DESC:FIELD OF THE INVENTION

The present disclosure relates generally to an AI application; an application based on image processing. In particular, the present disclosure relates to a method for removing duplicate images from Automatic Number Plate Recognition (ANPR) database for vehicle movement monitoring.

BACKGROUND

Intelligent Transport Management Systems (ITMS) which is part of Smart City Applications and Secure Vehicle Movement Monitoring Employs an Automatic Number Plate Recognition System (ANPR) for detection and identification of number plates. An IP based ANPR camera is installed in the system, streaming the data continuously to the ANPR system. The ANPR system detects number plates and identifies the characters. The acquired results are stored in the database for further analysis or processing. There are cases where duplicate entry or multiple entries of the same vehicle is stored in the database. In the case of the Secure Vehicle Movement Monitoring system deployed in defence bases, the entry and exit of the vehicles are monitored meticulously, to retain the controlled movement of the vehicle. In case of multiple entries of the same vehicle during the entry point, and a single entry during the exit point, the system creates a false alarm. As per the system, the vehicle never exited the premises, whereas the system exited earlier. Multiple entries of the same data not only result in redundant data storage but also the raising of false alarm generation in the case of secure vehicle movement monitoring. Due to the stern nature of the problem, a system or method is required to address this challenge. It would be more concerning if the false alarm is raised due to the multiple entries and makes the system unreliable.
DE202023101575U1 relates to the field of intelligent detection technologies for security purposes, and more particularly to an intelligent system using ANPR for campus security. Every educational institution attaches great importance to the safety of students on school premises. Schools have a duty of care to the young people they teach. The safety of students on school premises is vital. However, the number of crimes committed in schools is shocking. Because of these challenges faced by traditional systems, ANPR technology uses cameras to capture images of the license plates of vehicles entering and exiting campus. However, it is important to consider any potential privacy threats and to ensure that the system is being used in an ethical and responsible manner. In order to avoid false alarms and system failures, it is also necessary to ensure that the system is properly maintained and monitored.
US8339282B2 relates to a vehicle detection system for detecting the presence of at least part of vehicle image data, the system comprising: an interface configured to receive image data; an identifier module configured to identify a plurality of linear regions in an image represented by the image data; a comparator configured to compare at least one of the number, cumulative size, and density of the linear regions with a respective threshold value; and an output configured to issue a signal indicating the detection of a vehicle based on the results of the comparison.
KR20210154516A relates to a vehicle number plate recognition program using deep learning. More specifically, since the vehicle number plate recognition program can become foundation technology for smart tolling, the program is made to be versatility used, thereby solving an existing issue about difficulty in recognizing a new number plate or a number plate of an electric vehicle. In accordance with the present invention, the vehicle number plate recognition program comprises: a vehicle recognition deep learning model formed with YOLO v4, capable of recognizing a vehicle and its number plate in real time at the same time; and an OCR formed with a support vector machine to determine Korean alphabets and numbers of every number plate. Therefore, since the vehicle number plate recognition program can become foundation technology for smart tolling, the program is made to be versatility used, thereby bringing about an effect of solving an existing issue about difficulty in recognizing a new number plate or a number plate of an electric vehicle.
US11105649B2 relates to an approach for providing navigational assistance to a target location in a vehicle parking facility, a processor accesses stored vehicle registration numbers each mapped to a location of a vehicle in a parking facility in a database. A processor associates a user with a target vehicle registration number. A processor receives input of a reference vehicle registration number from a mobile user device at a current location of the user using an input component of the mobile user device and references the database of vehicle registration numbers. A processor identifies a current location in the parking facility by look up of the reference registration number; identifies a user's target location in the parking facility by look up of the target registration number associated with the user; and generates dynamic navigational instructions from the current location to the user's target location for sending to the mobile user device.
US11074434B2 relates to methods, systems, and computer programs are presented for detecting near-duplicate profile images of the users in a social network. One method includes operations for identifying an image in a profile of a user of the social network, determining a query feature vector for the image, the query feature vector comprising a set of features, and determining a dominant feature from the features, the dominant feature having a highest value from the values of the features. Further, the method includes operations for determining a bucket in a database of feature vectors based on the dominant feature, determining if the query feature vector is a near duplicate of any feature vector in the determined bucket, and determining if the profile of the user is a duplicate profile or a fake profile based on whether the query feature vector is a near duplicate of any feature vector in the determined bucket.
