Warehousing type pantograph three-dimensional imaging and detection method, system and device

Abstract

The invention discloses a warehousing type pantograph three-dimensional imaging and detection method, system and device, and belongs to the technical field of photoelectric detection of rail traffic safety. The method comprises the following steps: acquiring spatial information of upper and lower edges of a pantograph by adopting binocular stereoscopic vision, acquiring three-dimensional point cloud information of the upper surface of the pantograph by adopting a 3D line structured light three-dimensional vision sensor, and splicing three-dimensional point cloud of the upper and lower edges of the pantograph and three-dimensional point cloud data of the upper surface of the pantograph. According to the total point cloud data of the pantograph, an unsupervised anomaly detection algorithm is adopted to detect the defects of the pantograph, meanwhile, according to the distance between the lower surface and the upper surface of the pantograph, the abrasion and the remaining amount of a pantograph carbon contact strip are calculated, and pantograph detection is completed; according to the method, binocular stereoscopic vision is adopted to obtain the space information of the lower edge of the pantograph, the 3D line structured light three-dimensional vision sensor is adopted to obtain the three-dimensional point cloud information of the upper surface of the pantograph, and the abrasion, the surplus and the defect condition of the pantograph can be calculated through data splicing.

Description

Technical Field

[0001] The invention relates to a method, system and device for three-dimensional imaging and detection of a storage-type pantograph, and belongs to the technical field of photoelectric detection of rail transit safety. Background Art

[0002] As an important traction device for railway transportation, electric locomotives have the characteristics of long mileage, harsh operating environment, and complex maintenance. Among them, the pantograph is a key component of the train's energy source, and a good operating state is the key link to ensure the safe and efficient operation of the train. For a long time, the inspection of pantograph wear or surface defects has been mostly carried out by manual climbing, which has the disadvantages of low efficiency, poor accuracy, and high safety hazards, which seriously restricts the accuracy and efficiency of current train pantograph inspections.

[0003] At present, with the rapid development of photoelectric detection technology, it is widely used in rail transit safety detection due to its advantages of non-contact, high precision and large range, effectively improving detection efficiency and accuracy. However, due to the complex structure of locomotive pantographs, harsh working environment and various fault patterns, traditional photoelectric measurement devices cannot meet the detection needs of pantograph status parameters.

[0004] Therefore, realizing three-dimensional imaging of train pantographs and conducting dynamic visual inspection of their wear and defects is of great significance for ensuring the quality of train safety inspections and improving inspection efficiency. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide a method, system and device for three-dimensional imaging and detection of a storage-type pantograph, so as to realize three-dimensional imaging of a train pantograph and perform dynamic visual detection of its wear, defects, etc.

[0006] In order to achieve the above purpose / solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0007] A first aspect: A method for three-dimensional imaging and detection of a storage-type pantograph, the method comprising:

[0008] Collect the upper and lower edge data of the pantograph, process the upper and lower edge data of the pantograph, and obtain the three-dimensional point cloud of the upper and lower edges of the pantograph ;

[0009] Collect the upper surface image of the pantograph and obtain the 3D point cloud information of the pantograph through the 3D reconstruction algorithm;

[0010] Based on the pantograph 3D point cloud information, the pantograph 3D point cloud data is obtained by integrating the vehicle speed and time;

[0011] Process the pantograph 3D point cloud data to obtain the 3D point cloud data of the upper surface of the pantograph ;

[0012] The 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ;

[0013] The pantograph total point cloud data An unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated according to the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection.

[0014] Optionally, the upper and lower edge data of the pantograph are collected and processed to obtain a three-dimensional point cloud of the upper and lower edges of the pantograph. ,include:

[0015] Collect the upper and lower edge images of the pantograph, use image edge extraction to obtain the center coordinates of the edge image, and use the stereo vision sensor system parameters to achieve three-dimensional reconstruction of the upper and lower edges of the pantograph using the triangulation principle to obtain the three-dimensional point cloud of the upper and lower edges of the pantograph. .

