Railway foreign matter intrusion detection method and system based on motion process perception

Abstract

The invention provides a railway foreign matter intrusion detection method and system based on motion process perception, and belongs to the technical field of railway operation safety detection based on computer vision, and the method comprises the steps: obtaining two frames of railway monitoring video sequence images at a previous moment and a later moment; obtaining a foreground segmentation result of the image based on Gaussian filtering and binarization, image difference and morphological operation; constructing a historical foreground image based on a foreground segmentation result, and screening foreground regions; and according to the coordinates of the central point, distributing the single frame difference foreground result of the current frame to all historical foreground areas which are determined to move along the specific direction through screening. According to the invention, a brand new detection mode based on image foreground extraction is provided, comprehensive utilization of foreground results of each frame of image is realized in a motion process detection mode on the basis of foreground extraction, and analysis and screening of foreground target motion rationality are realized. False alarm caused by noise foreground is avoided while sensitive detection of all types of foreign matters is realized, the result reliability is high, and the practicability is high.

Description

Technical Field

[0001] The present invention relates to the technical field of railway operation safety detection based on computer vision, and in particular to a railway foreign object intrusion detection method and system based on motion process perception. Background Art

[0002] Safe operation is the first goal pursued by rail transit. Foreign objects intruding into railway limits is the main factor in the frequent occurrence of railway safety accidents. It poses a great threat to operational safety and also causes huge losses to the national economy and people’s property safety.

[0003] Traditional foreign body detection is done manually, which is time-consuming, labor-intensive and inefficient. In order to improve detection efficiency and reduce manual labor, a variety of automated foreign body detection technologies have been applied to the railway industry. According to the working principles of the detection system, the current railway foreign body intrusion detection technology can be divided into two types: contact and non-contact. The contact method mainly includes power grid detection, vibration fiber optic detection, etc., which requires foreign objects to contact the detection system to issue a foreign body intrusion alarm; the non-contact method includes infrared detection, laser radar detection, millimeter wave radar detection, video intelligent detection, etc., which do not require contact with foreign objects. The non-contact railway foreign body detection method has the advantages of simple deployment and a large detection range. Compared with other methods, the video intelligent detection method has the advantages of low cost, intuitive results, and good hardware foundation, and has become the main technical means for railway foreign body intrusion detection.

[0004] With the development of technology, image detection methods based on deep learning have shown the advantages of high accuracy and fast detection speed after sufficient training, and have become the mainstream method for railway foreign object intrusion detection. However, in actual use, methods based on deep learning also face challenges. As an open world, the types of foreign objects that may exist in railway scenes are theoretically infinite. When constructing a detection algorithm based on deep learning, the training samples can only cover as many characteristics of railway foreign objects as possible, making it difficult to achieve reliable detection of all possible intrusion targets in railway scenes.

[0005] The foreground extraction algorithm can detect all categories of foreign objects without training. Common methods include optical flow method, difference method, etc.

[0006] Zhang Lihua et al. proposed an improved ViBe algorithm to solve the problems of ghosting, dynamic background interference, and incomplete target detection results when the ViBe algorithm is used to detect moving targets. First, the pixel values ​​with large changes are eliminated in the process of initializing the background model using the mean method, so as to obtain a background image close to the real scene, which effectively suppresses the ghost phenomenon; then, the adaptive threshold radius is used instead of the fixed threshold radius to improve the algorithm's adaptability to dynamic backgrounds; finally, the median filter and area filling are used to filter out small noise points and fill the empty areas of the detection results respectively. A moving target tracking algorithm that integrates ViBe and CamShift is used to track moving targets in railway surveillance video images.

[0007] In view of the fact that jitter occurs in the vertical direction in railway scenes, Guo Baoqing et al. proposed a fast image de-jittering method combining one-dimensional grayscale projection with Gaussian filtering, which greatly improved the processing speed and achieved good de-jittering effect. In view of the complex and changeable background, a background update algorithm based on the statistical distribution of foreground targets was proposed. The target dispersion index was defined to determine the order of row and column projections, and background update was achieved by statistically analyzing the foreground target distribution. This improved the speed and solved the ghost problem that was difficult to solve with the traditional background update algorithm. Finally, the foreground target was obtained by background difference to realize the detection of foreign intrusion targets.

