may be configured to employ rule-based logic to implement predefined rules and / or thresholds for managing or adjusting the at least one configuration setting associated with the at least one camera. These predefined rules may be based on factors comprising the determined criticality level of the at least one camera and possible configuration settings associable with the at least one camera. The predefined rules may be defined as “if-then” statements, where the “if” condition specifies the criteria for triggering the corresponding rule, and the “then” action specifies the at least one configuration setting or a configuration change to be applied. For example, event-based rules can trigger actions like 202318179

[0156] 28 increasing resolution and / or frame rate of the at least one camera for critical events, while decreasing them for low-priority events. Further, resource-based rules may help optimize system performance by adjusting resolution, frame rate, and bitrate associated with the at least one camera based on the criticality level of the at least one camera, considering storage and bandwidth constraints.

[0157] Additionally, the processing unit 104 may employ a threshold-based logic which involves setting specific numerical values or conditions that trigger changes in the at least one configuration setting associated with the at least one camera. For example, if the number of detected events exceeds a predefined threshold within a specified time-period (say 15 minutes) and correspondingly the criticality level of the at least one camera is high, the processing unit 104 may cause the resolution or the frame rate of the at least one camera to be increased. The predefined threshold may be adjusted based on various factors, such as but not limited to, the availability of system resources and the desired level of sensitivity.

[0158] Further, the processing unit 104 may be configured to employ mapping techniques to establish relationships between criticality levels and configuration settings. Across embodiments, artificial intelligence algorithms may be employed to identify these relationships and / or rule-based logic may be employed to define specific rules for mapping the configuration settings to different criticality levels. By combining these approaches, the system 100 can dynamically manage configuration settings to optimise performance and resource allocation based on camera criticality.

[0159] Furthermore, in an embodiment, the processing unit 104 may be configured to adjust the at least one configuration setting associated with the at least one camera within one or more ranges to optimise camera performance. For example, exposure can be increased in low-light conditions, resolution can be reduced to conserve bandwidth, and frame rate can be adjusted to capture fastmoving objects. Such adjustments may be made within predefined limits to ensure optimal image quality and system efficiency.

[0160] The configurations of the processing unit 104 will be explained in further detail along with relevant examples in the following further description.

[0161] As aforementioned, the processing unit 104 is configured to analyse each obtained data stream to detect the one or more events. Analysing each data stream involves processing each data stream, which involves detection of significant occurrences or changes within the system 100 202318179

[0162] 29 and / or the environment. The one or more events may be manually configured by a user, automatically considered by the system 100 or the processing unit 104 based on abnormal, unusual, or anomalous behaviour, or may be a combination of manual configuration and automatic consideration. The one or more events may comprise at least one of: a change in the location and / or the field of view of the one or more cameras 102A, 102B-102N, a change in one or more ambient parameters in the surroundings of the one or more cameras 102A, 102B-102N, an occurrence of a predetermined time-period, and a detection of at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the field of view of the one or more cameras 102A, 102B-102N, and detection of an abnormal pattern in the fields of view of the one or more cameras 102A, 102B-102N.

[0163] The one or more events will now be explained with relevant examples, with reference to FIG 2 in conjunction with FIG 1 . FIG 2 illustrates exemplary events occurring in an exemplary environment 200, according to an embodiment of the present disclosure. The environment 200 may be a manufacturing industry, for example.

[0164] In an embodiment, the one or more events comprises a change in the location and / or the field of view of the one or more cameras 102A, 102B-102N. Change in the location and / or the field of view of the one or more cameras may occur when the one or more cameras are moved or repositioned. For example, if a camera is moved from one location to a different location, it may capture new areas of the environment 200, potentially revealing new information or detecting previously unseen events. The camera may also be configured to change its respective field of view in response to another event, such as but not limited to, a movement of one or more subjects or objects away from its initial field of view.

[0165] In an embodiment, a camera 102C is adapted to monitor a storeroom 202 in the environment 200. The initial position of the camera is represented by dotted lines (annotated as 102C’). When an individual 204 enters the storeroom 202 and moves in a direction away from the field of view of the camera in its initial position 102C’, the camera 102C may be oriented such that its field of view is suitably changed to capture the movement of the individual 204. In this scenario, change in orientation of the camera 102C is detected as an event.

[0166] In an embodiment, the one or more events comprises a change in one or more ambient parameters in the surroundings of the one or more cameras 102A, 102B-102N. Changes in one or more ambient parameters may also indicate significant events. The one or more ambient parameters may include temperature, humidity, light levels, noise levels, and the like. For 202318179

[0167] 30 example, a sudden increase in temperature might indicate a fire or other hazardous situation, which could warrant intervention activities in the environment 200. Likewise, other events associated with the change in the one or more ambient parameters could be linked with varied criticality levels of the one or more cameras 102A, 102B-102N.

[0168] In an embodiment, a camera 102D is adapted to monitor a machine shop 206 in the environment 200. The machine shop 206 may accommodate one or more machines 208. In the depicted example, the one or more machines 208 is a lathe. The machine shop 206 may be provided with a temperature measuring device 206A, such as a thermometer, temperature sensor, thermocouple, resistive temperature device (RTD), thermistor and the like. The temperature measuring device 206A may be positioned in the vicinity of the camera 102D. The temperature measuring device 206A may be adapted to continually measure the room temperature of the machine shop 206A. The temperature measuring device 206A may be communicably coupled to the processing unit 104 through one or more wired or wireless data communication techniques known in the art. When the processing unit 104 detects that the room temperature of the machine shop 206 has increased beyond a threshold temperature (for example, 40 °C), the increase in temperature may be detected as an event.

[0169] In an embodiment, the one or more events comprises an occurrence of a predetermined timeperiod. Occurrence of a predetermined time-period, such as specific hours of a day, may also be a trigger for other events of varied significance. For example, occurrence of a scheduled time of the day for performing a critical forensic experiment or for an important meeting can be an event detected by the processing unit 104. The processing unit 104 may be further configured to detect and alert on events that occur during the predetermined time-period.

[0170] In an embodiment, a camera 102E is adapted to monitor a research laboratory 210 in the environment 200. As an example, if a critical experiment is scheduled to be conducted between 12 pm and 1 pm in the research laboratory, the occurrence of the time-period from 12 pm to 1 pm may be an event. Appropriate monitoring measures may be configured for the camera 102E during the occurrence of time-period from 12 pm to 1 pm.

[0171] In an embodiment, the one or more events comprises the detection of at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the fields of view of the one or more cameras 102A, 102B-102N. The detection may involve the use of computer vision algorithms or machine learning models to identify and track specific objects, individuals or patterns within each obtained data stream. 202318179

[0172] 31

[0173] In an embodiment, the one or more events comprises the detection of the predefined identifier. The predefined identifier may refer to a unique code or a label assigned to an individual, an object, or other entities. Predefined identifiers may be used to track and identify specific items or persons in the environment 200. Examples of predefined identifiers include barcodes and QR codes that can be scanned by cameras and relayed to the processing unit 104 for identifying products, individuals or other objects, and identity cards (ID cards) worn by individuals that may contain unique identification information (such as name, ID number, or codes) etc.

[0174] In an embodiment, two cameras 102F, 102G are adapted to monitor an inventory room 212 in the environment 200. In an exemplary scenario as depicted, detection of a predefined identifier from an identity card 214 worn by an individual 216 from the data streams captured by the camera 102F by the processing unit 104 can be an event. Such an event detection may be helpful in identifying a certain individual in a specified region in the environment 200 where the individual is not supposed to or not authorised to enter.

