A rail detection system based on three-dimensional imaging

Abstract

The present invention relates to a rail inspection system based on three-dimensional imaging, comprising a mobile inspection device equipped with a central processing unit (CPU) and a three-dimensional imaging system, a positioning system, a laser ranging system, and a communication module, all connected to the CPU. The three-dimensional imaging system generates a three-dimensional model of the rail in real time as the mobile inspection device moves. Compared with existing technologies, the present invention utilizes the three-dimensional imaging system and the laser ranging system to detect a wider range of common rail faults, improving detection efficiency and significantly reducing personnel workload. Ultrasonic detection technology is combined to form a redundant system to determine the correctness of detection results, ensuring accurate and reliable detection results. A remote server is used to store rail inspection information in real time, and big data analysis is used to analyze rail status information and provide maintenance and repair recommendations.

Description

Technical Field

[0001] The present invention relates to the technical field of railway track detection, and in particular to a railway track detection system based on three-dimensional imaging. Background Art

[0002] In today's society, the subway has become an indispensable means of transportation, especially in large, populous cities. The development of the subway directly impacts the city's further development. Therefore, we must strengthen rail maintenance to extend its service life while ensuring people's personal safety. We must prioritize rail quality, regularly inspect it, and promptly address any problems we discover to ensure the proper operation of the subway.

[0003] Analysis of the main rail diseases and their causes: During subway operation, rails are the main part subject to stress, including vertical pressure, horizontal pressure, temperature stress, and longitudinal creep force. Under the action of these forces, rails are extremely susceptible to damage. The specific damaged parts are as follows:

[0004] 1. The first is the damage to the joints: The main problem at the joints is welding problems. The cause of this problem is that the welding process is affected by the construction process, which makes it easier to cause damage.

[0005] 2. Impact of rail wear: Wear is an unavoidable phenomenon during subway operation.

[0006] 3. Wear on the side of the rail.

[0007] 4. Rail fat edge: It is mainly caused by the outer side of the curved rail and the inner side of the straight rail top being crushed.

[0008] 5. Fish scale pattern: Fish scale damage is a fatigue damage caused by metal fragments similar to fish scales on the surface of the rail head.

[0009] 6. Damage to the switch core: The main damage to the switch core is the insufficient metal fatigue strength and the repeated action of the wheels, which eventually leads to fatigue damage.

[0010] As can be seen from the above, rail inspection is essential. To ensure normal operations, actual inspections are usually conducted in the early morning hours, with limited time and heavy workload, primarily through manual inspections. This requires workers to possess a certain level of fault identification ability. Given the varying abilities of personnel, various external factors can sometimes lead to misjudgment.

[0011] Numerous existing rail inspection methods exist, with ultrasonic flaw detection being the mainstream, and image acquisition and analysis also mentioned. However, in actual maintenance and inspection, manual inspection remains the primary method, with specialized instrumentation being relatively uncommon. This is primarily due to the following reasons: 1. The accuracy of existing technologies needs to be further improved. Given the critical importance of track inspection, failures caused by missed inspections can have catastrophic consequences, leading to a low prevalence of instrumentation. 2. The scope of existing technologies is incomplete. Due to the broad scope of track inspection, existing technologies may only offer a single detection function, requiring further improvement. 3. Existing inspection technologies are relatively conservative, with ultrasonic testing being the mainstream. Given the rapid pace of scientific and technological development, rail inspection technology needs further development and upgrading. Therefore, the development of an efficient and reliable inspection system is crucial. Summary of the Invention

[0012] The purpose of the present invention is to provide a rail detection system based on three-dimensional imaging in order to overcome the defects of the above-mentioned prior art.

[0013] The purpose of the present invention can be achieved by the following technical solutions:

[0014] A rail inspection system based on three-dimensional imaging includes a mobile inspection device equipped with a central processing unit and a three-dimensional imaging system, a positioning system, a laser ranging system, and a communication module, all connected to the central processing unit. The three-dimensional imaging system generates a three-dimensional model of the rail in real time as the mobile inspection device moves.