[0016] Optionally, the pantograph three-dimensional point cloud data is processed to obtain the three-dimensional point cloud data of the upper surface of the pantograph. ,include:

[0017] Collect the pantograph laser light strip image, and extract the center point coordinates of the light strip in the pantograph laser light strip image based on the Steger algorithm;

[0018] The coordinates of the center point of the light strip are used to obtain the pantograph interface contour data through the line structured light three-dimensional measurement principle;

[0019] Based on the pantograph interface contour data and the known vehicle speed and direction, the 3D point cloud data of the upper surface of the pantograph is obtained by integrating the distance in the passing direction. .

[0020] Optionally, the three-dimensional point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ,include:

[0021] Make a target with a planar light point, use a 3D structured light three-dimensional vision sensor and a binocular stereo vision sensor to synchronously collect the target, and calculate the external parameters between the binocular stereo vision sensor and the 3D structured light three-dimensional vision sensor. ;

[0022] The 3D point cloud of the upper and lower edges of the pantograph Through external parameters Convert to 3D point cloud of upper and lower edges of pantograph

[0023] The 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ;in, represents the rotation matrix, Represents a displacement vector.

[0024] Optionally, the three-dimensional point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph The formula for splicing is as follows:

[0025] .

[0026] Optionally, the pantograph total point cloud data The unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated based on the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection, including:

[0027] Collect a large number of normal pantograph 3D point cloud images;

[0028] The 3D point cloud information of the lower surface of the pantograph is used as the ground reference line of the carbon slide ; Perform multiple cross-section divisions on the upper surface of the pantograph to obtain multiple sets of cross-sectional profiles , obtained by taking the average value of multiple sets of profile data ,calculate Points and baseline The vertical distance between the two is used as the wear value curve of the pantograph carbon slide plate. ;

[0029] Wear value curve Integrate along the bottom surface of the pantograph carbon slide to obtain the remaining amount of the pantograph carbon slide;

[0030] According to the remaining amount of the pantograph carbon slide plate, automatic detection of pantograph anomalies is carried out, and abnormal patterns are automatically discovered from unlabeled 3D point cloud data.

[0031] A second aspect: A three-dimensional imaging and detection system for a storage-type pantograph, the system comprising:

[0032] The first image acquisition device is used to acquire the upper and lower edge data of the pantograph;

[0033] The first preprocessing module is used to process the upper and lower edge data of the pantograph to obtain the three-dimensional point cloud of the upper and lower edges of the pantograph ;

[0034] The second image acquisition device is used to acquire the upper surface image of the pantograph and obtain the three-dimensional point cloud information of the pantograph through a three-dimensional reconstruction algorithm;

[0035] Data processing module: used to obtain pantograph 3D point cloud data based on pantograph 3D point cloud information according to vehicle speed and time integration;

[0036] The second preprocessing module is used to process the pantograph 3D point cloud data to obtain the 3D point cloud data of the upper surface of the pantograph. ;

[0037] Data splicing module: used to stitch the 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ;

[0038] Calculation module: used to calculate the total point cloud data of the pantograph An unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated according to the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection.

[0039] The third aspect: A three-dimensional imaging and detection device for a storage-type pantograph, comprising a binocular stereo vision sensor and a 3D line structured light three-dimensional vision sensor arranged above the storage detection shed, and multiple groups of magnetic steel and photoelectric opposing beam switch devices arranged on the track, wherein the binocular stereo vision sensor is used to image and measure the lower edge of the pantograph, the 3D line structured light three-dimensional vision sensor is used to perform high-definition imaging of the pantograph to be detected, the magnetic steel is used to trigger and collect train arrival signals, and measure the train running speed, and the photoelectric opposing beam switch triggers the binocular stereo vision sensor.

[0040] Optionally, the binocular stereo vision sensor includes two high-resolution area array industrial cameras and a matching light source projector.

