A method and system for railway track intrusion detection based on periodic drone inspection

Abstract

This invention provides a railway track intrusion identification method and system based on periodic drone inspections. It involves data processing technology, controlling a drone to a designated location to collect track maps of the railway track; processing the dynamically open tracks and statically closed tracks in the track map according to a facility expansion strategy to generate a first processing area; identifying the intrusion trajectory and intrusion outline of the intruder in the first processing area; and expanding the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area. This allows for dynamic determination of the area to be identified based on the actual conditions of the railway track, thereby improving identification efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to data processing technology, and more particularly to a method and system for identifying railway track intrusions based on periodic unmanned aerial vehicle (UAV) inspections. Background Technology

[0002] As the core carrier of train operation, railway tracks are directly related to the safety of passenger lives and freight transport. Track intrusions (such as accidental entry by personnel, animals, or the accumulation of foreign objects) are significant hidden dangers that can cause train delays, derailments, and other accidents. With the development of drone technology, periodic drone inspections have become one of the important means of railway track safety monitoring. These drones collect high-definition track images covering the entire track, which are then input into a server for identification and processing to identify intrusion behaviors.

[0003] However, most existing technologies directly perform full-map intrusion identification on the complete track map collected by drones, resulting in massive data processing and a high risk of errors. For example, normal pedestrian crossings at railway crossings may be misidentified as intrusions, wasting computing power and easily leading to misjudgments due to data interference, thus affecting monitoring accuracy. Furthermore, they generally lack specificity in area-based processing. Human intrusions often exhibit patterns; for instance, people from nearby villages frequently cross railway tracks directly in relatively fixed areas for convenience, making targeted identification impossible. Conversely, if an area has no intrusion records for a long period, existing technologies still require continuous full-range identification, resulting in unnecessary computing power consumption.

[0004] Therefore, how to dynamically determine the area to be identified based on the actual conditions of the railway track, thereby improving identification efficiency and accuracy, has become an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a railway track intrusion identification method and system based on periodic UAV inspection, which can dynamically determine the area to be identified according to the actual situation of the railway track, thereby improving identification efficiency and accuracy.

[0006] A first aspect of the present invention provides a method for identifying railway track intrusions based on periodic unmanned aerial vehicle (UAV) inspections, comprising:

[0007] Control the drone to a designated location to collect track maps of the railway tracks;

[0008] Based on the facility expansion strategy, the dynamic open tracks and static closed tracks in the track diagram are processed to generate the first processing area;

[0009] Identify the intrusion trajectory and intrusion outline of the intruder in the first processing area, and perform regional expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area.

[0010] Optionally, in one possible implementation of the first aspect, the step of processing the dynamic open tracks and static closed tracks in the track diagram according to the facility expansion strategy to generate a first processing area includes:

[0011] Identify the dynamic level crossing area of ​​the level crossing control device in the track diagram, and treat the remaining area as a static closed area;

[0012] The track profile of the dynamic level crossing area is obtained as the dynamic open track, and the track profile of the static closed area is obtained as the static closed track.

[0013] The opening and closing times of the level crossing guidance device at the corresponding set location are retrieved, and the dynamic open track and static closed track are processed based on the opening and closing times to obtain the outer expansion area;

[0014] Identify the equipment outline of track-attached equipment in the track diagram, and process the outward expansion area based on the equipment outline to generate a first processing area.

[0015] Optionally, in one possible implementation of the first aspect, the processing of the dynamically open track and the statically closed track based on open and closed time periods respectively to obtain the outer expansion area includes:

[0016] Delete the dynamically open track during the open period, and move the outer contour of the static closed track to the outside of the railway track based on the preset outward expansion distance to obtain the open outward expansion area corresponding to the open period;

[0017] Based on a preset outward expansion distance, the outer contours of the dynamic open track and the static closed track within the closed period are simultaneously moved to the outside of the railway track to obtain the closed outward expansion area corresponding to the closed period.

[0018] The outer expansion area is obtained based on the open outer expansion area and the closed outer expansion area.

[0019] Optionally, in one possible implementation of the first aspect, the step of processing the outward expansion area based on the device outline to generate a first processing area includes:

[0020] The device outline is enlarged based on a set magnification to obtain the enlarged device outline.