[0175] In an embodiment, one or more events comprises the detection of a predefined pattern. Predefined patterns may refer to specific events or behaviours that indicate suspicious or anomalous activity. The predefined patterns may be identified using computer vision algorithms and machine learning techniques. Examples of predefined patterns include unusual movement patterns in a space, such as a person loitering in a restricted area or a vehicle driving erratically, and unusual object interactions, such as a person tampering with a piece of equipment or a vehicle approaching a restricted area.

[0176] In an exemplary scenario as depicted, detection of a running individual 218 inside the inventory room 212 from the video feed captured by the camera 102G may correspond to the detection of a predefined pattern. Running inside the inventory room 212 can be dangerous to the individual 218 as well as to the inventory. Inventory room 212 may contain shelves, boxes, and other obstacles that can be knocked over or tripped on by the running individual 218, which increases the likelihood of accidents and potentially leading to injuries. Further, running can potentially cause items to be dislodged from shelves or knocked over, resulting in damage or loss to the inventory, which may further lead to financial losses and operational disruptions. Therefore, it can be crucial to detect such predefined patterns in the inventory room 212 for record keeping and further intervention. 202318179

[0177] 32

[0178] In an embodiment, one or more events comprises the detection of a predefined individual. Predetermined individuals may refer to individuals identified as persons of interest or potential threats. These individuals may be known suspects, or individuals with a history of harmful or criminal activity or may be flagged for other reasons. Examples of predetermined individuals include individuals who have been identified as suspects in a crime or other incident, and individuals who have a previous criminal record or who are known to pose a security threat.

[0179] In an embodiment, a camera 102H is adapted to monitor a corridor 220 in the environment 200. When the camera 102H captures an individual 222 in the corridor 220, the processing unit 104 communicably coupled to the camera employs camera vision techniques or machine learning algorithms for facial recognition of the individual 222 to identify whether the individual 222 is a predefined individual such as a known suspect. Identification of the individual 222 as a known suspect may correspond to the detection of an event. Appropriate surveillance measures or actions may be undertaken based on the nature of threat posed by the individual 222.

[0180] In an embodiment, one or more events comprises the detection of a predetermined object. Predetermined objects may refer to objects that are considered to be potential security risks or threats. These objects may be unusual or out of place for the specific environment. Examples of predetermined objects may include suspicious packages and prohibited items (i.e., items that are prohibited in the particular environment, such as weapons and hazardous materials).

[0181] In an exemplary scenario as depicted, when the camera 102H captures an object 224 in the corridor 220, the processing unit 104 employs camera vision techniques, heat mapping, or machine learning algorithms for identifying whether the object 224 is a hazardous or a prohibited object. Identification of the object 224 as a hazardous / prohibited object may correspond to the detection of an event. Appropriate surveillance measures or actions may be undertaken based on the nature of threat posed by the object 224.

[0182] In an embodiment, the one or more events comprises detection of an abnormal pattern in the fields of view of the one or more cameras 102A, 102B-102N. In this context, the abnormal pattern may refer to a deviation from a statistically expected or an established norm or trend within the environment 200. This deviation may be indicative of an anomaly, an irregularity, or an event that is outside a normal range of parameters or behaviours.

[0183] In the environment 200, which may be considered an industrial environment, a normal pattern might be the consistent movement of machinery, workers, and materials in a predictable 202318179

[0184] 33 sequence or manner. However, an abnormal pattern could arise from a disruption to this routine. For example, detection of a person lying motionless on the factory floor, fall or collapse of an individual or equipment, and accidents within the environment could be considered as abnormal patterns. The system 100 may employ a vision based artificial intelligence model for the detection of such abnormal patterns. More particularly, the processing unit 104 may be configured to access the vision based artificial intelligence model for the detection of the abnormal patterns. Such abnormal patterns raise a red flag, prompting further investigation and / or immediate action. The occurrence of such abnormal events directly influences the criticality level or sensitivity level of the camera(s) monitoring such events.

[0185] A person of ordinary skill in the art would appreciate that the above-mentioned examples are merely illustrative and not intended to limit the scope of the invention. The invention encompasses a wide range of embodiments, and the specific examples provided are merely representative of some of the possible implementations.

[0186] Advantageously, the detection of the one or more events by the processing unit allows for proactive monitoring, enhanced situational awareness, and improved security of the environment 200. Furthermore, the detection of the one or more events provides valuable insights, enabling data-driven decision-making for optimization of surveillance strategies.

[0187] FIG 3 illustrates a block diagram of an exemplary architecture 300, in which an embodiment of the present disclosure can be implemented. The architecture 300 comprises a computer 302. As used herein, the term “computer” refers to a programmable electronic device capable of performing calculations, processing information, and executing instructions. This includes, but is not limited to, a general-purpose computer, a specialized computer, or any other device that can perform the functions described herein. The computer 302 may be an on-site computer located in the environment or an off-site computer operable by a remote user. The choice of location of the computer may depend on factors such as the size and complexity of the system 100, the need for local processing capabilities and network connectivity.

[0188] The computer 302 includes the processing unit 104, an accessible memory 304, a storage unit 306, a communication interface 308, an input-output unit 310, a network interface 312 and a bus 314.

[0189] The processing unit 104, as used herein, means any type of computational circuit, such as, but not limited to, a microprocessor unit, microcontroller, complex instruction set computing 202318179

[0190] 34 microprocessor unit, reduced instruction set computing microprocessor unit, very long instruction word microprocessor unit, explicitly parallel instruction computing microprocessor unit, graphics processing unit, digital signal processing unit, or any other type of processing circuit. The processing unit 104 may also include embedded controllers, such as generic or programmable logic devices or arrays, application specific integrated circuits, single-chip computers, and the like.

[0191] The memory 304 may be non-transitory volatile memory and non-volatile memory. The memory 304 may be coupled for communication with the processing unit 104, such as being a computer- readable storage medium. The processing unit 104 may execute machine-readable instructions and / or source code stored in the memory 304. A variety of machine-readable instructions may be stored in and accessed from the memory 304. The memory 304 may include any suitable elements for storing data and machine-readable instructions, such as read only memory, random access memory, erasable programmable read only memory, electrically erasable programmable read only memory, a hard drive, a removable media drive for handling compact disks, digital video disks, diskettes, magnetic tape cartridges, memory cards, and the like. In the present embodiment, the memory 304 includes a module package 316. The module package 316 is stored in the form of machine-readable instructions on any of the above-mentioned storage media and may be in communication with and executed by the processing unit 104.

[0192] The storage unit 306 may be a non-transitory storage medium configured for storing a database which comprises one or more user-defined criteria or patterns that can be utilized by the processing unit 104 for the detection of the one or more events.

[0193] The communication interface 308 is configured for establishing data communication between the one or more data sources and the processing unit 104 embodied in the computer 302.

[0194] The input-output unit 310 may include input devices such as a keypad, a touch-sensitive display, a gesture sensor etc. capable of receiving one or more inputs from the user to configure the one or more user-defined criteria or patterns that can be utilized by the processing unit 104 for the detection of the one or more events. The bus 314 acts as an interconnect between the processing unit 104, the memory 304, and the input-output unit 310.

[0195] The network interface 312 is configured to handle network connectivity, bandwidth, and network traffic with a communication network 318, a machine-learning model 320, and a semantic contextualization module 322. The machine-learning model 320 and the semantic contextualization module 322 are integral parts of the system 100. 202318179

[0196] 35

[0197] The communication network 318 serves as a communication channel between the machinelearning model 320, the semantic contextualization module 322, and the processing unit 104. The communication network 318 facilitates the exchange of data and control signals, enabling integration and operation of the machine-learning model 320, the semantic contextualization module 322, and the processing unit 104 within the system 100.