DE202023101575U1) relates to the field of intelligent detection technologies for security purposes, and more particularly to an intelligent system using ANPR for campus security. It discloses a facial recognition device for controlling school gates on campus. This system allows full automation and full transparency. It can monitor the vehicle access and need not be constantly monitored. The system can be implemented with ANPR bases security system to increase security on site. The system can be used to monitor compliance with rules for limited areas. The system allows security personnel to receive instant alerts on vehicles that may enter the premises and are flagged as stolen or locked. The warning unit which is part of this system triggers an alarm when a vehicle or person crosses the predetermined area on campus.
US8339282B2 depicts a vehicle detection system for detecting the presence of at least part of a vehicle in image data, the system comprising: an interface configured to receive image data; an identifier module configured to identify a plurality of linear regions in a represented by the image data. A comparator configured to compare at least on the number cumulative size and density of the linear regions with a respective threshold value; and an output configured to issue a signal indicating the detection of a vehicle based on the results of the comparison.
KR20210154516A comprises of a vehicle number plate recognition program using deep learning. The vehicle recognition deep learning model formed with YOLO v4 and OCR formed with Support vector machine to discriminate numbers of all license plates. This system promised to solve the demerits of the conventional license plate recognition program and had the ability to recognize the electric vehicle license plate.
US11105649B2 explores the method of providing navigational assistance to a target location in a vehicle parking facility. A processor receives input of a reference vehicle registration number at a current location of the user using an input component of the mobile user device. Processor connects to a remote server consists of a database of vehicle registration numbers mapped to locations in the parking facility. It generates dynamic navigational instructions based on the campus orientation and a current mode of transport of the user from the current location to the target location of the user for sending to the mobile user device.
US11074434B2 relates to the field of detecting near duplicate images in profiles. This method involves determination using neural network, a query feature vector for the image, the query feature vector comprising a value for each of a plurality of features. Further it determines the dominant feature from the plurality of features and determines in a bucket in a database of feature vectors based on the dominant feature. It determines whether the query feature vector is a near duplicate of any feature vector in the determined bucket.
Therefore, there is felt for a need for an invention which can remove duplicate number plates from the database on an automated secure vehicle movement monitoring system.
BRIEF DESCRIPTION OF THE DRAWINGS

These and other features, aspects, and advantages of the present invention will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:
Figure 1 illustrates an environment of a system for secure vehicle movement monitoring using an Automatic Number Plate Recognition (ANPR), according to an embodiment of the present disclosure;
Figure 2a illustrates a block diagram of the system, according to an embodiment of the present disclosure;
Figure 2b illustrates an overview flowchart of Automated Secure vehicle movement monitoring architecture, in accordance with an exemplary implementation of the present disclosure; and
Figure 3 illustrates a process flow for a method for secure vehicle movement monitoring using an Automatic Number Plate Recognition (ANPR) through the system, according to an embodiment of the present disclosure.
Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present invention. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present invention so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein.
DETAILED DESCRIPTION OF FIGURES

For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skilled in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.
For example, the term “some” as used herein may be understood as “none” or “one” or “more than one” or “all.” Therefore, the terms “none,” “one,” “more than one,” “more than one, but not all” or “all” would fall under the definition of “some.” It should be appreciated by a person skilled in the art that the terminology and structure employed herein is for describing, teaching, and illuminating some embodiments and their specific features and elements and therefore, should not be construed to limit, restrict, or reduce the spirit and scope of the present disclosure in any way.
For example, any terms used herein such as, “includes,” “comprises,” “has,” “consists,” and similar grammatical variants do not specify an exact limitation or restriction, and certainly do not exclude the possible addition of one or more features or elements, unless otherwise stated. Further, such terms must not be taken to exclude the possible removal of one or more of the listed features and elements, unless otherwise stated, for example, by using the limiting language including, but not limited to, “must comprise” or “needs to include.”
