1. A real-time intention-aware suspicious human behaviour detection system (100), the system (100) comprising: a capturing unit (102) configured to capture real-time videos and images of a human activity within a monitored environment; a communication network (104) configured to establish a communication link for data transmission within the system (100); a user device (106) connected to the capturing unit (102) via the communication network (104) and configured to upload real-time videos and images of human activity through a user interface (108); a processing unit (110) connected to the user device (106) via a communication unit (128), and configured to process real-time data for detection of intention-aware suspicious human behaviour, wherein the processing unit (110) further comprises: a data input module (112) configured to receive real-time data from the user device (106); a pre-processing module (114) configured to clean and normalize the received data to enhance data quality; a pose estimation module (116), configured to estimate the positions of a person’s body parts and generate a skeletal model; a feature extraction module (118) configured to extract human behaviour-related features from the skeletal model; a human behaviour detection module (120) configured to analyse the extracted human related features and detect the presence of human behaviour; a human behaviour classification module (122) configured to classify the detected human behaviour into multiple pre-defined categories; an alert generation module (124) configured to generate an alert based on the classified human behaviour; and an output module (126) configured to transmit extracted features, detected human behaviour, classified human behaviour and generated alert to the user device (106).

2. The system (100) as claimed in claim 1, wherein the system (100) further comprises a cloud data base (130) configured to store and manages processed data related to the real-time intention-aware human behaviour detection system (100) and provide real-time and historical access for monitoring and analysis.

3. The system (100) as claimed in claim 1, wherein the capturing unit (102) is a camera (132) configured to capture high-resolution videos and images of human activity within a monitored environment.

4. The system (100) as claimed in claim 1, wherein the pose estimation module (116) configured to estimate the positions of a person’s body parts including head, shoulders, elbows, hips, knees and many more.

5. The system (100) as claimed in claim 1, wherein the feature extraction module (118) configured to extract behaviour-related features including motion patterns, postures, and joint trajectories from the skeletal model.

6. The system (100) as claimed in claim 1, wherein the feature extraction module (118) utilizes convolutional neural network to extract behaviour-related features from the skeletal model.

7. The system (100) as claimed in claim 1, wherein the human behaviour detection module (120) utilizes transformer network to detect the presence of human behaviour.

8. The system (100) as claimed in claim 1, wherein the human behaviour classification module (122) configured to classify the detected human behaviour into multiple pre-defined categories including but not limit to walking, running, standing, loitering, falling, aggressive actions, and suspicious gestures using softmax.

9. The system (100) as claimed in claim 1 wherein the alert generation module (124) configured to generate an alert in the form of short service messages, notifications and emails upon detection of suspicious human behaviour.

10. A method (200) real-time intention-aware suspicious human behaviour detection system, the method (200) comprising: capturing real time videos and images of human activity within a monitored environment via a capturing unit (102); establishing a communication link for data transmission within the system (100) via a communication network (104); uploading real-time videos and images of human activity through a user interface (108) via a user device (106); processing real-time data for detection of intention-aware suspicious human behaviour via a processing unit (110), comprising several modules; receiving real-time data from the user device (106) via a data input module (112); cleaning and normalizing the received data to enhance data quality via a pre-processing module (114); estimating positions of a person’s body parts from the pre-processed data and generating a skeletal model via a pose estimation module (116); extracting human behaviour-related features from the skeletal model via a feature extraction module (118); analysing the extracted human behaviour-related features to detect the presence of human behaviour via a human behaviour detection module (120); classifying the detected human behaviour into multiple pre-defined categories via a human behaviour classification module (122); generating an alert based on the classified human behaviour via an alert generation module (124); transmitting extracted features, detected human behaviour, classified human behaviour and generated alert to the user device (106) via an output module (126); and displaying extracted features, detected human behaviour, classified human behaviour and generated alert on the user interface (108) of the user device (106).

Description:FIELD OF DISCLOSURE
[0001] The present disclosure relates to computer vision and automated visual behaviour analysis, more specifically, related to a real-time intention-aware suspicious human behaviour detection system.