[0041] Optionally, the 3D line structured light three-dimensional vision sensor includes a high-speed 3D camera and a line laser, wherein the high-speed 3D camera is used for imaging near-infrared laser, and the line laser is used for illuminating the high-speed 3D camera.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. The present invention proposes a storage-type pantograph 3D imaging and detection system and device, which adopts a combined installation layout mode of binocular stereo vision and 3D line structured light 3D vision sensor to ensure synchronous detection of the upper and lower edges and upper surface of the pantograph; an automatic vehicle receiving and vehicle speed calculation device composed of multiple groups of magnetic steels is used, supplemented by a photoelectric trigger device to trigger multiple groups of cameras to complete image data collection, which can realize all-weather and multi-model pantograph 3D imaging;

[0044] 2. The present invention proposes a calibration method for a binocular stereo vision sensor and a 3D line structured light three-dimensional vision sensor based on a combination of a spherical target and feature points. The 3D structured light three-dimensional vision sensor collects spatial target data, and the binocular stereo vision sensor collects feature point data, thereby realizing global calibration of the binocular stereo vision sensor and the 3D line structured light three-dimensional vision sensor, which is convenient for unifying spatial detection data for detecting pantograph wear and defects.

[0045] 3. The present invention proposes a wear and defect detection method based on pantograph three-dimensional data, which uses binocular stereo vision to obtain the spatial information of the lower edge of the pantograph, and uses a 3D line structured light three-dimensional vision sensor to obtain the three-dimensional point cloud information of the upper surface of the pantograph. The wear, remaining amount and defect of the pantograph can be calculated by data splicing;

[0046] 4. The detection device and method of the present invention have a certain degree of versatility. By designing different visual sensor sizes, matching different sensor shape layouts, and adopting different control devices and processes, the detection requirements in scenarios such as pantograph production can be met. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The figure is a schematic flow chart of the three-dimensional imaging and detection method of the storage type pantograph of the present invention;

[0048] Figure 2 Shown is a schematic diagram of the three-dimensional imaging and detection device for a storage-type pantograph of the present invention. DETAILED DESCRIPTION

[0049] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0050] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0051] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0052] Example 1

[0053] like Figure 1 As shown, the present invention discloses a three-dimensional imaging and detection method for a storage-type pantograph, the method comprising:

[0054] By arranging multiple sets of magnetic steel and photoelectric beam switching devices on the track, when the train approaches, the remote receiving magnetic steel triggers and collects the train arrival signal. When the locomotive passes through the measurement area, the speed measuring magnetic steel measures the train running speed. Synchronously, according to the set image acquisition resolution, the 3D structured light three-dimensional vision sensor is triggered at different trigger frequencies to collect images. The three-dimensional point cloud information of the pantograph is obtained through the three-dimensional reconstruction algorithm, and the complete three-dimensional point cloud data of the pantograph is obtained through the integration of vehicle speed and time. At the same time, the binocular stereo vision sensor is triggered by the photoelectric beam switch to collect the upper and lower edge data of the pantograph. The upper and lower edge information of the pantograph is obtained through binocular stereo vision reconstruction. When the train passes, the image data acquisition is stopped and the data integration and necessary data upload are completed.

[0055] Step 1: The binocular stereo vision sensor collects the upper and lower edge images of the pantograph, uses image edge extraction to obtain the center coordinates of the edge image, and uses the stereo vision sensor system parameters to achieve 3D reconstruction of the upper and lower edges of the pantograph using the triangulation principle to obtain its 3D point cloud. .

[0056] Step 2: The 3D structured light three-dimensional vision sensor collects the laser light strip image through a high-frequency trigger camera, extracts the coordinates of the center point of the light strip using the Steger algorithm, and obtains the pantograph interface contour data through the line structured light three-dimensional measurement principle. Based on the known vehicle speed and direction, the pantograph three-dimensional point cloud data is obtained by integrating the distance in the passing direction. .