[0021] Obtain the intersection point of the magnified device outline and the region line of the outer expansion area, and perform segmentation processing on the device outline based on the intersection point to obtain the deleted outline located within the outer expansion area and the retained outline located outside the outer expansion area;

[0022] Remove the area lines located within the enlarged device outline and delete the deleted outline to generate the first processing area.

[0023] Optionally, in one possible implementation of the first aspect, identifying the intrusion trajectory and intrusion outline of the intruder at the first processing area, and performing region expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area, includes:

[0024] Identify intrusion trajectories that intersect with the first processing area, as well as intrusion outlines of intruders that intersect with the first processing area;

[0025] The intermediate processing area is obtained by the union of the intrusion profile and the first processing area;

[0026] The intrusion trajectory that does not pass through the track outline within the intermediate processing area is regarded as an accidental intrusion trajectory, and the intrusion trajectory that passes through the track outline within the intermediate processing area is regarded as an intentional intrusion trajectory.

[0027] The intermediate processing area is expanded based on the accidental trajectory and the intentional trajectory to generate a second processing area.

[0028] Optionally, in one possible implementation of the first aspect, the step of performing region expansion processing on the intermediate processing area based on the accidental trajectory and the intentional trajectory respectively to generate a second processing area includes:

[0029] The trajectory diagram is converted to coordinates.

[0030] Based on the erroneous trajectory located in the intermediate processing area, construct an extended circle corresponding to the erroneous trajectory;

[0031] Based on the intermediate connecting line of the intentional trajectory, construct an extended channel area corresponding to the intentional trajectory;

[0032] A second processing area is generated based on the union of the extended circle and the extended channel area with the intermediate processing area.

[0033] Optionally, in one possible implementation of the first aspect, constructing an extended circle corresponding to the intrusive trajectory based on the intrusive trajectory located within the intermediate processing area includes:

[0034] Obtain the coordinates of the intersection point between the erroneous trajectory and the area line of the intermediate processing zone, and calculate the average of the intersection point coordinates to obtain the center coordinates;

[0035] The farthest distance from the center coordinates to the erroneous trajectory within the intermediate processing area is determined as the expansion radius. An expansion circle is constructed based on the expansion radius and with the center coordinates as the center.

[0036] Optionally, in one possible implementation of the first aspect, constructing an extended channel region corresponding to the intentional trajectory based on the intermediate connecting line of the intentional trajectory includes:

[0037] Connect the two endpoints of the intentional trajectory to generate the middle connecting line;

[0038] The intermediate connecting line is shifted to both sides until it intersects with the intentional trajectory at only one point, thus generating an extended channel area.

[0039] Optionally, in one possible implementation of the first aspect, it also includes:

[0040] By statistically analyzing the intrusion trajectories adjacent to the intrusion contour, a set of identification trajectories for each intrusion contour is obtained;

[0041] The collection time of each intrusion trajectory in the identification trajectory set is used as the cutoff time;

[0042] The previous acquisition time corresponding to the cutoff time is retrieved as the backtracking time. Based on the backtracking time and the cutoff time, the backtracking time period of each intrusion trajectory is obtained.

[0043] The intrusion trajectories in the identification trajectory set are sorted based on the backtracking time period to obtain the identification trajectory sequence.

[0044] A second aspect of the present invention provides a railway track intrusion identification system based on periodic unmanned aerial vehicle (UAV) inspection, comprising:

[0045] The control module is used to control the drone to a set location to collect track maps of the railway tracks;

[0046] The generation module is used to process the dynamic open tracks and static closed tracks in the track map according to the facility expansion strategy, and generate the first processing area;

[0047] An extension module is used to identify the intrusion trajectory and intrusion outline of the intruder in the first processing area, and to perform regional expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area.

[0048] A third aspect of the present invention provides a storage medium storing a computer program, which, when executed by a processor, is used to implement the first aspect of the present invention and various methods possibly involved in the first aspect.

[0049] The beneficial effects of this invention are as follows:

[0050] 1. This invention periodically collects trajectory maps using drones, generating a first processing area through dynamic open and static closed trajectories. A second processing area is then generated by identifying intrusion trajectories and outlines, covering both the intruder itself and potential risk areas. This avoids the low identification efficiency and wasted computational power of traditional methods that require analyzing all data, thus improving identification efficiency while maintaining accuracy.

