Method and device for processing dangerous points along railway

Abstract

The invention relates to a method and device for processing dangerous points along a railway, and the method comprises the steps: obtaining multi-temporal remote sensing image data along the railway, and carrying out the preprocessing of the multi-temporal remote sensing image data; preparing railway line ground feature training sample data with mark information, and marking and preprocessing five types of main ground feature elements; training a deep learning surface feature detection model by using the surface feature training sample data, extracting various surface feature elements along the railway, and converting the surface feature elements into vector data matched with geographic coordinates; dangerous point elements are obtained through change detection, the distance between the change elements and the railway is calculated to divide danger levels, and whether field detection and hidden danger elimination are needed or not is judged. Therefore, the problems that an existing railway line dangerous point detection method is low in detection precision and efficiency, and the detection requirement of large railway line dangerous points cannot be met are solved.

Description

Technical Field

[0001] The present application relates to the field of satellite remote sensing monitoring technology, and in particular to a method and device for handling dangerous points along a railway. Background Art

[0002] High-resolution remote sensing images have a wide coverage, high resolution, and strong real-time performance. Compared with other technical means such as drone aerial photography and Google Maps, they have low cost and strong periodicity, providing an effective technical means for quickly, dynamically, and objectively detecting dangerous points along the railway.

[0003] At present, remote sensing technology is rarely used to detect dangerous points along railways. Instead, it is mainly based on manual inspection by railway staff. There are shortcomings such as long detection time, low efficiency, large consumption of manpower and material resources, and interference from subjective factors. It is difficult to conduct a comprehensive inspection of the entire railway network. With the rapid development and expansion of railway networks, railway project routes have become longer and pass through multiple areas with complex terrain. Many places are difficult to reach due to inconvenient transportation. Traditional manual inspections can no longer meet the detection needs of dangerous points along large railways, and omissions may occur.

[0004] In summary, the existing methods for detecting dangerous points along railways have low detection accuracy and efficiency, which cannot meet the detection needs of dangerous points along large railways and need to be solved urgently. Summary of the invention

[0005] The present application provides a method and device for handling dangerous points along railways, so as to solve the problems that the existing methods for detecting dangerous points along railways have low detection accuracy and efficiency and cannot meet the detection needs of dangerous points along large railways.

[0006] The first aspect of the present application provides a method for processing dangerous points along a railway, comprising the following steps: acquiring multi-temporal remote sensing image data along a target railway, and performing preprocessing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; collecting remote sensing image data along the target railway, and annotating the remote sensing image data to construct a ground object training sample data set along the target railway, and training a pre-constructed ground object detection model through the ground object training sample data set; cropping the standard image data to obtain multiple target size standard image data, and inputting the multiple target size standard image data into In the trained object detection model, object prediction results of multiple target objects corresponding to each target size standard image data in the multiple target size standard image data are generated; coordinate matching and data conversion operations are performed on the object prediction results to obtain object vector data corresponding to the object prediction results, and change detection processing is performed on the object vector data to determine at least one change element corresponding to the object vector data, and the distance between each change element of the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level.

[0007] Optionally, in one embodiment of the present application, the multi-temporal remote sensing image data along the target railway is acquired, and the multi-temporal remote sensing image data is preprocessed and data aligned to obtain standard image data corresponding to the multi-temporal remote sensing image data, including: based on a preset detection interval, cropping the multi-temporal remote sensing image data to obtain the multi-temporal remote sensing image data within the detection interval; performing data processing and radiation correction operations on the multi-temporal remote sensing image data within the detection interval to obtain multi-temporal remote sensing processed data corresponding to the multi-temporal remote sensing image data; performing data alignment operations on the multi-temporal remote sensing processed data to generate the standard image data.

[0008] Optionally, in one embodiment of the present application, the remote sensing image data along the target railway is collected and annotated to construct a ground object training sample data set along the target railway, and a pre-constructed ground object detection model is trained through the ground object training sample data set, including: annotating multiple target ground objects in the remote sensing image data through a preset deep learning label making tool to obtain label data corresponding to the remote sensing image data, wherein the multiple target ground objects include buildings, vegetation, bare land, roads and water bodies along the target railway; constructing a ground object training sample data set along the target railway according to the label data and the remote sensing image data; respectively cropping the label data and the remote sensing image data in the ground object training sample data set into sub-region data of multiple target sizes, and performing a naming format unification operation on the label data and the sub-region data of multiple target sizes of the remote sensing image data; and training the pre-constructed ground object detection model through the sub-region data of multiple target sizes after performing the naming format unification operation.

[0009] Optionally, in one embodiment of the present application, the coordinate matching and data conversion operations are performed on the feature prediction results to obtain feature vector data corresponding to the feature prediction results, and the feature vector data is subjected to change detection processing to determine at least one change element corresponding to the feature vector data, and the distance between each of the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level, including: obtaining the actual geographic coordinates and pixel coordinates corresponding to the feature prediction results, matching the actual geographic coordinates and the pixel coordinates to generate coordinate matching data, and converting the coordinate matching data into vector data format to obtain the feature vector data; perform vector splicing on the feature vector data corresponding to each target size standard image data to obtain spliced ​​image vector data; perform differential operation on the spliced ​​image vector data to obtain the differential result corresponding to the spliced ​​image vector data, so as to determine the at least one change element according to the differential result; calculate the distance between each change element and the target railway line, and set multiple danger buffer zones according to the distance, so as to determine the danger level of each change element through the multiple danger buffer zones; based on the preset railway safety operation requirements and the danger level, determine the target danger point element in the at least one change element that has a railway safety operation risk, and perform corresponding danger investigation operations on the target danger point element.