Whether or not a certain feature or element was limited to being used only once, it may still be referred to as “one or more features” or “one or more elements” or “at least one feature” or “at least one element.” Furthermore, the use of the terms “one or more” or “at least one” feature or element do not preclude there being none of that feature or element, unless otherwise specified by limiting language including, but not limited to, “there needs to be one or more...” or “one or more element is required.”
Unless otherwise defined, all terms and especially any technical and/or scientific terms, used herein may be taken to have the same meaning as commonly understood by a person ordinarily skilled in the art.
Reference is made herein to some “embodiments.” It should be understood that an embodiment is an example of a possible implementation of any features and/or elements of the present disclosure. Some embodiments have been described for the purpose of explaining one or more of the potential ways in which the specific features and/or elements of the proposed disclosure fulfil the requirements of uniqueness, utility, and non-obviousness.
Use of the phrases and/or terms including, but not limited to, “a first embodiment,” “a further embodiment,” “an alternate embodiment,” “one embodiment,” “an embodiment,” “multiple embodiments,” “some embodiments,” “other embodiments,” “further embodiment”, “furthermore embodiment”, “additional embodiment” or other variants thereof do not necessarily refer to the same embodiments. Unless otherwise specified, one or more particular features and/or elements described in connection with one or more embodiments may be found in one embodiment, or may be found in more than one embodiment, or may be found in all embodiments, or may be found in no embodiments. Although one or more features and/or elements may be described herein in the context of only a single embodiment, or in the context of more than one embodiment, or in the context of all embodiments, the features and/or elements may instead be provided separately or in any appropriate combination or not at all. Conversely, any features and/or elements described in the context of separate embodiments may alternatively be realized as existing together in the context of a single embodiment.
Any particular and all details set forth herein are used in the context of some embodiments and therefore should not necessarily be taken as limiting factors to the proposed disclosure.
Embodiments of the present disclosure will be described below in detail with reference to the accompanying drawings.
Figure 1 illustrates an environment of a system for secure vehicle movement monitoring using an Automatic Number Plate Recognition (ANPR), according to an embodiment of the present disclosure. Figure 2a illustrates a block diagram of the system, according to an embodiment of the present disclosure.
In an embodiment, the environment 100 provides the system 104 (alternatively referred to as the ANPR system 104 for secure vehicle movement monitoring using the Automatic Number Plate Recognition (ANPR). The ANPR system 104 is implemented within an image-capturing device 102 strategically placed at an entry gate 110a and an exit gate 110b of a secured premise. Alternatively the ANPR system 104 may also reside in a cloud server 106 or any other processing machine placed on the entry gate 110a and the exit gate 110b.
In an embodiment, the ANPR system 104 is responsible for capturing and processing data of the vehicles 108 to ensure accurate tracking of movements within the premises. The ANPR system 104 is further connected to the remote server 106 that performs data processing, duplicate entry detection, and long-term data storage to enhance security and prevent unauthorized access.
In an embodiment, the ANPR system 104 is configured to capture, via the image capturing device 102, a video stream of a vehicle 108 entering the secured premise and transmitting a video stream using a RTSP protocol. The captured data (i.e., the video stream) is received by a data receipt module within the ANPR system 104, which pushes the received data into a data queue for further processing.
In an embodiment, a data pre-processing module extracts a sub-frame containing an area of interest, specifically focusing on the license plate, while discarding the remaining portion of a video frame in the video stream. This extracted sub-frame is then sent to a license plate detection module, which utilizes an SSD Inception V2 model to detect the presence of a license plate on the vehicle 108. If a license plate is detected, the license plate detection module extracts its coordinates, which are subsequently used to generate a license plate image. The generated image is processed by a character recognition module employing a convolutional neural network (CNN)-based OCR model to recognize and extract the characters on the license plate.
In an embodiment, once the license plate image, extracted characters, timestamp, and gate ID are obtained, they are transmitted to a server module (alternatively residing in the remote server 106) via a data transfer module for further analysis and storage. The server module receives the transmitted data (e.g., the license plate image, extracted characters, timestamp, and gate ID) and performs duplicate entry detection to determine whether the license plate has been previously recorded in a database. This is achieved using a data similarity check module that applies an image hashing algorithm to compute a similarity score between the received license plate image and a set of stored license plate images. The similarity score is then compared against a predefined similarity score threshold. If the similarity score meets or exceeds the predefined similarity score threshold, indicating that the vehicle’s 108 license plate has already been recorded, the existing record in the database is updated with the latest entry. If the similarity score falls below the predefined similarity score threshold, a new entry is created in the database, advantageously ensuring that duplicate entries are prevented.