BACKGROUND OF THE DISCLOSURE
[0002] Video surveillance systems monitor human activity in public and secured environments including transportation hubs, workplaces, commercial facilities, and urban infrastructure. These systems continuously capture large volumes of video data that require automated analysis to identify behaviours indicating security threats and abnormal events. The effectiveness of surveillance operations depends on the system’s ability to accurately interpret human behaviour from video streams in real time. Accurate detection of suspicious and abnormal behaviour enables timely alert generation, supports informed decision-making by human operators, and facilitates rapid response to potential incidents. Limitations in behaviour analysis lead to false alarms and missed detections, which reduce situational awareness and delay intervention. High false-alarm rates increase operator workload and reduce confidence in automated systems, while undetected suspicious activities create safety risks. These limitations directly affect the reliability and practical usability of video surveillance systems.
[0003] Conventional video based anomaly detection and human activity recognition systems are widely used in surveillance applications to monitor human behaviour. These systems typically rely on handcrafted features that analyse spatial visual appearance and temporal motion patterns extracted from video sequences. While such approaches help in detecting visible deviations in movement and activity, they primarily focus on how an action looks rather than why it occurs. Existing systems do not incorporate an understanding of human intention, body posture and contextual interactions between individuals, objects and the surrounding environment. As a result, activities that are normal but visually uncommon, such as running in crowded areas and sudden movements during emergencies are often misclassified as suspicious, leading to a high number of false alarms. The absence of interpretability reduces operator trust, limits accountability, and makes it difficult to verify and justify system decisions. This lack of transparency significantly restricts the deployment of such systems in safety critical environments, including public surveillance, transportation hubs, and secure facilities where reliable explainable, and context-aware decision-making is essential.
[0004] The present invention improves conventional video-based human activity analysis by including intention awareness to better detect abnormal behaviour. The system analyses human behaviour using visual information captured from existing video cameras, without requiring additional sensors and wearable devices. The invention operates in real time and is adaptable to diverse surveillance environments. By modelling human pose, motion patterns, and contextual scene information, the system infers behavioural intent rather than relying solely on motion irregularities. Further, the system enables transparent and accountable detection of suspicious human behaviour. The present invention thereby improves detection accuracy, reduces false alarms, and enhances trustworthiness in safety-critical surveillance applications.
SUMMARY OF THE DISCLOSURE
[0005] The following is a summary description of illustrative embodiments of the invention. It is provided as a preface to assist those professional to more rapidly assimilate the detailed design discussion which ensues and is not intended in any way to limit the scope of the claims which are appended hereto in order to particularly point out the invention.
[0006] According to illustrative embodiments, the present disclosure focuses on a real-time intention-aware suspicious human behaviour detection system which overcomes the above-mentioned disadvantages and provides the users with a useful and commercial choice.
[0007] An objective of the present disclosure is to provide a system for real-time detection of suspicious human behaviour from video data.
[0008] Another objective of the present disclosure is to provide a behavioural intent that captures both spatial and temporal relationships.
[0009] Another objective of the present disclosure is to reduce false alarms by distinguishing between normal activities and genuinely suspicious behaviour.
[0010] Another objective of the present disclosure is to enhance decision transparency through human understandable explanations of detected human behaviour.
[0011] Another objective of the present disclosure is to accurately detect suspicious human behaviour by analysing human pose and motion patterns.
[0012] Yet another objective of the present disclosure is to enable reliable operation of the system in safety critical surveillance environments by improving detection accuracy, trustworthiness, and operational efficiency.
[0013] In light of the above, in one aspect of the present disclosure, a real-time intention-aware suspicious human behaviour detection system is disclosed herein. The system comprises a capturing unit configured to capture real-time videos and images of a human activity within a monitored environment. The system includes a communication network configured to establish a communication link for data transmission within the system. The system further includes a user device connected to the capturing unit via the communication network and configured to upload real-time videos and images of human activity through the user interface. The system also includes a processing unit connected to the user device via a communication unit and configured to process real-time data for detection of intention-aware suspicious human behaviour, wherein the processing unit further comprises a data input module configured to receive real-time data from the user device, a pre-processing module configured to clean and normalize the received data to enhance data quality, a pose estimation module configured to estimate the positions of a person’s body parts and generate a skeletal model, a feature extraction module configured to extract human behaviour-related features from the skeletal model, a human behaviour detection module configured to analyse the extracted human related features and detect the presence of human behaviour, a human behaviour classification module configured to classify the detected human behaviour into multiple pre-defined categories, an alert generation module configured to generate an alert based on the classified human behaviour, an output module configured to transmit extracted features, detected human behaviour, classified human behaviour and generated alert to the user device.
[0014] In one embodiment, the system further comprises a cloud data base configured to store and manages processed data related to the real time intention aware human behaviour detection module and provides real-time and historical access for monitoring and analysis.
