Claims:WE CLAIM:

1. A method (300) of detecting and classifying vehicles, the method (300) comprising steps of:
detecting (302), by a first image acquisition module (14a), presence of vehicle in its field of view;
capturing (302), by the first image acquisition module (14a), an image of the vehicle upon detecting the vehicle;
activating (304), by a processor (102), a second image acquisition module (14b) upon detecting the vehicle by the first image acquisition module (14a);
capturing (308), by second image acquisition module (14b), an image of the axle and wheels of the vehicle; and
processing (314), by the processor (102), the images from the first image acquisition module (14a) and the second image acquisition module (14b) to classifying the vehicle, wherein the step of processing comprising:
identifying, by the processor (102), the size of the vehicle from the image captured by the first image acquisition module (14a);
identifying, by the processor (102), the number of wheels and axles of the vehicle from the image captured by the second image acquisition module (14b); and
mapping, by the processor (102), the size and number of wheels and axles of the vehicle using a deep learning model for classifying the type of vehicle.

2. The method (300) as claimed in claim 1, comprising:
capturing (312), by a third image acquisition module (14c), a licence plate image;
extracting (318) and converting (318), by the processor (102), alphanumeric characters from the licence plate image captured into text strings using an Optical Character Recognition (OCR) technique for identifying a licence plate number of the vehicle.

3. The method (300) as claimed in claim 1, comprising:
cropping (318), by the processor (102), a licence plate image from the image of the vehicle;
extracting (318), by the by the processor (102), alphanumeric characters from the licence plate image into text strings using Optical Character Recognition (OCR); and
converting, by the processor (102), the alphanumeric characters into text strings.

4. The method (300) as claimed in claim 3, comprising:
displaying, by the processor (102), the image of the vehicle, the licence plate image, time stamp of the image of the vehicle and the text strings corresponding to the licence plate number of the vehicle.

5. The method (300) as claimed in claim 1, comprising:
sub-classifying (316), by the processor (102), the vehicles based on the shape, size of the vehicles and wheels and axles of the vehicle identified from the image of the vehicle.

6. The method (300) as claimed in claim 1, comprising:
comparing, by the processor (102), the image of the vehicle with pre-stored images of the vehicles to classify the type of the vehicle.

7. A system (12) for detecting and classifying vehicles, the system (12) comprising:
a processor (102); and
a memory (104) coupled to the processor (102), wherein the processor (102) is configured to execute program instructions (108) stored in the memory (104), to:
activate a first image acquisition module (14a) to capture an image of the vehicle in its field of view;
activate a second image acquisition module (14b) to capture an image of the axle and wheels of the vehicle upon detecting the vehicle by the first image acquisition module (14a); and
process the images from the first image acquisition module (14a) and the second image acquisition module (14b) to classify the vehicle, by:
identifying the size of the vehicle from the image captured by the first image acquisition module (14a);
identifying the number of wheels and axles of the vehicle from the image captured by the second image acquisition module (14b); and
mapping the size and number of wheels and axles of the vehicle using a deep learning model for classifying the type of vehicle.

8. The system (12) as claimed in claim 7, wherein the processor (102) executes the program instructions (108) to:
crop a licence plate image from the image of the vehicle;
extract alphanumeric characters from the licence plate image into text strings using Optical Character Recognition (OCR); and
convert the alphanumeric characters into text strings.

9. The system (12) as claimed in claim 8, wherein the processor (102) executes the program instructions (108) to:
display the image of the vehicle, the licence plate image, time stamp of the image of the vehicle and the text strings corresponding to the licence plate number of the vehicle.

10. The system (12) as claimed in claim 7, wherein the processor (102) executes the program instructions (108) to:
sub-classify the vehicles based on the shape, size of the vehicles and wheels and axles of the vehicle identified from the image of the vehicle.

11. A system (702) for detecting and classifying vehicles, the system (702) comprising:
a processor (102); and
a memory (104) coupled to the processor (102), wherein the processor (102) is configured to execute program instructions (108) stored in the memory (104), to:
activate a first sensor (706a) to detect presence of a vehicle;
activate a second sensor (706b) to capture an image of the axle and wheels of the vehicle upon detecting the vehicle by the sensor (706a), wherein the second sensor (706b) captures a vehicle image and a licence plate image of the vehicle; and
process the images captured by the second sensor (706b) to classify the vehicle, by:
identifying the size of the vehicle from the image captured by the second sensor (706b);
identifying the number of wheels and axles of the vehicle from the image captured by the second sensor (706b); and
mapping the size and number of wheels and axles of the vehicle using a deep learning model for classifying the type of vehicle.

12. The system (702) as claimed in claim 11, wherein the processor (102) executes the program instructions (108) to:
extract alphanumeric characters from the licence plate image into text strings using Optical Character Recognition (OCR); and
convert the alphanumeric characters into text strings.

13. The system (702) as claimed in claim 11, wherein the first sensor (706a) is one of a camera, an infrared sensor, a laser-based sensor, a passive infrared and an ultrasound sensor, and a loop detector, and
wherein the second sensor (706b) is a camera.
, Description:A SYSTEM AND A METHOD OF DETECTING AND CLASSIFYING VEHICLES

FIELD OF INVENTION

[01] The present invention generally relates to a field of classifying vehicles. More specifically, the present invention relates to a system and a method of detecting and classifying vehicles.