[0010] The second aspect of the present application provides a device for processing dangerous points along a railway, including: a preprocessing module, which is used to obtain multi-phase remote sensing image data along the target railway, and preprocess and data registration operations on the multi-phase remote sensing image data to obtain standard image data corresponding to the multi-phase remote sensing image data; a training module, which is used to collect remote sensing image data along the target railway, and annotate the remote sensing image data to construct a ground object training sample data set along the target railway, and train a pre-constructed ground object detection model through the ground object training sample data set; a prediction module, which is used to crop the standard image data to obtain multiple target size standard image data, and annotate the multiple target size standard image data. The image data is input into the trained object detection model to generate object prediction results of multiple target objects corresponding to each target size standard image data in the multiple target size standard image data; the analysis module is used to perform coordinate matching and data conversion operations on the object prediction results to obtain object vector data corresponding to the object prediction results, and perform change detection processing on the object vector data to determine at least one change element corresponding to the object vector data, and calculate the distance between each change element of the at least one change element and the target railway line, and determine the danger level of each change element according to the distance, so as to execute the danger handling operation corresponding to each change element based on the danger level.

[0011] Optionally, in one embodiment of the present application, the preprocessing module includes: a first cropping unit, used to crop the multi-temporal remote sensing image data based on a preset detection interval to obtain the multi-temporal remote sensing image data within the detection interval; a data processing unit, used to perform data processing and radiation correction operations on the multi-temporal remote sensing image data within the detection interval to obtain multi-temporal remote sensing processed data corresponding to the multi-temporal remote sensing image data; a data registration unit, used to perform data registration operations on the multi-temporal remote sensing processed data to generate the standard image data.

[0012] Optionally, in one embodiment of the present application, the training module includes: a labeling unit, used to label multiple target objects in the remote sensing image data through a preset deep learning label making tool to obtain label data corresponding to the remote sensing image data, wherein the multiple target objects include buildings, vegetation, bare land, roads and water bodies along the target railway; a construction unit, used to construct a training sample data set of objects along the target railway based on the label data and the remote sensing image data; a second cropping unit, used to respectively crop the label data and the remote sensing image data in the object training sample data set into sub-region data of multiple target sizes, and perform a naming format unification operation on the label data and the sub-region data of multiple target sizes of the remote sensing image data; a model processing unit, used to train the pre-constructed object detection model by performing the naming format unification operation on the sub-region data of multiple target sizes.

[0013] Optionally, in one embodiment of the present application, the analysis module includes: a matching unit, used to obtain the actual geographic coordinates and pixel coordinates corresponding to the feature prediction result, and match the actual geographic coordinates and the pixel coordinates to generate coordinate matching data, and convert the coordinate matching data into a vector data format to obtain the feature vector data; a splicing unit, used to perform vector splicing on the feature vector data corresponding to each target size standard image data to obtain spliced ​​image vector data; a differential unit, used to perform a differential operation on the spliced ​​image vector data to obtain a differential result corresponding to the spliced ​​image vector data, so as to determine the at least one change element according to the differential result; a calculation unit, used to calculate the distance between each change element and the target railway line, and set a plurality of danger buffer zones according to the distance, so as to determine the danger level of each change element through the plurality of danger buffer zones; a determination unit, used to determine the target danger point element with railway safety operation risk in the at least one change element based on the preset railway safety operation requirements and the danger level, and perform corresponding danger screening operations on the target danger point element.

[0014] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for handling dangerous points along a railway as described in the above embodiment.

[0015] The fourth aspect of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above-mentioned method for handling dangerous points along the railway.

[0016] The fifth aspect of the present application provides a computer program product, including a computer program, which is executed to implement the above-mentioned method for handling dangerous points along the railway.

[0017] Therefore, the embodiments of the present application have the following beneficial effects:

[0018] The embodiments of the present application can obtain multi-temporal remote sensing image data along the target railway, and perform pre-processing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; collect remote sensing image data along the target railway, and annotate the remote sensing image data to construct a ground object training sample data set along the target railway, and train a pre-constructed ground object detection model through the ground object training sample data set; crop the standard image data to obtain a plurality of target size standard image data, and input the plurality of target size standard image data into the trained ground object detection model to generate ground object prediction results of a plurality of target ground objects corresponding to each target size standard image data in the plurality of target size standard image data; perform coordinate matching and data conversion operations on the ground object prediction results to obtain ground object vector data corresponding to the ground object prediction results, and perform change detection processing on the ground object vector data to determine at least one change element corresponding to the ground object vector data, and calculate the distance between each change element in the at least one change element and the target railway, and determine the danger level of each change element according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level. This application can significantly improve the efficiency and accuracy of dangerous point detection along the railway, thereby providing important technical support for ensuring the safety of railway operation and playing a positive role in protecting people's lives and property. Thus, it solves the problems that the existing dangerous point detection methods along the railway have low detection accuracy and efficiency and cannot meet the detection needs of dangerous points along large railways.