[0057] Step 3: Make a target with a planar light point, so that the 3D structured light three-dimensional vision sensor and the binocular stereo vision sensor can collect the target synchronously, and calculate the external parameters between the binocular stereo vision sensor and the 3D structured light three-dimensional vision sensor. , realize the splicing of the above two types of data to obtain the total point cloud data , which is used to identify pantograph defects. The specific formula is as follows:

[0058] .

[0059] Based on the 3D point cloud data of the upper surface of the pantograph generated in step 3 and the upper and lower edges, the unsupervised anomaly detection algorithm is used to detect the pantograph defects, and the wear and remaining amount of the pantograph carbon slide plate are calculated according to the distance between the lower surface and the upper surface. It is mainly divided into the following four steps:

[0060] Step 1: After the equipment is installed, collect a large number of normal pantograph 3D point cloud images;

[0061] Step 2: Measurement of the wear and remaining amount of the pantograph carbon slide. The three-dimensional point cloud information of the lower surface of the carbon slide obtained in the above steps is used as the ground reference line of the carbon slide. ; Perform multiple cross-section divisions on the upper surface of the pantograph to obtain multiple sets of cross-sectional profiles , obtained by taking the average value of multiple sets of profile data ,calculate Points and baseline The vertical distance between the two is used as the wear value curve of the pantograph carbon slide plate. At the same time, the wear value curve The remaining amount of the pantograph carbon slide board is obtained by integrating along the bottom surface of the pantograph carbon slide board.

[0062] Step 4: Automatic detection of pantograph anomalies. Unsupervised anomaly detection algorithms aim to automatically discover abnormal patterns from unlabeled 3D point cloud data. These algorithms usually identify abnormal points by analyzing the local or global features of the point cloud and the distribution characteristics of the point cloud data. Use an autoencoder to encode and decode the 3D point cloud data and calculate the reconstruction error. The reconstruction error of abnormal points is usually high due to the large difference in distribution between abnormal points and normal data. Unsupervised anomaly detection algorithms based on 3D point clouds identify abnormal points by extracting features, estimating density, clustering analysis, calculating distance or reconstruction error, etc. These algorithms do not require pre-labeled abnormal samples for training.

[0063] Through the above four steps, the pantographs entering the warehouse are automatically inspected, and the pantograph abnormalities can be accurately identified. This method does not require a large number of negative samples, but only needs to collect positive samples after the vehicle passes. At the same time, it has low requirements on vehicle models and environments, has certain versatility, and can be extended to the pantograph inspection of trucks, buses, EMUs, and subway trains.

[0064] The present invention proposes a three-dimensional imaging and detection system and device for a storage-type pantograph. Through system calibration, image acquisition, data conversion and intelligent recognition, supplemented by a linear array high-definition camera image acquisition module and an electrical control device, three-dimensional imaging of the storage-type pantograph can be achieved, and the size of the upper and lower slides of the pantograph can be calculated. The pantograph anomaly detection algorithm is used to achieve automatic and intelligent detection of pantographs of multiple models at all times of the day, and the relevant detection results are sent to the train inspection personnel in a timely manner to guide the vehicle reception inspection and maintenance.

[0065] Example 2

[0066] A storage-type pantograph three-dimensional imaging and detection system, the system comprising:

[0067] The first image acquisition device is used to acquire the upper and lower edge data of the pantograph;

[0068] The first preprocessing module is used to process the upper and lower edge data of the pantograph to obtain the three-dimensional point cloud of the upper and lower edges of the pantograph ;

[0069] The second image acquisition device is used to acquire the upper surface image of the pantograph and obtain the three-dimensional point cloud information of the pantograph through a three-dimensional reconstruction algorithm;

[0070] Data processing module: used to obtain pantograph 3D point cloud data based on pantograph 3D point cloud information according to vehicle speed and time integration;

[0071] The second preprocessing module is used to process the pantograph 3D point cloud data to obtain the 3D point cloud data of the upper surface of the pantograph. ;

[0072] Data splicing module: used to stitch the 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ;

[0073] Calculation module: used to calculate the total point cloud data of the pantograph An unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated according to the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection.