In an embodiment, the ANPR system 104 is further configured to perform a continuous monitoring process via an alarm generation module to track vehicle presence within the secured premise. The alarm generation module retrieves vehicle data where the exit status is marked as “no” indicating that the vehicle has not yet left the premises. The alarm generation module is configured to compute a vehicle premise time based on determining the difference between the current time and the stored entry timestamp of the vehicle 108. The computed vehicle premise time is then compared against the permitted stay duration. If the vehicle premise time exceeds the permitted stay duration, the alarm generation module generates an alarm notification containing details such as the gate ID, vehicle ID, and permitted time. The alarm notification is then transmitted to the ANPR system 104 or a client ANPR system to notify security personnel or trigger necessary actions.
In an embodiment, the ANPR system 104 is further configured to facilitate exit monitoring, including a process for checking the exit status of vehicles. When the vehicle 108 arrives at the exit gate 110b, the image capturing device 102 captures the video stream of the vehicle 108. The ANPR system 104 processes the captured video to extract the license plate information and retrieves the most recent stored entry corresponding to the detected license plate number.
The exit image is then compared against the stored entry using the image hashing algorithm, and the similarity score is computed. If the similarity score meets or exceeds the predefined similarity score threshold, the ANPR system 104 updates the exit status, indicating that the vehicle has lawfully exited the secured premise. Once the exit status is updated, the corresponding vehicle record is removed from active monitoring, advantageously ensuring that only unauthorized or overstayed vehicles trigger alerts.
In an embodiment, the ANPR system 104 is further configured to enhance security and prevent unauthorized modifications, based on including encryption and secure data transmission mechanisms. In the embodiment, before transmitting the license plate image, extracted characters, timestamp, and gate ID to the remote server 106 or the server module, the data is encrypted to ensure confidentiality. Upon receiving the encrypted data, the server module decrypts it before processing and storing it in a secure repository. Additionally, an integrity verification process is performed to detect unauthorized data modifications. If any tampering is identified, the alert notification is generated and transmitted to the relevant authorities, advantageously ensuring the integrity and reliability of the ANPR system 104.
In an advantageous aspect, the ANPR system 104 provides a robust and automated technique for secure vehicle movement monitoring by integrating the ANPR system 104 with the remote server 106 to enable real-time data capture, processing, and storage. Thus, based on leveraging advanced image processing techniques, including SSD Inception V2-based license plate detection, CNN-based character recognition, and image hashing-based similarity detection, the ANPR system 104 ensures accurate and efficient tracking of vehicles entering and exiting secured premises. The continuous monitoring via the alarm generation module further enhances security by detecting and flagging vehicles that exceed their permitted stay duration. Additionally, the incorporation of encryption and integrity verification mechanisms ensures secure data transmission and prevents unauthorized access or manipulation. Advantageously, the ANPR system 104 offers a highly reliable and automated solution for vehicle access control in secured environments, reducing the need for manual monitoring and enhancing overall security efficiency.
The ANPR system 104 may be connected with the image capturing device 102 or the remote server 106 or the database using a wireless network. The wireless network as appeared throughout the present disclosure may be a zig-bee network, a cellular telephone network such as 4G, 5G, an 802.11, 802.16, 802.20, 802.1Q, Wi-Fi, or a WiMax network. Further, the network may be a public network, such as the Internet, a private network, such as an intranet, or combinations thereof, and may utilize a variety of networking protocols now available or later developed including, but not limited to TCP/IP based networking protocols.
Referring to Figure 2a the ANPR system 104 may include, but not limited to, a processor 202, memory 204, modules 206, and data 208. The modules 206 and the memory 204 may be coupled to the processor 202.
The processor 202 can be a single processing unit or several units, all of which could include multiple computing units. The processor 202 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor 302 is adapted to fetch and execute computer-readable instructions and data stored in the memory 304.
The memory 204 may include any non-transitory computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read-only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.