[0019] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0021] Figure 1 A flowchart of a method for handling dangerous points along a railway provided according to an embodiment of the present application;

[0022] Figure 2 A schematic diagram of setting a buffer using the ArcGIS Pro buffer tool is provided for one embodiment of the present application;

[0023] Figure 3 A cropped image of the detection area in 2021 provided for one embodiment of the present application;

[0024] Figure 4 A schematic diagram of the execution logic of a method for handling dangerous points along a railway provided by an embodiment of the present application;

[0025] Figure 5 A schematic diagram of a sample making tool using EasyFeature is provided for one embodiment of the present application;

[0026] Figure 6 A schematic diagram of a building mask for 2021 processed by python is provided for one embodiment of the present application;

[0027] Figure 7 A schematic diagram of a python code of relevant parameters of a selected deep learning neural network provided for one embodiment of the present application;

[0028] Figure 8 A schematic diagram of a python code for a cropping operation provided for an embodiment of the present application;

[0029] Fig. 9 A schematic diagram of a python code for matching a prediction result with an actual geographic coordinate provided in one embodiment of the present application;

[0030] Fig.10 A schematic diagram of a 2021 vector partial effect matching image coordinates provided for an embodiment of the present application;

[0031] Fig.11 A schematic diagram of a python code for concatenating prediction result vectors and restoring them to original size vectors is provided for one embodiment of the present application;

[0032] Fig.12 A vector difference schematic diagram of an ArcGIS Pro clipping tool provided for one embodiment of the present application;

[0033] Fig.13 A schematic diagram of building change detection in a small area provided for one embodiment of the present application;

[0034] Fig.14 A schematic diagram of calculating the distance from a feature to a railroad track using a tool for generating a neighbor table using ArcGIS Pro is provided as an embodiment of the present application;

[0035] Fig.15 A property diagram of a building change vector from 2021 to 2022 provided for one embodiment of the present application;

[0036] Fig.16 A schematic diagram of detecting some newly built building elements in a vacant lot provided in one embodiment of the present application;

[0037] Fig.17 A schematic diagram of a verification result of a dangerous point in a newly built part of a building element detected in a certain vacant land provided by an embodiment of the present application;

[0038] Fig.18 This is an example diagram of a device for handling dangerous points along a railway according to an embodiment of the present application;

[0039] Fig.19 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0040] Among them, 10-dangerous point processing device along the railway, 100-preprocessing module, 200-training module, 300-prediction module, 400-analysis module, 1901-memory, 1902-processor, 1903-communication interface. DETAILED DESCRIPTION

[0041] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0042] The following describes a method and device for processing dangerous points along railways according to an embodiment of the present application with reference to the accompanying drawings. In response to the problems mentioned in the above background technology, the present application provides a method for processing dangerous points along railways, in which multi-temporal remote sensing image data along the target railway is acquired, and pre-processing and data registration operations are performed on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; remote sensing image data along the target railway is collected, and the remote sensing image data is annotated to construct a ground object training sample data set along the target railway, and a pre-constructed ground object detection model is trained with the ground object training sample data set; the standard image data is cropped to obtain standard image data of multiple target sizes, and multiple target size labels are annotated. The quasi-image data is input into the trained object detection model to generate object prediction results for multiple target objects corresponding to each target size standard image data in multiple target size standard image data; coordinate matching and data conversion operations are performed on the object prediction results to obtain object vector data corresponding to the object prediction results, and change detection processing is performed on the object vector data to determine at least one change element corresponding to the object vector data, and the distance between each change element in the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level. The present application can significantly improve the efficiency and accuracy of dangerous point detection along the railway, thereby providing important technical support for ensuring the safety of railway operation and playing a positive role in protecting people’s lives and property. As a result, the existing dangerous point detection methods along the railway have low detection accuracy and efficiency, and cannot meet the detection needs of dangerous points along large railways.

[0043] Specifically, Figure 1 A flowchart of a method for handling dangerous points along a railway provided in an embodiment of the present application.

[0044] like Figure 1 As shown, the method for handling dangerous points along the railway includes the following steps:

[0045] In step S101, multi-temporal remote sensing image data along the target railway are acquired, and pre-processing and data registration operations are performed on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data.

[0046] The embodiments of the present application can firstly obtain multi-temporal remote sensing image data along the target railway, and according to actual needs, crop the multi-temporal remote sensing image data including the area along the railway along the detection interval to obtain the remote sensing image within the detection range, so as to reduce the amount of data to be processed and improve the processing efficiency; secondly, by pre-processing and aligning the acquired multi-temporal remote sensing image data, its spatial consistency is ensured for subsequent differential processing.

[0047] Optionally, in one embodiment of the present application, multi-temporal remote sensing image data along the target railway is obtained, and the multi-temporal remote sensing image data is preprocessed and data registration operations are performed on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data, including: based on a preset detection interval, cropping the multi-temporal remote sensing image data to obtain the multi-temporal remote sensing image data within the detection interval; performing data processing and radiation correction operations on the multi-temporal remote sensing image data within the detection interval to obtain multi-temporal remote sensing processed data corresponding to the multi-temporal remote sensing image data; performing data registration operations on the multi-temporal remote sensing processed data to generate standard image data.