[0051] 2. This invention can automatically adjust the expansion range to ensure that the first processing area neither misses key monitoring points around the equipment nor includes normal activity areas such as road crossings. This makes the analysis in the first processing area more targeted and improves processing efficiency.

[0052] 3. This invention enables dynamic processing of risk areas. By comprehensively considering both trajectory and foreign object contours, the monitoring range can be adaptively adjusted according to the actual intrusion situation, preventing wasted computing power due to over-expansion and improving processing efficiency. Simultaneously, the expanded second processing area includes trajectory timing information and contour features, allowing the processing end to quickly determine the nature of the intrusion (e.g., static foreign object / dynamic target), improving the efficiency of response decisions. Attached Figure Description

[0053] Figure 1 This is a schematic diagram illustrating the application scenario of the technical solution provided by the present invention;

[0054] Figure 2 A flowchart of a railway track intrusion identification method based on periodic unmanned aerial vehicle (UAV) inspection provided by the present invention;

[0055] Figure 3 A schematic diagram of an open outer expansion region provided by the present invention;

[0056] Figure 4 This is a schematic diagram of the structure of a railway track intrusion identification system based on periodic drone inspection provided by the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0059] It should be understood that in the various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0060] It should be understood that in this invention, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0061] It should be understood that in this invention, "multiple" refers to two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "Contains A, B, and C", "Contains A, B, and C" means that all three A, B, and C are contained; "Contains A, B, or C" means that one of A, B, and C is contained; "Contains A, B, and / or C" means that any one, two, or three of A, B, and C are contained.

[0062] It should be understood that in this invention, "B corresponding to A", "B corresponding to A", "A and B correspond", or "B and A correspond" means that B is associated with A, and B can be determined based on A. Determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information. Matching A and B is defined as a similarity between A and B that is greater than or equal to a preset threshold.

[0063] Depending on the context, "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection."

[0064] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0065] like Figure 1The diagram illustrates a scenario of the technical solution provided by this invention. This application scenario includes a server and a drone, with the server and drone communicating with each other. When the railway department needs to conduct periodic track safety inspections (daily early morning non-operational hours, quarterly special track protection inspections), the server first obtains the coordinates of key monitoring points along the railway line (i.e., pre-configured collection points by personnel) through the drone's positioning module (e.g., GPS). The server then sends flight control commands to the drone, directing it to fly along a pre-set inspection route to each pre-configured location. The drone collects railway track maps of the corresponding areas using high-definition cameras and lidar sensors mounted on its fuselage. Subsequently, the track is expanded according to a facility expansion strategy, identifying key track areas requiring monitoring and generating a first processing area. Later, the intrusion trajectory formed by tracking the movement of intruders on the track map is used. Based on this trajectory, the first processing area is customized and expanded to obtain a second processing area, which is then sent to the server for further processing.

[0066] This invention provides a railway track intrusion identification method based on periodic unmanned aerial vehicle (UAV) inspection, such as... Figure 2 As shown, steps S1-S4 are included:

[0067] S1 controls the drone to a set location to collect track maps of the railway tracks.

[0068] The set location refers to the pre-planned drone inspection point, which can be determined by latitude and longitude, route mileage markers, or relative orbital position, or it can be a data collection point set in advance by humans.

[0069] S2, based on the facility expansion strategy, processes the dynamic open tracks and static closed tracks in the track diagram to generate the first processing area.

[0070] In some embodiments, step S2 (processing the dynamic open tracks and static closed tracks in the track diagram according to the facility expansion strategy to generate a first processing area) includes S21-S24:

[0071] S21, identify the dynamic level crossing area of ​​the level crossing control device in the track diagram, and treat the remaining area as a static closed area.

[0072] It should be noted that there are areas along the railway track that can be temporarily opened, such as level crossings. These areas allow vehicles or pedestrians to pass through during open hours, but are closed during closed hours. For example, when the signal light is green, the barrier is open to allow people and vehicles to pass through, and when it is red, the barrier is lowered to obstruct people and vehicles from passing through.