[0198] In an embodiment, the machine-learning model 320 may be utilized by the processing unit 104 to analyse the dynamically each obtained data streams for detecting the one or more events. In other words, the processing unit 104 is configured to analyse each dynamically obtained data stream by execution of the machine-learning model 320 for detecting the one or more events. The machine-learning model 320 can be trained on extensive labelled data to learn patterns and correlations that may be indicative of the one or more events. Once trained, the machine-learning model 320 can be applied to new data streams to detect similar events in real-time. Advantageously, the machine-learning model 320 can achieve higher accuracy rates than conventional rule-based methods for event detection. The machine-learning mode 320 may also be configured to handle large volumes of data and can advantageously analyse data streams from a scalable number of cameras.

[0199] In a further embodiment, the machine-learning model 320 is a multimodal machine-learning model. That is, in the embodiment, the processing unit 104 is configured to analyse each dynamically obtained data stream by execution of the multimodal machine-learning model for the detection of the one or more events. The multimodal machine-learning model may be well suited for dynamic surveillance management as the model can process multiple formats of data such as image feed from the one or more cameras 102A, 102B-102N, location data associated with the one or more cameras 102A, 102B-102N, sensory data etc. and provide more accurate and reliable event detection especially in a noisy environment. Additionally, the multimodal machine learning model may advantageously detect patterns and relationships that may be difficult to identify using a single-modal machine-learning model.

[0200] In a further embodiment, the machine-learning model 320 is a multi-modal Retrieval-Augmented Generation (RAG) based Generative Artificial Intelligence (GenAI) model (i.e., RAG system). The RAG system is trained using a comprehensive dataset which comprises a plurality of annotated images, a plurality of video feeds, and a plurality of sensor data from a plurality of three- dimensional environments. The training data set comprises labelled examples of various objects a plurality of object classifications, one or more spatial positions, and dynamic changes over time. 202318179

[0201] 36

[0202] For instance, consider a case of video surveillance of a building interior. The training data might include labelled examples of objects such as people, furniture, and doors. Spatial positions could be represented as coordinates within the video frame, and dynamic changes over time could be captured by tracking the movement of objects. The RAG system is trained by application of supervised learning techniques on a plurality of neurons of the RAG system. The RAG system may be configured to ingest data streams obtained from the one or more data sources (such as the one or more cameras 102A, 102B-102N), detect the one or more events in each data stream, generate structured queries, retrieve relevant information, and generate responses.

[0203] Advantageously, the RAG system can leverage comprehensive training data to provide accurate and relevant responses based on the specific context of the analysed data streams. Furthermore, the RAG system can generate human-readable summaries in natural language, making it convenient for users to understand to understand and appropriately respond to the detected events.

[0204] In another embodiment, the processing unit 104 is configured to implement a Convolutional Neural Network (CNN) model for detecting the one or more events in the data streams captured by the one or more cameras 102A, 102B-102N. CNNs are well-suited for image processing and can be used to detect objects, faces, and visual patterns in the footages captured by the one or more cameras 102A, 102B-102N. In yet another embodiment, the processing unit 104 is configured to implement a Recurrent Neural Networks (RNN) model for detecting the one or more events associated with time-series data, such as traffic data within the environment. RNNs are well-suited for processing sequential or time-series data, making them suitable for detecting anomalies or patterns over time, such as unusual behaviour or changes in activity levels in the respective fields of view being monitored.

[0205] The processing unit 104 is configured to dynamically determine a criticality level of at least one camera of the one or more cameras 102A, 102B-102N based on the detection of the one or more events in the corresponding data stream that is analysed. In some embodiments, the processing unit 104 may implement a rule-based logic, a pre-trained machine learning model, or employ a hybrid approach of utilizing a combination of rule-based logic and machine learning model to dynamically determine the criticality level of the at least one camera. In an embodiment, the processing unit 104 may be configured to dynamically determine the respective criticality levels of each the plurality of cameras 102A, 102B-102N as depicted in FIG 1 which illustrates an exemplary embodiment of the present disclosure. 202318179

[0206] 37

[0207] Dynamically determining the criticality level of the at least one camera refers to the process of continuously assessing and updating the criticality or the importance of the at least one camera based on the analysis of the one or more data streams. In an embodiment, the criticality level of the at least one camera can be reassessed at regular intervals, for example, every two or three seconds. This approach accounts for periodic updates in the criticality level based on the detection of the one or more events during the specified interval. In an embodiment, the criticality level of the at least one camera can be updated in real-time as the data streams are analysed. This approach allows for rapid adjustments to the criticality level of the at least one camera based on current conditions. In an embodiment, in addition to periodic or real-time analysis, the processing unit 104 may be configured to analyse multiple data streams in parallel. Advantageously, such parallel processing enables the system to process data more efficiently when dealing with large volumes of data.

[0208] The criticality level of the at least one camera can be determined using various metrics or classifications. For example, a numerical rating system may be employed, where at least one camera is assigned a criticality level on a scale of 1 to 10, with 1 representing the lowest level of criticality and 10 representing the highest level of criticality. Alternatively, a categorial system can be used for classifying the at least one camera into categories such as “low”, “medium”, “high” or “severe” criticality. Alternatively, a percentage-based rating system can be used to quantify the criticality level of the at least one camera, with higher percentages indicating higher criticality.

[0209] In an embodiment, if the one or more events are deterministic in nature and can be classified into predefined categories the processing unit 104 may utilise a lookup table to map the detected one or more events to corresponding criticality levels. Advantageously this approach can simplify the determination of criticality level and enable consistent application of rule-based logic. However, it is noteworthy that real-world events may not always fit neatly into predefined categories, and the lookup table may need to be updated periodically to account for such real-world events.

[0210] In an embodiment, the processing unit 104 is configured to dynamically determine the criticality level of the at least one camera by determining, by execution of the semantic contextualization module 322, at least one label associable with the at least one camera based on the detection of the one or more events in the corresponding data stream. The processing unit 104 is further configured to tag the at least one label to the at least one camera. The at least one label is associated with the criticality level of the at least one camera. 202318179

[0211] 38

[0212] As used herein, the semantic contextualization module 322 corresponds to a machine-readable set of instructions accessible by the processing unit 104. The semantic contextualization module 322 may leverage various techniques, including rule-based methods, statistical models, and machine-learning algorithms to extract meaningful and contextual information from the data streams processed by execution of the machine-learning model 320. The semantic contextualization module 322 may also incorporate knowledge graphs or ontologies for provide additional context and understanding of the data streams.

[0213] In an exemplary embodiment, the semantic contextualization module 322 works in conjunction with the machine-learning model 320 being the RAG system. In the embodiment, the machinelearning model 320 (RAG system) processes raw data, such as video streams captured by the at least one camera, sensor data, etc. to extract relevant features and prepare for analysis. The machine-learning model 320 then applies its learned patterns and algorithms to identify the one or more events within the corresponding data stream being analysed. The semantic contextualization module 322 then analyses the detected one or more events to extract their meaning and context. The semantic contextualization module 322 may identify keywords, phrases, and relationships between concepts to gain a deeper understanding of the detected one or more events. The semantic contextualization module 322 may use the extracted semantic information to classify the detected one or more events into specific categories or types. The semantic contextualization module 322 may also provide context for the detected one or mor events, helping to understand their significance and potential implications. This context may then be used to determine the criticality level of the at least one camera.

[0214] Once the semantic contextualization module 322 has analysed the corresponding data stream and the detected one or more events, the semantic contextualization module 322 determines at least one label associable with the at least one camera. The at least one label may be a descriptive term or a category that represents the camera’s importance or sensitivity. For example, the at least one label might include “high-priority”, “medium-priority”, or “low-priority”.

