MACHINE VISION-BASED RAILWAY COLLISION PREVENTION FRAMEWORK

Abstract

ABSTRACT “MACHINE VISION-BASED RAILWAY COLLISION PREVENTION FRAMEWORK” The present invention addresses the challenge of enhancing railway safety by introducing a novel computer vision-based approach. The problem at hand involves the risk of head-on collisions resulting from multiple trains sharing the same track. Conventional safety measures often face limitations in accurately detecting and preventing such collisions, particularly in challenging environmental conditions such as fog, darkness, and rain. The need for proactive interventions and real-time monitoring to stop accidents before they happen further exacerbates the issue. By bridging the gap between traditional safety measures and emerging technological advancements, the proposed framework utilizing the latest machine vision techniques strives to significantly enhance railway safety. Figure 1

Specification

Description:TECHNICAL FIELD
[0001] The present invention relates to the field of automated systems for railway, and more particularly, the present invention relates to the Machine vision-based railway collision prevention framework.
BACKGROUND ART
[0002] The following discussion of the background of the invention is intended to facilitate an understanding of the present invention. However, it should be appreciated that the discussion is not an acknowledgment or admission that any of the material referred to was published, known, or part of the common general knowledge in any jurisdiction as of the application's priority date. The details provided herein the background if belongs to any publication is taken only as a reference for describing the problems, in general terminologies or principles or both of science and technology in the associated prior art.
[0003] The history of safety mechanisms in Indian Railways, including the introduction of the Kavach system, is marked by a continuous evolution aimed at enhancing the security and safety of passengers and railway assets. Here is an overview of the key developments:
[0004] Early Safety Measures (19th Century): When Indian Railways was first established in the 19th century, safety measures were rudimentary. Over time, basic practices such as the use of signals, brakes, and track inspections were introduced to prevent accidents and ensure safer operations.
[0005] Automatic Signaling (1920s): The 1920s witnessed the introduction of automatic signaling systems to manage train movements. This technology improved the control and coordination of trains, minimizing the risk of collisions.
[0006] Anti-Collision Devices (ACDs): In the late 1990s, Indian Railways introduced Anti-Collision Devices (ACDs). These innovative systems were designed to automatically apply brakes and stop trains in case of potential collisions, helping reduce accidents significantly.
[0007] Kavach Security System (2020s): Indian Railways Kavach is a modern security system introduced in recent years. This system leverages cutting-edge technology, including advanced CCTV cameras, sensors, and artificial intelligence, to monitor railway premises for potential security threats. Kavach is primarily focused on preventing unauthorized intrusions and detecting suspicious activities, providing real-time alerts and enhancing overall railway security.
[0008] Train Protection and Warning System (TPWS): TPWS, implemented in select sections, is another significant safety feature introduced in Indian Railways. It helps prevent accidents caused by human errors by automatically applying brakes when a train exceeds permissible speed limits.
[0009] Upgrade in Safety Standards: Besides these specific systems, Indian Railways has continually improved its safety standards, emphasizing better track maintenance, signaling technology, and safety awareness among railway personnel.
[0010] The history of safety mechanisms in Indian Railways is characterized by a gradual evolution from basic safety practices to highly advanced and sophisticated systems like Kavach. Indian Railways Kavach is a state-of-the-art, non-plagiarized security system that has been developed to enhance the safety and security of the extensive Indian railway network. This innovative system integrates modern technology and surveillance equipment to safeguard passengers, staff, and assets. Kavach combines elements like sensors, and satellite information to monitor and identify potential threats, including unauthorized intrusions, unusual movement activities. The introduction of Indian Railways Kavach exemplifies the commitment of the Indian Railways to passenger safety and security, while simultaneously modernizing and advancing the entire railway system.
[0011] However, Three trains crashed in the Balasore district of the eastern Indian state of Odisha on June 2, 2023. Near the Bahanaga Bazar railway station, the Coromandel Express entered the passing loop at full speed rather than the main line and collided with a goods train. The Coromandel Express was traveling at a high speed when it derailed, and three of its coaches ended up crashing into the approaching SMVT Bengaluru-Howrah Superfast Express on the nearby track. On October 11, this year, at around 9:30 p.m., a second North East Express disaster occurred in Buxar, Bihar, close to Raghunathpur station. At least four people were killed and numerous others were injured.