[0015] In one embodiment, the capturing unit is a camera configured to capture high-resolution videos and images of human activity within a monitored environment.
[0016] In one embodiment, the pose estimation module configured to estimate the positions of a person’s body parts including head, shoulders, elbows, hips, knees and many more.
[0017] In one embodiment, the feature extraction module configured to extract behaviour-related features including motion patterns, postures, and joint trajectories from the skeletal model.
[0018] In one embodiment, the feature extraction module utilizes convolutional neural network to extract behaviour-related features from the skeletal model.
[0019] In one embodiment, the human behaviour detection module utilizes transformer network to detect the presence of human behaviour.
[0020] In one embodiment, the human behaviour classification module configured to classify the detected human behaviour into multiple pre-defined categories including but not limit to walking, running, standing, loitering, falling, aggressive actions, and suspicious gestures using softmax.
[0021] In one embodiment, the alert generation module configured to generate an alert in the form of short service messages, notifications and emails upon detection of suspicious human behaviour.
[0022] In light of the above, in one aspect of the present disclosure, a method for real-time intention-aware suspicious human behaviour detection system is disclosed herein. The method comprises capturing real-time videos and images of human activity within a monitored environment via a capturing unit. The method includes establishing a communication link for data transmission within the system via a communication network. The method further includes uploading real-time videos and images of human activity through a user interface via a user device. The method also includes processing real-time data for detection of intention-aware suspicious human behaviour via a processing unit, comprising several modules. The method further includes receiving real-time data from the user device via a data input module. Furthermore, the method includes cleaning and normalizing the received data to enhance data quality via a pre-processing module. Moreover, the method includes estimating positions of a person’s body parts and generating a skeletal model corresponding to human activity via a pose estimation module. The method further includes extracting human behaviour-related features from the skeletal model via a feature extraction module. The method also includes analysing the extracted human behaviour-related features and detecting the presence of human behaviour via a human behaviour detection module. The method also includes classifying the detected human behaviour into predefined categories via a human behaviour classification module. Furthermore, the method includes generating an alert based on the classified human behaviour via an alert generation module. Moreover, the method includes transmitting extracted features, detected human behaviour, classified human behaviour and generated alert to the user device via an output module. At last, the method includes displaying extracted features, detected human behaviour, classified human behaviour and generated alert on the user interface of the user device.
[0023] These and other advantages will be apparent from the present application of the embodiments described herein.
[0024] The preceding is a simplified summary to provide an understanding of some embodiments of the present invention. This summary is neither an extensive nor exhaustive overview of the present invention and its various embodiments. The summary presents selected concepts of the embodiments of the present invention in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the present invention are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below.
[0025] These elements, together with the other aspects of the present disclosure and various features are pointed out with particularity in the claims annexed hereto and form a part of the present disclosure. For a better understanding of the present disclosure, its operating advantages, and the specified object attained by its uses, reference should be made to the accompanying drawings and descriptive matter in which there are illustrated exemplary embodiments of the present disclosure.
[0026] These and other advantages will be apparent from the present application of the embodiments described herein.
[0027] The preceding is a simplified summary to provide an understanding of some embodiments of the present invention. This summary is neither an extensive nor exhaustive overview of the present invention and its various embodiments. The summary presents selected concepts of the embodiments of the present invention in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the present invention are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below.
[0028] These elements, together with the other aspects of the present disclosure and various features are pointed out with particularity in the claims annexed hereto and form a part of the present disclosure. For a better understanding of the present disclosure, its operating advantages, and the specified object attained by its uses, reference should be made to the accompanying drawings and descriptive matter in which there are illustrated exemplary embodiments of the present disclosure.
BRIEF DESCRIPTION OF THE DRAWINGS
[0029] To describe the technical solutions in the embodiments of the present disclosure or in the prior art more clearly, the following briefly describes the accompanying drawings required for describing the embodiments or the prior art. Apparently, the accompanying drawings in the following description merely show some embodiments of the present disclosure, and a person of ordinary skill in the art can derive other implementations from these accompanying drawings without creative efforts. All of the embodiments or the implementations shall fall within the protection scope of the present disclosure.