BACKGROUND OF THE INVENTION

[02] Vehicles include, but not limited to, bicycles, motorcycles, auto rickshaws, cars, pickups, minivans, Sports Utility Vehicles (SUV), buses, trucks, etc. Typically, sensors are installed at designated places such as toll booths, tunnels, bridges, roadways, etc., for collecting information of the vehicles. The information collected is then used for optimizing traffic management and law enforcement. In addition, the information collected is used for detecting the presence of a vehicle in a particular zone, counting the number of vehicles on the roadway, such as the volume of vehicles on the roadway, determining the lane position, classifying the vehicles, counting the number of axles, determining the direction of the vehicle(s), estimating the occupancy and determining the speed of vehicles, etc.

[03] Generally, the sensors include cameras, radar-based sensors, laser-based sensors, passive infrared and ultrasound sensors and so on. State of the art vehicle classification systems communicatively connect to the sensors and classify the vehicles based on their size. Further, some of the vehicle classification systems rely on side views and other features of the vehicles to classify the vehicles.

[04] Several attempts have been made in the past for classifying vehicles using images of the vehicles. One such example is disclosed in a United States granted patent No. 6897789, entitled “System for determining kind of vehicle and method therefor” (“the ‘789 Patent”). The ‘789 Patent discloses a system for determining a kind of vehicle and a method therefore, including a vehicle detection unit for detecting a vehicle which reaches to a vehicle detection region on a roadway, a wheel shaft number counting unit for counting a number of wheel shafts of the detected vehicle, an image photographing unit for photographing a front or rear image of the detected vehicle and a vehicle kind determination unit for yielding distances and widths of the tires of the detected vehicle on the basis of the photographed image from the image photographing unit and determining the kind of the vehicle on the basis of the number of wheel shafts detected from the wheel shaft counting unit and the yielded distance and width values can precisely determine the kind of vehicle traveling the roadway.

[05] Another example is disclosed in a United States granted patent No. 7136828, entitled “System for determining kind of vehicle and method therefor” (“the ‘828 Patent”). The ‘828 Patent discloses an intelligent vehicle identification system that uses inductive loop technology to profile and classify a vehicle. In a tolling industry application, classification of the vehicle is made prior to the vehicle arriving at a payment point in a toll lane in which the vehicle travels. A predetermined fare associated with the classification is then solicited from an operator of the vehicle without efforts from a toll attendant. In a preferred embodiment, the system also includes an intelligent queue loop that verifies the vehicle at the payment point to prevent misclassification due to a second vehicle, e.g., a motorcycle, that changes from a different lane to the toll lane in question.

[06] Although the above-discussed disclosures disclose several methods of classifying the vehicles based on the shape and other features of the vehicles, they have few problems. For example, the state of the art vehicle classification systems does not accurately classify the vehicles on a multi-lane road. Further, they use side profile or front profile either alone or in combination to classify the vehicles. Use of side profile or front profile is not sufficient to accurately classify the vehicles.

[07] Therefore, there is a need for an improved system that utilizes multiple cameras capturing images of a variety of features of the vehicle and their arrangements for classifying the vehicles in a robust manner with a high degree of accuracy.

SUMMARY OF THE INVENTION

[08] The problems in the existing art are met by a system and a method of detecting and classifying vehicles.

[09] Accordingly, it is an object of the present invention to provide a system and a method for classifying vehicles with similar appearances or profile into the correct categories based on their axle size, axle count, vehicle type among others in real-time.

[010] It is another object of the present invention to provide a system for identifying the licence plate number of the vehicle with a high degree of accuracy.

[011] In order to achieve one or more of the objects as stated above, the present invention provides a system detecting and classifying vehicles. The system comprises a first image acquisition module. The first image acquisition module detects and captures an image of the vehicle in its field of view. The system comprises a second image acquisition module. The second image acquisition module captures an image of the axle and wheels of the vehicle upon detecting the vehicle by the first image acquisition module. The system processes the images from the first image acquisition module and the second image acquisition module to classify the vehicle. The system classifies the vehicle by identifying the size of the vehicle from the image captured by the first image acquisition module, identifying the number of wheels and axles of the vehicle from the image captured by the second image acquisition module, and mapping the size and number of wheels and axles of the vehicle using a deep learning model for classifying the type of vehicle.

[012] In one aspect of the present invention, the system further comprises a third image acquisition module. The third image acquisition module captures a licence plate image of the vehicle upon detecting the vehicle by the first image acquisition module. The system extracts and converts alphanumeric characters from the licence plate image captured into text strings using an Optical Character Recognition (OCR) technique for identifying a licence plate number of the vehicle. The system further displays the image of the vehicle, the licence plate image, time stamp of the image of the vehicle and the text strings corresponding to the licence plate number of the vehicle.

[013] In one advantageous feature of the present invention, the system identifies, reads, and extracts the characters on the license plate of a vehicle – regardless of the vehicle being static or in motion. This ensures the system for use in a wide range of applications such as traffic management, security related applications, separating commercial and passenger vehicles among others.