[0012] It motivates the scientific communities to further innovate and come up with additional surveillance mechanism. In this direction, we have proposed a machine vision-based railway collision prevention framework. It is not an alternative to the existing systems, but an augmentation of smart surveillance across the railway networks
[0013] In light of the foregoing, there is a need for Machine vision-based railway collision prevention framework that overcomes problems prevalent in the prior art associated with the traditionally available method or system, of the above-mentioned inventions that can be used with the presented disclosed technique with or without modification.
[0014] All publications herein are incorporated by reference to the same extent as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. Where a definition or use of a term in an incorporated reference is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies, and the definition of that term in the reference does not apply.
OBJECTS OF THE INVENTION
[0015] The principal object of the present invention is to overcome the disadvantages of the prior art by providing Machine vision-based railway collision prevention framework.
[0016] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that combines machine vision, real-time data analysis, and advanced warning mechanisms to proactively prevent train collisions.
[0017] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that has ability to accurately detect and predict potential collision scenarios between trains.
[0018] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that continuously monitors train movements, tracks occupancy, and identifies potential collision risks in real-time.
[0019] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that ensures timely alerts and proactive interventions to prevent accidents before they occur.
[0020] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that incorporates advanced data analysis capabilities, allowing for the seamless integration of multiple data sources such as train schedules, track maintenance records, and signaling system information.
[0021] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that can identify patterns, predict potential collision scenarios, and make intelligent recommendations to optimize train schedules and track utilization.
[0022] Another object of the present invention is to provide Machine vision-based railway collision prevention framework that contributes to improved operational efficiency, reduced delays, and enhanced overall railway performance.
[0023] The foregoing and other objects of the present invention will become readily apparent upon further review of the following detailed description of the embodiments as illustrated in the accompanying drawings.
SUMMARY OF THE INVENTION
[0024] The present invention relates to a machine vision-based railway collision prevention framework.
[0025] Railway transportation serves as a crucial artery of modern society, facilitating the efficient movement of people and goods across vast distances. However, the safety of railway operations remains a constant concern due to the potential risks associated with train collisions. Recent accidents, including the tragic collision in Odisha, underscore the urgent need for innovative solutions to prevent such incidents. This paper addresses the challenge of enhancing railway safety by introducing a novel computer vision-based approach. The problem at hand involves the risk of head-on collisions resulting from multiple trains sharing the same track. Conventional safety measures often face limitations in accurately detecting and preventing such collisions, particularly in challenging environmental conditions such as fog, darkness, and rain. The need for proactive interventions and real-time monitoring to stop accidents before they happen further exacerbates the issue. By bridging the gap between traditional safety measures and emerging technological advancements, the proposed framework utilizing the latest machine vision techniques strives to significantly enhance railway safety.
[0026] The proposed framework combines machine vision, real-time data analysis, and advanced warning mechanisms to proactively prevent train collisions. The key contribution of our proposed invention lies in its ability to accurately detect and predict potential collision scenarios between trains. By leveraging computer vision algorithms, including YOLOv5 object detection techniques, our framework continuously monitors train movements, tracks occupancy, and identifies potential collision risks in real-time. This ensures timely alerts and proactive interventions to prevent accidents before they occur. Furthermore, our invention incorporates advanced data analysis capabilities, allowing for the seamless integration of multiple data sources such as train schedules, track maintenance records, and signaling system information. By harnessing the power of data analytics and machine learning algorithms, the proposed framework can identify patterns, predict potential collision scenarios, and make intelligent recommendations to optimize train schedules and track utilization. This contributes to improved operational efficiency, reduced delays, and enhanced overall railway performance.
[0027] While the invention has been described and shown with reference to the preferred embodiment, it will be apparent that variations might be possible that would fall within the scope of the present invention.
BRIEF DESCRIPTION OF DRAWINGS
[0028] So that the manner in which the above-recited features of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may have been referred by embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.
[0029] These and other features, benefits, and advantages of the present invention will become apparent by reference to the following text figure, with like reference numbers referring to like structures across the views, wherein:
[0030] Figure 1 Schematic diagram of the proposed framework;
[0031] Figure 2 Used devices;
[0032] Figure. 3 Changes in IN, OUT and WARN variables based on different cases;
[0033] Figure 4 Flowchart representation of our proposed framework;
[0034] Figure 5 Output results of different cases from the proposed framework; and
[0035] Figure 6 Sample user interfaces (dashboards) for different cases generated by the proposed smart framework and real-time railway data.