[0215] The processing unit 104 is configured to tag the determined at least one label to the at least one camera. In an embodiment, a dedicated database can be created to store information about each camera, including its unique identifier, location, and associated labels. The database can be queried to retrieve the criticality level of the corresponding camera. In an embodiment, a custom data structure can be defined to store information about each camera 102A, 102B-102N, including its unique identifier, location, and associated labels. This data structure can be integrated into the 202318179

[0216] 39 system architecture and accessed by the processing unit 104 to determine the criticality of each camera 102A, 102B-102N.

[0217] Advantageously, determination of the criticality level provides valuable data that can be used for informed decision-making to optimise surveillance strategies. By identifying criticality trends for the cameras 102A, 102B-102N, an enterprise an allocate resources and attention to most important areas in the environment, ensuring that critical data is protected and monitored effectively.

[0218] Advantageously, the system 100 can effectively tag cameras 102A, 102B-102N with appropriate labels and associate the cameras 102A, 102B-102N with the corresponding criticality levels. This information can then be used for prioritizing monitoring and maintenance activities, allocating resources in an efficient manner, and improve overall system performance.

[0219] The processing unit 104 is further configured to dynamically manage at least one configuration setting associated with the at least one camera based on the determined criticality level. The at least one configuration setting is indicative of the criticality level of the at least one camera in relation to the detection of the one or more events.

[0220] Advantageously, by tailoring configuration setting associated with the cameras 102A, 102B-102N to their respective criticality levels, the system 100 can ensure that the critical data is captured. Further, the system 100 is capable of adjusting the configuration settings associated with the cameras 102A, 102B-102N in real-time to optimize their performance and security based on the nature of the one or more events and the perceived criticality level. Furthermore, the ability of the system 100 to dynamically adjust configuration settings provides greater flexibility and adaptability in response to evolving threats and changing environmental conditions.

[0221] FIG 4 is a block diagram of an exemplary embodiment of the system 100 for dynamic surveillance management incorporating indication devices 402A, 402B-402N, 404.

[0222] In an embodiment, the system 100 comprises one or more indication devices. The one or more indication devices may include one or more of illumination devices 402A, 402B-402N, sound devices, visual displays, handheld devices, and the like. In the depicted embodiment, the illumination devices 402A, 402B-402N may be light lamps, light emitting diodes (LEDs), or projection systems. In an embodiment, the one or more indication devices may include a handheld device or a terminal device such as a mobile phone 404 or a tablet that provides a user interface. 202318179

[0223] 40

[0224] In one embodiment, there could be a single indication device associate with multiple cameras 102A, 102B-102N. For example, the single indication device could be a central control panel, or a dedicated monitoring station embodied in the mobile phone 404. In an embodiment, each camera 102A, 102B-102N has its own dedicated indication device. In the depicted embodiment, each camera 102A, 102B-102N is provided with a corresponding illumination device 402A, 402B- 402N. This could provide a more localized and specific information about the status of individual cameras 102A, 102B-102N. In the depicted embodiment, the cameras 102A, 102B-102N are provided with a common indication device - the mobile phone 404 as well as dedicated illumination devices 402A, 402B-402N. The choice of indication devices depends on factors such as but not limited to, size and complexity of the system 100 or the environment the desired level of detail in indication, and the preferences of the users.

[0225] In the embodiment, the one or more indication devices 402A, 402B-402N, 404 are communicably coupled to the processing unit 104. The one or more indication devices 402A, 402B-402N, 404 may be communicably coupled to the processing unit through a variety oof wired or wireless communication techniques known in the art. The specific method for such communication may be selected based on factors such as the distance between the one or more indication devices 402A, 402B-402N, 404 and the one or more cameras 102A, 102B-102N, the required data transmission rate, and the security requirements of the system 100.

[0226] As already mentioned, the processing unit 104 is configured to dynamically manage at least one configuration setting associated with the at least one camera based on the determined criticality level. In an embodiment, the at least one configuration setting comprise an indication of the criticality level of the at least one camera by at least one indication device of the one or more indication devices 402A, 402B-402N, 404. The at least one indication device is associated with the at least one camera.

[0227] Advantageously, managing the indication of the criticality levels of the cameras 102A, 102B-102N provides users / operators with real-time information about the criticality status of the cameras 102A, 102B-102N. By indicating the criticality levels of the cameras 102A, 102B-102N, users can make informed decisions pertaining to incident response. Further, managing the indication of the criticality levels of the cameras 102A, 102B-102N by the indication devices 402A, 402B-402N, 404 may help in reducing response times by alerting users of critical situations as soon as they occur. Furthermore, a record of the indications being managed by the system 100 can be used for accountability and compliance purpose. Moreover, cameras with a high or severe criticality 202318179

[0228] 41 level can be scheduled for maintenance during off-peak hours or when they are not capturing sensitive data, thereby reducing the risk of data loss or corruption.

[0229] In a preferred embodiment, the at least one indication device is an illumination device, such as an LED, configured to emit light at a predefined wavelength (or colour) corresponding to the determined criticality level of the at least one camera. The illumination device may be positioned on, adjacent to, or proximal to the at least one camera. This provides a clear and visually intuitive indication of the criticality status of the at least one camera. In the depicted embodiment, the illumination device 402A is associated with the first camera 102A and is configured to indicate the criticality level of the first camera 102A. Similarly, illumination device 402B is associated with the second camera 102B and is configured to indicate the criticality level of the second camera 102B.

[0230] A colour coding scheme may be employed to indicate the criticality level / status of the cameras 102A, 102B-102N. For example, the indication devices 402A, 402B-402N, 404 could be configured to emit red light for cameras with severe criticality level, orange light for cameras with a high criticality level, yellow light for cameras with nominal or normal criticality level, and green light for cameras with a low criticality level. Such a colour coding scheme allows operators to quickly and easily identify the criticality levels of individual cameras based on the colour of the emitted light.

[0231] In another embodiment, blinking rate of the illumination devices 402A, 402B-402N may be dynamically adjusted by the processing unit to visually indicate the criticality level of the associated cameras 102A, 102B-102N. For example, a rapid, flashing pattern of illumination by the illumination devices 402A, 402B-402N may correspond to severe criticality of the associated cameras, which may indicate a severe-priority threat, such as a breach or intrusion. As another example, a slow and steady blinking pattern may correspond to low criticality of the associated cameras 102A, 102B-102N, which may indicate a less urgent event, such as a minor traffic incident or a routine activity in the environment.

[0232] By visually indicating the criticality levels of the cameras 102A, 102B-102N, operators can prioritize maintenance activities and avoid disrupting critical data streams. For example, cameras with a high or severe criticality level can be scheduled for maintenance during off-peak hours or when they are not capturing sensitive data, thereby reducing the risk of data loss or corruption.

[0233] In one embodiment, the one or more indication devices comprise a mobile phone 404 or a terminal device that supports an application. The application includes a map of the environment displaying 202318179

[0234] 42 locations of the one or more cameras 102A, 102B-102N and corresponding icons (for example, LED icons) that indicate the criticality levels of the one or more cameras 102A, 102B-102N. Additionally, the application may support pop-up notifications that provide real-time information about the criticality level of each camera. This allows users to simultaneously monitor the status of multiple cameras and identify any areas of concern.

[0235] In an embodiment, some indication devices could be configured to indicate a traffic level of the data to the users. This would provide the users with information about the volume and intensity of the data being processed by the system. Advantageously, this information can then be used to make informed decision about resource allocation, such as adjusting the number of cameras being monitored or allocating additional processing power. By understanding the volume of data being processed, the users can identify potential bottlenecks and allocate resources accordingly. This can help to ensure that the system 100 is running efficiently, and that critical data is not being lost or delayed.