[0074] Example 3

[0075] like Figure 2 As shown, a three-dimensional imaging and detection device for a storage-type pantograph is disclosed. The device is hoisted above the storage inspection shed, wherein the visual sensor includes two high-resolution area array industrial cameras and a matching light source projector, forming a binocular stereo vision sensor, which mainly images and measures the lower edge of the pantograph; the 3D line structured light three-dimensional vision sensor is composed of a high-speed 3D camera and a line laser, equipped with a filter and a mechanical mounting. Assisted by power supply, trigger and data acquisition cables, high-definition imaging of the pantograph to be inspected is achieved. The industrial camera has high photosensitivity and high sensitivity, and the 3D camera, assisted by a narrow-band filter, can achieve imaging of near-infrared lasers. The laser projector is mainly used for lighting. It has uniform and stable light output and can work continuously for a long time, meeting the requirements of outdoor all-weather imaging. By arranging multiple sets of magnetic steel and photoelectric switching devices on the track, when a train approaches, the remote receiving magnet triggers and collects the train arrival signal. When the locomotive passes through the measurement area, the speed measuring magnet measures the train speed. Synchronously, according to the set image acquisition resolution, the 3D structured light three-dimensional vision sensor is triggered at different trigger frequencies to collect images.

[0076] This device uses binocular stereo vision to obtain spatial information of the lower edge of the pantograph, and uses a 3D line structured light three-dimensional vision sensor to obtain three-dimensional point cloud information of the upper surface of the pantograph. The wear, remaining amount and defects of the pantograph can be calculated through data stitching. By designing different visual sensor sizes, matching different sensor shape layouts, and using different control devices and processes, the detection needs in scenarios such as pantograph production can be met.

[0077] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A three-dimensional imaging and detection method for a storage-type pantograph, characterized in that: The method comprises: Collect the upper and lower edge data of the pantograph, process the upper and lower edge data of the pantograph, and obtain the three-dimensional point cloud of the upper and lower edges of the pantograph ; Collect the upper surface image of the pantograph and obtain the 3D point cloud information of the pantograph through the 3D reconstruction algorithm; Based on the pantograph 3D point cloud information, the pantograph 3D point cloud data is obtained by integrating the vehicle speed and time; Process the pantograph 3D point cloud data to obtain the 3D point cloud data of the upper surface of the pantograph ; The 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ; The pantograph total point cloud data An unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated according to the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection.

2. The method for three-dimensional imaging and detection of a warehouse-type pantograph according to claim 1, characterized in that: The pantograph upper and lower edge data are collected and processed to obtain the pantograph upper and lower edge three-dimensional point clouds. ,include: Collect the upper and lower edge images of the pantograph, use image edge extraction to obtain the center coordinates of the edge image, and use the stereo vision sensor system parameters to achieve three-dimensional reconstruction of the upper and lower edges of the pantograph using the triangulation principle to obtain the three-dimensional point cloud of the upper and lower edges of the pantograph. .

3. The method for three-dimensional imaging and detection of a storage-type pantograph according to claim 1, characterized in that: The three-dimensional point cloud data of the pantograph is processed to obtain the three-dimensional point cloud data of the upper surface of the pantograph. ,include: Collect the pantograph laser light strip image, and extract the center point coordinates of the light strip in the pantograph laser light strip image based on the Steger algorithm; The coordinates of the center point of the light strip are used to obtain the pantograph interface contour data through the line structured light three-dimensional measurement principle; Based on the pantograph interface contour data and the known vehicle speed and direction, the 3D point cloud data of the upper surface of the pantograph is obtained by integrating the distance in the passing direction. .