[0008] The existing railway foreign object intrusion detection method based on foreground extraction has improved the detection accuracy, but it is still easy to produce a large number of noisy detection results in complex railway scenarios, which increases the workload of manual review and makes it difficult to achieve the goal of automated detection. Summary of the invention

[0009] The purpose of the present invention is to provide a railway foreign object intrusion detection method and system based on motion process perception that overcomes sample limitations and achieves reliable detection of all categories of railway foreign object intrusion, so as to solve at least one technical problem existing in the above-mentioned background technology.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions:

[0011] In a first aspect, the present invention provides a railway foreign body intrusion detection method based on motion process perception, comprising:

[0012] Acquire two frames of railway monitoring video sequence images at the previous and next moments;

[0013] Based on Gaussian filtering and binarization, image difference, and morphological operations, the foreground segmentation results of the image are obtained;

[0014] Based on the foreground segmentation results, a historical foreground map is constructed to screen the foreground area;

[0015] For a certain foreground area, based on the center point coordinates, the foreground result with the smallest difference in image width coordinates between the center point and the boundary of the foreground area outer frame in the direction of motion and less than the pixel threshold is taken as the target positioning result of the foreground area in the current frame.

[0016] Optionally, when the image size is smaller than a preset threshold size, the input image is synchronously enlarged to the threshold size, and a preliminary foreground segmentation result is obtained after completing Gaussian filtering and binarization and completing image difference. The morphological processing of first image corrosion and then image expansion is used to eliminate small noise foregrounds while reducing possible holes in the foreground, thereby obtaining a higher quality foreground segmentation result.

[0017] Optionally, the construction of the historical foreground map includes: adding the current frame difference result to the past frame difference result, and the foreground area in the historical foreground map corresponding to the moving target continues to extend as the foreground results accumulate, and the development direction of the foreground area represents the position and direction of the moving target.

[0018] Optionally, the screening of the foreground area includes: data association based on the foreground area external frame in the historical background image of the previous and next frames and motion feature judgment, wherein data association refers to using IOU to characterize the position difference of each foreground area, and implementing the association of the previous and next frame data based on the Hungarian algorithm, thereby establishing or updating the historical foreground area spatial information trajectory of the target at each moment.

[0019] Optionally, set the result D of the foreground area of ​​the current frame h The tracking result f of the trajectory corresponding to the foreground area in the previous frame t Each contains U1 foreground area bounding box information and V1 trajectory box information. To achieve the correlation matching between the two, first calculate each bounding box result D h (i)(i∈[1,U1]) and each trajectory result f t The IOU value of the target box corresponding to (j)(j∈[1,V1]) is used as the similarity metric, and then the current frame detection result and trajectory result are matched based on the Hungarian algorithm.

[0020] Optionally, feature analysis is performed on the successfully matched foreground area external frame and the matched foreground area spatial information trajectory at the four corner points of the tracking result of the previous frame to determine the target motion mode and the corresponding motion situation.

[0021] In a second aspect, the present invention provides a railway foreign object intrusion detection system based on motion process perception, comprising:

[0022] An acquisition module is used to acquire two frames of railway monitoring video sequence images at previous and next moments;

[0023] The segmentation module is used to obtain the foreground segmentation result of the image based on Gaussian filtering and binarization, image difference, and morphological operations;

[0024] The screening module is used to construct a historical foreground map based on the foreground segmentation results and screen the foreground area;

[0025] The matching module is used to take, for a certain foreground area, based on the coordinates of the center point, the foreground result with the smallest difference in image width coordinates between the center point and the boundary of the foreground area outer frame in the direction of motion and less than the pixel threshold as the target positioning result of the foreground area in the current frame.

[0026] In a third aspect, the present invention provides a non-transitory computer-readable storage medium, which is used to store computer instructions. When the computer instructions are executed by a processor, the railway foreign object intrusion detection method based on motion process perception as described in the first aspect is implemented.

[0027] In a fourth aspect, the present invention provides a computer device comprising a memory and a processor, wherein the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the railway foreign object intrusion detection method based on motion process perception as described in the first aspect.

[0028] In a fifth aspect, the present invention provides an electronic device comprising: a processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the railway foreign object intrusion detection method based on motion process perception as described in the first aspect.