The modules 206, amongst other things, include routines, programs, objects, components, data structures, etc., which perform particular tasks or implement data types. The modules 206 may also be implemented as, signal processor(s), state machine(s), logic circuitries, and/or any other device or component that manipulate signals based on operational instructions.
Further, the modules 206 can be implemented in hardware, instructions executed by a processing unit, or by a combination thereof. The processing unit can comprise a computer, a processor, such as the processor 302, a state machine, a logic array, or any other suitable devices capable of processing instructions. The processing unit can be a general-purpose processor which executes instructions to cause the general-purpose processor to perform the required tasks or, the processing unit can be dedicated to performing the required functions. In another embodiment of the present disclosure, the modules 206 may be machine-readable instructions (software) which, when executed by a processor/processing unit, perform any of the described functionalities.
Figure 2b illustrates an overview flowchart of the automated secure vehicle movement monitoring architecture of the ANPR system 104, in accordance with an exemplary implementation of the present disclosure.
The various embodiments of the present disclosure relate generally to an AI application or an application based on image processing. In particular, the present disclosure relates to a method for removing duplicate images from the ANPR database in the context of automated secure vehicle movement monitoring.
Figure 2b illustrates a flowchart of the ANPR system 104 which consists of the data receipt module, the data processing module, the duplication detection module and the alarm generation module. A camera module (may be in the image capturing device 102) at step 202 is installed in the pole (area of capture), which captures and streams the data through the RTSP protocol.
In an embodiment, the data receipt module at step 204 part of Automatic Number Plate Recognition (ANPR) system 100 receives the data from the camera module and pushes the data to the data queue.
In an embodiment, the data-pre-processing module receives the frame at step 206 from the data queue and extract a the sub-frame (i.e., area of interest) from the original frame and discards the rest of the frame. The extracted frame is then sent to the license plate detection module.
In an embodiment, the license plate detection module at step 208 uses the SSD Inception V2 model to detect whether the license plate is present in the sub-frame or not. If the license plate is present in the sub-frame, the coordinates of the license plate are extracted. The license plate image is created using the extracted coordinates and sent to the character recognition module. The character recognition module at step 210 uses the convolutional neural network (CNN) model to extract the characters from the detected license plate image.
In an embodiment, at step 212, the data transfer module to the ANPR server, takes the license plate image and the extracted character, date time, and gate_id and sends to the ANPR server module for further processing.
At step 214, the data receipt module in the ANPR server module receives data from the ANPR client and sends it to the data pre-processing module.
At steps 216 and 218, the data pre- processing module receives the data, and checks the received license plate image with the existing data for similarity score by calling the similarity check module.
At step 220, once the similarity score is received and is checked against the similarity score threshold (e.g., the predefined similarity score threshold).
At step 222, if the similarity score exceeds the similarity score threshold, then no entry is made in the database (e.g., the ANPR database).
At step 224, if the similarity score is less than the similarity score threshold, then the data is stored in the database.
In an embodiment, at step 226, the alarm generation module generates the alarm (i.e., the alarm notification) when the vehicle 108 is present in the premise beyond the time of permit and sends it to the client ANPR systems periodically. The similarity score check module uses a perceptual hasher algorithm to find whether the image is similar or not. Further, the alarm generation module is configured to check for the similarity score, if the similarity score exceeds the similarity score threshold, then it is of a similar image.
The alarm generation module runs as a background thread continuously monitoring the exit status of the vehicle. When the vehicle is entered into the gate (i.e., the entry gate), an entry is made in the database. The entered data includes license plate image, license plate number, time of entry, permitted time of stay, exit status etc., The value is kept as no initially against its exit status. The thread continuously checks the exit status of the vehicle where the value is set as no. It takes the vehicle data, and calculates its vehicle_premise_time based on the following equation (1):
?vehicle?_(?premise?_time ) = ?current ?_time - ?vehicle?_(?entry?_time ) …(1)
If vehicle_premise_time exceeds the threshold time, generate an alarm and send it to the client ANPR system.
The present disclosure focuses on the setup which consists of an ANPR application combined with secure vehicle movement monitoring and alarm generation.