[0073] Among them, the level crossing control device refers to the equipment used to control the opening and closing of level crossings, such as barrier gates. A level crossing is the intersection of a railway and a highway. The area blocked by barriers is defined as the dynamic level crossing area. The dynamic level crossing area refers to the track area controlled by the level crossing control device that can be opened during specific periods.

[0074] For dynamic level crossing areas, intrusion is allowed during open periods but not during closed periods. Therefore, the tracks need to be adapted to different states. The remaining track areas are areas where intrusion is not allowed, i.e., static closed areas.

[0075] S22, obtain the track profile of the dynamic level crossing area as the dynamic open track, and the track profile of the static closed area as the static closed track.

[0076] It is understandable that the track profile in the dynamic level crossing area is considered as the dynamic open track, and the track profile in the static closed area is considered as the static closed track.

[0077] S23, retrieve the opening and closing time periods of the level crossing guidance device at the corresponding set location, and process the dynamic open track and static closed track based on the opening and closing time periods respectively to obtain the outer expansion area.

[0078] Among them, the level crossing guidance device refers to the equipment used to indicate the open and closed status of the level crossing. It can be a traffic signal light or a control barrier device that sets the permitted and prohibited time periods for passage. The open period refers to the time period during which vehicles or pedestrians are allowed to pass through the level crossing; the closed period refers to the time period during which passage is prohibited.

[0079] In some embodiments, step S23 (processing the dynamic open track and the static closed track based on the open period and the closed period respectively to obtain the outer expansion area) includes S231-S233:

[0080] S231, delete the dynamically open track during the open period, and move the outer contour of the static closed track to the outside of the railway track based on the preset expansion distance to obtain the open expansion area corresponding to the open period.

[0081] It should be noted that during open hours, dynamically open tracks (such as level crossing areas) must allow legal passage for vehicles and pedestrians. Including these areas in the extended monitoring range would frequently lead to misjudgments of legal passage, increasing the identification of invalid data and the amount of data processing required. On the other hand, statically closed tracks are always prohibited from intrusion, and extended monitoring is necessary to ensure the safety of the area surrounding the track. Therefore, we will dynamically determine the extended monitoring area based on the actual situation.

[0082] The preset outward expansion distance is a distance that is pre-set by the user to expand outward, and can be dynamically set according to the actual situation.

[0083] It's easy to understand that during open hours, because dynamically open tracks require legal passage, the server directly deletes the outward expansion processing requirements for that area, excluding it from the monitoring scope. Next, for statically closed tracks, the server extracts the coordinate data of its outer contour and, according to a preset outward expansion distance (e.g., 1.5 meters), moves the outer contour a corresponding distance away from the track's centerline, forming a new outward expansion boundary, thus obtaining the open expansion zone. This ensures that during open hours, monitoring is only conducted on the prohibited intrusion areas surrounding the statically closed tracks. See [link to relevant documentation]. Figure 3 The outer contour of the track is moved and extended, that is, the area between the new outer boundaries.

[0084] S232, based on a preset outward expansion distance, simultaneously move the outer contours of the dynamically open track and the statically closed track to the outside of the railway track during the closed period to obtain the closed outward expansion area corresponding to the closed period.

[0085] Similar to the logic of S231, since personnel are not allowed to enter either the dynamically open track or the statically closed track during the closed period, they can be treated as a whole and expanded outwards simultaneously to obtain the closed expansion area corresponding to the closed period.

[0086] S233, the outer expansion area is obtained based on the open outer expansion area and the closed outer expansion area.

[0087] S24, Identify the equipment outline of track auxiliary equipment in the track diagram, process the outward expansion area based on the equipment outline, and generate the first processing area.

[0088] It should be noted that there are various auxiliary devices along the track (such as switches, signals, etc.), which are also related to the safety of train operation and are areas where intrusion is prohibited.

[0089] Among them, track-related facilities refer to other railway facilities other than the main track.

[0090] Therefore, the server will identify the equipment outline of the track-attached equipment in the track diagram, and then expand the outer area based on the equipment outline to generate the first processing area.

[0091] In some embodiments, step S24 (processing the outward expansion area based on the device outline to generate a first processing area) includes S241-S243:

[0092] S241, the device outline is enlarged based on a set magnification to obtain the enlarged device outline.