[0048] In the actual implementation process, the embodiment of the present application can obtain images of a section of the Wuxian Railway in Wuhan, China in three phases from 2021 to 2023, such as Figure 2 As shown, taking the 2021 image as an example, the embodiment of the present application can first use the buffer tool in the ArcGIS Pro software to set a buffer zone of one kilometer away from the rail vector, and then obtain images of about one kilometer along the left and right sides of the rail after registration, denoising, and clipping according to the buffer zone, as shown in FIG. Figure 3 The specific steps are as follows:

[0049] Step 1. The embodiment of the present application uses the buffer tool in ArcGIS Pro based on the obtained rail line feature Shp file, sets the input feature to the rail line feature, sets the buffer distance to a linear unit of one kilometer, and outputs a strip buffer layer;

[0050] Step 2: Based on the specific conditions of the image, use the georeferencing tool in ArcGIS Pro to georeference the multi-temporal image to prevent the overall deviation of the geographical location from interfering with subsequent change detection; then use the filter tool to denoise the image to improve the image quality and facilitate subsequent deep learning interpretation. If the image needs radiation correction, ENVI or Erdas Imagine can be used for correction;

[0051] Step 3. Finally, use the Extract by Mask tool in ArcGIS Pro, set the input raster to the multi-temporal remote sensing image, and the feature mask data to the strip buffer layer obtained in step 1 to crop the image of the one-kilometer detection area next to the railway track.

[0052] Afterwards, the embodiments of the present application can perform pre-processing such as image denoising, enhancement, and radiation correction on the cropped multi-temporal remote sensing image data, and then align the multi-temporal remote sensing image data from different times and sources, thereby providing reliable data guidance and basis for the subsequent detection of dangerous points along the railway.

[0053] In step S102, remote sensing image data along the target railway is collected and annotated to construct a ground object training sample data set along the target railway, and a pre-constructed ground object detection model is trained by the ground object training sample data set.

[0054] Furthermore, the embodiments of the present application also need to use professional deep learning labeling tools such as EasyFeature to label five major types of land features along the railway, including buildings, vegetation, bare land, roads, and water bodies, with a focus on land features that may pose a danger to railway operations, in order to construct a land feature training sample data set along the railway; secondly, a series of preprocessing is performed on the training sample data set: both the images and labels are cropped into small areas of 512×512 pixels, and the image and label files of each area are named in a unified format; thereafter, the embodiments of the present application can reasonably divide the training data set into a training set, a validation set, and a test set for model training, validation, and evaluation.

[0055] Furthermore, if Figure 4 As shown, the embodiments of the present application can select a suitable deep learning network model (i.e., a land object detection model) and pre-train the land object detection model, input the pre-processed and divided training sample data set into the land object detection model for training, adjust the land object detection model parameters according to the training status of the validation set and the test set, and optimize the structure and performance of the land object detection model.

[0056] Optionally, in one embodiment of the present application, remote sensing image data along the target railway is collected and annotated to construct a ground object training sample data set along the target railway, and a pre-constructed ground object detection model is trained through the ground object training sample data set, including: annotating multiple target ground objects in the remote sensing image data through a preset deep learning label making tool to obtain label data corresponding to the remote sensing image data, wherein the multiple target ground objects include buildings, vegetation, bare land, roads and water bodies along the target railway; constructing a ground object training sample data set along the target railway based on the label data and the remote sensing image data; respectively cropping the label data and the remote sensing image data in the ground object training sample data set into sub-region data of multiple target sizes, and performing a naming format unification operation on the sub-region data of multiple target sizes of the label data and the remote sensing image data; and training the pre-constructed ground object detection model through the sub-region data of multiple target sizes after performing the naming format unification operation.

[0057] It should be noted that the embodiments of the present application can prepare training sample data of features along the railway with marking information, including the annotation of five major features, and pre-process the sample data; Figure 5As shown, the embodiments of the present application can use EasyFeature3 software to draw training samples of five substrates; Figure 6 As shown, the embodiment of the present application can use Python related library functions to extract the mask of the corresponding feature, organize and sort the data, and divide the data set to construct a training sample data set of features along the railway. The specific implementation steps are as follows:

[0058] Step 1: Use professional deep learning labeling software to obtain training sample data, and the embodiment of the present application can use EasyFeature3 software to draw training sample vectors, and use a variety of vector editing tools such as drawing polygons, semi-automatic collection, local boundary redrawing, polygon splitting, etc. to draw vectors to obtain five types of features in the 2021 image;

[0059] Step 2: Perform a series of preprocessing on the training sample data: Use functions such as Python's OpenCV library and GDAL library to obtain the binary ground object mask, and crop the remote sensing image data and labels into small areas of 512×512 pixels, and name the image and label files of each area in a unified format.

[0060] Step 3: Use the built-in dataset partitioning function of pytorch to partition the training dataset. The ratio of training set, test set, and validation set can be set to 8:1:1.

[0061] Furthermore, the embodiments of the present application can train a deep learning object detection model and extract various object elements along the railway, including model selection, model training, object extraction, matching geographic coordinates and converting vectors. Figure 7 As shown, the embodiment of the present application can select the D-LinkNet34 semantic segmentation neural network as the ground object extraction model, and use the model pre-trained with 0.1m high-resolution data of the Changgan Railway for pre-training, and then input the preprocessed and divided training sample data set into the ground object detection model for training, adjust the model parameters according to the training situation of the validation set and the test set, and optimize the ground object detection model, thereby greatly ensuring the detection reliability and generalization performance of the ground object detection model.

[0062] In step S103, the standard image data is cropped to obtain a plurality of target size standard image data, and the plurality of target size standard image data is input into the trained object detection model to generate object prediction results for a plurality of target objects corresponding to each target size standard image data in the plurality of target size standard image data.