[0236] Referring to FIG 4 and / or FIG 1 , in an embodiment, the at least one configuration setting comprises at least one of: one or more imaging parameters of the at least one camera, one or more network parameters associated with the at least one camera, one or more data compression parameters associated with the one or more streams captured by the at least one camera, one or more security parameters associated with the at least one camera, and an alignment and / or an orientation of the at least one camera. References to “at least one camera” may refer to any of the depicted cameras 102A, 102B-102N, considered either individually or in any combination.

[0237] In an embodiment, the at least one configuration setting comprises one or more imaging parameters of the at least one camera. The one or more imaging parameters are dynamically configured based on the determined criticality level of the at least one camera, ensuring that the captured data is of sufficient quality and relevance for specific surveillance scenario. The one or more imaging parameters may include, but not be limited to, frame rate, resolution, focus setting, zoom level, noise level, colour correction, exposure compensation, and the like.

[0238] For example, with an increase in criticality level of the at least one camera, the frame rate configuration of the at least one camera may be increased. Advantageously, higher frame rates can capture more details and facilitate the analysis of fast-moving objects or events, which may be necessary when the criticality level of the at least one camera is high. As another example, with an increase in criticality level of the at least one camera, the processing unit 104 may cause the resolution of the at least one camera to be increased. Accordingly, the at least one camera 202318179

[0239] 43 can monitor the respective field of view with a higher level of detail with an increase in the corresponding criticality level. Furthermore, based on the criticality level of the at least one camera, the focus setting of the at least one camera may be suitably adjusted to ensure that objects or subjects within the field of view are in sharp focus, providing clearer and more accurate information.

[0240] With an increase in criticality level of the camera, the precision and / or the resolution associated with the zoom level of the at least one camera may be increased, allowing for better control by a remote operator. The extent of noise reduction in a footage captured by the at least one camera can be increased with an increase in the criticality level of the at least one camera, thereby improving footage quality. Additionally, based on the criticality level of the at least one camera, the extent of colour correction and exposure compensation on the image / video feed captured by the at least one camera may be suitably adjusted to give a clearer image for further analysis and intervention as may be necessary to ensure security.

[0241] Advantageously, by dynamically adjusting the imaging parameters based on the criticality of the cameras 102A, 102B-102N, the system 100 can optimize the captured data for specific use cases, ensuring that the data is of sufficient quality and relevance for analysis and decision making.

[0242] In an embodiment, the at least one configuration setting comprises one or more network parameters associated with the at least one camera. The one or more network parameters associated with the at least one camera may include, but not be limited to, data transmission rate of the data streams captured by the at least one camera and network priority associated with the at least one camera. The one or more network parameters may be dynamically adjusted based on the determined criticality level of the at least one camera, ensuring that the at least one camera has the necessary resources and the bandwidth to transmit data effectively. When the criticality level of the at least one camera is high, the processing unit 104 can dynamically adjust the network parameters to increase the rate of data transmission from the at least one camera to the processing unit. This can involve allocating more bandwidth to the at least one camera, assigning it a higher network priority, and optimizing network routing to minimize latency.

[0243] When the criticality level of the at least one camera is low, the processing unit 104 can dynamically adjust the network parameters to reduce resource consumption and prioritise other cameras with higher criticality levels. This could involve decreasing the data transmission rate from the at least one camera to the processing unit 104 to conserve bandwidth and network resources, assigning 202318179

[0244] 44 a lower priority to the at least one camera’s network traffic, and optimizing network routing such that the data from critical cameras is prioritised.

[0245] Advantageously, by dynamically adjusting the network parameters based on the criticality of the cameras 102A, 102B-102N, the system 100 can ensure that critical data is transmitted efficiently and reliably, even during periods of high network load or congestion. Consequently, the system 100 can also ensure that resources are allocated efficiently, and that critical or sensitive data is prioritised. This can help prevent data loss from the critical cameras and enhance the overall performance of the system 100.

[0246] In an embodiment, the at least one configuration setting comprises one or more data compression parameters associated with the one or more data streams captured by the at least one camera. The one or more data compression parameters may include compression ratio. When the criticality level of the at least one camera is high, the processing unit 104 configures a lower compression ratio for the one or more data streams captured by the at least one camera, to preserve more details and ensure that the captured data is sufficient for analysis and decisionmaking. Conversely, when the criticality level of the at least one camera is low, the processing unit 104 configures a higher compression ratio for the one or more data streams captured by the at least one camera, to reduce storage requirements and bandwidth consumption.

[0247] Advantageously, by dynamically adjusting data compression parameters based on the criticality level of the cameras 102A, 102B-102N, the system 100 can optimize the balance between data quality and resource utilization. Consequently, this helps to ensure that critical data is captured and transmitted effectively, while also minimizing the impact on system performance and costs.

[0248] In an embodiment, the at least one configuration setting comprises one or more security parameters associated with the at least one camera. The one or more security parameters may include, but not be limited to, access control settings and encryption settings. Access control settings regulate who can access and control the at least one camera and its captured data. If the criticality level of the at least one camera is high, the processing unit 104 may configure a more stringent access control setting to prevent unauthorized access and data breaches. Correspondingly, the processing unit 104 may configure stronger encryption levels for the network associated with the at least one camera to help protect the data captured by the at least one camera from unauthorized access and intervention. Conversely, for cameras with lower criticality levels, less stringent security measures may be sufficient. This can help reduce the computational 202318179

[0249] 45 overhead and resource requirements associated with security measures, while still maintaining a reasonable level of protection.

[0250] Advantageously, by dynamically adjusting security parameters based on camera criticality, the system 100 can ensure that sensitive data is adequately protected while minimizing the impact on system performance and resource consumption.

[0251] In an embodiment, the at least one configuration setting comprises an alignment and / or an orientation of the at least one camera. This configuration setting may be particularly useful in situations where a suspect or a person of interest moves beyond the initial field of view of the at least one camera. For example, consider a case wherein the at least one camera is monitoring a hallway. If a suspect is captured by the at least one camera, the criticality level of the camera may be assigned as high. If the suspect who is initially captured by the at least one camera then moves out of its field of view, the processing unit 104 may cause the alignment and / or the orientation of the at least one camera to be dynamically adjusted to capture the suspect’s continued movement.

[0252] Advantageously, by dynamically adjusting the alignment and / or orientation of the one or more cameras 102A, 102B-102N, a wider range of area may be captured when the one or more events are detected, ensuring that important events are captured to the extent possible. Furthermore, operators can gain a better understanding of the overall situation within the environment, thereby enabling them to address the situation.

[0253] Referring FIG 4, in an embodiment, the one or more cameras 102A, 102B-102N comprises a plurality of cameras with overlapping fields of view adapted to monitor a common area or region in the environment. In the embodiment, the processing unit 104 is configured to determine a relative priority of each camera based on one or more criticality levels associated with each camera determined over a predetermined period. For example, the predetermined period could be thirty minutes. The processing unit 104 is further configured to activate the one or more indication devices 402A, 402B-402N, 404 associated with the plurality of cameras 102A, 102B- 102N to indicate the determined relative priority of each camera to the user. Thereby, the processing unit provides a recommendation of order of maintenance activity for the plurality of cameras 102A, 102B-102N that minimizes the potential for data loss due to interruption in the corresponding data streams.

[0254] FIG 5 illustrates a scene within an exemplary environment 500, in accordance with an embodiment of the present disclosure. In an exemplary embodiment, the environment 500 202318179

[0255] 46 comprises a ballroom 502. The ballroom 502 has an entrance 502A and an exit 502B. The system 100 comprises a first camera 102A adapted for monitoring the entrance 502A of the ballroom 502 and a second camera adapted for monitoring the exit 502B of the ballroom 502. The first camera 102A is provided with a first indication device 402A. In the embodiment, the first indication device 402A is an LED. The second camera 102B is provided with a second indication device 402B. In the embodiment, the second indication device 402B is an LED. As depicted, the first camera 102A and the second camera 102B have overlapping fields of view, meaning they can capture some of the same areas within the ballroom 502. In other words, there is at least a partial overlap in the respective fields of view of the first camera 102A and the second camera 102B. As an example, the overlapping field of view might be around 20% of the total space monitored by the first camera 102A and the second camera 102B, indicating that the first camera 102A and the second camera 102B share a significant portion of their coverage.

