CROSS-REFERENCE TO RELATED APPLICATIONS AND PRIORITY
[001] The present application does not claim priority from any patent application.
TECHNICAL FIELD
[002] The present disclosure in general relates to the field of image processing, machine learning, and deep learning procedures. More particularly, the present invention relates to a system and method to processing images for detecting hurdles on highways and traffic roads.
BACKGROUND
[003] Traveling through hilly and landslide prone zones have been a nightmare to travellers for a long time. If the roads are suddenly blocked due to natural disaster like landslide, fallen trees, debris or by herd/wildlife movement, there is no such predictive mechanism which can notify the travellers much before they reach to the actual point of disaster. Recent events about the landslide and increasing number of wildlife animals due to highways widening and forest cutting has attracted as much as 1 million roads killing per day of animals all over the world.
[004] Highways going through canopy and wildlife crossing, hilly areas and landslide prone zones are the highest accidental points. This problem becomes more severe if it occurs in night when the visibility is poor. If warning systems can alert travellers for such events in advance, the travellers can adopt an alternative route or halts until the event is clear.
SUMMARY
[005] Before the present systems and method for real-time detection of one or more hurdles in a geographical area is illustrated. It is to be understood that this application is not limited to the particular system, and methodologies described, as there can be multiple possible embodiments that are not expressly illustrated in the present disclosure. It is also to be understood that the terminology used in the description is for the purpose of describing the particular versions or embodiments only, and is not intended to limit the scope of the present application. This summary is provided to introduce concepts related to systems and method for real-time detection of one or more hurdles in a geographical area. This summary is not intended to identify essential features of the claimed subject matter nor is it intended for use in determining or limiting the scope of the claimed subject matter.
3
[006] In another implementation, a system for real-time detection of one or more hurdles in a geographical area. The system comprises a memory and a processor coupled to the memory, further the processor is configured to execute programmed instructions stored in the memory. In one embodiment, the processor may execute programmed instructions stored in the memory for processing a set of sample video frames, using machine learning algorithm, to generate a trained network model. The trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area, wherein the set of sample video frames may comprise one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time. In one embodiment, the processor may execute programmed instructions stored in the memory for receiving a set of video frames, captured in real-time, by one or more video surveillance cameras placed in the geographical area. The set of video frames may be associated with one or more roads in the geographical area, wherein the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time. In one embodiment, the processor may execute programmed instructions stored in the memory for analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area, wherein deep learning algorithm is configured to process the set of video frames based on the trained network model. In one embodiment, the processor may execute programmed instructions stored in the memory for identifying one or more users in the geographical area that are affected by the hurdle, wherein the one or more users are identified by analysing the hurdle based on a set of predefined parameters. In one embodiment, the processor may execute programmed instructions stored in the memory for transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle.
[007] In one implementation, a method for real-time detection of one or more hurdles in a geographical area is illustrated. The method may comprise steps for processing a set of sample video frames, using machine learning algorithm, to generate a trained network model. The trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area, wherein the set of sample video frames may comprise one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and one or more sample video frames, corresponding to each
4
hurdle from the set of hurdles, captured during night time. The method may further comprise steps for receiving a set of video frames, captured in real-time, by one or more video surveillance cameras placed in the geographical area. The set of video frames may be associated with one or more roads in the geographical area, wherein the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time. The method may further comprise steps for analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area, wherein deep learning algorithm is configured to process the set of video frames based on the trained network model. The method may further comprise steps for identifying one or more users in the geographical area that are affected by the hurdle, wherein the one or more users are identified by analysing the hurdle based on a set of predefined parameters. The method may further comprise steps for transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle.