[0063] In step S104, coordinate matching and data conversion operations are performed on the feature prediction results to obtain feature vector data corresponding to the feature prediction results, and change detection processing is performed on the feature vector data to determine at least one change element corresponding to the feature vector data, and the distance between each of the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level.

[0064] Furthermore, after the training accuracy of the object detection model reaches the expected level, the embodiment of the present application may first crop the multi-temporal remote sensing image data along the railway into 512×512 pixels (i.e., the target size) to obtain multiple target size standard image data, and then input the multiple target size standard image data into the trained object detection model for prediction to extract the corresponding five major types of objects; match the pixel coordinates of the prediction results with the actual geographic coordinates to ensure the accuracy of the geographic location of the prediction results, and convert the matched geographic coordinate data into a vector data format to facilitate subsequent analysis and processing.

[0065] It can be understood that since the multi-temporal remote sensing image data along the railway are cropped to 512×512 pixels during prediction, the embodiments of the present application can first perform vector stitching, merge the vectors of the objects in the scattered small areas into the vector of the entire area, and perform a differential operation on the stitched vector data to compare the differences between the objects in the two time periods to identify new, disappeared or changed objects. Afterwards, the embodiments of the present application post-process the detected change spots according to the differential results to remove noise and irrelevant information to obtain accurate change elements.

[0066] In addition, the embodiments of the present application can use geographic information system tools or distance calculation algorithms to calculate the distance from each change element to the nearest railway; according to the distance between the change element and the railway, buffer zones of different radii are set to represent the range of different danger levels, and the change elements are divided into different danger levels according to the distance from the railway track, such as high, medium, and low danger levels; finally, the embodiments of the present application can combine the railway safety operation requirements and the land feature category of the change element itself to determine whether the change element may pose a threat to the safety of railway operation;

[0067] Therefore, the embodiments of the present application adopt the multi-temporal remote sensing image data acquired along the railway, and use deep learning technology to extract and detect changes of various landforms along the line, so as to identify dangerous points that may pose a danger to the safety of railway operation, and provide a reference for the railway department to maintain railway safety. It largely solves the problems of traditional manual inspection and detection, such as long time consumption, low efficiency, high consumption of manpower and material resources, and interference from subjective factors. It provides important technical support for ensuring the safety of railway operation, plays a positive role in protecting people’s lives and property, saves resource costs, and promotes social stability.

[0068] Optionally, in one embodiment of the present application, coordinate matching and data conversion operations are performed on the feature prediction results to obtain feature vector data corresponding to the feature prediction results, and change detection processing is performed on the feature vector data to determine at least one change element corresponding to the feature vector data, and the distance between each of the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level, including: obtaining the actual geographic coordinates and pixel coordinates corresponding to the feature prediction results, matching the actual geographic coordinates and pixel coordinates to generate coordinate matching data, and converting the coordinate matching data into vector data. According to the format, to obtain the ground object vector data; vector stitching is performed on the ground object vector data corresponding to each target size standard image data to obtain the stitching image vector data; a differential operation is performed on the stitching image vector data to obtain the differential result corresponding to the stitching image vector data, so as to determine at least one change element according to the differential result; the distance between each change element and the target railway line is calculated, and a plurality of danger buffer zones are set according to the distance, so as to determine the danger level of each change element through the plurality of danger buffer zones; based on the preset railway safety operation requirements and danger levels, a target danger point element with railway safety operation risks in at least one change element is determined, and a corresponding danger investigation operation is performed on the target danger point element.

[0069] Specifically, Figure 8 As shown, the embodiment of the present application can perform the above-mentioned image data cropping operation on the remote sensing images of 2022 and 2023; Fig. 9 As shown in the figure, the multi-period data is intelligently interpreted through the trained object detection model, the pixel coordinates of the predicted results are matched with the actual geographic coordinates, and the matched geographic coordinate data are converted into vector data format, and finally the multi-period object element vector matching the image coordinates is obtained, as shown in Fig.10 shown.

[0070] After the training accuracy of the object detection model reaches the expected level, the embodiment of the present application can use the cv2.copyMakeBorder function in opencv to first crop the image data of 2022 and 2023 to 512×512 pixels, and then use the trained object detection model for prediction to extract the corresponding five major types of objects; further, the embodiment of the present application can use the GetRasterBand function and GetGeoTransform function in the Gdal library to match the pixel coordinates of the prediction results with the actual geographic coordinates to ensure the accuracy of the geographic location of the prediction results, and convert the matched geographic coordinate data into vector data format.

[0071] Furthermore, the embodiments of the present application can obtain dangerous point elements by change detection such as vector splicing, vector difference and change spot post-processing; Fig.11 As shown, the embodiment of the present application can splice the 512×512 predicted vectors to the original size; Fig.12 As shown, the embodiment of the present application uses a clipping tool to obtain a change element pattern by differentiating the two phases of vectors to be compared, and then removes the pseudo changes caused by angles, light or boundaries through post-processing to obtain a true ground feature change vector (i.e., ground feature vector data); Fig.13 Taking the building change detection in a small area as an example, the steps of obtaining the real ground object change vector (i.e. ground object vector data) are as follows:

[0072] Step 1: Use functions such as paste and crop in the opencv library to loop through and stitch the predicted vectors into the same vector file;

[0073] Step 2: Load the obtained multi-phase vectors of various types of features into ArcGIS Pro, use the clipping tool to take the two phases of vectors to be compared as input features and clipping features, and output different change feature patches;

[0074] Step 3: Use the field search tool in the attribute table to set an area threshold to eliminate smaller changes, that is, to filter out pseudo changes caused by angles, light, boundaries, etc., and obtain the real ground feature change vector.