[0030] The advantages and features of the present disclosure will become better understood with reference to the following detailed description taken in conjunction with the accompanying drawing, in which:
[0031] FIG. 1 illustrates a block diagram of a real time intention aware suspicious human behaviour detection system, in accordance with an embodiment of the present disclosure; and
[0032] FIG. 2 illustrates a flowchart of a method, outlining the sequential steps for the real-time intention-aware suspicious human behaviour detection system, in accordance with an embodiment of the present disclosure.
[0033] Like reference, numerals refer to like parts throughout the description of several views of the drawing.
[0034] The real-time intention-aware suspicious human behaviour detection system is illustrated in the accompanying drawings, which like reference letters indicate corresponding parts in the various figures. It should be noted that the accompanying figure is intended to present illustrations of exemplary embodiments of the present disclosure. This figure is not intended to limit the scope of the present disclosure. It should also be noted that the accompanying figure is not necessarily drawn to scale.
DETAILED DESCRIPTION OF THE DISCLOSURE
[0035] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments given in such detail as to communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the present disclosure.
[0036] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It may be apparent to one skilled in the art that embodiments of the present disclosure may be practiced without some of these specific details.
[0037] Various terms as used herein are shown below. To the extent a term is used, it should be given the broadest definition persons in the pertinent art have given that term as reflected in printed publications and issued patents at the time of filing.
[0038] The terms “a” and “an” herein do not denote a limitation of quantity but rather denote the presence of at least one of the referenced items.
[0039] The terms “having”, “comprising”, “including”, and variations thereof signify the presence of a component.
[0040] Referring now to FIG. 1 and FIG. 2 to describe various exemplary embodiments of the present disclosure. FIG. 1 illustrates a block diagram of a real-time intention-aware suspicious human behaviour detection system 100, in accordance with an embodiment of the present disclosure.
[0041] The system 100 may include a capturing unit 102. The system 100 may include a communication network 104. The system 100 may include a user device 106. The system 100 may include a processing unit 110 which further comprises a data input module 112, a pre-processing module 114, a pose estimation module 116, a feature extraction module 118, a human behaviour detection module 120, a human behaviour classification module 122, an alert generation module 124, and an output module 126. The system 100 may include a cloud database 130.
[0042] The capturing unit 102 configured to capture real-time videos and images of a human activity within a monitored environment. The video streams and image frames representing human presence, posture, motion, and activity occurring within the monitored environment.
[0043] In one embodiment of the present invention, the capturing unit 102 is a camera 132 configured to capture high-resolution videos and images of human activity within a monitored environment.
[0044] In one embodiment of the present invention, the capturing unit 102 enables accurate identification of posture, movement dynamics, and interaction patterns necessary for downstream behavioural analysis and classification processes.
[0045] In one embodiment of the present invention, the capturing unit 102 operates under diverse environmental conditions including low light, and variable illumination. The capturing unit 102 may be deployed in a combination to provide multi-angle coverage and improved robustness of human activity monitoring.
[0046] The communication network 104 configured to establish a communication link for data transmission within the system 100. The communication network 104 may include but not limited to wired and wireless networks.
[0047] In one embodiment of the present invention, the communication network 104 includes wireless networks, including but not limited to Bluetooth and wireless fidelity.
[0048] In one embodiment of the present invention, the system 100 further comprises the cloud data base 130 configured to store and manages processed data related to the real-time intention-aware human behaviour detection system 100 and provide real-time and historical access for monitoring and analysis.
[0049] The user device 106 connected to the capturing unit 102 via the communication network 104 and configured to upload real-time videos and images of human activity through a user interface 108.
[0050] In embodiment of the present invention, the user device 106 includes but not limited to a smartphone, laptop, computer and any other wearable device capable of data processing and user interaction. The user device 106 receives system outputs, display detected human behaviour and alerts, and allow a user to monitor and control the operation of the system 100.
[0051] In one embodiment of the present invention, the user device 106 receives outputs from the system 100 including detected human behaviour results, behaviour classification information and alert notifications related to suspicious activities.
[0052] The processing unit 110 connected to the user device 106 via a communication unit 128, and configured to process real-time data for detection of intention-aware suspicious human behaviour.
[0053] In one embodiment of the present invention, the processing unit 110 includes but not limited to a microcontroller, a microprocessor, a central processing unit and many more.
[0054] In one embodiment of the present invention, the processing unit 110 is the central brain of the system 100, receiving, coordinating, analysing, interpreting, and transmitting data, controlling all modules, and enabling real-time intention-aware detection of suspicious behaviour.