[008] In yet another implementation, a computer program product having embodied computer program for real-time detection of one or more hurdles in a geographical area is disclosed. The program may comprise a program code for processing a set of sample video frames, using machine learning algorithm, to generate a trained network model. The trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area, wherein the set of sample video frames may comprise one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time. The program may comprise a program code for receiving a set of video frames, captured in real-time, by one or more video surveillance cameras placed in the geographical area. The set of video frames may be associated with one or more roads in the geographical area, wherein the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time. The program may comprise a program code for analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area, wherein deep learning algorithm is configured to process the set of video frames based on the trained network model. The program may comprise a program code for identifying one or more users in the geographical area that are affected by the hurdle, wherein the one or more users are
5
identified by analysing the hurdle based on a set of predefined parameters. The program may comprise a program code for transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle.
BRIEF DESCRIPTION OF DRAWINGS
[009] The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to refer like features and components.
[0010] Figure 1 illustrates a network implementation of a system configured for real-time detection of one or more hurdles in a geographical area using deep learning and a convolution neural network, in accordance with an embodiment of the present subject matter.
[0011] Figure 2 illustrates the system configured for real-time detection of one or more hurdles in a geographical area using deep learning and a convolution neural network, in accordance with an embodiment of the present subject matter.
[0012] Figure 3 illustrates a method for real-time detection of one or more hurdles in a geographical area using deep learning and a convolution neural network, in accordance with an embodiment of the present subject matter.
DETAILED DESCRIPTION
[0013] Some embodiments of the present disclosure, illustrating all its features, will now be discussed in detail. The words “receiving”, “processing”, “analysing”, “identifying”, “transmitting”, and other forms thereof, are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms "a", "an" and "the" include plural references unless the context clearly dictates otherwise. Although any systems and methods similar or equivalent to those described herein can be used in real-time detection of one or more hurdles in a geographical area, the exemplary, systems and method for real-time detection of one or more hurdles is
6
now described. The disclosed embodiments of the system and method for real-time detection of one or more hurdles in a geographical area using deep learning and convolution neural network are merely exemplary of the disclosure, which may be embodied in various forms.
[0014] Various modifications to the embodiment will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. However, one of ordinary skill in the art will readily recognize that the present disclosure for real-time detection of one or more hurdles in a geographical area is not intended to be limited to the embodiments illustrated, but is to be accorded the widest scope consistent with the principles and features described herein.
[0015] The system enables analysis of images/ videos captured from highway surveillance system and perform predictive analysis for detection of hazards like landslides, tree falling, blockage of roads due to movement of wildlife or blocked by debris. Using image processing based computer vision and deep learning algorithms, these hazards can be detected much before a traveller reach to the point of destination. The proposed system detects such blockages in day time as well as during night time. If there is a movement of herd and animals, that can also be detected with the proposed system and as per their estimated time of arrival to the crossroads. Such affected zones can be flagged and travellers falling under such zones is sent a precautionary alarm to take preventive actions like an alternative path to follow or halt their journey till the events are clear. The proposed system makes the highways safe for both wildlife and human being in day and night, by using the existing powerful surveillance resources and available bandwidth with system.
[0016] In one embodiment, high-resolution surveillance cameras are facilitated on the highway. Camera captures the movements of herd, debris, landslides or blockage events on the highway roads and sends this image to the system for analysis. The system at remote location or on cloud is enabled with deep learning algorithms and image processing techniques which are capable in identifying the wildlife movement, debris, landslides, objects, and animals on the roads. Once the surveillance system identifies such movements it generates and sends a warning to the travellers which are in closed vicinity of such zones.
[0017] In one embodiment, the surveillance cameras are enabled with 360° views, night vision and able to read the number plates with reasonable good accuracy. The number plates
7
can also be used for identifying the passengers traveling through the geographical area. The surveillance cameras are connected to a video surveillance system through high speed internet services to send the images to the system. The existing surveillance system and network may also be used for capturing the images/ videos in real-time.