In an embodiment, the present disclosure envisages a system and a methodology to remove duplicate number plates while storing the data in the remote server 106 at real-time on a secure vehicle movement monitoring premises.
In an embodiment, the present disclosure focuses on an automated vehicle monitoring system which consists of client and server modules which include the data similarity check module, alarm generation module and database module.
In an embodiment, the automated vehicle monitoring system client module, receives the data from the camera, identifies the number plate and recognizes the character in the identified number plate and sends it to the centralized server database module.
In an embodiment, the ANPR server module receives the data and sends it to the data similarity check module to check for the earlier entry (data replication) of data in the database.
In an embodiment, the Data similarity check module uses the image hashing algorithm to calculate the similarity score and a data threshold module to calculate whether the vehicle 108 presence is beyond the permitted time.
In an embodiment, the image hashing algorithm generates the 256 bit hash values of the images to be compared. It calculates the hamming distance between the two hashes or the similarity between the hashes using the equation (2):
Similarity Score = (Same Bits)/(Total Bits) * 100 …(2)
In an embodiment, the similarity score value is calculated and if the value is above 95% means data is already available in the database, a threshold time is calculated.
In an embodiment, the data threshold module calculates the data threshold time. It is measured against the difference of time between the received data and the existing entry.
In an embodiment, the similarity score and the data threshold module decide whether the data is an existing one or a newer one. In case of an existing entry, the data has to be updated with the latest entry. In case of a new entry, it has to be added to the database.
In an embodiment, the alarm generation module continuously runs in the background to find whether any vehicle is present in the defence premise beyond the time of permit. If any vehicles are present on the premises beyond a time limit, it generates the alert including gate id, vehicle id, permitted time etc.
In an advantageous aspect, the present disclosure provides a Smart city application which includes the Automatic Number Plate Recognition (ANPR) system 104 to detect the license plates and recognize it as a method of traffic policing or law enforcement. The present disclosure involves in variety of policing operations such as speed limit enforcement, toll enforcement, and secure movement of vehicles in defence bases etc., ANPR Zone is the zone which is covered by the IP cameras installed in the tower and the captured data is streamed to the Edge AI devices. The algorithm running in the edge AI devices (e.g. the image capturing device 102) receives the data from a camera (a part of the image capturing device 102) and processes it frame by frame. The presence of the license plate is checked in the frame and if it is present, it is sent to the character recognition module for further processing. Later, the identified registration number along with the license plate image is sent to the remote server 106 for storage. There are cases where the license plate of the same vehicle is detected multiple times and stored in the database. Advantageously, the identification and removal of repeated data are necessary to avoid an ineffective system. Particularly, in cases such as secured vehicle movement zones like defence bases where the entry and exit of vehicles is monitored seriously. The presence of vehicles in the secured zone beyond the authorized time limit triggers the alarm. If duplicate entries of the same plates are not removed, the system triggers a false alarm. In traditional applications, the values can be compared to identify whether it is duplicates or not. However, in an image processing application like ANPR, the values of the license plate number cannot be used to compare, as the very own value which has to differentiate the number plates are different for the same license plate. This invention proposes a method to identify the duplicate entry of the data and remove the same to improve the effectiveness of the ANPR system 104.
Figure 3 illustrates a process flow for a method for secure vehicle movement monitoring using the Automatic Number Plate Recognition (ANPR) through the system 104, according to an embodiment of the present disclosure. The method 300 may be operated via the system 104. For the sake of brevity, the operational methodology explained in Figure 1, Figure 2a, and Figure 2b is not repeated in the description of Figure 3.
At step 302, the method 300 may include capturing, via the image capturing device 102 installed in the pole, the video stream of the vehicle 108 entering the secured premise and transmitting the video stream using RTSP protocol.
At step 304, the method 300 may include receiving, via the data receipt module, the video stream and pushing the received data to a data queue.
At step 306, the method 300 may include extracting, via the data pre-processing module, the sub-frame indicating an area of interest comprising a license plate, from a video frame of the video stream, and discarding the remaining portion of the frame.
At step 308, the method 300 may include detecting, via the license plate detection module, the license plate in the sub-frame using the SSD Inception V2 model, and extracting coordinates of the license plate.