[0029] The beneficial effects of the present invention are as follows: a brand-new detection mode based on image foreground extraction is proposed, and comprehensive utilization of foreground results of each frame image is achieved by motion process detection on the basis of foreground extraction. The analysis and screening of the rationality of foreground target motion is realized without the need for model training. While realizing sensitive detection of all categories of foreign objects, false alarms caused by noise foreground are avoided, and the results are highly reliable and practical.

[0030] Additional advantages of the present invention will be more clearly given in the following description or learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0032] Figure 1 This is a schematic diagram of the overall process of the railway foreign object intrusion detection method based on motion process perception described in an embodiment of the present invention.

[0033] Figure 2 This is a functional structure diagram of the foreground extraction module described in an embodiment of the present invention.

[0034] Figure 3 This is an overall functional structure diagram of the motion process perception module described in an embodiment of the present invention.

[0035] Figure 4 This is a schematic diagram of a method for screening foreground region motion relations according to an embodiment of the present invention.

[0036] Figure 5 This is a schematic diagram of a method for locating a target corresponding to a foreground area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The embodiments of the present invention 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 with the same or similar functions. The embodiments described below by the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0038] It should be understood by those skilled in the art that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.

[0039] It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have a meaning consistent with that in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless as defined herein.

[0040] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or groups thereof.

[0041] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" 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 invention. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. Different embodiments or examples described in this specification and features of different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

[0042] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0043] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0044] Example 1

[0045] In this embodiment 1, a railway foreign object intrusion detection system based on motion process perception is first provided, including: an acquisition module, used to acquire two frames of railway monitoring video sequence images at previous and subsequent moments; a segmentation module, used to acquire the foreground segmentation result of the image based on Gaussian filtering and binarization, image difference, and morphological operations; a screening module, used to construct a historical foreground map based on the foreground segmentation result, and to screen the foreground area; a matching module, used to, for a certain foreground area, based on the coordinates of the center point, take the foreground result with the smallest difference in the image width coordinates between the center point and the boundary of the foreground area outer frame in the motion direction and less than the pixel threshold as the target positioning result of the foreground area in the current frame.

[0046] In this embodiment 1, the above-mentioned system is used to implement a railway foreign object intrusion detection method based on motion process perception, which includes: obtaining two frames of railway monitoring video sequence images at previous and subsequent moments; obtaining the foreground segmentation result of the image based on Gaussian filtering and binarization, image difference, and morphological operations; constructing a historical foreground map based on the foreground segmentation result, and screening the foreground area; according to the center point coordinates, assigning the current frame single frame differential foreground result to all historical foreground areas that have been screened and determined to be moving in a specific direction. For a certain foreground area, according to the center point coordinates, the foreground result with the smallest difference between the image width coordinates of the center point and the boundary of the foreground area external frame in the direction of motion and less than the pixel threshold is used as the target positioning result of the foreground area in the current frame.

[0047] Among them, when the image size is smaller than the preset threshold size, the input image is synchronously enlarged to the threshold size. After completing Gaussian filtering and binarization and completing image difference, the preliminary foreground segmentation result is obtained. Through the morphological processing of image corrosion and then image expansion, the small noise foreground is eliminated while reducing the possible holes in the foreground, and a higher quality foreground segmentation result is obtained.

[0048] The construction of the historical foreground map includes: adding the current frame difference result with the past frame difference result. The foreground area in the historical foreground map corresponding to the moving target continues to extend as the foreground results accumulate. The development direction of the foreground area represents the position and direction of the moving target.

[0049] The construction process of the historical prospect map is as follows:

[0050] R_his t =add(R_his t-1 ,R t )

[0051] Among them, R_his t 、R_his t-1 is the historical foreground image at the t and t-1 frame time, R t It is the difference result of a single frame at time t, and the add operation is the addition operation of the foreground image.