[014] In another advantageous feature of the present invention, the system receives the image of the vehicle, and reads the license plate number in any language using deep learning methods. This in addition to wheel and axle information helps to capture and record relevant details with very high accuracy of above 95%.

[015] Further advantages and examples of the invention will be brought out in the following part of the specification, wherein detailed description is for the purpose of fully disclosing the invention without placing limitations thereon.

BRIEF DESCRIPTION OF THE DRAWINGS

[016] Further features and advantages of the present subject matter will become apparent from the following detailed description, taken in combination with the appended drawings, in which:

[017] FIG. 1 illustrates an environment of a system for detecting and classifying vehicles, in accordance with one embodiment of the present invention;

[018] FIG. 2 illustrates a block diagram of the system; in accordance with one embodiment of the present invention;

[019] FIG. 3 shows a front view illustrating how image acquisition modules implements at a toll booth for capturing images of vehicles, in accordance with one embodiment of the present invention;

[020] FIG. 4 illustrates a block diagram of a user device, in accordance with one embodiment of the present invention;

[021] FIG. 5 illustrates a method of detecting and classifying vehicles, in accordance with one embodiment of the present invention;

[022] FIG. 6 illustrates an environment of detecting the presence of vehicle at a toll booth, in accordance with one exemplary embodiment of the present invention;

[023] FIG. 7 illustrates an environment of a second image acquisition module capturing vehicle wheel and axle information, in accordance with one exemplary embodiment of the present invention;

[024] FIG. 8 illustrates a screenshot presenting the information of the vehicles on the user device, in accordance with one exemplary embodiment of the present invention.

[025] FIG. 9 illustrates an environment of a system for detecting and classifying vehicles, in accordance with another embodiment of the present invention; and

[026] FIG. 10 illustrates a method of detecting and classifying vehicles, in accordance with one embodiment of the present invention.

[027] It will be noted that throughout the appended drawings, like features are identified by like reference numerals.

DETAILED DESCRIPTION OF THE INVENTION

[028] The following detailed description is susceptible to various modifications and alternative forms, specific embodiments thereof will be described in detail and shown by way of example. It should be understood, however, that there is no intent to limit example embodiments of the present invention to the particular forms disclosed. Conversely, example embodiments are to cover all modifications, equivalents, and alternatives falling within the scope of the invention.

[029] It should be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements are not limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the present invention.

[030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the terms “comprises,” “comprising,” “includes,” “including,” and/or “having” specify the presence of stated features, integers, steps, operations, elements, and/or components when used herein, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

[031] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. It should be further understood that terms, e.g., those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and are not to be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[032] It should also be noted that in some alternative implementations, functions/acts noted in a specific block may occur out of the order noted in a flowchart. For example, two blocks shown in succession may in fact be executed substantially concurrently or may sometimes be executed in a reverse order depending upon functionality or acts involved.

[033] It should be understood that the present subject matter describes a system and method of detecting and classifying vehicles. The system comprises image acquisition modules such as a first image acquisition module, a second image acquisition module and a third image acquisition module arranged to capture images of vehicles from multiple viewpoints. The first image acquisition module mounts at the top and captures vehicle size and other characteristics of the vehicle. The second image acquisition module captures wheel and axle information of the vehicle. The third image acquisition module captures the number plate of the vehicle. The image acquisition modules transmit the images of the vehicles to the system. The system identifies and recognizes objects present in the images based on visual cues by incorporating neural network models. Subsequently, the system classifies the vehicles based on pre-defined criteria or learnt patterns using a neural network or machine learning model. In one example, the system classifies the vehicles based on their load carrying capacity. In another example, the system classifies the vehicles based on their shape and size.

[034] Various features and embodiments of a system and a method for detecting and classifying vehicles are explained in conjunction with the description of FIGs. 1-10.

[035] FIG. 1 shows an environment 10 in which a system 12 for detecting and classifying vehicles. implements, in accordance with one exemplary embodiment of the present invention. The system 12 includes a server or a database comprising an application to execute functions for detecting and classifying vehicles. In one embodiment, the system 12 operates as a standalone device or connects to other (e.g., networked) systems. A person skilled in the art appreciates that the system 12 implements in any different computing systems, environments, and/or configurations such as a workstation, an electronic device, a mainframe computer, a laptop, and so on. In a networked deployment, the system 12 operates in the capacity of a server or a client machine in server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. In one example, a toll operator, law enforcement, traffic management authority or any specific entity manages the system 12 for detecting and classifying vehicles depending on the need.

[036] FIG. 2 shows a diagrammatic representation of the system 12, in accordance with one embodiment of the present invention. The system 12 comprises a first processor 102 (e.g., a central processing unit (CPU), a graphics processing unit (GPU) or both), a first memory 104, which communicates with at least one other via a bus 106. Each of the first processor 102 and the first memory 104 comprises instructions 108 stored therein. The system 12 further comprises a display 110 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)). The system 12 further comprises an alphanumeric input device (e.g., a keyboard) and/or a touchscreen 112, a user interface (UI) navigation device 114 (e.g., a mouse), a disk drive unit 116, a signal generation device 118 (e.g., a speaker) and a network interface device 120.