[0055] The data input module 112 configured to receive real-time data from the user device 106. The Data input module 112 serves as the entry point of the system 100, receiving real-time data from the user device 106 and acting as the initial interface for all incoming information before it is forwarded to subsequent modules for processing.
[0056] In one embodiment of the present invention, the data input module 112 further configured to organize, buffer, and forward the received data to the pre-processing module 114 for cleaning and normalization, ensuring that the data is in a suitable format and quality for subsequent analysis by the system 100.
[0057] In one embodiment of the present invention, the data input module 112 organizes the received visual information by arranging the incoming video frames and images into a structured format.
[0058] The pre-processing module 114 configured to clean and normalize the received data to enhance data quality. The pre-processing module 114 performs tasks such as noise reduction, resolution adjustment, frame alignment, and data formatting to ensure consistency and reliability.
[0059] In one embodiment of the present invention, the pre-processing module 114 reduces noise, adjusts brightness and contrast, resizes images to a predefined format, and standardizes the visual information to ensure consistent input for subsequent modules.
[0060] The pose estimation module 116, configured to estimate the positions of a person’s body parts and generate a skeletal model. The pose estimation module 116 thereby ensures precise, real-time monitoring of human activity, facilitating intention-aware behaviour analysis and supporting timely alert generation in the monitored environment.
[0061] In one embodiment of the present invention, the pose estimation module 116 configured to estimate the positions of a person’s body parts including head, shoulders, elbows, hips, knees and many more.
[0062] In one embodiment of the present invention, the post estimation module 116 analyses the processed visual information to detect human joint locations and determine spatial coordinates of multiple body landmarks across successive frames. The post estimation module 116 further enables precise tracking of human movements and provides reliable input data for higher-level behavioural interpretation and suspicious activity detection within the monitored environment.
[0063] The feature extraction module 118 configured to extract human behaviour-related features from the skeletal model. The feature extraction module 118 serves as a critical component for subsequent behaviour analysis.
[0064] In one embodiment of the present invention, the feature extraction module 118 configured to extract behaviour-related features including motion patterns, postures, and joint trajectories from the skeletal model.
[0065] In one embodiment of the present invention, the feature extraction module 118 utilizes convolutional neural network to extract behaviour-related features from the skeletal model.
[0066] In one embodiment of the present invention, the extracted features are then provided to the human behaviour detection module 120 and subsequent modules for real-time classification, prediction, and alert generation, supporting accurate, intention-aware detection of suspicious human behaviour.
[0067] The human behaviour detection module 120 configured to analyse the extracted human related features and detect the presence of human behaviour.
[0068] In one embodiment of the present invention, the human behaviour detection module 120 utilizes transformer network to detect the presence of human behaviour.
[0069] In one embodiment of the present invention, the human behaviour detection module 120 evaluates the continuity, consistency, and progression of body movements over time to identify and confirm the occurrence of various behavioural events. The human behaviour detection module 120 ensures that sufficient movement data and posture consistency are available before forwarding the detected behaviour to the human behaviour classification module 122.
[0070] The human behaviour classification module 122 configured to classify the detected human behaviour into multiple pre-defined categories.
[0071] In one embodiment of the present invention, the human behaviour classification module 122 configured to classify the detected human behaviour into multiple pre-defined categories including but not limit to walking, running, standing, loitering, falling, aggressive actions, and suspicious gestures using softmax.
[0072] In one embodiment of the present invention, the classified behaviour is then forwarded to the alert generation module 124 and output module 126 for real-time notification and display on the user device 106.
[0073] The alert generation module 124 configured to generate an alert based on the classified human behaviour.
[0074] In one embodiment of the present invention, the alert generation module 124 configured to generate an alert in the form of short service messages, notifications and emails upon detection of suspicious human behaviour.
[0075] In one embodiment of the present invention, the alert generation module 124 generates an alert when the classified human behaviour corresponds to a predefined suspicious activity, including aggressive actions, unauthorized entry into restricted areas, abnormal movement, falling incidents and loitering.
[0076] The output module 126 configured to transmit extracted features, detected human behaviour, classified human behaviour and generated alert to the user device 106.
[0077] In one embodiment of the present invention, the output module 126 operates to deliver the final processed results of the system 100 to the user device 106. The output module 126 enables the user to monitor human activity effectively, take appropriate actions, and respond promptly to potential security threats within the monitored environment.
[0078] FIG. 2 illustrates a flowchart of a method 200, outlining the sequential steps for real-time intention-aware suspicious human behaviour detection system 100, in accordance with an embodiment of the present disclosure.