[0015] Preferably, the central processing unit is connected to the server via the communication module.

[0016] Preferably, it further includes an ultrasonic detection system, which is connected to the central processing unit.

[0017] Preferably, the three-dimensional imaging system includes cameras arranged in three directions: upward, left, and right. The three cameras are all connected to an image processor, and the image processor is connected to the central processing unit.

[0018] Preferably, the image processor includes a protocol decoding module, a clock domain conversion and RGB format conversion module, a DDR module, an image preprocessing module and a three-dimensional image synthesis module connected in sequence, and the DDR module and the image preprocessing module are respectively provided with three corresponding to the three cameras.

[0019] Preferably, the three-dimensional image synthesis module adopts phase interference technology.

[0020] Preferably, the positioning system includes an electronic tag reader / writer, and electronic tags storing track position information are provided on the rails.

[0021] Preferably, the positioning system further includes a speed sensor.

[0022] Preferably, it also includes a mobile terminal wirelessly connected to the central processing unit.

[0023] Preferably, it also includes a display connected to the central processing unit.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] 1. Through the three-dimensional imaging system and laser ranging system, more common rail faults can be detected, which improves detection efficiency and greatly reduces personnel workload. The three-dimensional image can provide a more intuitive understanding of the rail status.

[0026] 2. Use the server to store rail inspection information in real time. Through big data analysis, the rail status information can be analyzed and maintenance and repair suggestions can be given. By accumulating historical data and analyzing the characteristics of common faults, technical improvements can be made to the locations where faults often occur to eliminate the faults.

[0027] 3. Combine ultrasonic detection technology to form a redundant system to determine the correctness of the identification test results, making the test results accurate and reliable.

[0028] 4. Speed sensor and electronic tag reader are used to locate the position of the rail where the system is located, ensuring positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Schematic diagram of the overall structure of the system of the present invention;

[0030] Figure 2 Schematic diagram of the structure of the three-dimensional imaging system in the system of the present invention. DETAILED DESCRIPTION

[0031] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0032] Example

[0033] like Figure 1 As shown, this application proposes a track inspection system based on 3D imaging. This system primarily utilizes 3D imaging technology to diagnose track conditions, while also incorporating ultrasonic detection technology to create a redundant system. This system can quickly identify various track fault or hidden danger patterns, thereby ensuring efficient and accurate maintenance.

[0034] This system includes a mobile inspection device, typically a patrol cart, equipped with a central processing unit (CPU) 1 and connected to it: a 3D imaging system 2, a positioning system 3, a laser ranging system 4, and a communication module 6. The 3D imaging system 2 generates a 3D model of the railroad track in real time as the mobile inspection device moves. The communication module 6 includes a wireless communication module. The CPU 1 is located on a motherboard with peripheral circuits.

[0035] The central processing unit 1 is connected to a display 8 and a mobile terminal 9. In this embodiment, the mobile terminal 9 is a mobile phone. The display 8 can be a local display or a remote display. The local display is directly connected to the central processing unit 1, while the remote display is connected to the central processing unit 1 via a network. The central processing unit 1 can transmit the judgment results to the mobile terminal 9 or the display 8 in real time via the wireless network. The fault location is marked and the staff is prompted to conduct maintenance. The display 8 displays a three-dimensional model of the track. The user can manually flip the model to observe the track status.

[0036] The following describes in detail the various systems connected to the central processing unit 1.

[0037] 3D Imaging System 2: In this embodiment, 3D Imaging System 2 includes cameras positioned in three directions: upward, left, and right. These cameras capture real-time images of the railroad tracks from these three directions (upward, left, and right). The captured data is fed into an image processor, which processes and outputs 3D images of the tracks. This information is then uploaded to the CPU 1. The CPU 1 stores the received information on the server 7 and displays the processed results on a display 8 and a mobile phone. Alternatively, 3D Imaging System 2 can be implemented using other existing technologies, such as laser 3D imaging technology.