[037] The disk drive unit 116 includes a machine-readable medium 122 on which is stored one or more sets of instructions and data structures (e.g., software 108) embodying or utilized by any one or more of the methodologies or functions described herein. It should be understood that the term “machine-readable medium” might be taken to include a single medium or multiple medium (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more sets of instructions. The term “machine-readable medium” may also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present invention, or that is capable of storing, encoding or carrying data structures utilized by or associated with such a set of instructions. The term “machine-readable medium” may accordingly be taken to include, but not be limited to, solid-state memories, optical and magnetic media, and carrier wave signals.

[038] The instructions 108 may also reside, completely or at least partially, within the memory 104 and/or within the first processor 102 during execution thereof by the system 12, the first memory 104 and the first processor 102 also constituting machine-readable media. The instructions 108 may further be transmitted or received over a network 16 via the network interface device 120 utilizing any one of a number of well-known transfer protocols.

[039] The network 16 may be a wireless network, a wired network or a combination thereof. Network 16 can be implemented as one of the different types of networks, such as intranet, local area network (LAN), wide area network (WAN), the internet, and the like. The network 16 may either be a dedicated network or a shared network. The shared network represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), and the like, to communicate with one another. Further the network 16 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, and the like.

[040] Referring back to FIG. 1, the system 12 communicatively connects to a plurality of image acquisition modules such a first image acquisition module 14a, a second image acquisition module 14b, a third image acquisition module 14c, etc., collectively called as image acquisition modules 14, or simply an image acquisition module 14 when referred to a single image acquisition module. In the present description, the term “image acquisition module” is interchangeably used with “camera”, “image capturing module” or “image capturing unit”. The image acquisition modules 14, include but not limited to, cameras, radar-based sensors, laser-based sensors, passive infrared and ultrasound sensors, loop detectors and so on. The image acquisition modules 14 detect the presence of a vehicle in their field of view and capture one or more images of the vehicle. For example, consider the image acquisition modules 14 are cameras, then the image acquisition modules 14 continuously monitor the roadway for changes in their respective field of views indicating the presence of a vehicle. Once the image acquisition modules 14 detect presence of the vehicle, then image acquisition modules 14 capture the image and store in the memory or transmit to a device situated in a remote location via various communication protocols. The image acquisition modules 14 include analog or digital, and capture one or more images of the vehicle. The image acquisition modules 14 help to record images of vehicles that are moving or in static and help in identifying the vehicle. For instance, the image acquisition modules 14 help to capture images of vehicles and to identify their characteristics. In addition, the image acquisition modules 14 can be used to generate a vehicle identifier such as a vehicle license number based on an image of a license plate.

[041] FIG. 3 shows a front view illustrating how image acquisition modules 14 implements at a toll booth 18 for capturing images of vehicles 20, in accordance with one exemplary embodiment of the present invention. A person skilled in the art understands that the image acquisition modules 14 mounted at the toll booth 18 is presented for illustrative purpose only, and image acquisition modules 14 can be mounted at designated places such as tunnels, bridges, roadways, check points, entrance and exit points of a structure/building, etc., for collecting information of the vehicles without departing from the scope of the present invention.

[042] As can be seen in FIG. 3, the first image acquisition module 14a mounts at a distance from the toll booth 18. The first image acquisition module 14a mounts at a height (i.e., at the top) from the ground such that the first image acquisition module 14a captures the vehicle crossing a reference line or vehicle detection zone for triggering capturing of the images of the vehicle 20 that is passing the toll booth 18. Here, the first image acquisition module 14a captures an image of the vehicle 20 from a certain angle depending on its position where it is mounted at the toll booth 18. In one example, the first image acquisition module 14a captures the shape and size of the vehicle 20. The second image acquisition module 14b mounts at the front and/or rear side of toll booth 18 such that second image acquisition module 14b captures axle and wheel information of the vehicle 20. Here, the second image acquisition module 14b captures the wheel and axle information as the vehicle 20 crosses the reference line or vehicle detection zone. Specifically, the second image acquisition module 14b captures front and rear of the vehicle 20 and counts the number of wheels and axles at the front and rear of the vehicle 20. The third image acquisition module 14c mounts at the front or rear of the toll booth 18 at a suitable height from the ground. The third image acquisition module 14c captures multiple images, say 3 to 5 images of the licence plate number. The third image acquisition module 14c identifies, reads and extracts the characters from the images corresponding to the license plate of the vehicle 20 regardless of whether the vehicle 20 is static or in motion.