DETAILED DESCRIPTION OF THE INVENTION
[0036] While the present invention is described herein by way of example using embodiments and illustrative drawings, those skilled in the art will recognize that the invention is not limited to the embodiments of drawing or drawings described and are not intended to represent the scale of the various components. Further, some components that may form a part of the invention may not be illustrated in certain figures, for ease of illustration, and such omissions do not limit the embodiments outlined in any way. It should be understood that the drawings and the detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents, and alternatives falling within the scope of the present invention as defined by the appended claim.
[0037] As used throughout this description, the word "may" is used in a permissive sense (i.e. meaning having the potential to), rather than the mandatory sense, (i.e. meaning must). Further, the words "a" or "an" mean "at least one" and the word "plurality" means "one or more" unless otherwise mentioned. Furthermore, the terminology and phraseology used herein are solely used for descriptive purposes and should not be construed as limiting in scope. Language such as "including," "comprising," "having," "containing," or "involving," and variations thereof, is intended to be broad and encompass the subject matter listed thereafter, equivalents, and additional subject matter not recited, and is not intended to exclude other additives, components, integers, or steps. Likewise, the term "comprising" is considered synonymous with the terms "including" or "containing" for applicable legal purposes. Any discussion of documents, acts, materials, devices, articles, and the like are included in the specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention.
[0038] In this disclosure, whenever a composition or an element or a group of elements is preceded with the transitional phrase "comprising", it is understood that we also contemplate the same composition, element, or group of elements with transitional phrases "consisting of", "consisting", "selected from the group of consisting of, "including", or "is" preceding the recitation of the composition, element or group of elements and vice versa.
[0039] The present invention is described hereinafter by various embodiments with reference to the accompanying drawing, wherein reference numerals used in the accompanying drawing correspond to the like elements throughout the description. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiment set forth herein. Rather, the embodiment is provided so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those skilled in the art. In the following detailed description, numeric values and ranges are provided for various aspects of the implementations described. These values and ranges are to be treated as examples only and are not intended to limit the scope of the claims. In addition, several materials are identified as suitable for various facets of the implementations. These materials are to be treated as exemplary and are not intended to limit the scope of the invention.
[0040] The present invention relates to a machine vision-based railway collision prevention framework.
[0041] The proposed system will use an array of cameras (installed on pole at a heightened place) and deep learning-based object detection techniques as form of a framework. A camera keeps track of the railway-tracks at a specific location (with fixed latitude and longitude). It also detects trains/obstacles and computes whether there is any probability of an accident. If the propose framework learns that accident is probable, it sends the information (including the range) to the concerned stakeholders. A schematic diagram of the proposed framework has been shown in Figure 1. The diagram has two railway stations (SA and SB), four cameras (CN1T1, CN2T2, CM1T1, and CM1T2), the coverage ranges (D1, D2; where, D1=D2). Here, we have assumed that the proposed framework is working between SA and SB. Each camera covers D/2 (D1 or D2). If a train exits from the coverage of a camera, it will get into the coverage of another camera. CM1T1 and CM2T1 covers total D distance (D=D1+D2) and T1 indicates railway track number.
[0042] The proposed system will also include a user-friendly dashboard for the authorities that display the real-time train location and any obstacles on the railway track. Here, the authorities include station masters, signal officers of stations A and B, and drivers of incoming and outgoing trains. It will help the concerned parties take the necessary and preventive steps to stop or avoid derailment or collision. The system was validated through simulation and demonstrated its reliability in any environmental situation, such as fog, rain, or darkness. The proposed framework has been designed to add another layer to the currently used collision prevention mechanism, keeping the feasibility of implementation in mind.
[0043] It is done via the Raspberry Pi 4 module and a 5-megapixel (MP) native resolution sensor Pi camera (that can capture static images of dimention 2592 pixel × 1944 pixel). The continuous video feed is captured by the Pi camera in the Rasberry Pi 4 module. The open-source YOLOv51 detects and ByteTrack2 the trains or any vehicles on the track at railway crossings. The non-vehicles are detected by using traditional image processing on the railway tracks. Different cases are tested using real-world video analysis and dummy simulations. The devices are shown in Figure 2 (a) and (b).
[0044] The frames of the video feed are converted from a BGR image to an RGB image, and the initialized holistic model is then used to make predictions. The model's predictions are stored in the list variable to examine the following conditions:
- Train is detected in normal day light
- Train is detected in night (only light source is train-engine head-light)
- Train is detected in rain
- Train is detected in fog or smog (mostly in winter or pollution)
- Obstacle (both vehicles and non-vehicles) is detected on railway track
[0045] The cameras CN1T2 and CN2T2 work accordingly on the railway track number T2. A train enters camera coverage, the IN and OUT counters is set to 1 and 0, respectively. Again, a train exits camera coverage, the IN and OUT counters are set to 0 and 1, respectively (see Figure 3a). Similarly, WARN variable has been set (1) or reset (default: 0) if the camera detects any obstacle on the same railway track (refer to Figure 3b). The complete working principle has been shown in Figure 4 as a flowchart.