[0018] In one embodiment, the system is enabled with machine learning and deep learning algorithms. The system is configured to receive the image and video data from the surveillance system and process the images using Deep learning and Machine Learning trained models. The Machine learning algorithms enables a trained network models based on a set of training images. These trained network models are capable of animal detection, herd detection and movement detection in input images and videos. Based on the detection, the machine learning algorithms can give the predictive analysis. Based on the prediction, the system may generate alerts to warn the travellers expecting to reach such locations within the geographical area.
[0019] In one embodiment, the network model is trained based on day time images and night time images. In order to train the network model for night time image recognition, the model is trained based on shape of the objects, as the night vision mode, doesn’t enable the prominent features of an object like texture, colour and surface localization, so the local features are not reliable for object recognition. Hence, the training database is built over the shape based parameters, where only high-level abstraction of shapes is feed into the network for training.
[0020] In one embodiment, machine learning algorithms provided the prior knowledge of the shapes which ease the decision making of a deep learning algorithm and improves the final score of detection and classification. The possible shapes are provided the bios factor based on the geographies and existence of an object in a region. For example, if a forest region is expecting an Elephants in the region, and Deep Learning algorithms are detection with the score say X, then an additional bios factor B is provided to make the detection with the accuracy of X+B.
[0021] In one embodiment, for real-time prediction of hazards, the system may be implemented at the video surveillance system. In another embodiment, the position of the cloud server machine/ host machine and image processing algorithms are deployed at the virtual machine environment or the remote host.
8
[0022] In one embodiment, for real time prediction of hazards the system is configured to detect movement of wildlife, debris, landslides and obstacles on highways. In case of Animal/Herd Detection, the surveillance camera (A) is installed along the Highway. A herd of animals (B) is approaching to the highway. Following activities are performed by system:
- Using the Machine Learning methods, the herd of animals is detected. The estimated animal count and type of animals are extracted from the Herd.
- Using the Image Processing techniques, the distance (distance-x) from the camera is calculated.
- Using Image processing technique, the drift velocity of the herd towards the crossroads is measured.
- Using velocity and distance the prediction about estimated time of arrival (ETA1) of herd to the crossroads boundaries is calculated.
[0023] In one embodiment, in case of debris or landslide detection, the following activities are performed by system:
- The system stores the image of highway when there is no debris/landslide.
- System samples the images with the configurable scan-rate and compare the present images with the stored image.
- Both the images, sampled and stored, are compared and difference is measured using image processing techniques.
- Once the difference characteristics found in images are met with the problem scenarios, the alarm is triggered from the system.
- The alarm is communicated to the closed vicinity group of camera to create the alert zone.
- Vehicles inside the alert-zone are communicated about the blockage on the highways.
[0024] In one embodiment, notifications are sent to the affected travellers. The machine learning algorithms calculate the estimated time of arrival of the herd/animal based on the predictive analysis. Further, the system starts communication with its neighbouring cameras based on the calculation of the distance between the traveller and the herd/ animal. An alert message is posted for the travellers under the alert zone. Further, the network implementation of system configured for real-time detection of one or more hurdles in a geographical area is illustrated with Figure 1.
9
[0025] Referring now to Figure 1, a network implementation 100 of a system 102 for real-time detection of one or more hurdles in a geographical area is disclosed. Although the present subject matter is explained considering that the system 102 is implemented on a server, it may be understood that the system 102 may also be implemented in a variety of computing systems, such as a laptop computer, a desktop computer, a notebook, a workstation, a mainframe computer, a server, a network server, and the like. In one implementation, the system 102 may be implemented over a server. Further, the system 102 may be implemented in a cloud network. The system 102 may further be configured to communicate with a video surveillance system 108. The video surveillance system 108 may be configured to receive one or more videos/ images captured by one or more video surveillance cameras 110. Further the video surveillance system 108 may transmit the one or more images to the system 102 for further processing. In one embodiment, the system 102 may be part of the video surveillance system 108.