[0075] In addition, the embodiment of the present application also needs to calculate the distance from the changing element to the railway, and classify the danger level according to the distance to determine whether on-site inspection and elimination of hidden dangers are required. Fig.14 As shown, the embodiment of the present application can generate a neighbor table tool to calculate the distance from each change element in the feature change vector to the rail, and then use the buffer tool to set buffers of different danger levels; finally, the embodiment of the present application combines the railway safety operation requirements and the feature category of the change element itself to determine whether the change element may pose a threat to the safety of railway operation; Fig.15As shown in the NEAR_DIST column of the change element attribute table, each change element is arranged in ascending order according to the distance from the railroad track, which can intuitively understand the danger level of the danger point. Fig.15 It can be seen that the distance between a certain dangerous element and the railroad track is 53.7m, and the danger level is medium. Click the element in the attribute table to locate the location of the change element in the ArcGIS Pro view; Fig.16 As shown, the embodiment of the present application detects that some building elements have been newly built on the vacant land, and compares the images to see that Fig.17 As shown, it can be seen intuitively that the dangerous factor is the newly built color steel house.

[0076] Specifically, the specific process of determining the danger point and the danger level according to the distance in the embodiment of the present application is as follows:

[0077] 1. Load the rail vector and the change feature vector into ArcGIS Pro software, then use the Generate Nearest Table tool to calculate the vertical distance from each vector in the change feature vector to the rail, and add this distance to the attribute field of the vector;

[0078] 2. Use the buffer tool to set 0-20m, 20-50m, and 50-100m as high, medium, and low danger levels, and calculate the risk level of each changing element based on the coordinates;

[0079] 3. Combined with the requirements for safe railway operation and the type of the changing elements themselves, compare the two images to determine whether the changing elements may pose a threat to the safety of railway operation. If so, they are dangerous points (i.e., target dangerous point elements), and perform corresponding danger inspection operations on the target dangerous point elements.

[0080] To sum up, the embodiments of the present application can not only clearly and intuitively display the location of the dangerous point and the distance from the dangerous point to the rail, but also provide clear boundary vectors and attribute information; in addition, the embodiments of the present application can also classify the possible dangerous points into dangerous levels, the detection method is true and reliable, highly operational, and the results of ground object interpretation and change detection are good, thereby verifying the effectiveness and reliability of the method for handling dangerous points along the railway of the present application.

[0081] According to the method for processing dangerous points along railways proposed in the embodiment of the present application, multi-temporal remote sensing image data along the target railway is obtained, and pre-processing and data registration operations are performed on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; remote sensing image data along the target railway is collected, and the remote sensing image data is annotated to construct a ground object training sample data set along the target railway, and a pre-constructed ground object detection model is trained with the ground object training sample data set; the standard image data is cropped to obtain multiple target size standard image data, and the multiple target size standard image data are input into the training data set. In the trained object detection model, object prediction results of multiple target objects corresponding to each target size standard image data in multiple target size standard image data are generated; coordinate matching and data conversion operations are performed on the object prediction results to obtain object vector data corresponding to the object prediction results, and change detection processing is performed on the object vector data to determine at least one change element corresponding to the object vector data, and the distance between each change element in at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level. The present application can significantly improve the efficiency and accuracy of dangerous point detection along the railway, thereby providing important technical support for ensuring the safety of railway operation and playing a positive role in protecting people's lives and property.

[0082] Secondly, the device for handling dangerous points along the railway proposed in accordance with the embodiment of the present application is described with reference to the accompanying drawings.

[0083] Fig.18 It is a block diagram of a device for handling dangerous points along a railway according to an embodiment of the present application.

[0084] like Fig.18 As shown, the railway dangerous point processing device 10 includes: a preprocessing module 100, a training module 200, a prediction module 300 and an analysis module 400.

[0085] The preprocessing module 100 is used to obtain multi-temporal remote sensing image data along the target railway, and perform preprocessing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data.

[0086] The training module 200 is used to collect remote sensing image data along the target railway and annotate the remote sensing image data to construct a ground object training sample data set along the target railway, and train a pre-constructed ground object detection model through the ground object training sample data set.

[0087] The prediction module 300 is used to crop the standard image data to obtain multiple target size standard image data, and input the multiple target size standard image data into the trained object detection model to generate object prediction results for multiple target objects corresponding to each target size standard image data in the multiple target size standard image data.

[0088] The analysis module 400 is used to perform coordinate matching and data conversion operations on the feature prediction results to obtain feature vector data corresponding to the feature prediction results, and to perform change detection processing on the feature vector data to determine at least one change element corresponding to the feature vector data, and to calculate the distance between each of the at least one change element and the target railway line, and to determine the danger level of each change element according to the distance, so as to execute the danger handling operation corresponding to each change element based on the danger level.

[0089] Optionally, in one embodiment of the present application, the preprocessing module 100 includes: a first cropping unit, a data processing unit and a data registration unit.

[0090] The first cropping unit is used to crop the multi-temporal remote sensing image data based on a preset detection interval to obtain the multi-temporal remote sensing image data within the detection interval.

[0091] The data processing unit is used to perform data processing and radiation correction operations on the multi-temporal remote sensing image data within the detection interval to obtain multi-temporal remote sensing processed data corresponding to the multi-temporal remote sensing image data.