[0052] The screening of the foreground area includes: data association based on the external frame of the foreground area in the historical background image of the previous and next frames and motion feature judgment. Data association refers to the use of IOU to characterize the position difference of each foreground area, and the association of the previous and next frame data is realized based on the Hungarian algorithm, so as to establish or update the historical foreground area spatial information trajectory of the target at each moment. The formula is as follows:

[0053]

[0054] Let the result D of the external frame of the foreground area of ​​the current frame h The tracking result f of the trajectory corresponding to the foreground area in the previous frame t Each contains U1 foreground area bounding box information and V1 trajectory box information. To achieve the correlation matching between the two, first calculate each bounding box result D h (i)(i∈[1,U1]) and each trajectory result f t The IOU value of the target box corresponding to (j)(j∈[1,V1]) is used as the similarity metric, and then the current frame detection result and trajectory result are matched based on the Hungarian algorithm.

[0055] The four corner points of the tracking result of the successfully matched foreground area external frame and the spatial information trajectory of the matched foreground area are analyzed for features in the previous frame to determine the target motion mode and the corresponding motion situation. For a target moving to the left, if the result of the external frame of its historical foreground image in the current frame and its corresponding trajectory in the previous frame satisfies that at least one corner point on the right side coincides, and the coordinate difference of the left corner point in the width direction of the corresponding image is greater than the threshold, then it is considered that the movement of the target in the last two frames conforms to the motion process, so that the leftward motion flag bit of the historical foreground area corresponding to the target is increased by one. If this flag bit exceeds 3, it is marked that this foreground area has passed the screening and its corresponding moving target is a leftward moving target. The screening method for the historical foreground area corresponding to the rightward moving target is consistent with the aforementioned leftward motion discrimination method.

[0056] Example 2

[0057] like Figure 1 As shown, this embodiment 2 provides a railway foreign object intrusion detection system based on motion process perception, including three parts: foreground extraction module 1, motion process perception module 2, and target positioning module 3. The target tracking and foreground extraction related methods are integrated to construct a full-category foreign object intrusion detection algorithm suitable for railways, and a foreground extraction algorithm based on inter-frame difference is proposed; based on the foreground segmentation results and combined with the target tracking data association, the perception of the motion process of the intruding target and the further screening of the foreground information are realized, thereby realizing the detection of all categories of railway intrusion targets while avoiding noise foreground interference. The railway foreign object intrusion detection system based on motion process perception proposed in this embodiment 2 can issue an information warning for the intruding foreign object after detecting the railway intrusion target, so as to prompt the staff to deal with and repair it in time.

[0058] like Figure 2 As shown, the foreground extraction module includes four steps: image size judgment and small-size image enlargement, Gaussian filtering and binarization, image difference, and morphological operation. The input is two frames of railway monitoring video sequence images at the previous and next moments. Considering that the target size in the railway scene is small in the image, difference and post-processing at this time may cause the loss of the foreground. Therefore, when the image size is less than (1280, 720), the input image is synchronously enlarged to this size. After completing Gaussian filtering and binarization and completing image difference, a preliminary foreground segmentation result can be obtained, but the image may still contain some small noise at this time. In this embodiment, the morphological processing of first image corrosion and then image expansion is used to eliminate the small noise foreground while reducing the possible holes in the foreground as much as possible, so as to obtain higher quality foreground information. In particular, after the image is enlarged, the size of the structural element of the image corrosion is synchronously changed from (5, 5) to (3, 3), and the size of the structural element of the image expansion is always maintained at (9, 9).

[0059] like Figure 3As shown, the input of the motion process perception module is the result of the foreground extraction module, which includes two steps: obtaining the historical foreground map and screening the foreground area.

[0060] The process of motion perception is based on the construction of the historical foreground map. That is, in the process of real-time processing of railway surveillance video, the current frame difference result is added to the past frame difference result. The foreground area in the historical foreground map corresponding to the moving target continues to extend with the accumulation of foreground results. The development direction of the foreground area represents the position and direction of the moving target.

[0061] The construction process of the historical prospect map is as follows:

[0062] R_his t =add(R_his t-1 ,R t )

[0063] Among them, R_his t 、R_his t-1 For the t , the historical foreground image at frame t-1, R t It is the difference result of a single frame at time t, and the add operation is the addition operation of the foreground image.