[043] FIG. 4 shows a block diagram of the image acquisition module 14 i.e., each of the first image acquisition module 14a, the second image acquisition module 14b and third image acquisition module 14c, in accordance with one exemplary embodiment of the present invention. The image acquisition module 14 includes an image sensor 202, a second processor 204, a second memory 206, and a transceiver 208. The image sensor 202 detects the presence of an object (e.g., vehicle) and provides a signal that triggers the image acquisition module 14 to capture one or more images of the object. The second processor 204 converts the images captured by the image sensor 202 into digital form. If the image sensor 202 is an analog camera, then the image sensor 202 connects to separate digitizing hardware. The hardware includes a dedicated processing device for analog-to-digital conversion or may be based on an input device installed in a general-purpose computer, which may perform additional functions such as image processing. The second processor 204 processes a single image for each vehicle or multiple images of each vehicle, depending on the functionality of the image sensor 202 and/or business requirements (e.g., accuracy, jurisdictional requirements). If multiple images are used, each image may be processed, and the results may be compared or combined to enhance the accuracy of the process. For example, more than one image of a rear license plate, or images of both front and rear license plates, may be processed and the results are compared to determine the most likely registration number and/or confidence level. Image processing may include identifying the distinguishing features of a vehicle (e.g., the license plate of a vehicle, shape of the vehicle) within the image, and analysing those features. Analysis may include optical character recognition (OCR), template matching, or other analysis techniques.

[044] The second memory 206 stores the images processed by the second processor 204. The second processor 204 communicates with the transceiver 208 to send or receive data from other devices such as the system 12, for example.

[045] Referring back to FIG. 1, the system 12 further communicatively connects to a plurality of user devices i.e., a first user device 22.a, a second user device 22.b, etc., collectively referred to as user devices 22, or simply user device 22 when referred to a single user device. Each of the system 12 and the plurality of user devices 22 communicatively connect via the network 16. The user device 22 includes, but not limited to, a mobile phone, a laptop, a desktop computer, a tablet, a wrist watch and other electronic devices. In one example, a user such as a law enforcement officer or toll operator operates the user device 22 to access the data stored and processed by the system 12.

[046] FIG. 5 shows a method 300 for detecting and classifying vehicles is explained, in accordance with one exemplary embodiment of the present invention. The method 300 may be described in the general context of computer executable instructions. Generally, computer executable instructions may include routines, programs, objects, components, data structures, procedures, modules, functions, etc., that perform particular functions or implement particular abstract data types. The method 300 may also be practiced in a distributed computing environment where functions are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, computer executable instructions may be located in both local and remote computer storage media, including memory storage devices.

[047] The order in which the method 300 is described and is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method 300 or alternate methods. Additionally, blocks may be deleted from the method 300 without departing from the scope of the invention described herein. Furthermore, the method may be implemented in any suitable hardware, software, firmware, or combination thereof. However, for ease of explanation, in the embodiments described below, the method 300 may be implemented in the above-described system 12.

[048] At step 302, the system 12 employs the first image acquisition module 14a to detect presence of the vehicle 20. Here, the system 12 activates the first image acquisition module 14a. Once activated, the first image acquisition module 14a continuously checks for presence of vehicle(s) in its field of view, as shown at step 304. If the first image acquisition module 14a does not detect the presence of the vehicle, then method 300 moves back to step 302. If the first image acquisition module 14a detects the presence of the vehicle at the toll booth, then the method moves to steps 306 and 310. FIG. 6 shows an environment 400 illustrating detecting the presence of vehicles at a toll booth, in accordance with one exemplary embodiment of the present invention. When a vehicle 402 approaches a toll booth 404 and crosses a reference line 406, then the system 12 activates the first image acquisition module 14a. Here, the first image acquisition module 14a activates and creates a bounding box 408 to capture the shape of vehicle 402. A person skilled in the art understands that when the vehicle 402 comes in proximity to the toll booth 404, the first image acquisition module 14a captures the details of the vehicle 402 such as vehicle position with respect to its own position. Further, the first image acquisition module 14a captures the images comprising the full frame information such as shape and size of the vehicle 402.

[049] Once the first image acquisition module 14a captures the image of the vehicle, the first image acquisition module 14a transmits the images to the system 12. Upon receiving the images from the first image acquisition module 14a, the system 12 activates the second image acquisition module 14b (step 306) and the third image acquisition module 14c (step 310). Here, the first image acquisition module 14a acts as a trigger device upon detecting the presence of the vehicle 20 crossing the reference line or vehicle detection zone and transmits a signal to the system 12. When triggered, the second image acquisition module 14b gets activated (step 306) and captures vehicle wheel and axle information such as number of wheels and number of axles present on the vehicle 20 and their size. In one example, the second image acquisition module 14b captures the images of the wheels and axle at the front of the vehicle 20 as the vehicle 20 passes the reference line. FIG. 7 shows an environment or field of view 500 of the second image acquisition module 14b capturing vehicle wheel and axle information, in accordance with one exemplary embodiment of the present invention. Here, the second image acquisition module 14b creates one or more bounding boxes 502 around wheels and/or axles 504 of the vehicle. Subsequently, the second image acquisition module 14b captures the number of wheels and axle at the rear of the vehicle 20 as the vehicle 20 passes the reference line. This way, the second image acquisition module 14b accurately captures the size, wheels and axle information of the vehicle.

[050] After capturing the size, wheels and axle information of the vehicle, the method 300 moves to step 308. At step 308, the system 12 calculates or counts the number of axles or wheels using a deep learning model. Here, the system 12 utilises the information corresponding to the size and shape of the vehicle 20 detected by the first image acquisition module 14a and identifies the type of vehicle. Further, the system 12 maps the number of wheels detected by the second image acquisition module 14b with the type of vehicle. This helps to classify the type of vehicle such as a truck, bus, van, car, etc.