[0026] Further, it will be understood that the system 102 may be accessed by multiple users through one or more user devices 104-1, 104-2…104-N, collectively referred to as user device 104 hereinafter, or applications residing on the user device 104. Examples of the user device 104 may include, but are not limited to, a portable computer, a personal digital assistant, a handheld device, and a workstation. The user device 104 may be communicatively coupled to the system 102 through a network 106.
[0027] In one implementation, the network 106 may be a wireless network, a wired network or a combination thereof. The network 106 may 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 106 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), Hypertext Transfer Protocol Secure(HTTPS), File Transfer Protocol(FTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), and the like, to communicate with one another. Further, the network 106 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, and the like. In one embodiment, the system 102 may be configured to receive one or more images from the video surveillance system 108. Once the system 102 receives the one or
10
more images, the system 102 is configured to process the one or more images as described with respect to figure 2.
[0028] Referring now to figure 2, the system 102 is configured for real-time detection of one or more hurdles in a geographical area is illustrated in accordance with an embodiment of the present subject matter. In one embodiment, the system 102 may include at least one processor 202, an input/output (I/O) interface 204, and a memory 206. The at least one 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, at least one processor 202 may be configured to fetch and execute computer-readable instructions stored in the memory 206.
[0029] The I/O interface 204 may include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like. The I/O interface 204 may allow the system 102 to interact with the user directly or through the user device 104. Further, the I/O interface 204 may enable the system 102 to communicate with other computing devices, such as web servers and external data servers (not shown). The I/O interface 204 may facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. The I/O interface 204 may include one or more ports for connecting a number of devices to one another or to another server.
[0030] The memory 206 may include any 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 memory 206 may include modules 208 and data 210.
[0031] The modules 208 may include routines, programs, objects, components, data structures, and the like, which perform particular tasks, functions or implement particular abstract data types. In one implementation, the module 208 may include a data pre-processing module 212, a video surveillance module 214, a hurdle detection module 216, a user identification module 218, a alert generation module 220, and other modules 222.
11
The other modules 222 may include programs or coded instructions that supplement applications and functions of the system 102.
[0032] The data 210, amongst other things, serve as a repository for storing data processed, received, and generated by one or more of the modules 208. The data 210 may also include a central data 228, and other data 230. In one embodiment, the other data 230 may include data generated as a result of the execution of one or more modules in the other modules 220. In one implementation, a user may access the system 102 via the I/O interface 204. The user may be registered using the I/O interface 204 in order to use the system 102. In one aspect, the user may access the I/O interface 204 of the system 102 for obtaining information, providing input information or configuring the system 102. The functioning of all the modules in the system 102 is described as below:
DATA PREPROCESSING MODULE 212
[0033] In one embodiment, the data pre-processing module 212 may be configured processing a set of sample video frames, using machine learning algorithm, to generate a trained network model. The trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area. The set of sample video frames comprise one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time. In one embodiment, the set of hurdles at least comprises road blocks, landslides, debris, objects, animal migration, and herds.
VIDEO SURVEILLANCE MODULE 214
[0034] In one embodiment, the video surveillance module 214 is configured for receiving a set of video frames, captured in real-time, by one or more video surveillance cameras placed in the geographical area. The set of video frames are associated with one or more roads in the geographical area. In one embodiment, the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time.
12
HURDLE DETECTION MODULE 216
[0035] In one embodiment, the hurdle detection module 216 is configured for analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area. The deep learning algorithm is configured to process the set of video frames based on the trained network model. In one embodiment, the first subset of video frames is analysed by using the deep learning algorithm based on a texture, a colour and a surface localization of the hurdle on the geographical area or by using the generative adversarial networks to determine the hurdle. The generative adversarial networks is a type of neural network. Further, the second set of video frames is analysed by the deep learning algorithm based on shape of the hurdle in the geographical area to determine the hurdle.