[0093] The magnification factor is a pre-set outline magnification ratio based on the actual situation, such as 1.2 times. It can be pre-set by the personnel based on the actual situation. A corresponding magnification factor can be configured for each type of equipment. For example, a larger magnification factor is set for equipment of high importance, and a smaller magnification factor is set for equipment of low importance. This will not be elaborated on here.

[0094] Understandably, the server first extracts the original equipment outlines of each track accessory from the track diagram and obtains the set magnification factor for each device; then, using the geometric center of the equipment outline as a reference, the outline is magnified proportionally according to the set factor to generate the magnified equipment outline.

[0095] S242, obtain the intersection point of the magnified device outline and the region line of the outer expansion area, and perform segmentation processing on the device outline based on the intersection point to obtain the deleted outline located in the outer expansion area and the retained outline located outside the outer expansion area.

[0096] We will not only monitor the track for intrusions and generate dynamic zones, but also monitor important equipment around the track and further expand the area outwards through the equipment.

[0097] Therefore, the intersection points of the magnified device outline and the outer expansion area line are obtained; then, based on these intersection points, the device outline is divided into several segments to determine which parts are located within the outer expansion area and which parts are located outside the outer expansion area; the parts located within the outer expansion area are marked as deleted outlines, and the parts located outside the outer expansion area are marked as retained outlines.

[0098] S243, remove the area lines located within the enlarged device outline and delete the deleted outline to generate the first processing area.

[0099] Understandably, the process involves first deleting the area lines located in the inner and outer expansion areas of the enlarged device outline, and then deleting all the deleted outlines. This allows the enlarged device outline to merge with the area lines of the expansion area, thus expanding the expansion area with the device as the first expansion.

[0100] S3, identify the intrusion trajectory and intrusion outline of the intruder in the first processing area, and perform regional expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area.

[0101] In some embodiments, step S3 (identifying the intrusion trajectory and intrusion outline of the intruder in the first processing area, and performing region expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area) includes S31-S34:

[0102] S31, identify the intrusion trajectory that intersects with the first processing area, and the intrusion profile of the intruder that intersects with the first processing area.

[0103] It should be noted that we will customize and expand the first processing area based on the personnel's intrusion trajectory and the intrusion objects dropped in the track, so that subsequent customized monitoring and processing can be carried out for areas where personnel frequently intrude and areas where items are dropped.

[0104] Therefore, it will identify intrusion trajectories that intersect with the first processing area, as well as intrusion outlines of intruders that intersect with the first processing area. The intrusion trajectory refers to the walking trajectory of the intruder at the track. It can be the movement path formed in consecutive frame images when an intruder is detected, and the trajectory is then updated in the image. It can also be the traces left by the person walking. This is existing technology and will not be elaborated here. The intruder can be a dropped item.

[0105] S32, an intermediate processing area is obtained based on the union of the intrusion profile and the first processing area.

[0106] Understandably, the intrusion profile is merged with the first processing area to obtain the first processing area containing the complete range of the intruder and the original monitoring area, thus obtaining the intermediate processing area.

[0107] S33, the intrusion trajectory that does not pass through the track outline in the intermediate processing area is regarded as an erroneous trajectory, and the intrusion trajectory that passes through the track outline in the intermediate processing area is regarded as an intentional trajectory.

[0108] Understandably, each intrusion trajectory is compared with the track outline in the intermediate processing area for judgment. If the trajectory does not intersect with the track outline, it is judged as an accidental entry trajectory, that is, the person was just moving around the track and accidentally entered the monitoring area, but did not intrude into the rails. If the trajectory crosses the track outline, it is judged as an intentional trajectory, that is, the person crossed the rails after entering the monitoring area.

[0109] S34, perform region expansion processing on the intermediate processing area according to the accidental entry trajectory and the intentional trajectory respectively, and generate a second processing area.

[0110] In some embodiments, step S34 (performing region expansion processing on the intermediate processing area according to the accidental trajectory and the intentional trajectory respectively, to generate a second processing area) includes S341-S344:

[0111] S341, perform coordinate processing on the trajectory diagram.

[0112] S342, based on the erroneous trajectory located in the intermediate processing area, construct an extended circle corresponding to the erroneous trajectory.

[0113] It should be noted that stray trajectories typically involve briefly entering the edge of the monitoring area or lingering in the vicinity, posing a relatively low threat. To effectively track these trajectories and prevent them from approaching the orbit further, a moderately extended monitoring area needs to be constructed.