[0038] In this embodiment, the image processor adopts FPGA. Figure 2 , including a protocol decoding module, a clock domain conversion and RGB format conversion module, a DDR module, an image preprocessing module, and a 3D image synthesis module, which are connected in sequence. The DDR module and the image preprocessing module each have three corresponding cameras. The image processor in this application uses existing image processing technology, and its computer program is all existing technology, without any method improvements. The specific processing process is as follows:

[0039] The camera first feeds the video stream data into the protocol decoding module. Since the output video format of the camera selected in this embodiment is based on the ITU-T 656 protocol, it needs to be converted to an RGB format for easier processing. Therefore, the useful data must first be captured according to the protocol. The captured data then enters the clock domain conversion and RGB format conversion module. This module's primary function is to arrange the captured data into RGB format and unify the three received data channels under a single clock, facilitating subsequent data synchronization. After the RGB format conversion, the data stream is first buffered in the DDR module due to the large amount of data. After buffering, the buffered data from each DDR module is simultaneously fed into the corresponding image preprocessing module. The image preprocessing module primarily performs the following functions: 1. Image extraction: Extracts useful images, discards unused images, and retains only images near the rails. This avoids processing useless data and improves processing efficiency. 2. Image filtering: Due to external environmental factors and the camera system itself, images can contain a lot of noise. Therefore, it is necessary to remove this interference to prevent system misjudgments caused by noise interference. 3. Image enhancement: Although there will be a flash to fill in the light when shooting, due to the complex on-site environment, some details of the captured image are still unclear and do not meet the requirements, so it is necessary to enhance the image details. In this application, histogram stretching technology is used to achieve this goal. 4. Image detail judgment: In order to achieve the purpose of identifying faults, it is necessary to identify the specific details in the image captured by each camera. For example, when it is necessary to judge whether the joint is damaged, considering the actual situation, the damage to the joint is often caused by welding problems, and the specific manifestation is the presence of gray spots. Therefore, the system will extract image features based on the image characteristics of the gray spots. When it meets the image gray spot characteristics, it is judged that gray spots exist and marked. It can also be combined with multiple existing algorithms to identify various fault modes such as fat edges of rails. After the image preprocessing module processes the corresponding data, it will send the processing results to the three-dimensional image synthesis module. In this embodiment, the three-dimensional image synthesis module uses phase interference technology to synthesize three images into a three-dimensional image and output it to the central processor 1.

[0040] Positioning System 3: To establish a one-to-one correspondence between track images and track positions, Positioning System 3 is introduced. Its primary function is to locate the entire system's position, i.e., its track coordinates. Positioning System 3 consists of two main components: an electronic tag reader / writer and a speed sensor. The functions of these two devices are described below.

[0041] Electronic tag reader / writer: To improve positioning accuracy, a specially numbered electronic tag is affixed to the inside of the track. The electronic tag contains track position information. When the tag reader / writer scans and reads the tag, the central processing unit 1 matches the read tag position number with the data in the database to obtain the current position information. Speed sensor: The main purpose of the speed sensor is to detect real-time position changes in the system and transmit this change back to the central processing unit 1. It should be noted that if the speed sensor produces an error, the electronic tag will eliminate this error, thereby ensuring positioning accuracy.

[0042] Laser ranging system 4: The main function of the laser ranging system 4 is to measure the distance and height difference between the two tracks, and upload the measurement information to the central processor 1 to determine whether it meets the requirements.

[0043] In this embodiment, the detection system also includes an ultrasonic detection system 5 connected to the central processing unit 1. This system utilizes ultrasonic principles for flaw detection, thus serving as a redundant detection system. The ultrasonic detection system 5 performs the same functions as conventional ultrasonic track status detection systems: detection information is uploaded to the central processing unit 1, which stores the information on a server 7 and displays the results on a display 8 and a mobile phone.