[0046] The proposed framework combines machine vision, real-time data analysis, and advanced warning mechanisms to proactively prevent train collisions. The key contribution of our proposed invention lies in its ability to accurately detect and predict potential collision scenarios between trains. By leveraging computer vision algorithms, including YOLOv5 object detection techniques, our framework continuously monitors train movements, tracks occupancy, and identifies potential collision risks in real-time. This ensures timely alerts and proactive interventions to prevent accidents before they occur. Furthermore, our invention incorporates advanced data analysis capabilities, allowing for the seamless integration of multiple data sources such as train schedules, track maintenance records, and signaling system information. By harnessing the power of data analytics and machine learning algorithms, the proposed framework can identify patterns, predict potential collision scenarios, and make intelligent recommendations to optimize train schedules and track utilization. This contributes to improved operational efficiency, reduced delays, and enhanced overall railway performance. Some simulation results are displayed in Figure 5. The real-time data analytics can be realized through Figure 6 UIs.
[0047] The proposed research presents several novel contributions to the field of railway safety and accident prevention. The key novelties of this work are outlined below:
- 1. Integration of Computer Vision and Advanced Algorithms: This research integrates computer vision techniques, specifically leveraging the YOLOv object detection framework, with advanced tracking algorithms such as Bytetracker. This combination enables real-time and accurate detection of trains, as well as robust tracking across consecutive video frames. By integrating these cutting-edge technologies, our proposed system surpasses traditional methods of train detection and tracking, significantly improving the overall effectiveness and reliability of railway safety systems.
- 2. Comprehensive Train Track Labeling: The novel approach introduced in this research involves labeling each track with the number of incoming and outgoing trains, denoted as "IN=x" and "OUT=y," respectively, before and after a threshold line. This detailed track labeling provides a precise understanding of train movements and enables the system to issue timely alerts and prevent head-on clashes between trains. The track labeling mechanism enhances the granularity of train track monitoring, contributing to a more proactive and efficient accident prevention system.
- 3. Adaptable Detection in Challenging Conditions: This research addresses the challenge of detecting trains in challenging environmental conditions, including darkness, fog, and other adverse scenarios. By developing and implementing specialized computer vision algorithms, our proposed system exhibits enhanced robustness and adaptability. It ensures reliable train detection in various lighting and weather conditions, significantly improving the overall effectiveness and reliability of the accident prevention system.
- 4. Proactive Alert System: The incorporation of a proactive alert system within the proposed solution represents a novel contribution to the field. By accurately tracking the number of incoming and outgoing trains on each track and comparing it against predefined thresholds, our system generates timely alerts when the limit is exceeded. This proactive approach allows for the swift implementation of preventive measures, minimizing the risk of accidents and ensuring the safety of passengers and railway personnel.
- 5. Real-time Monitoring and Decision Support: The proposed system provides real-time monitoring and decision support capabilities. By continuously analyzing train movements, track occupancy, and other relevant data, the system facilitates informed decision-making for railway authorities. The integration of data analytics and machine learning algorithms enables the system to identify patterns, predict collision risks, and optimize train schedules and track utilization. This proactive decision support contributes to enhanced operational efficiency, reduced delays, and improved overall railway performance.
[0048] In summary, the proposed research presents several novel contributions to the field of railway safety. These include the integration of advanced computer vision algorithms, comprehensive train track labeling and adaptability to challenging conditions, a proactive alert system, and real-time monitoring with decision support capabilities. These novelties collectively advance the state-of-the-art in train collision prevention systems, offering an innovative and effective approach to enhancing railway safety and preventing accidents.
[0049] The proposed system provides a real-time comprehensive approach to enhance railway safety through the development of a computer vision-based train collision prevention system. By leveraging advanced algorithms, such as the YOLOv5 object detection framework and ByteTrack for tracking, our system offers real-time train detection, accurate track labeling, and proactive alert mechanisms. The integration of specialized computer vision algorithms ensures reliable train detection even in challenging environmental conditions, such as darkness and fog.