At step 310, the method 300 may include generating, via the character recognition module, the license plate image using the coordinates and recognizing license plate characters using the convolutional neural network (CNN)-based OCR model.
At step 312, the method 300 may include transmitting, via a data transfer module, the license plate image, extracted characters, timestamp, and gate ID to a server module.
At step 314, the method 300 may include receiving, via the server module, the transmitted data and performing duplicate entry detection based on comparing the license plate image with a set of stored license plate images using an image hashing algorithm.
At step 316, the method 300 may include determining, via the similarity score module, the similarity score between the received and stored license plate images and comparing the similarity score against the predefined similarity score threshold.
At step 318, the method 300 may include storing the received data in the database if the similarity score is below the predefined similarity score threshold or updating the existing record if the similarity score meets or exceeds the predefined similarity score threshold.
At step 320, the method 300 may include monitoring, via the alarm generation module, vehicle presence in the premise based on continuously checking the exit status of vehicles recorded in the database.
At step 322, the method 300 may include generating, via the alarm generation module, the alarm notification when the vehicle premise time exceeds a predefined threshold time and transmitting the alarm notification to a client ANPR system.
Thus, at least some of the technical advantages provided by the present disclosure include: focusing on a method to monitor secure vehicle movement in the restricted premises; generating an alarm when any vehicle is present in the secure premise such as defence premise beyond the time of permit; removes duplicate entry of data based on near duplicate image search algorithm using image hashing algorithm; combines the ANPR system and the duplicate image detection algorithm to avoid the generation of false alarm; method can be generalized with any number of entry/exit points where the results are stored and processed in the centralized server; can be embedded in a docker container and can be deployed; can be deployed in the edge ANPR devices which ensure the prediction in real-time.
The foregoing description has been set merely to illustrate the invention and is not intended to be limiting. Since modifications of the disclosed embodiments incorporating the substance of the invention may occur to a person skilled in the art, the invention should be construed to include everything within the scope of the invention.
While specific language has been used to describe the present subject matter, any limitations arising on account thereto, are not intended. As would be apparent to a person in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein. The drawings and the foregoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment.
,CLAIMS:1. A method for secure vehicle movement monitoring using an Automatic Number Plate Recognition (ANPR), the method comprising:
capturing, via an image capturing device installed in a pole, a video stream of a vehicle entering a secured premise and transmitting the video stream using RTSP protocol;
receiving, via a data receipt module, the video stream and pushing the received data to a data queue;
extracting, via a data pre-processing module, a sub-frame indicating an area of interest comprising a license plate, from a video frame of the video stream, and discarding the remaining portion of the frame;
detecting, via a license plate detection module, the license plate in the sub-frame using an SSD Inception V2 model, and extracting coordinates of the license plate;
generating, via a character recognition module, a license plate image using the coordinates and recognizing license plate characters using a convolutional neural network (CNN)-based OCR model;
transmitting, via a data transfer module, the license plate image, extracted characters, timestamp, and gate ID to a server module;
receiving, via a server module, the transmitted data and performing duplicate entry detection based on comparing the license plate image with a set of stored license plate images using an image hashing algorithm;
determining, via a similarity score module, a similarity score between the received and stored license plate images and comparing the similarity score against a predefined similarity score threshold;
storing the received data in a database if the similarity score is below the threshold or updating the existing record if the similarity score meets or exceeds the predefined similarity score threshold;
monitoring, via an alarm generation module, vehicle presence in the secured premises based on continuously checking an exit status of vehicles recorded in the database; and
generating, via the alarm generation module, an alarm notification when a vehicle premise time exceeds a predefined threshold time and transmitting the alarm notification to a client ANPR system.

2. The method as claimed in claim 1, comprising:
generating, via a data similarity check module, a 256-bit hash value for the license plate image using the image hashing algorithm;
computing, via the similarity score module, the similarity score between the license plate image and the stored images;
determining, via the similarity score module, whether the similarity score meets or exceeds the predefined similarity score threshold; and
preventing, via the similarity score module, duplicate database entries based on updating the existing record if the similarity score meets or exceeds the predefined similarity score threshold.