[0064] Based on the historical foreground map, in this embodiment, the concept of historical foreground area is proposed, which is defined by the foreground parts and their bounding boxes in the historical foreground map. The screening process of the foreground area is based on the data association and motion feature judgment of the foreground area bounding box in the historical background map of the previous and next frames, where data association refers to the use of IOU to characterize the position difference of each foreground area, and the data association of the previous and next frames is realized based on the Hungarian algorithm, so as to establish or update the spatial information trajectory of the historical foreground area of ​​the target at each moment, and the formula is as follows:

[0065]

[0066] Let the result D of the external frame of the foreground area of ​​the current frame h The tracking result f of the trajectory corresponding to the foreground area in the previous frame t Each contains U1 foreground area bounding box information and V1 trajectory box information. To achieve the correlation matching between the two, first calculate each bounding box result D h (i)(i∈[1,U1]) and each trajectory result f t The IOU value of the target box corresponding to (j)(j∈[1,V1]) is used as the similarity metric. Then the current frame detection result and the trajectory result are matched based on the Hungarian algorithm.

[0067] The foreground area bounding box results of the previous and next frames that satisfy the spatial matching relationship are identified as belonging to the same moving target. In the process of real-time processing of the image sequence, the successfully matched foreground area bounding box and the matching foreground area spatial information trajectory are subjected to feature analysis at the four corner points of the tracking result of the previous frame, such as Figure 4 As shown, two motion modes and four motion conditions of leftward motion and rightward motion are constructed based on the image width direction. For a leftward moving target, if the bounding box of its historical foreground image in the current frame and its corresponding trajectory in the previous frame satisfy that at least one corner point on the right side coincides, and the coordinate difference of the left corner point in the corresponding image width direction is greater than the threshold, then the motion of the target in the last two frames is considered to be consistent with the motion process, so that the leftward motion flag bit of the historical foreground area corresponding to the target is increased by one. If this flag bit exceeds 3, it is marked that this foreground area has passed the screening and its corresponding moving target is a leftward moving target. The screening method for the rightward moving target corresponding to the historical foreground area is consistent with the aforementioned leftward motion discrimination method.

[0068] like Figure 5 As shown in , after completing the motion process detection and foreground area screening, the target positioning module distributes the current frame single frame differential foreground results to all historical foreground areas that have been screened and determined to be moving in a specific direction according to the center point coordinates. Figure 5 As shown, for a certain foreground area, the foreground result with the smallest difference between the image width coordinates of the center point and the boundary of the foreground area outer frame in the moving direction and less than the pixel threshold is used as the target positioning result of the foreground area in the current frame. In this embodiment, the pixel threshold is 20 pixels.

[0069] Example 3

[0070] This embodiment 3 provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the railway foreign object intrusion detection method based on motion process perception as described above is implemented. The method includes:

[0071] Acquire two frames of railway monitoring video sequence images at the previous and next moments;

[0072] Based on Gaussian filtering and binarization, image difference, and morphological operations, the foreground segmentation results of the image are obtained;

[0073] Based on the foreground segmentation results, a historical foreground map is constructed to screen the foreground area;

[0074] For a certain foreground area, based on the center point coordinates, the foreground result with the smallest difference in image width coordinates between the center point and the boundary of the foreground area outer frame in the direction of motion and less than the pixel threshold is taken as the target positioning result of the foreground area in the current frame.

[0075] Example 4

[0076] This embodiment 4 provides a computer device, including a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the above-mentioned railway foreign object intrusion detection method based on motion process perception, the method comprising:

[0077] Acquire two frames of railway monitoring video sequence images at the previous and next moments;

[0078] Based on Gaussian filtering and binarization, image difference, and morphological operations, the foreground segmentation results of the image are obtained;

[0079] Based on the foreground segmentation results, a historical foreground map is constructed to screen the foreground area;

[0080] For a certain foreground area, based on the center point coordinates, the foreground result with the smallest difference in image width coordinates between the center point and the boundary of the foreground area outer frame in the direction of motion and less than the pixel threshold is taken as the target positioning result of the foreground area in the current frame.

[0081] Example 5

[0082] This embodiment 5 provides an electronic device, including: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes instructions for implementing the railway foreign object intrusion detection method based on motion process perception as described above, the method comprising:

[0083] Acquire two frames of railway monitoring video sequence images at the previous and next moments;

[0084] Based on Gaussian filtering and binarization, image difference, and morphological operations, the foreground segmentation results of the image are obtained;

[0085] Based on the foreground segmentation results, a historical foreground map is constructed to screen the foreground area;

[0086] For a certain foreground area, based on the center point coordinates, the foreground result with the smallest difference in image width coordinates between the center point and the boundary of the foreground area outer frame in the direction of motion and less than the pixel threshold is taken as the target positioning result of the foreground area in the current frame.