USER IDENTIFICATION MODULE 218
[0036] In one embodiment, the user identification module 218 is configured for identifying one or more users in the geographical area that are affected by the hurdle. In one embodiment, the one or more users are identified by analysing the hurdle based on a set of predefined parameters. The set of predefined parameters comprise, speed on the user, distance between the hurdle and the user, direction of movement of the user, and type of hurdle. Furthermore, the one or more users in the vicinity of the hurdle are identified based on GPS signal received from the user’s electronic/ mobile device 104, image processing of the set of set of video frames captured in real-time, or network signal received from the user’s mobile device 104.
ALERT GENERATION MODULE 220
[0037] Finally, the alert generation module 220 is configured for transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle. The alerts may be in the form of an SMS, warning call, notifications, and the like. Further, method for real-time detection of one or more hurdles in a geographical area is illustrated with respect to figure 3.
[0038] Referring now to figure 3, a method 300 for real-time detection of one or more hurdles in a geographical area, is disclosed in accordance with an embodiment of the
13
present subject matter. The method 300 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, functions, and the like, 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.
[0039] The order in which the method 300 is described 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, individual blocks may be deleted from the method 300 without departing from the spirit and scope of the subject matter described herein. Furthermore, the method 300 can 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 considered to be implemented in the above described system 102.
[0040] At block 302, the data pre-processing module 212 may be configured processing a set of sample video frames, using machine learning algorithm, to generate a trained network model. The trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area. The set of sample video frames comprise one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time. In one embodiment, the set of hurdles at least comprises road blocks, landslides, debris, animal migration, and herds.
[0041] At block 304, the video surveillance module 214 is configured for receiving a set of video frames, captured in real-time, by one or more video surveillance cameras placed in the geographical area. The set of video frames are associated with one or more roads in the geographical area. In one embodiment, the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time.
14
[0042] At block 306, the hurdle detection module 216 is configured for analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area. The deep learning algorithm is configured to process the set of video frames based on the trained network model. In one embodiment, the first subset of video frames is analysed by using the deep learning algorithm based on a texture, a colour and a surface localization of the hurdle on the geographical area or by using the generative adversarial networks to determine the hurdle. The generative adversarial networks is a type of neural network. Further, the second set of video frames is analysed by the deep learning algorithm based on shape of the hurdle in the geographical area to determine the hurdle.
[0043] At block 308, the user identification module 218 is configured for identifying one or more users in the geographical area that are affected by the hurdle. In one embodiment, the one or more users are identified by analysing the hurdle based on a set of predefined parameters. The set of predefined parameters comprise, speed on the user, distance between the hurdle and the user, direction of movement of the user, and type of hurdle. Furthermore, the one or more users in the vicinity of the hurdle are identified based on GPS signal received from the user’s electronic/ mobile device 104, image processing of the set of set of video frames captured in real-time, or network signal received from the user’s mobile device 104.
[0044] At block 310, the alert generation module 220 is configured for transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle. The alerts may be in the form of an SMS, warning call, notifications, and the like.
[0045] Although implementations for systems and methods for real-time detection of one or more hurdles in a geographical area has been described, it is to be understood that the appended claims are not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed as examples of implementations for real-time detection of one or more hurdles in a geographical area.

WE CLAIM:
1. A system for real-time detection of one or more hurdles in a geographical area, the system comprising:
a memory; and
a processor coupled to the memory,
wherein the processor is configured to execute programmed instructions stored in the memory for:
processing a set of sample video frames, using machine learning algorithm, to generate a trained network model, wherein the trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area, wherein the set of sample video frames comprise
one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and
one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time;
receiving a set of video frames, captured in real-time, by one or more video surveillance cameras, wherein one or more video surveillance cameras are placed in the geographical area, wherein the set of video frames are associated with one or more roads in the geographical area, wherein the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time;
analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area, wherein deep learning algorithm is configured to process the set of video frames based on the trained network model;
identifying one or more users in the geographical area that are affected by the hurdle, wherein the one or more users are identified by analysing the hurdle based on a set of predefined parameters; and
16
transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle.