[0092] The data registration unit is used to perform data registration operations on multi-temporal remote sensing processing data to generate standard image data.

[0093] Optionally, in one embodiment of the present application, the training module 200 includes: a labeling unit, a construction unit, a second cropping unit and a model processing unit.

[0094] Among them, the labeling unit is used to label multiple target objects in the remote sensing image data through a preset deep learning label making tool to obtain label data corresponding to the remote sensing image data, wherein the multiple target objects include buildings, vegetation, bare land, roads and water bodies along the target railway.

[0095] The construction unit is used to construct a ground feature training sample dataset along the target railway according to the label data and the remote sensing image data.

[0096] The second cropping unit is used to crop the label data and remote sensing image data in the ground object training sample data set into sub-region data of multiple target sizes, and perform a naming format unification operation on the sub-region data of multiple target sizes of the label data and the remote sensing image data.

[0097] The model processing unit is used to train a pre-built ground object detection model by executing sub-region data of multiple target sizes after a naming format unification operation.

[0098] Optionally, in one embodiment of the present application, the analysis module 400 includes: a matching unit, a splicing unit, a difference unit, a calculation unit and a determination unit.

[0099] Among them, the matching unit is used to obtain the actual geographic coordinates and pixel coordinates corresponding to the object prediction results, and match the actual geographic coordinates and pixel coordinates to generate coordinate matching data, and convert the coordinate matching data into a vector data format to obtain the object vector data.

[0100] The stitching unit is used to perform vector stitching on the ground object vector data corresponding to each target size standard image data to obtain stitching image vector data.

[0101] The differential unit is used to perform a differential operation on the spliced ​​image vector data to obtain a differential result corresponding to the spliced ​​image vector data, so as to determine at least one change element according to the differential result.

[0102] The calculation unit is used to calculate the distance between each change element and the target railway line, and set a plurality of danger buffer zones according to the distance, so as to determine the danger level of each change element through the plurality of danger buffer zones.

[0103] The determination unit is used to determine a target dangerous point element with a railway safe operation risk in at least one variable element based on preset railway safe operation requirements and danger levels, and perform corresponding danger screening operations on the target dangerous point element.

[0104] It should be noted that the above explanation of the embodiment of the method for handling dangerous points along the railway is also applicable to the device for handling dangerous points along the railway of this embodiment, and will not be repeated here.

[0105] The device for processing dangerous points along a railway proposed in an embodiment of the present application includes a preprocessing module for acquiring multi-temporal remote sensing image data along a target railway, and performing preprocessing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; a training module for collecting remote sensing image data along a target railway, and annotating the remote sensing image data to construct a ground object training sample data set along the target railway, and training a pre-constructed ground object detection model through the ground object training sample data set; a prediction module for cropping the standard image data to obtain a plurality of target size standard image data, and annotating the plurality of target size standard image data. The image data is input into the trained object detection model to generate object prediction results for multiple target objects corresponding to each target size standard image data in multiple target size standard image data; the analysis module is used to perform coordinate matching and data conversion operations on the object prediction results to obtain object vector data corresponding to the object prediction results, and perform change detection processing on the object vector data to determine at least one change element corresponding to the object vector data, and calculate the distance between each change element in at least one change element and the target railway line, and determine the danger level of each change element based on the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level. The present application can significantly improve the efficiency and accuracy of dangerous point detection along the railway, thereby providing important technical support for ensuring the safety of railway operation and playing a positive role in protecting people's lives and property.

[0106] Fig.19 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0107] Memory 1901 , processor 1902 , and a computer program stored in the memory 1901 and executable on the processor 1902 .

[0108] When the processor 1902 executes the program, the method for handling dangerous points along the railway provided in the above embodiment is implemented.

[0109] Furthermore, the electronic device further comprises:

[0110] The communication interface 1903 is used for communication between the memory 1901 and the processor 1902 .

[0111] The memory 1901 is used to store computer programs that can be executed on the processor 1902 .

[0112] The memory 1901 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0113] If the memory 1901, the processor 1902 and the communication interface 1903 are implemented independently, the communication interface 1903, the memory 1901 and the processor 1902 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.19 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0114] Optionally, in a specific implementation, if the memory 1901, the processor 1902 and the communication interface 1903 are integrated on a chip, the memory 1901, the processor 1902 and the communication interface 1903 can communicate with each other through an internal interface.

[0115] The processor 1902 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0116] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for handling dangerous points along the railway is implemented as described above.

[0117] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned method for handling dangerous points along the railway.

[0118] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0119] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0120] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0121] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0122] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0123] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0124] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0125] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for handling dangerous points along a railway, characterized in that: The following steps are involved: Acquire multi-temporal remote sensing image data along the target railway, and perform preprocessing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; Collecting remote sensing image data along the target railway, and annotating the remote sensing image data to construct a ground object training sample data set along the target railway, and training a pre-constructed ground object detection model through the ground object training sample data set; The standard image data is cropped to obtain a plurality of target size standard image data, and the plurality of target size standard image data is input into a trained ground object detection model to generate ground object prediction results of a plurality of target ground objects corresponding to each target size standard image data in the plurality of target size standard image data; Coordinate matching and data conversion operations are performed on the feature prediction result to obtain feature vector data corresponding to the feature prediction result, and change detection processing is performed on the feature vector data to determine at least one change element corresponding to the feature vector data, and the distance between each of the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform a danger handling operation corresponding to each change element based on the danger level.