[0050] The key contributions of this research lie in the integration of computer vision and advanced algorithms, the comprehensive track labeling mechanism, the adaptability to challenging conditions, the implementation of a proactive alert system, and the provision of real-time monitoring and decision support capabilities. These novelties collectively push the boundaries of railway safety systems, addressing the critical issue of train collisions and striving to prevent accidents before they occur. By implementing the proposed train collision prevention system, railway authorities can proactively monitor train tracks, identify potential collision risks, and take timely preventive actions. This will significantly enhance the safety of passengers and railway personnel, minimize the risk of head-on clashes between trains, and contribute to a safer and more efficient railway transportation system.
[0051] It is important to note that the success of this proposed system relies on collaboration with railway authorities, continuous testing and evaluation, and incorporation of feedback from industry experts. Further research and development efforts are necessary to refine and optimize the system, considering the specific requirements and operational constraints of different railway networks. Ultimately, the potential impact of this research extends beyond the prevention of train collisions. By improving railway safety, the proposed system can foster public confidence in rail travel, promote efficient and reliable transportation, and contribute to sustainable development by encouraging modal shifts from road to rail.
[0052] The proposed framework, it can pave the way for a safer, more reliable, and efficient railway system that prioritizes the well-being and security of passengers, personnel, and the communities it serves.
[0053] Various modifications to these embodiments are apparent to those skilled in the art from the description and the accompanying drawings. The principles associated with the various embodiments described herein may be applied to other embodiments. Therefore, the description is not intended to be limited to the 5 embodiments shown along with the accompanying drawings but is to be providing the broadest scope consistent with the principles and the novel and inventive features disclosed or suggested herein. Accordingly, the invention is anticipated to hold on to all other such alternatives, modifications, and variations that fall within the scope of the present invention and appended claims. , Claims:CLAIMS
We Claim:
1) An automated railway collision prevention system, the system comprising:
- an array of cameras mounted at elevated positions along railway tracks at predetermined intervals, each camera configured to monitor a specific railway track segment, detect obstacles and/or trains, and capture real-time video data;
- a deep learning-based object detection module configured to process the video data from each camera to identify trains and obstacles on the railway tracks, assess their movements, and predict potential collision scenarios;
- a data processing unit comprising a trained deep learning model that detects and classifies objects within the video feed, calculates a probability of collision, and transmits alerts if a collision is predicted;
- a communication module configured to relay real-time alerts, including the location and range of the potential collision, to stakeholders, including railway authorities, station personnel, and train drivers;
- a user interface that displays real-time location data and alerts on a dashboard accessible to authorized railway personnel, enabling preventive actions to avoid collisions.
2. The system as claimed in claim 1, wherein the cameras are spaced such that the coverage of adjacent cameras overlaps by a predetermined distance to ensure uninterrupted surveillance of the railway track.
3. The system as claimed in claim 1, wherein each camera covers a distance range, D/2D/2D/2, on each side, and adjacent cameras cover the complete range DDD in total, ensuring continuous monitoring as a train moves out of the coverage area of one camera and into another.
4. The system as claimed in claim 1, wherein the system further comprising a Raspberry Pi 4 module and a 5-megapixel camera sensor configured to capture high-resolution images at 2592 x 1944 pixels, with the system configured to process continuous video feeds for object detection using a YOLOv5 model and ByteTrack tracking algorithms.
5. The system as claimed in claim 1, wherein the system further comprising a counter for each track segment to record train entries and exits, with variables 'IN' and 'OUT' representing entry and exit events, respectively, and a 'WARN' variable to indicate detected obstacles on the track.
6. The system as claimed in claim 1, wherein the object detection module operates in a variety of environmental conditions, including low-light, fog, and rain, and the trained deep learning model is capable of detecting trains using train engine headlights as the primary light source at night.
7. The system as claimed in claim 1, wherein the user interface displays comprehensive information regarding train locations, track occupancy, and real-time alerts for potential collisions, enabling railway personnel to make informed, real-time decisions.
8. The system as claimed in claim 1, wherein the system further comprising a data analytics module that integrates multiple data sources, including train schedules, track maintenance records, and signaling system data, to predict potential collision risks and optimize train schedules and track utilization.
9. A method for railway collision prevention, the method comprising:
- capturing real-time video feeds of railway tracks using an array of cameras positioned along the track;
- detecting trains and obstacles within the captured video using deep learning-based object detection algorithms;
- calculating collision probabilities based on the detected objects and their movements;
- transmitting real-time alerts to stakeholders if a collision probability threshold is met; and
- displaying real-time data on a user dashboard accessible to railway personnel to support proactive collision prevention and efficient decision-making.

Documents