[0114] In some embodiments, step S342 (constructing an extended circle corresponding to the intrusive trajectory based on the intrusive trajectory located in the intermediate processing area) includes:

[0115] Obtain the coordinates of the intersection point between the erroneous trajectory and the area line of the intermediate processing zone, and calculate the average value of the intersection point coordinates to obtain the center coordinates.

[0116] It should be noted that the activity range of a stray trajectory often revolves around the edge of the intermediate processing area. If selected randomly, the expanded circle may fail to accurately cover the key areas of the trajectory, either missing the trajectory's activity range or over-expanding, increasing the risk of misjudgment. Determining the center of the circle by the intersection of the trajectory and the boundary of the intermediate processing area allows the expanded circle to anchor the key locations where the trajectory enters and exits the monitoring area. Subsequently, using the radius of the circle's center to the farthest distance the trajectory intrudes into the area can ensure that the expanded circle completely covers all activity points of the trajectory within the intermediate processing area, avoiding both insufficient coverage and meaningless expansion of the range.

[0117] It's easy to understand that since it's a case of accidental entry, people usually leave the monitoring area after entering. Therefore, there are usually multiple intersection coordinates, and the center coordinate can be calculated. If there is only one, then that coordinate is used directly as the center coordinate. For example, if someone accidentally enters, they immediately return along the same route.

[0118] The farthest distance from the center coordinates to the erroneous trajectory within the intermediate processing area is determined as the expansion radius. An expansion circle is constructed based on the expansion radius and with the center coordinates as the center.

[0119] It is understandable that this circular area is the extended circle corresponding to the current erroneous trajectory, ensuring that the activity range of the erroneous trajectory within the intermediate processing area can be covered relatively completely.

[0120] S343, construct an extended channel area corresponding to the intentional trajectory based on the intermediate connecting line of the intentional trajectory.

[0121] It should be noted that deliberate trajectories that pass directly through the track area pose a high threat and are often the result of nearby residents making regular intrusions for convenience. Therefore, it is necessary to predict their possible paths and expand the monitoring range. By constructing a channel-like extended area along the track direction, the possible approach paths of the intruders can be effectively covered.

[0122] In some embodiments, step S343 (constructing an extended channel region corresponding to the intentional trajectory based on the intermediate connecting line of the intentional trajectory) includes:

[0123] Connect the two endpoints of the intentional trajectory to generate the intermediate connecting line.

[0124] It should be noted that intentional trajectories often exhibit a clear directionality, pointing directly to the track area. Therefore, a channel-like extended area should be constructed to cover its possible path of travel.

[0125] The intermediate connecting line is shifted to both sides until it intersects with the intentional trajectory at only one point, thus generating an extended channel area.

[0126] Therefore, the starting and ending coordinates of the intentional trajectory are first extracted from the coordinated trajectory map, and an intermediate connecting line is generated connecting these two points. Then, this intermediate connecting line is translated to both sides in parallel, and the number of intersections between the translation line and the intentional trajectory is detected after each translation. When the translation line intersects the intentional trajectory at exactly one point, the translation is stopped, and the translation line at this time is the boundary of one side of the channel. The same operation is performed on the other side to obtain the boundary of the other side of the channel. Finally, the area enclosed by the two boundary lines and the line connecting the trajectory endpoints is the extended channel area. This area is close to the outermost edge of the intentional trajectory and can better cover the possible movement path of the intruder.

[0127] S344, a second processing area is generated based on the union of the extended circle and the extended channel area and the intermediate processing area.

[0128] Understandably, the process begins by merging the extended circles corresponding to all intrusive trajectories with the extended channel areas corresponding to all intentional trajectories. Then, the merged area is combined with the intermediate processing area to form a continuous space containing the original monitoring area and all extended areas, generating the final second processing area. This area fully covers the original monitoring range as well as the extended range for different threat trajectories.

[0129] Based on the above embodiments, it also includes:

[0130] By statistically analyzing the intrusion trajectories adjacent to the intrusion contour, a set of identification trajectories for each intrusion contour is obtained.