2. The method according to claim 1, characterized in that The step of acquiring multi-temporal remote sensing image data along the target railway and performing preprocessing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data includes: Based on a preset detection interval, cropping the multi-temporal remote sensing image data to obtain the multi-temporal remote sensing image data within the detection interval; Performing data processing and radiation correction operations on the multi-temporal remote sensing image data within the detection interval to obtain multi-temporal remote sensing processed data corresponding to the multi-temporal remote sensing image data; A data registration operation is performed on the multi-temporal remote sensing processed data to generate the standard image data.

3. The method according to claim 2, characterized in that The collecting of remote sensing image data along the target railway and labeling of the remote sensing image data to construct a ground object training sample data set along the target railway, and training a pre-constructed ground object detection model by using the ground object training sample data set, includes: Using a preset deep learning labeling tool, labeling multiple target objects in the remote sensing image data to obtain label data corresponding to the remote sensing image data, wherein the multiple target objects include buildings, vegetation, bare land, roads and water bodies along the target railway; Constructing a ground object training sample data set along the target railway according to the label data and the remote sensing image data; Respectively cutting the label data and the remote sensing image data in the ground object training sample data set into sub-region data of multiple target sizes, and performing a naming format unification operation on the sub-region data of multiple target sizes of the label data and the remote sensing image data; The pre-built ground object detection model is trained by performing a naming format unification operation on the sub-region data of multiple target sizes.

4. The method according to claim 1, characterized in that: The coordinate matching and data conversion operations are performed on the feature prediction results to obtain feature vector data corresponding to the feature prediction results, and the feature vector data is subjected to change detection processing to determine at least one change element corresponding to the feature vector data, and the distance between each of the at least one change element and the target railway line is calculated, and the danger level of each change element is determined according to the distance, so as to perform the danger handling operation corresponding to each change element based on the danger level, including: Acquire the actual geographic coordinates and pixel coordinates corresponding to the object prediction result, match the actual geographic coordinates and the pixel coordinates to generate coordinate matching data, and convert the coordinate matching data into a vector data format to obtain the object vector data; Performing vector stitching on the ground object vector data corresponding to each target size standard image data to obtain stitching image vector data; Performing a differential operation on the stitched image vector data to obtain a differential result corresponding to the stitched image vector data, and determining the at least one change element according to the differential result; Calculating the distance between each of the change elements and the target railway line, and setting a plurality of danger buffer zones according to the distance, so as to determine the danger level of each of the change elements through the plurality of danger buffer zones; Based on the preset railway safety operation requirements and the hazard level, a target hazard point element that has a railway safety operation risk in the at least one variable element is determined, and a corresponding hazard screening operation is performed on the target hazard point element.

5. A device for handling dangerous points along a railway, characterized in that: include: A preprocessing module, used for acquiring multi-temporal remote sensing image data along the target railway, and performing preprocessing and data registration operations on the multi-temporal remote sensing image data to obtain standard image data corresponding to the multi-temporal remote sensing image data; A training module, used for collecting remote sensing image data along the target railway, and annotating the remote sensing image data to construct a ground object training sample data set along the target railway, and training a pre-constructed ground object detection model through the ground object training sample data set; A prediction module, used for cropping the standard image data to obtain a plurality of target size standard image data, and inputting the plurality of target size standard image data into a trained object detection model to generate object prediction results of a plurality of target objects corresponding to each target size standard image data in the plurality of target size standard image data; The analysis module is used to perform coordinate matching and data conversion operations on the feature prediction results to obtain feature vector data corresponding to the feature prediction results, and to perform change detection processing on the feature vector data to determine at least one change element corresponding to the feature vector data, and to calculate the distance between each of the at least one change element and the target railway line, and to determine the danger level of each change element according to the distance, so as to execute the danger handling operation corresponding to each change element based on the danger level.

6. The device according to claim 5, characterized in that The preprocessing module comprises: A first cropping unit, configured to crop the multi-temporal remote sensing image data based on a preset detection interval to obtain the multi-temporal remote sensing image data within the detection interval; A data processing unit, used for performing data processing and radiation correction operations on the multi-temporal remote sensing image data within the detection interval to obtain multi-temporal remote sensing processed data corresponding to the multi-temporal remote sensing image data; The data registration unit is used to perform data registration operation on the multi-temporal remote sensing processed data to generate the standard image data.

7. The device according to claim 6, characterized in that The training module includes: A labeling unit, used to label multiple target objects in the remote sensing image data by using a preset deep learning label making tool to obtain label data corresponding to the remote sensing image data, wherein the multiple target objects include buildings, vegetation, bare land, roads and water bodies along the target railway; A construction unit, used to construct a ground object training sample data set along the target railway according to the label data and the remote sensing image data; A second cropping unit is used to crop the label data and the remote sensing image data in the ground object training sample data set into sub-region data of multiple target sizes, and perform a naming format unification operation on the sub-region data of multiple target sizes of the label data and the remote sensing image data; The model processing unit is used to train the pre-built ground object detection model by executing the sub-region data of multiple target sizes after the naming format unification operation.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for handling dangerous points along a railway as described in any one of claims 1 to 4.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for handling dangerous points along a railway as described in any one of claims 1 to 4.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed to implement the method for handling dangerous points along a railway as described in any one of claims 1 to 4.

Images