[0079] At step 202, the real time videos and images of human activity is captured within a monitored environment via a capturing unit 102.
[0080] At step 204, the communication link for data transmission is established within the system 100 via a communication network 104.
[0081] At step 206, the real-time videos and images of human activity is uploaded through a user interface 108 via a user device 106.
[0082] At step 208, the real-time data for detection of intention-aware suspicious human behaviour is processed via a processing unit 110 comprising several modules.
[0083] At step 210, the real-time data is received from the user device 106 via a data input module 112.
[0084] At step 212, the received data is cleaned and normalized to enhance data quality via a pre-processing module 114.
[0085] At step 214, the positions of a person’s body parts are estimated from the pre-processed data and generate a skeletal model via a pose estimation module 116.
[0086] At step 216, the human behaviour-related features are extracted from the skeletal model via a feature extraction module 118.
[0087] At step 218, the extracted human behaviour-related features are analysed and detect the presence of human behaviour via a human behaviour detection module 120.
[0088] At step 220, the detected human behaviour is classified into multiple pre-defined categories via a human behaviour classification module 122.
[0089] At step 222, the alert is generated based on the classified human behaviour via an alert generation module 124.
[0090] At step 224, the extracted features, detected human behaviour, classified human behaviour and generated alert is transmitted to the user device 106 via an output module 126.
[0091] At step 226, the extracted features, detected human behaviour, classified human behaviour and generated alert is displayed to the user interface 108 of the user device 106.
[0092] In the best mode of operation of the present invention, the real-time intention-aware suspicious human behavior detection system 100 operates in a seamless, end-to-end manner to monitor, detect, classify, and alert on human activities in a monitored environment. The workflow begins with the capturing unit 102, typically a camera 132, which continuously captures real-time videos and images of human activity within the environment. These captured data streams are transmitted through the communication network 104 to the user device 106, where the user interface 108 allows for uploading, monitoring, and interacting with the data. Once received, the data is processed by the processing unit 110. The first step involves the data input module 112, which receives the uploaded data from the user device 106. Next, the pre-processing module 114 cleans and normalizes the data, ensuring high quality by removing noise, standardizing formats, and enhancing visual clarity. This ensures that subsequent modules operate efficiently and accurately. The pose estimation module 116 then analyzes the pre-processed images and video frames to identify and estimate the positions of key body parts such as the head, shoulders, elbows, hips, and knees, generating a skeletal model of the human subject. These skeletal models, along with raw data, are fed into the feature extraction module 118, which employs a convolutional neural network to extract behavior-related features including motion patterns, postures, and joint trajectories. This feature-rich representation provides a detailed understanding of human activity. The human behavior detection module 120 uses a transformer network to analyze the extracted features and detect the presence of human activity. Once detected, the human behavior classification module 122 classifies the activity into multiple pre-defined categories, such as walking, running, standing, loitering, falling, aggressive actions, and suspicious gestures using softmax-based classification. Based on the classified behavior, the alert generation module 124 generates real-time alerts if suspicious activity is detected. These alerts are in the form of messages, notifications, and emails, ensuring immediate notification to security personnel. The output module 126 transmits extracted features, detected behaviors, classified actions, and generated alerts back to the user device 106. The user interface 108 of the user device 106 provides real-time visualization of the transmitted information, facilitating continuous monitoring and prompt intervention. In addition, the system 100 employs the cloud database 130 to store processed data, thereby enabling historical analysis, behavioral trend detection, and long-term monitoring, which collectively facilitate continuous system improvement and support data-driven insights.
[0093] While the invention has been described in connection with what is presently considered to be the most practical and various embodiments, it will be understood that the invention is not to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims.
[0094] A person of ordinary skill in the art may be aware that, in combination with the examples described in the embodiments disclosed in this specification, units and algorithm steps may be implemented by electronic hardware, computer software, or a combination thereof.
[0095] The foregoing descriptions of specific embodiments of the present disclosure have been presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the present disclosure to the precise forms disclosed, and many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described to best explain the principles of the present disclosure and its practical application, and to thereby enable others skilled in the art to best utilize the present disclosure and various embodiments with various modifications as are suited to the particular use contemplated. It is understood that various omissions and substitutions of equivalents are contemplated as circumstances may suggest or render expedient, but such omissions and substitutions are intended to cover the application or implementation without departing from the scope of the present disclosure.