[0131] It should be noted that we can associate historical trajectories adjacent to the intrusion outline to obtain the complete movement process, thereby determining who caused the intrusion to fall. The trajectories of multiple objects to be identified can be put into the same set, and then viewed and confirmed sequentially. The direction can be selected to backtrack, and personnel do not need to view the entire video.

[0132] It is worth mentioning that the intrusion trajectory is adjacent to the intrusion outline. Here, "adjacent" can be within a preset range, such as within 1 meter. This will not be elaborated on here.

[0133] Therefore, the intrusion trajectories adjacent to the intrusion contour are counted to obtain the identification trajectory set of each intrusion contour.

[0134] The collection time of each intrusion trajectory in the identification trajectory set is used as the cutoff time.

[0135] It is understandable that the collection time of each intrusion trajectory in the identification trajectory set is used as the cutoff time.

[0136] The previous acquisition time corresponding to the cutoff time is retrieved as the backtracking time. Based on the backtracking time and the cutoff time, the backtracking time period of each intrusion trajectory is obtained.

[0137] It is easy to understand that the previous collection time corresponding to the cutoff time is retrieved as the retrospective time. Based on the retrospective time and the cutoff time, the retrospective time period of each intrusion trajectory is obtained, that is, the trajectory from zero to time, and the movement trajectory and behavior of each person during the retrospective time period are viewed.

[0138] The intrusion trajectories in the identification trajectory set are sorted based on the backtracking time period to obtain the identification trajectory sequence.

[0139] It is easy to understand that the investigation is carried out in chronological order. Therefore, by tracing back the time period, the intrusion trajectories in the identification trajectory set can be sorted to obtain the identification trajectory sequence, which facilitates subsequent backtracking.

[0140] See Figure 4 This is a schematic diagram of a railway track intrusion detection system based on periodic drone inspection, provided by an embodiment of the present invention. The railway track intrusion detection system based on periodic drone inspection includes:

[0141] The control module is used to control the drone to a set location to collect track maps of the railway tracks;

[0142] The generation module is used to process the dynamic open tracks and static closed tracks in the track map according to the facility expansion strategy, and generate the first processing area;

[0143] An extension module is used to identify the intrusion trajectory and intrusion outline of the intruder in the first processing area, and to perform regional expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area.

[0144] The present invention also provides a readable storage medium storing a computer program, which, when executed by a processor, is used to implement the methods provided in the various embodiments described above.

[0145] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0146] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0147] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A railway track intrusion identification method based on periodic unmanned aerial vehicle (UAV) inspection, characterized in that, include: Control the drone to a designated location to collect track maps of the railway tracks; Based on the facility expansion strategy, the dynamic open tracks and static closed tracks in the track diagram are processed to generate the first processing area, which includes: Identify the dynamic level crossing area of ​​the level crossing control device in the track diagram, and treat the remaining area as a static closed area; The track profile of the dynamic level crossing area is obtained as the dynamic open track, and the track profile of the static closed area is obtained as the static closed track. The opening and closing times of the level crossing guidance device at the corresponding set location are retrieved, and the dynamic open track and static closed track are processed based on the opening and closing times to obtain the outer expansion area; Identify the equipment outline of track-attached equipment in the track diagram, and process the outward expansion area based on the equipment outline to generate a first processing area; Identify the intrusion trajectory and intrusion outline of the intruder in the first processing area, and perform region expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area, including: Identify intrusion trajectories that intersect with the first processing area, as well as intrusion outlines of intruders that intersect with the first processing area; The intermediate processing area is obtained by the union of the intrusion profile and the first processing area; The intrusion trajectory that does not pass through the track outline within the intermediate processing area is regarded as an accidental intrusion trajectory, and the intrusion trajectory that passes through the track outline within the intermediate processing area is regarded as an intentional intrusion trajectory. The intermediate processing area is expanded based on the accidental trajectory and the intentional trajectory to generate a second processing area.

2. The method according to claim 1, characterized in that, The process of processing the dynamically open track and the statically closed track based on the open period and the closed period respectively yields the outer expansion area, including: Delete the dynamically open track during the open period, and move the outer contour of the static closed track to the outside of the railway track based on the preset outward expansion distance to obtain the open outward expansion area corresponding to the open period; Based on a preset outward expansion distance, the outer contours of the dynamic open track and the static closed track during the closed period are simultaneously moved to the outside of the railway track to obtain the closed outward expansion area corresponding to the closed period. The outer expansion area is obtained based on the open outer expansion area and the closed outer expansion area.