4. The method for three-dimensional imaging and detection of a storage-type pantograph according to claim 1, characterized in that: The three-dimensional point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ,include: Make a target with a planar light point, use a 3D structured light three-dimensional vision sensor and a binocular stereo vision sensor to synchronously collect the target, and calculate the external parameters between the binocular stereo vision sensor and the 3D structured light three-dimensional vision sensor. ; The 3D point cloud of the upper and lower edges of the pantograph Through external parameters Convert to 3D point cloud of upper and lower edges of pantograph The 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ;in, represents the rotation matrix, Represents a displacement vector.

5. The method for three-dimensional imaging and detection of a warehouse-type pantograph according to claim 4, characterized in that: The three-dimensional point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph The formula for splicing is as follows: 。 6. The method for three-dimensional imaging and detection of a storage-type pantograph according to claim 1, characterized in that: The pantograph total point cloud data The unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated based on the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection, including: Collect a large number of normal pantograph 3D point cloud images; The 3D point cloud information of the lower surface of the pantograph is used as the ground reference line of the carbon slide ; Perform multiple cross-section divisions on the upper surface of the pantograph to obtain multiple sets of cross-sectional profiles , obtained by taking the average value of multiple sets of profile data ,calculate Points and baseline The vertical distance between the two is used as the wear value curve of the pantograph carbon slide plate. ; Wear value curve Integrate along the bottom surface of the pantograph carbon slide to obtain the remaining amount of the pantograph carbon slide; According to the remaining amount of the pantograph carbon slide plate, automatic detection of pantograph anomalies is carried out, and abnormal patterns are automatically discovered from unlabeled 3D point cloud data.

7. A three-dimensional imaging and detection system for a storage-type pantograph, characterized in that: The system comprises: The first image acquisition device is used to acquire the upper and lower edge data of the pantograph; The first preprocessing module is used to process the upper and lower edge data of the pantograph to obtain the three-dimensional point cloud of the upper and lower edges of the pantograph ; The second image acquisition device is used to acquire the upper surface image of the pantograph and obtain the three-dimensional point cloud information of the pantograph through a three-dimensional reconstruction algorithm; Data processing module: used to obtain pantograph 3D point cloud data based on pantograph 3D point cloud information according to vehicle speed and time integration; The second preprocessing module is used to process the pantograph 3D point cloud data to obtain the 3D point cloud data of the upper surface of the pantograph. ; Data splicing module: used to stitch the 3D point cloud of the upper and lower edges of the pantograph 3D point cloud data of the upper surface of the pantograph Splice to get the total point cloud data of the pantograph ; Calculation module: used to calculate the total point cloud data of the pantograph An unsupervised anomaly detection algorithm is used to detect pantograph defects. At the same time, the wear and remaining amount of the pantograph carbon slide plate are calculated according to the distance between the lower and upper surfaces of the pantograph to complete the pantograph detection.

8. A three-dimensional imaging and detection device for a storage-type pantograph, characterized in that: It includes a binocular stereo vision sensor and a 3D line structured light three-dimensional vision sensor arranged above the storage inspection shed, and multiple groups of magnetic steel and photoelectric beam switching devices arranged on the track. The binocular stereo vision sensor is used to image and measure the lower edge of the pantograph, the 3D line structured light three-dimensional vision sensor is used to perform high-definition imaging of the pantograph to be detected, the magnetic steel is used to trigger and collect train arrival signals and measure the train running speed, and the photoelectric beam switch triggers the binocular stereo vision sensor.

9. The three-dimensional imaging and detection device for a storage-type pantograph according to claim 8, characterized in that: The binocular stereo vision sensor includes two high-resolution area array industrial cameras and a matching light source projector.

10. The three-dimensional imaging and detection device for a storage-type pantograph according to claim 8, characterized in that: The 3D line structured light three-dimensional vision sensor includes a high-speed 3D camera and a line laser. The high-speed 3D camera is used for imaging near-infrared lasers, and the line laser is used for illuminating the high-speed 3D camera.

Images