[051] As specified above, upon receiving the images from the first image acquisition module 14a, the system 12 activates the second image acquisition module 14b (step 306) and the third image acquisition module 14c (step 310). At step 310, the system 12 activates the third image acquisition module 14c. Once activated. The third image acquisition module 14c obtains metadata of the vehicle from the licence number plate or in the form of a Quick Response (QR) code in the vehicle number plate or on the windshield of the vehicle, as shown at step 312. The metadata of the vehicle includes, but not limited to, vehicle type, customer classification, dates, locations, licence plate number, etc. The third image acquisition module 14c recognises the QR code and/or the licence number plate and obtains the metadata of the vehicle. From the metadata, the system 12 identifies the type of vehicle such as truck, bus, van, car, etc., as shown at step 314. Here, the system 12 maps the data corresponding to the number of axles or wheels identified (step 308) and sub-classifies the type of vehicles, as shown at step 316. Consider that the system 12 identifies the vehicle as a car (i.e., four-wheeled vehicle) at steps 306 and 312. The system 12 employs a neural network or a machine learning model for classifying the type of vehicles based on the pre-stored data. For the above example, the system 12 employs the neural network and sub-classifies the vehicle as car, jeep, auto rickshaw based on the shape, number of axles and metadata of the vehicle identified from each of the first image acquisition module 14a, the second image acquisition module 14b and the third image acquisition module 14c.

[052] At step 318, the system 12 employs the third image acquisition module 14c to detect or recognize the vehicle number plate of the vehicle 20 (bounding box 410 in FIG. 6). In order to recognize the vehicle number plate, the third image acquisition module 14c employs OCR, template matching, or other analysis techniques to identify, read and extract the characters on the license plate of the vehicle 20 regardless whether the vehicle 20 is static or in motion. In one implementation, the first processor 102 processes the image captured by the third image acquisition module 14c to recognize the licence plate number of the vehicle. In order to recognize the licence plate number of the vehicle, the first processor 102 presents an integral function of detecting a license plate image at virtually any viewing angle and under a multitude of conditions. As known, a typical license plate is rectangular in shape. However, the license plate image may appear like a parallelogram or trapezoid depending on several factors such as the angle of the license plate image captured, for example.  In accordance with the present invention, the first processor 102 dewarps the license plate image that is in parallelogram and trapezoid and converts them into a rectangular image. For example, the first processor 102 dewarps the four vertices of the quadrilateral and the 4 vertices of an un-rotated rectangle with an aspect ratio of 2:1 (width: height), or any other suitable license plate aspect ratio into a perspective transform depending on the need. The perspective transform is applied around the quadrilateral region and the 2:1 aspect ratio object image is cropped out. The cropped image is used for recognizing the alphanumeric characters. In one example, the first processor 102 dewarps the license plate image to resize or rescale the license plate image to a suitable size for identifying the alphanumeric characters.

[053] After dewarping, the first processor 102 performs OCR text extraction on the license plate image. In one example, the first processor 102 utilizes specialized or a commercial OCR software application installed for accurate extraction of the license plate number. Here, the first processor 102 extracts the text and converts into text strings. The first processor 102 employs a text translator either on its own or using a third party software application to identify and convert the text from any language. In one example, the first processor 102 performs OCR text extraction to obtain a logo i.e., symbol made up of text and images that identifies a manufacturer of the vehicle. The logo helps to identify/categorize the vehicle based on the manufacturer, for example.

[054] Consider that the licence plate image is unclear or unregistered/blank. In such a case, the first processor 102 cannot extract text from the licence plate image. Here, the first processor 102 triggers an alert to concerned authorities to manually verify the vehicle and take required action.

[055] At this point of time, the system 12 has all the information corresponding to each of the vehicles. The information includes, but not limited to, image of the vehicle, vehicle number information, number of wheels and axle of the vehicle, timestamp, colour, location of the toll booth, etc. The system 12 presents the vehicle snapshot/image, image of the licence plate captured by the first image acquisition module 14a, and the type of vehicle, axle count and date and time stamp identified for display on the user device 22 (step 320). FIG. 8 shows an exemplary screenshot 600 displayed on the user device 22 by the system 12, in accordance with one embodiment of the present invention. The screenshot 600 presents a first section 602 showing images of first vehicle 602a, second vehicle 602b, and third vehicle 602c. Further, the screenshot 600 presents a second section 604 showing the number plate image captured by the third image acquisition module 14c and a third section 606 showing the licence number plate recognized (LPR) corresponding to each of first vehicle 602a, second vehicle 602b, and third vehicle 602c. Further, the system 12 displays the time stamp of the image captured corresponding to the detection of the vehicle at the toll booth in a fourth section 608.