[0256] Reference is now made to FIG 5 in conjunction with FIG 4. Over a predetermined period (for example, 30 minutes), the system 100, particularly the processor 104 of the system 100, would analyse the data streams captured by both the first camera 102A and the second camera 102B to determine their respective criticality levels. Based on factors such as but not limited to the frequency of the detected events, the sensitivity of the monitored area, and the potential consequences of data loss, the first camera 102A may be assigned a higher criticality level than the second camera 102B. The first camera 102A may be considered more critical due to its potential for capturing important security footage, such as individuals 504A entering the ballroom 502 from the entrance 502A. The second camera 102B may be considered less critical, as it primarily captures footage of individuals 504B leaving the ballroom 502 from the exit 502B.

[0257] The processing unit 104 of the system 100 would then activate the indication devices 402A, 402B associated with both the cameras 102A, 102B to indicate their relative priorities. For example, the first indication device 402A associated with the first camera 102A may emit a red light for indicating severe criticality of the first camera 102A, and the second indication device 402B associated with the second camera 102B might emit a green light for indicating low criticality of the second camera 102B. If a situation requiring maintenance of both cameras 102A, 102B should arise, which would not affect immediate functioning of both the cameras 102A, 102B, the second camera 102B could be maintained first as it streams less critical data in comparison to the first camera 102A. Once the second camera 102B is maintained, the first camera 102A could be maintained. Advantageously, the system 100 of the present disclosure is adapted to prioritize maintenance of the cameras based on their relative criticality so as to minimize the potential for data loss. 202318179

[0258] 47

[0259] In an embodiment, the second camera 102B is located within a predetermined proximity (for example, 10 m) to the first camera 102A. The processing unit 104 is configured to dynamically obtain one or more data streams associated with the first camera 102A (such as, footage captured by the first camera 102A). The processing unit 104 is configured to analyse each obtained data stream associated with the first camera 102A to detect the one or more events. The processing unit 104 is configured to dynamically determine the criticality level of the first camera 102A based on the detection of the one or more events in the corresponding data stream. The processing unit 104 is further configured to dynamically manage at least one configuration setting associated with at least the second camera 102B based on the determined criticality level of the first camera 102A. In a further embodiment, the processing unit 104 may dynamically manage configuration setting(s) associated with the first camera 102A as well as the second camera 102B. The at least one configuration setting is indicative of the criticality level of the first camera 102A in relation to the detection of the one or more events.

[0260] For example, consider the depicted scenario of a vehicle 506 in a restricted area 502C of the ballroom 502 being captured by the first camera 102A, which constitutes the detection of an event. Based on the nature of the event, the first camera 102A whose footage initially captured the vehicle 506 may be assigned a high criticality level. Additionally, since the second camera 102B is located proximate to the first camera 102A, it may also be crucial to monitor the movement and activities of the vehicle 506 within the field of view of the second camera 102B, which may occur shortly after the vehicle 506 is being captured by the first camera 102A. Therefore, the processing unit 104 could assign a high criticality level to the second camera 102A and adjust its configuration setting(s) accordingly. This could involve increasing the resolution, frame rate, or encryption parameters for the second camera 102B to ensure that it can capture and transmit relevant data effectively. Advantageously, by dynamically managing the configuration settings as described herein, the system 100 can ensure that critical events are captured and analysed effectively even when subjects or objects move between the fields of view of proximate cameras. This can contribute towards improving the overall security and effectiveness of the system 100.

[0261] FIG 6 is a flow diagram of a method 600 for dynamic surveillance management in the environment, accordance with an embodiment of the present disclosure.

[0262] At step 602, the method 600 involves dynamically obtaining, by the processing unit 104, one or more data streams from one or more data sources. In an embodiment, the one or more data streams are associated with at least one of the one or more cameras 102A, 102B-102N and the 202318179

[0263] 48 environment. In an embodiment, the one or more data streams comprises at least one of: the data captured by the one or more cameras 102A, 102B-102N, data pertaining to location and / or field of view of the one or more cameras 102A, 102B-102N within the environment, and ambient data associated with the surroundings of the one or more cameras 102A, 102B-102N. The ambient data may include, but not be limited to, information pertaining to temperature, humidity, light levels, noise levels, air quality, and the like.

[0264] In an embodiment, the one or more data sources are the one or more cameras 102A, 102B-102N. In an embodiment, the processing unit 402 may be configured to directly receive data streams from the one or more cameras 102A, 102B-102N, either through a wired or a wireless network connection. In an embodiment, the one or more data sources may comprise of one or more intermediary servers. In another embodiment, the one or more data sources may comprise a remote server. The data streams may be routed to the processing unit from a remote server by a remote vendor or operator, such as a cloud-based surveillance service provider. In yet another embodiment, the one or more data sources can be a combination of the aforementioned data sources.

[0265] At step 604, the method 600 involves analysing, by the processing unit 104, each obtained data stream to detect one or more events. Detecting the one or more events may comprise detecting at least one of: change in the location and / or the field of view of the one or more cameras 102A, 102B-102N, a change in one or more ambient parameters in the surroundings of the one or more cameras 102A, 102B-102N, an occurrence of a predetermined time-period, and at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the field of view of the one or more cameras 102A, 102B-102N, and an abnormal pattern in the fields of view of the one or more cameras 102A, 102B-102N .

[0266] At step 606, the method 600 comprises dynamically determining, by processing unit 104, a criticality level of at least one camera of one or more cameras 102A, 102B-102N, based on the detection of the one or more events in the corresponding data stream. In an embodiment, the method 600 comprises analysing, by the processing unit 104, each dynamically obtained data stream by execution of a machine learning model 320 for detecting the one or more events. The method further comprises dynamically determining the criticality level of the at least one camera, by the processing unit 104, by: determining, by execution of a semantic contextualization module 322 on the corresponding data stream, at least one label associable with the at least one camera based on the detection of the one or more events in the corresponding data stream; and tagging 202318179

[0267] 49 the at least one label to the at least one camera. The at least one label is associated with the criticality level of the at least one camera.

[0268] At step 608, the method 600 comprises dynamically managing, by the processing unit 104, at least one configuration setting associated with the at least one camera based on the determined criticality level. The at least one configuration setting is indicative of the criticality level of the at least one camera in relation to the detection of the one or more events. In an embodiment, dynamically managing the at least one configuration setting associated with the at least one camera comprises managing at least one of: an indication of the criticality level of the at least one camera by at least one indication device associated with the at least one camera; one or more imaging parameters of the at least one camera; one or more network parameters associated with the at least one camera; one or more data compression parameters associated with the one or more data streams captured by the at least one camera; one or more security parameters associated with the at least one camera; an alignment and / or an orientation of the at least one camera.

[0269] In an embodiment wherein when the one or more cameras 102A, 102B-102N comprises a first camera 102A and a second camera 102B located within a predetermined proximity to the first camera 102A, the method 600 comprises dynamically obtaining, by the processing unit 104, one or more data streams associated the first camera. The method 600 further comprises analysing, by the processing unit 104, each obtained data stream associated with the first camera 102A to detect the one or more events. The method 600 further comprises dynamically determining, by the processing unit 104, the criticality level of the first camera 102A based on the detection of the one or more events in the corresponding data stream. The method 600 further comprises dynamically managing, by the processing unit 104, at least one configuration setting associated with the at least the second camera 102B based on the determined criticality level of the first camera 102A. In a further embodiment, the method 600 may involve dynamically manage configuration setting(s) associated with the first camera 102A as well as the second camera 102B by the processing unit 104. The at least one configuration setting is indicative of the criticality level of the first camera 102A in relation to the detection of the one or more events.