[0044] In this embodiment, the detection system further includes a remote server 7, whose main function is to store various historical information received by the central processing unit 1, including track image information. When the central processing unit 1 needs big data for track status diagnosis and prediction, it will extract relevant data from the server 7.

[0045] The system works as follows:

[0046] As the inspection vehicle moves, the speed sensor and the electronic tag reader / writer jointly locate the position of the rail on which this detection system is located. The three-dimensional imaging system 2 generates a three-dimensional model of the rail in real time as the inspection vehicle moves. The central processing unit 1 system simultaneously receives this information and matches the rail position coordinates with the three-dimensional image. At the same time, the central processing unit 1 system also receives information from the ultrasonic detection system 5 and the laser ranging system 4. The laser ranging system 4 measures the width of the rail. The main function of the ultrasonic detection system 5 is also to detect the status of the rail. It and the three-dimensional imaging system 2 are two independent detection systems, thus forming a redundant detection system. They can also monitor each other for faults to ensure the accuracy of the detection results. The central processing unit 1 sends the judgment results to the mobile terminal 9 or display 8 in real time.

[0047] Alarm modes are categorized into three types for different rail faults: emergency, general, and noteworthy. The display interface displays different colors for different alarm states, and a corresponding prompt box appears at the fault location. Workers can view the rail inspection status on their mobile phones and zoom in, out, and flip the 3D rail model.

Claims

1. A rail inspection system based on three-dimensional imaging, comprising a mobile inspection device, characterized in that: The mobile inspection device is provided with a central processing unit and a three-dimensional imaging system, a positioning system, a laser ranging system and a communication module connected to the central processing unit. The three-dimensional imaging system generates a three-dimensional model of the rail in real time as the mobile inspection device moves. The three-dimensional imaging system includes cameras arranged in three directions: upper, left, and right. The three cameras are connected to an image processor, which is connected to the central processing unit. The image processor includes a protocol decoding module, a clock domain conversion and RGB format conversion module, a DDR module, an image preprocessing module, and a three-dimensional image synthesis module connected in sequence. The DDR module and the image preprocessing module are each provided with three corresponding to the three cameras. The three-dimensional image synthesis module adopts phase interferometry technology. The positioning system includes an electronic tag reader / writer, an electronic tag storing track position information is provided on the rail, and the positioning system also includes a speed sensor; The image processor is used to perform the following processes: The camera sends the output video stream data to the protocol decoding module, and the protocol decoding module parses the video stream data according to the ITU656 protocol to extract the first data; The first data is transmitted to a clock domain conversion and RGB format conversion module, which is used to convert the first data into RGB format image data and unify it into a clock domain; the RGB format image data is sent to a DDR module for buffering; the buffered image data is respectively input into an image preprocessing module, which includes the following processes: extracting an image area near the rail and eliminating images of irrelevant areas; filtering out noise interference in the image; enhancing image details using a histogram stretching method; performing image feature extraction on the image characteristics of gray spots to obtain gray spot image features; when the buffered image data meets the gray spot image features, it is determined that the image data has gray spots, and the image data is marked and identified as a damaged joint. The image data processed by the image preprocessing module is input to the three-dimensional image synthesis module, which uses phase interference technology to synthesize images from multiple cameras into a three-dimensional image and sends the synthesized image to the central processing unit.

2. The rail detection system based on three-dimensional imaging according to claim 1, characterized in that: The central processing unit is connected to the server via the communication module.

3. The rail detection system based on three-dimensional imaging according to claim 1, characterized in that: It also includes an ultrasonic detection system, which is connected to the central processing unit.

4. The rail detection system based on three-dimensional imaging according to claim 1, characterized in that: Also included is a mobile terminal wirelessly connected to the central processor.

5. The rail detection system based on three-dimensional imaging according to claim 1, characterized in that: Also included is a display connected to the central processing unit.

Images