[056] In one alternate embodiment of the present invention, the system 12 receives the information from each of the first image acquisition module 14a, the second image acquisition module 14b and third image acquisition module 14c and checks the information with a pre-stored data in the memory 104. Here, the system 12 employs the first processor 102 for checking the information received from the first image acquisition module 14a, the second image acquisition module 14b and third image acquisition module 14c with that of the pre-stored data. The first processor 102 employs a neural network or a machine learning model for detecting, identifying and classifying the vehicles based on the pre-stored data. Specifically, the first processor 102 identifies and recognizes objects present in the images captured and transmitted by each of the first image acquisition module 14a and the second image acquisition module 14b. For example, the first processor 102 identifies and recognizes the objects such as vehicle’s appearance as captured from the first image acquisition module 14a and the second image acquisition module 14b including number of axles, load carrying capacity, etc. In another example, the first processor 102 checks the images for identifying the distinguishing features of the vehicle 20 (e.g., shape of the vehicle or the license plate of a vehicle) within the image, and analyses the features. The first processor 102 analyzes the images using optical character recognition (OCR), template matching, or other analysis techniques. This enables the first processor 102 to identify the number plate written in various languages.

[057] By checking the information with the pre-stored data, the first processor 102 obtains subtle variations in the images captured by the first image acquisition module 14a and the second image acquisition module 14b and classifies the vehicles 20 using a variety of criteria. The first processor 102 employs the neural network or machine learning model to classify the vehicles 20. This helps to perform a multi-layered inspection to decide on the class labels (classification categories) for each category based on pre-defined criteria. For example, the first processor 102 checks the images with the pre-stored data to classify the vehicles 20 into light commercial vehicle, heavy commercial vehicle, multi-axle commercial vehicles, etc. based on their load carrying capacity (pre-defined criteria). In another example, the first processor 102 classifies the vehicles 20 into sedan, pickup, minivan, Sports Utility Vehicles (SUV), bus, truck classes based on the shape and size of vehicles 20. In addition, the first processor 102 analyzes the image of the vehicle captured by the first image acquisition module 14a and compares with the pre-stored image of the vehicle and identifies a vehicle model corresponding to the captured image for classifying a current vehicle as the identified vehicle model. Similarly, the first processor 102 analyzes the wheel and axle information of the vehicle captured by the second image acquisition module 14b and compares with the pre-stored image of the vehicle and identifies a vehicle model corresponding to the captured image for classifying a current vehicle as the identified vehicle model. As the first processor 102 checks minute details obtained from the images, the system 12 classifies the vehicles 20 based on their load carrying capacity and eliminates false positives and duplicates while increasing the accuracy of the results.

[058] Although the above description is explained considering that the first image acquisition module 14a captures the images of the vehicle, the second image acquisition module 14b captures the images of the axle/wheel and the third image acquisition module 14c captures the number plate image, it is possible to utilise the single image acquisition module, say first image acquisition module to capture images of the vehicle and number plate image. This way, the presently disclosed system can be used with the help of first image acquisition module 14a and the second image acquisition module 14b, without departing from the scope of the present invention.

[059] FIG. 9 shows an exemplary environment 700 in which a system 702 for detecting and classifying implements, in accordance with another embodiment of the present invention. In the present embodiment, the system 702 operates similar to the system 12 explained above. Here, the system 702 communicatively connects to a first sensor 706a and a second sensor 706b via a network 704. The network 704 functions similar to the network 16. In the present invention, the first sensor 706a comprises one of a presence sensor, a motion sensor, image capturing unit such as a camera, infrared sensor, laser-based sensor, passive infrared and ultrasound sensor, loop detector and so on. The second sensor 706b includes image capturing unit such as a camera. In one example, the first sensor 706a mounts at the top of a toll booth 708. Here, the first sensor 706a detects presence of a vehicle 710 in its field of view or predefined area or line crossing depending on the need. Further, the second sensor 706b mounts at an inclination such that the second sensor 706b can capture images of the axle and/wheels of the vehicle 710 and the image of vehicle number plate.

[060] Now referring to FIG. 10, a method 800 for detecting and classifying vehicles is explained, in accordance with one exemplary embodiment of the present invention. The method 800 may be described in the general context of computer executable instructions. Generally, computer executable instructions may include routines, programs, objects, components, data structures, procedures, modules, functions, etc., that perform particular functions or implement particular abstract data types. The method 800 may also be practiced in a distributed computing environment where functions are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, computer executable instructions may be located in both local and remote computer storage media, including memory storage devices.

[061] The order in which the method 800 is described and is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method 800 or alternate methods. Additionally, blocks may be deleted from the method 800 without departing from the scope of the invention described herein. Furthermore, the method may be implemented in any suitable hardware, software, firmware, or combination thereof. However, for ease of explanation, in the embodiments described below, the method 800 may be implemented in the above-described system 702.