[0270] Another aspect of the present disclosure relates to a computer program product. The computerprogram product comprises machine-readable instructions stored therein, which when executed by the processing unit 104, cause the processing unit 104 to perform the method 600. In an embodiment, computer program product comprises machine-readable instructions embodied in a non-transitory computer readable storage medium, such as the memory 304 (See FIG 2). The 202318179

[0271] 50 processing unit 104 of the computer 302 retrieves these computer executable instructions and executes them. When the computer executable instructions are executed by the processing unit 104, the computer executable instructions cause the processing unit 104 to perform the steps of the method 600.

[0272] The advantages of the system 100 and method 600 of the present disclosure are manifold. Advantageously, determination of the criticality levels of the cameras provides invaluable data that can be used for informed decision-making to optimize surveillance strategies. By identifying criticality trends of the cameras, an enterprise can allocate resources and attention to most important areas in the environment, ensuring that critical data is protected and monitored effectively. Moreover, by tailoring configuration settings associated with the cameras to the respective criticality levels, the system and method can ensure that critical or sensitive data is captured.

[0273] Further, in scenario where maintenance of a plurality of cameras is required in a situation where such maintenance is not detrimental to surveillance, prioritizing the maintenance of less critical cameras helps to minimize potential data loss and maintain the overall integrity of the system.

[0274] Further, the system and method of the present disclosure are capable of adjusting the configuration settings associated with the cameras in real-time to optimize their performance, resource utilization, and security based on the nature of the one or more events and the perceived criticality level. Furthermore, the system’s ability to dynamically adjust configuration settings provides greater flexibility and adaptability in response to evolving threats and changing environmental conditions.

[0275] Further, the system and the method are capable of accommodating one or more events that may be manually configured by the user and / or automatically configured by the processing unit based on abnormal, unusual, or anomalous behaviour. Advantageously, the system and the method allow for managing the at least one configuration setting in relation to the criticality level of the at least one camera based on the occurrence of the one or more events. For example, in a scenario where the at least one camera captures an event involving a trusted individual (trust being a predefined criteria set for the particular individual) loitering in a corridor area, the criticality level of the at least one camera may be set as low or normal. Correspondingly, the configuration setting(s) associated with the at least one camera may be managed accordingly, such as maintaining the same network priority and the frame rate of the at least one camera. Further, an indication device such as an LED associated with the at least one camera may be configured to 202318179

[0276] 51 emit a green light. In a scenario where the at least one camera captures an event involving a suspect individual loitering in the corridor area, quality level of the at least one camera can be set as high or severe. Correspondingly the configuration setting associated with the at least one camera may be managed accordingly, such as increasing the network priority and the frame rate of the at least one camera to capture essential surveillance data. Further, an indication device such as an LED associated with the at least one camera may be configured to emit a red light to alert the user.

[0277] The ability of the system and the method of the present disclosure to detect events and assess camera criticality enables proactive maintenance, reducing the likelihood of unexpected failures or disruptions. From a scalability perspective, the system can be adapted to various environments and surveillance needs, making it suitable for a wide range of applications.

[0278] Furthermore, the system of the present disclosure employs criticality-based configuration management that can significantly reduce the occurrence of false alarms. That is, the system can dynamically adjust camera configuration settings based on the respective criticality levels, reducing the likelihood of false alarms caused by inappropriate configurations. Further, the system can dynamically adjust configuration settings associated with cameras to ensure that resources are used efficiently and effectively. By minimising false alarms and optimising resource allocation, the system of the present disclosure also helps reduce operational costs associated with surveillance management.

[0279] The present disclosure may take the form of a computer program product comprising program modules accessible from computer-usable or computer-readable medium storing program code for use by or in connection with one or more computers, processing units, or instruction execution systems. For the purpose of this description, a computer-usable or computer-readable medium is any apparatus that may contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The medium may be electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device) or a propagation medium. Both processors and program code for implementing each aspect of the technology may be centralized or distributed (or a combination thereof).

[0280] The present disclosure may take the form of a computer-readable storage medium comprising instructions which, when executed by the processing unit, cause the processing unit to perform the method steps described hereinabove. The computer-readable storage medium may include, but not be limited to, a portable computer diskette, a hard disk, a random-access memory (RAM) 202318179

[0281] 52 device, a read-only memory (ROM) device, an erasable programmable read-only memory (EPROM or Flash memory) device, a portable compact disc read-only memory (CDROM), an optical storage device, and a magnetic storage device. While the present disclosure has been described in detail with reference to certain embodiments, it should be appreciated that the present disclosure is not limited to those embodiments. The foregoing examples have been provided merely for the purpose of explanation and are in no way to be construed as limiting of the present disclosure disclosed herein. Those skilled in the art, having the benefit of the teachings of this specification, may effect numerous modifications thereto and changes may be made without departing from the scope of the disclosure in its aspects.

[0282] 202318179

[0283] List of References

[0284] 100 System for dynamic surveillance management

[0285] 102A, 102B-102N One or more cameras

[0286] 104 Processing unit

[0287] 106 Network

[0288] 200 Exemplary environment

[0289] 202 Storeroom

[0290] 102C Camera adapted to monitor the storeroom

[0291] 1020 Initial position of the camera 102C

[0292] 204 Individual

[0293] 206 Machine shop

[0294] 102D Camera adapted to monitor the machine shop

[0295] 206A Temperature measuring device

[0296] 208 One or more machines

[0297] 210 Research laboratory

[0298] 102E Camera adapted to monitor the research laboratory

[0299] 212 Inventory room

[0300] 102F, 102G Cameras adapted to monitor the inventory room

[0301] 214 Identity card

[0302] 216 Individual

[0303] 218 Running individual

[0304] 220 Corridor

[0305] 102H Camera adapted to monitor the corridor

[0306] 222 Individual

[0307] 224 Object

[0308] 300 Exemplary architecture

[0309] 302 Computer

[0310] 304 Memory

[0311] 316 Module package

[0312] 306 Storage unit

[0313] 308 Communication interface

[0314] 310 Input-output unit

[0315] 312 Network interface

[0316] 314 Bus

[0317] 318 Communication network

[0318] 320 Machine-learning model

[0319] 322 Semantic contextualization module 202318179

[0320] 402A, 402B-402N, 404 One or more indication devices

[0321] 402A, 402B-402N One or more illumination devices

[0322] 404 Mobile phone

[0323] 500 Exemplary environment

[0324] 502 Ballroom

[0325] 502A Entrance of the ballroom

[0326] 502B Exit of the ballroom

[0327] 504A Individuals entering the ballroom

[0328] 504B Individuals leaving the ballroom

[0329] 502C Restricted area

[0330] 102A First camera

[0331] 402A Indication device associated with the first camera

[0332] 102B Second camera

[0333] 402B Indication device associated with the second camera

[0334] 506 Vehicle

[0335] 600 Method for dynamic surveillance management

[0336] 602, 604, 606, 608 Method steps

Claims

20231817955Claims1. A system (100) for dynamic surveillance management in an environment, the system (100) comprising:- one or more cameras (102A, 102B-102N), each capable of capturing data within a corresponding field of view in the environment; and- a processing unit (104) communicably coupled to the one or more cameras (102A, 102B- 102N), the processing unit (104) being configured to: dynamically obtain one or more data streams from one or more data sources; analyse each obtained data stream to detect one or more events; dynamically determine a criticality level of at least one camera of the one or more cameras (102A, 102B-102N) based on the detection of the one or more events in the corresponding data stream; and dynamically manage at least one configuration setting associated with the at least one camera based on the determined criticality level, the at least one configuration setting being indicative of the criticality level of the at least one camera in relation to the detection of the one or more events.