2. The system of claim 1, wherein the set of hurdles at least comprises road blocks, landslides, debris, objects, animal migration, and herds.
3. The system of claim 1, wherein the first subset of video frames is analysed by using the deep learning algorithm based on a texture, a colour and a surface localization of the hurdle on the geographical area or by using the generative adversarial networks to determine the hurdle.
4. The system of claim 1, wherein the second set of video frames is analysed by the deep learning algorithm based on shape of the hurdle in the geographical area to determine the hurdle.
5. The system of claim 1, wherein the one or more users in the vicinity of the hurdle are identified based on GPS signal received from the user’s electronic device, image processing of the set of set of video frames captured in real-time, or network signal received from the user’s mobile device.
6. The system of claim 1, wherein the set of predefined parameters comprise, speed on the user, distance between the hurdle and the user, direction of movement of the user, and type of hurdle.
7. A method for real-time detection of one or more hurdles in a geographical area, the method comprising:
processing, by a processor, a set of sample video frames, using machine learning algorithm, to generate a trained network model, wherein the trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area, wherein the set of sample video frames comprise
one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and
17
one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time;
receiving, by the processor, a set of video frames captured in real-time by one or more video surveillance cameras, wherein one or more video surveillance cameras are placed in the geographical area, wherein the set of video frames are associated with one or more roads in the geographical area, wherein the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time;
analysing, by the processor, the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area, wherein deep learning algorithm is configured to process the set of video frames based on the trained network model;
identifying, by the processor, one or more users in the geographical area that are affected by the hurdle, wherein the one or more users are identified by analysing the hurdle based on a set of predefined parameters; and
transmitting, by the processor, at least one alert to one or more users, in the geographical area, that are affected by the hurdle.
8. The method of claim 7, wherein the set of hurdles at least comprises road blocks, landslides, debris, objects, animal migration, and herds.
9. The method of claim 7, wherein the first subset of video frames is analysed by using the deep learning algorithm based on a texture, a colour and a surface localization of the hurdle on the geographical area or by using the generative adversarial networks to determine the hurdle.
10. The method of claim 7, wherein the second set of video frames is analysed by the deep learning algorithm based on shape of the hurdle in the geographical area to determine the hurdle.
18
11. The method of claim 7, wherein the one or more users in the vicinity of the hurdle are identified based on GPS signal received from the user’s electronic device, image processing of the set of set of video frames captured in real-time, or network signal received from the user’s mobile device.
12. The method of claim 7, wherein the set of predefined parameters comprise, speed on the user, distance between the hurdle and the user, direction of movement of the user, and type of hurdle.
13. A computer program product having embodied thereon a computer program for real-time detection of one or more hurdles in a geographical area, the computer program product comprises:
a program code for processing a set of sample video frames, using machine learning algorithm, to generate a trained network model, wherein the trained network model enables detection of one or more hurdles from a set of hurdles, in a geographical area, wherein the set of sample video frames comprise
one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during day time, and
one or more sample video frames, corresponding to each hurdle from the set of hurdles, captured during night time;
a program code for receiving a set of video frames captured in real-time by one or more video surveillance cameras, wherein one or more video surveillance cameras are placed in the geographical area, wherein the set of video frames are associated with one or more roads in the geographical area, wherein the set of video frames comprise at least one of a first subset of video frames captured during day time, and a second subset of video frames captured during night time;
a program code for analysing the set of video frames, using one or more deep learning algorithm, to identify at least one hurdle from the set of hurdles in the geographical area, wherein deep learning algorithm is configured to process the set of video frames based on the trained network model;
19
a program code for identifying one or more users in the geographical area that are affected by the hurdle, wherein the one or more users are identified by analysing the hurdle based on a set of predefined parameters; and
a program code transmitting at least one alert to one or more users, in the geographical area, that are affected by the hurdle.