3. The method as claimed in claim 1, wherein the alarm generation module monitors the vehicle presence comprises:
retrieving, a vehicle data where the exit status is marked indicating a forbidden entry;
computing the vehicle premise time as the difference between the current time and a stored entry timestamp;
comparing the vehicle premise time a the permitted stay duration; and
generating the alarm notification if the vehicle premise time exceeds the permitted stay duration and transmitting the notification to the client ANPR system.

4. The method as claimed in claim 1, wherein checking the exit status of the vehicles comprises:
capturing a video stream of a vehicle at an exit gate;
retrieving a latest stored entry corresponding to the license plate number;
comparing, an exit image with a corresponding record of the exit gate using an image hashing algorithm;
updating the exit status indicating a permitted if the similarity score is equal or exceeds the predefined similarity score threshold; and
removing a vehicle record from active monitoring if the exit status is updated.

5. The method as claimed in claim 1, comprising:
encrypting, the license plate image, extracted characters, timestamp, and gate ID before transmission;
decrypting the encrypted data before processing;
storing the decrypted data in a secure repository; and
detecting unauthorized data modifications based on generating the alert notification when tampering is identified.

6. A system for secure vehicle movement monitoring using an Automatic Number Plate Recognition (ANPR), the system comprising:
a memory;
at least one processor in communication with the memory, the at least one processor configured to:
capture, via an image capturing device installed in a pole, a video stream of a vehicle entering a secured premise and transmitting the video stream using RTSP protocol;
receive, via a data receipt module, the video stream and pushing the received data to a data queue;
extract, via a data pre-processing module, a sub-frame indicating an area of interest comprising a license plate, from a video frame of the video stream, and discarding the remaining portion of the frame;
detect, via a license plate detection module, the license plate in the sub-frame using an SSD Inception V2 model, and extracting coordinates of the license plate;
generate, via a character recognition module, a license plate image using the coordinates and recognizing license plate characters using a convolutional neural network (CNN)-based OCR model;
transmit, via a data transfer module, the license plate image, extracted characters, timestamp, and gate ID to a server module;
receive, via a server module, the transmitted data and performing duplicate entry detection based on comparing the license plate image with a set of stored license plate images using an image hashing algorithm;
determine, via a similarity score module, a similarity score between the received and stored license plate images and comparing the similarity score against a predefined similarity score threshold;
store the received data in a database if the similarity score is below the threshold or updating the existing record if the similarity score meets or exceeds the predefined similarity score threshold;
monitor, via an alarm generation module, vehicle presence in the premise based on continuously checking an exit status of vehicles recorded in the database; and
generate, via the alarm generation module, an alarm notification when a vehicle premise time exceeds a predefined threshold time and transmitting the alarm notification to a client ANPR system.

7. The system as claimed in claim 6, comprising:
generate, via a data similarity check module, a 256-bit hash value for the license plate image using the image hashing algorithm;
compute, via the similarity score module, the similarity score between the license plate image and the stored images;
determine, via the similarity score module, whether the similarity score meets or exceeds the predefined similarity score threshold; and
prevent, via the similarity score module, duplicate database entries based on updating the existing record if the similarity score meets or exceeds the predefined similarity score threshold.

8. The system as claimed in claim 6, wherein the alarm generation module monitors the vehicle presence comprises:
retrieve, a vehicle data where the exit status is marked indicating a forbidden entry;
compute the vehicle premise time as the difference between the current time and a stored entry timestamp;
compare the vehicle premise time a the permitted stay duration; and
generate the alarm notification if the vehicle premise time exceeds the permitted stay duration and transmitting the notification to the client ANPR system.

9. The system as claimed in claim 6, wherein checking the exit status of the vehicles comprises:
capture a video stream of a vehicle at an exit gate;
retrieve a latest stored entry corresponding to the license plate number;
compare, an exit image with a corresponding record of the exit gate using an image hashing algorithm;
update the exit status indicating a permitted if the similarity score is equal or exceeds the predefined similarity score threshold; and
remove a vehicle record from active monitoring if the exit status is updated.

10. The system as claimed in claim 6, comprising:
encrypt, the license plate image, extracted characters, timestamp, and gate ID before transmission;
decrypt the encrypted data before processing;
store the decrypted data in a secure repository; and
detect unauthorized data modifications based on generating the alert notification when tampering is identified.