3. The method according to claim 1, characterized in that, The process of processing the outward expansion area based on the device outline to generate a first processing area includes: The device outline is enlarged based on a set magnification to obtain the enlarged device outline. Obtain the intersection point of the magnified device outline and the region line of the outer expansion area, and perform segmentation processing on the device outline based on the intersection point to obtain the deleted outline located within the outer expansion area and the retained outline located outside the outer expansion area; Remove the area lines located within the enlarged device outline and delete the deleted outline to generate the first processing area.

4. The method according to claim 1, characterized in that, The step of expanding the intermediate processing area based on the accidental trajectory and the intentional trajectory to generate a second processing area includes: The trajectory diagram is converted to coordinates. Based on the erroneous trajectory located in the intermediate processing area, construct an extended circle corresponding to the erroneous trajectory; Based on the intermediate connecting line of the intentional trajectory, construct an extended channel area corresponding to the intentional trajectory; A second processing area is generated based on the union of the extended circle and the extended channel area with the intermediate processing area.

5. The method according to claim 4, characterized in that, The step of constructing an extended circle corresponding to the mis-entry trajectory based on the mis-entry trajectory located within the intermediate processing area includes: Obtain the coordinates of the intersection point between the erroneous trajectory and the area line of the intermediate processing zone, and calculate the average of the intersection point coordinates to obtain the center coordinates; The farthest distance from the center coordinates to the erroneous trajectory within the intermediate processing area is determined as the expansion radius. An expansion circle is constructed based on the expansion radius and with the center coordinates as the center.

6. The method according to claim 4, characterized in that, The step of constructing an extended channel area corresponding to the intentional trajectory based on the intermediate connecting line of the intentional trajectory includes: Connect the two endpoints of the intentional trajectory to generate the middle connecting line; The intermediate connecting line is shifted to both sides until it intersects with the intentional trajectory at only one point, thus generating an extended channel area.

7. The method according to claim 1, characterized in that, Also includes: By statistically analyzing the intrusion trajectories adjacent to the intrusion contour, a set of identification trajectories for each intrusion contour is obtained; The collection time of each intrusion trajectory in the identification trajectory set is used as the cutoff time; The previous acquisition time corresponding to the cutoff time is retrieved as the backtracking time. Based on the backtracking time and the cutoff time, the backtracking time period of each intrusion trajectory is obtained. The intrusion trajectories in the identification trajectory set are sorted based on the backtracking time period to obtain the identification trajectory sequence.

8. A railway track intrusion identification system based on periodic unmanned aerial vehicle (UAV) inspection, characterized in that, include: The control module is used to control the drone to a set location to collect track maps of the railway tracks; The generation module is used to process the dynamic open tracks and static closed tracks in the track map according to the facility expansion strategy, and generate the first processing area, including: Identify the dynamic level crossing area of ​​the level crossing control device in the track diagram, and treat the remaining area as a static closed area; The track profile of the dynamic level crossing area is obtained as the dynamic open track, and the track profile of the static closed area is obtained as the static closed track. The opening and closing times of the level crossing guidance device at the corresponding set location are retrieved, and the dynamic open track and static closed track are processed based on the opening and closing times to obtain the outer expansion area; Identify the equipment outline of track-attached equipment in the track diagram, and process the outward expansion area based on the equipment outline to generate a first processing area; An extension module is used to identify the intrusion trajectory and intrusion outline of the intruder at the first processing area, and to perform region expansion processing on the first processing area based on the intrusion trajectory and intrusion outline to generate a second processing area, including: Identify intrusion trajectories that intersect with the first processing area, as well as intrusion outlines of intruders that intersect with the first processing area; The intermediate processing area is obtained by the union of the intrusion profile and the first processing area; The intrusion trajectory that does not pass through the track outline within the intermediate processing area is regarded as an accidental intrusion trajectory, and the intrusion trajectory that passes through the track outline within the intermediate processing area is regarded as an intentional intrusion trajectory. The intermediate processing area is expanded based on the accidental trajectory and the intentional trajectory to generate a second processing area.

Images