[062] At step 802, the system 702 employs the first sensor 706a to detect presence of the vehicle 710. Here, the system 702 activates the first sensor 706a. Once activated, the first sensor 706a continuously checks for presence of vehicle(s) 710 in its field of view or vehicles crossing a reference line, as shown at step 804. If the first sensor 706a does not detect the presence of the vehicle, then method 800 moves back to step 802. If the first sensor 706a detects the presence of the vehicle at the toll booth 708, then the method moves to steps 806. At step 806, the system 702 activates the second sensor 706b. As presented above, the second sensor 706b is a camera mounted at the front side of the toll booth 708 and captures images of the vehicle, axles and the vehicle number plate (step 808). Based on the number of image of the vehicle 710, the system 702 identifies the type of the vehicle 710 (step 810). Further, the system 702 counts the number of axles and/or wheels from the images (step 812). Here, the system 702 maps the data corresponding to the number of axles or wheels identified (step 812) and sub-classifies the type of vehicles, as shown at step 816. Consider that the system 702 identifies the vehicle as a car (i.e., four-wheeled vehicle) at steps 810 and 812. The system 702 employs a neural network or a machine learning model for classifying the type of vehicles based on the pre-stored data. For the above example, the system 12 employs the neural network and sub-classifies the vehicle as car, jeep, auto rickshaw based on the shape, number of axles identified by the second sensor 706b.

[063] At step 814, the system 702 employs the second sensor 706b to detect or recognize the vehicle number plate of the vehicle 710. The system 702 performs OCR text extraction on the license plate image as explained above. Further, the method 800 moves to step 818. At step 818, the system 702 presents the vehicle snapshot/image, image of the licence plate captured by second sensor 706b and the type of vehicle, axle count and date and time stamp identified for display on the user device, as explained above.

[064] A person skilled in the art understands that the system described herein provides several advantages and can be used in several applications. In one example, the system can be used for classifying vehicles on the road. In one example, the system can be used for separating commercial and passenger vehicles. In one example, the system can be used for identifying commercial and passenger vehicles for automatic category based tolling. In one example, the system can be used for detecting lane restriction violations for law enforcement. In one example, the system can be used for traffic analysis. In one example, the system can be used for security related applications ensuring limited/controlled access to vehicles.

[065] In addition, the system can be used for allocating parking for vehicles in a parking garage based on their size and available space. Further, the system can be used for vehicle speed measurement for law enforcement. Furthermore, the system can be used for classifying the vehicles from front, rear and side views.

[066] The system helps to categorize the vehicles into different categories as per established laws of different states or countries. Based on the vehicle’s appearance from various angles, number of axles, load carrying capacity and brands, the system classifies the vehicle based on their size, shape, load carrying capacity and any other required parameters.

[067] The presently disclosed system can be used in several applications. In one example, the system can be deployed at a toll booth to count the number of vehicles passing in a single direction and/or both directions. Further, the system can be used to categorize the vehicles passing by the toll booth. In addition, the system helps to determine the size of the vehicle using the axle/wheel count. Based on the size of the vehicle, the tariff can be charged for the vehicle.

[068] In another example, the system can be deployed at a structure/building, say at entry and exit points of a shopping mall or apartment. The system can be used to obtain the size of the vehicle entering the structure. After obtaining the size details of the vehicle, a designated parking space can be assigned for the vehicle. In addition, the system helps to determine the number of vehicles and type of vehicles present in the structure at any given point of time. Further, the time of entry and exit can be used to calculate tariff for parking the vehicle in the structure.

[069] In yet another example, the system can be deployed at entry and exit points of a manufacturing or mining entity or airport or seaport or any construction site. The system can be used to obtain the number of vehicles and type of vehicles present in the port, for example. The information corresponding to the type of vehicles is used to assign vehicles for carrying loads based on their size or load-carrying capacity.

[070] In yet another example, the system can be deployed for managing traffic on road or at a junction where two or more roads intersect. The system can be used to activate and capture an image of a vehicle when the vehicle passes when a stop signal/sign is ON and trigger an alert to concerned authorities. The system can be used to capture the licence plate image of the vehicle and extract the licence plate number as explained above. The images captured can be used to raise a ticket/charge to the owner of the vehicle.

[071] The present invention has been described in particular detail with respect to various possible embodiments, and those of skill in the art will appreciate that the invention may be practiced in other embodiments. First, the particular naming of the components, capitalization of terms, the attributes, data structures, or any other programming or structural aspect is not mandatory or significant, and the mechanisms that implement the invention or its features may have different names, formats, or protocols. Further, the system may be implemented via a combination of hardware and software, as described, or entirely in hardware elements. Also, the particular division of functionality between the various system components described herein is merely exemplary, and not mandatory; functions performed by a single system component may instead be performed by multiple components, and functions performed by multiple components may instead be performed by a single component.

[072] Some portions of the above description present the features of the present invention in terms of algorithms and symbolic representations of operations on information. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. These operations, while described functionally or logically, should be understood as being implemented by computer programs.

[073] Further, certain aspects of the present invention include process steps and instructions described herein in the form of an algorithm. It should be noted that the process steps and instructions of the present invention could be embodied in software, firmware or hardware, and when embodied in software, could be downloaded to reside on and be operated from different platforms used by real time network operating systems.

[074] The algorithms and operations presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may also be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these systems will be apparent to those of skill in the, along with equivalent variations. In addition, the present invention is not described with reference to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any references to specific languages are provided for disclosure of enablement and best mode of the present invention.

[075] It should be understood that components shown in FIGUREs are provided for illustrative purposes only and should not be construed in a limited sense. A person skilled in the art will appreciate alternate components that may be used to implement the embodiments of the present invention and such implementations will be within the scope of the present invention.