[0096] Disjunctive language such as the phrase “at least one of X, Y, Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and/or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each is present.
[0097] In a case that no conflict occurs, the embodiments in the present disclosure and the features in the embodiments may be mutually combined. The foregoing descriptions are merely specific implementations of the present disclosure but are not intended to limit the protection scope of the present disclosure. Any variation, replacement readily figured out by a person skilled in the art within the technical scope disclosed in the present disclosure shall fall within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
, C , Claims:I/We Claim:
1. A real-time intention-aware suspicious human behaviour detection system (100), the system (100) comprising:
a capturing unit (102) configured to capture real-time videos and images of a human activity within a monitored environment;
a communication network (104) configured to establish a communication link for data transmission within the system (100);
a user device (106) connected to the capturing unit (102) via the communication network (104) and configured to upload real-time videos and images of human activity through a user interface (108);
a processing unit (110) connected to the user device (106) via a communication unit (128), and configured to process real-time data for detection of intention-aware suspicious human behaviour, wherein the processing unit (110) further comprises:
a data input module (112) configured to receive real-time data from the user device (106);
a pre-processing module (114) configured to clean and normalize the received data to enhance data quality;
a pose estimation module (116), configured to estimate the positions of a person’s body parts and generate a skeletal model;
a feature extraction module (118) configured to extract human behaviour-related features from the skeletal model;
a human behaviour detection module (120) configured to analyse the extracted human related features and detect the presence of human behaviour;
a human behaviour classification module (122) configured to classify the detected human behaviour into multiple pre-defined categories;
an alert generation module (124) configured to generate an alert based on the classified human behaviour; and
an output module (126) configured to transmit extracted features, detected human behaviour, classified human behaviour and generated alert to the user device (106).
2. The system (100) as claimed in claim 1, wherein the system (100) further comprises a cloud data base (130) configured to store and manages processed data related to the real-time intention-aware human behaviour detection system (100) and provide real-time and historical access for monitoring and analysis.
3. The system (100) as claimed in claim 1, wherein the capturing unit (102) is a camera (132) configured to capture high-resolution videos and images of human activity within a monitored environment.
4. The system (100) as claimed in claim 1, wherein the pose estimation module (116) configured to estimate the positions of a person’s body parts including head, shoulders, elbows, hips, knees and many more.
5. The system (100) as claimed in claim 1, wherein the feature extraction module (118) configured to extract behaviour-related features including motion patterns, postures, and joint trajectories from the skeletal model.
6. The system (100) as claimed in claim 1, wherein the feature extraction module (118) utilizes convolutional neural network to extract behaviour-related features from the skeletal model.
7. The system (100) as claimed in claim 1, wherein the human behaviour detection module (120) utilizes transformer network to detect the presence of human behaviour.
8. The system (100) as claimed in claim 1, wherein the human behaviour classification module (122) configured to classify the detected human behaviour into multiple pre-defined categories including but not limit to walking, running, standing, loitering, falling, aggressive actions, and suspicious gestures using softmax.
9. The system (100) as claimed in claim 1 wherein the alert generation module (124) configured to generate an alert in the form of short service messages, notifications and emails upon detection of suspicious human behaviour.
10. A method (200) real-time intention-aware suspicious human behaviour detection system, the method (200) comprising:
capturing real time videos and images of human activity within a monitored environment via a capturing unit (102);
establishing a communication link for data transmission within the system (100) via a communication network (104);
uploading real-time videos and images of human activity through a user interface (108) via a user device (106);
processing real-time data for detection of intention-aware suspicious human behaviour via a processing unit (110), comprising several modules;
receiving real-time data from the user device (106) via a data input module (112);
cleaning and normalizing the received data to enhance data quality via a pre-processing module (114);
estimating positions of a person’s body parts from the pre-processed data and generating a skeletal model via a pose estimation module (116);
extracting human behaviour-related features from the skeletal model via a feature extraction module (118);
analysing the extracted human behaviour-related features to detect the presence of human behaviour via a human behaviour detection module (120);
classifying the detected human behaviour into multiple pre-defined categories via a human behaviour classification module (122);
generating an alert based on the classified human behaviour via an alert generation module (124);
transmitting extracted features, detected human behaviour, classified human behaviour and generated alert to the user device (106) via an output module (126); and
displaying extracted features, detected human behaviour, classified human behaviour and generated alert on the user interface (108) of the user device (106).