2. The system (100) according to claim 1 , wherein the processing unit (104) is configured to dynamically obtain the one or more data streams associated with at least one of the one or more cameras (102A, 102B-102N) and the environment, and wherein the one or more data streams comprises at least one of: data captured by the one or more cameras (102A, 102B-102N), data pertaining to location and / or field of view of the one or more cameras (102A, 102B-102N), and ambient data associated with the surroundings of the one or more cameras (102A, 102B-102N).

3. The system (100) according to claim 1 or claim 2, wherein the processing unit (104) is configured to analyse each obtained data stream to detect the one or more events comprising at least one of: a change in the location and / or the field of view of the one or more cameras (102A, 102B- 102N); a change in one or more ambient parameters in the surroundings of the one or more cameras (102A, 102B-102N);- an occurrence of a predetermined time-period;20231817956- detection of at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the fields of view of the one or more cameras (102A, 102B-102N); and detection of an abnormal pattern in the fields of view of the one or more cameras (102A, 102B-102N).

4. The system (100) according to any of the preceding claims, wherein the processing unit (104) is configured to: analyse each dynamically obtained data stream by execution of a machine learning model (320) for detecting the one or more events; and dynamically determine the criticality level of the at least one camera by: determining, by execution of a semantic contextualization module (322) on the corresponding data stream, at least one label associable with the at least one camera based on the detection of the one or more events in the corresponding data stream, and tagging the at least one label to the at least one camera, whereby the at least one label is associated with the criticality level of the at least one camera.

5. The system (100) according to any of the preceding claims, wherein the system (100) further comprises one or more indication devices (402A, 402B-402N, 404) associated with the one or more cameras (102A, 102B-102N), the one or more indication devices (402A, 402B-402N, 404) being communicatively coupled to the processing unit (104), wherein the processing unit (104) is configured to dynamically manage the at least one configuration setting associated with the at least one camera, the at least one configuration setting comprising an indication of the criticality level of the at least one camera by at least one indication device of the one or more indication devices (402A, 402B-402N), said at least one indication device being associated with the at least one camera, and wherein said at least one indication device is an illumination device configured to emit light of a predefined wavelength corresponding to the determined criticality level of the at least one camera.

6. The system (100) according to claim 5, wherein the one or more cameras (102A, 102B-102N) comprises a plurality of cameras with overlapping fields of view adapted to monitor a common region in the environment, and wherein the processing unit (104) is further configured to:20231817957 determine a relative priority of each camera based on one or more criticality levels associated with each camera determined over a predetermined period; and activate the one or more indication devices (402A, 402B-402N, 404) associated with the plurality of cameras to indicate the determined relative priority of each camera to the user, thereby providing a recommendation of order of maintenance activity for the plurality of cameras that minimizes the potential for data loss due to interruption in the corresponding data streams.

7. The system (100) according to any of the preceding claims, wherein the processing unit (104) is configured to dynamically manage the at least one configuration setting associated with the at least one camera relative to the criticality level of the at least one camera, the at least one configuration setting comprising at least one of:- one or more imaging parameters of the at least one camera;- one or more network parameters associated with the at least one camera;- one or more data compression parameters associated with the one or more data streams captured by the at least one camera;- one or more security parameters associated with the at least one camera; and- an alignment and / or an orientation of the at least one camera.

8. The system (100) according to any of the preceding claims, wherein the one or more cameras (102A, 102B-102N) comprises a first camera (102A) and a second camera (102B) located within a predetermined proximity to the first camera, and wherein the processing unit (104) is configured to: dynamically obtain one or more data streams associated with the first camera (102A); analyse each obtained data stream associated with first camera (102A) to detect the one or more events; dynamically determine the criticality level of the first camera (102A) based on the detection of the one or more events in the corresponding data stream; and dynamically manage at least one configuration setting associated with at least the second camera (102B) based on the determined criticality level of the first camera (102A), the at least one configuration setting being indicative of the criticality level of the first camera (102A) in relation to the detection of the one or more events.

9. A method (600) for dynamic surveillance management in an environment, the method (600) comprising:20231817958 dynamically obtaining, by a processing unit (104), one or more data streams from one or more data sources; analysing, by the processing unit (104), each obtained data stream to detect one or more events; dynamically determining, by the processing unit (104), a criticality level of at least one camera of one or more cameras (102A, 102B-102N), based on the detection of the one or more events in the corresponding data stream; and dynamically managing, by the processing unit (104), at least one configuration setting associated with the at least one camera based on the determined criticality level, the at least one configuration setting being indicative of the criticality level of the at least one camera in relation to the detection of the one or more events.

10. The method (600) according to claim 9, wherein the one or more data streams are associated with at least one of the one or more cameras (102A, 102B-102N) and the environment, and wherein the one or more data streams comprises at least one of: data captured by the one or more cameras (102A, 102B-102N), data pertaining to location and / or field of view of the one or more cameras (102A, 102B-102N), and ambient data associated with the surroundings of the one or more cameras (102A, 102B-102N).11 . The method (600) according to claims 9 or claim 10, wherein detecting the one or more events comprises detecting at least one of: a change in the location and / or the field of view of the one or more cameras (102A, 102B- 102N); a change in one or more ambient parameters in the surroundings of the one or more cameras (102A, 102B-102N); an occurrence of a predetermined time-period (102A, 102B-102N);- at least one of a predefined identifier, a predefined pattern, a predetermined individual, and a predetermined object in the fields of view of the one or more cameras (102A, 102B- 102N); and an abnormal pattern in the fields of view of the one or more cameras (102A, 102B-102N).

12. The method (600) according to any of claims 9-11 , comprising: analysing, by the processing unit (104), each dynamically obtained data stream by execution of a machine learning model (320) for detecting the one or more events; and dynamically determining the criticality level of the at least one camera, by the processing unit (104), by:20231817959 determining, by execution of a semantic contextualization module (322) on the corresponding data stream, at least one label associable with the at least one camera based on the detection of the one or more events in the corresponding data stream, and tagging the at least one label to the at least one camera, whereby the at least one label is associated with the criticality level of the at least one camera.

13. The method (600) according to any of claims 9-12, wherein dynamically managing the at least one configuration setting associated with the at least one camera comprises managing at least one of:- an indication of the criticality level of the at least one camera by at least one indication device (402A, 402B-402N, 404) associated with the at least one camera;- one or more imaging parameters of the at least one camera;- one or more network parameters associated with the at least one camera;- one or more data compression parameters associated with the one or more data streams captured by the at least one camera;- one or more security parameters associated with the at least one camera; and- an alignment and / or an orientation of the at least one camera.

14. The method (600) according to any of claims 9-13, wherein when the one or more cameras (102A, 102B-102N) comprises a first camera (102A) and a second camera (102B) located within a predetermined proximity to the first camera, the method comprises: dynamically obtaining, by the processing unit (104), one or more data streams associated the first camera (102A); analysing, by the processing unit (104), each obtained data stream associated with the first camera (102A) to detect the one or more events; dynamically determining, by the processing unit (104), the criticality level of the first camera (102A) based on the detection of the one or more events in the corresponding data stream; and dynamically managing, by the processing unit (104), at least one configuration setting associated with at least the second camera (102B) based on the determined criticality level of the first camera (102A), the at least one configuration setting being indicative of the criticality level of the first camera (102A) in relation to the detection of the one or more events.2023181796015. A computer-program product having machine-readable instructions stored therein, that when executed by a processing unit (104), cause the processing unit (104) to perform a method according to any of claims 9-14.

Images