[0087] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0089] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0091] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative work on the basis of the technical solution disclosed in the present invention should be included in the scope of protection of the present invention.

Claims

1. A railway foreign object intrusion detection method based on motion process perception, It is characterized in that include: Acquire two frames of railway monitoring video sequence images at the previous and next moments; Based on Gaussian filtering and binarization, image difference, and morphological operations, the foreground segmentation results of the image are obtained; Based on the foreground segmentation results of each frame, a historical foreground map is constructed to screen the foreground area; For a certain foreground area, based on the coordinates of the center point of the foreground result in the current frame, the foreground result with the smallest difference between the image width coordinates of the center point and the boundary of the foreground area's external frame in the direction of motion and less than the set pixel threshold is used as the target positioning result of the foreground area in the current frame.

2. According to the method for detecting railway foreign body intrusion based on motion process perception according to claim 1, It is characterized in that When the image size is smaller than the preset threshold size, the input image is synchronously enlarged to the threshold size. After completing Gaussian filtering, binarization and image difference, the preliminary foreground segmentation result is obtained. Through the morphological processing of image corrosion and then image expansion, the small noise foreground is eliminated while reducing the possible holes in the foreground, and a higher quality foreground segmentation result is obtained.

3. According to the method for detecting railway foreign body intrusion based on motion process perception according to claim 1, It is characterized in that The construction of the historical foreground map includes: adding the current frame difference result with the past frame difference result. The foreground area in the historical foreground map corresponding to the moving target continues to extend as the foreground results accumulate. The development direction of the foreground area represents the position and direction of the moving target.

4. According to the method for detecting railway foreign body intrusion based on motion process perception according to claim 1, It is characterized in that The screening of the foreground area includes: data association based on the external frame of the foreground area in the historical background image of the previous and next frames and motion feature judgment, where data association refers to the use of IOU to characterize the position difference of each foreground area, and the association of the previous and next frame data is realized based on the Hungarian algorithm, so as to establish or update the spatial information trajectory of the historical foreground area of ​​the target at each moment.

5. According to claim 4, the railway foreign body intrusion detection method based on motion process perception, It is characterized in that Let the result D of the external frame of the foreground area of ​​the current frame h The tracking result f of the trajectory corresponding to the foreground area in the previous frame t Each contains U1 foreground area bounding box information and V1 trajectory box information. To achieve the correlation matching between the two, first calculate each bounding box result D h (i)(i∈[1,U1]) and each trajectory result f t The IOU value of the target box corresponding to (j)(j∈[1,V1]) is used as the similarity metric, and then the current frame detection result and trajectory result are matched based on the Hungarian algorithm.

6. According to claim 5, the railway foreign body intrusion detection method based on motion process perception, It is characterized in that The successfully matched foreground area external frame and the matched foreground area spatial information trajectory are subjected to feature analysis at the four corner points of the tracking result of the previous frame to determine the target motion mode and the corresponding motion situation.

7. A railway foreign object intrusion detection system based on motion process perception, It is characterized in that include: An acquisition module is used to acquire two frames of railway monitoring video sequence images at previous and next moments; The segmentation module is used to obtain the foreground segmentation result of the image based on Gaussian filtering and binarization, image difference, and morphological operations; The screening module is used to construct a historical foreground map based on the foreground segmentation results of each frame and screen the foreground area; The matching module, for a certain foreground area, takes the foreground result with the smallest difference between the image width coordinates of the center point and the boundary of the foreground area's external frame in the direction of motion and less than the set pixel threshold as the target positioning result of the foreground area in the current frame based on the coordinates of the center point of the foreground result in the current frame.

8. A non-transitory computer-readable storage medium, It is characterized in that The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by the processor, the railway foreign object intrusion detection method based on motion process perception as described in any one of claims 1-6 is implemented.

9. A computer device, It is characterized in that It includes a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the railway foreign object intrusion detection method based on motion process perception as described in any one of claims 1-6.

10. An electronic device, It is characterized in that include: A processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the railway foreign object intrusion detection method based on motion process perception as described in any one of claims 1-6.

Images