Coordinate system-based identification method and system, and electronic device

Abstract

The present disclosure provides a coordinate system-based identification method and system, wherein the identification method comprises: identifying objects in a target range; establishing an identification coordinate system with a target object in the target range as an origin, the identification coordinate system including objects around the target object; and comparing the established identification coordinate system with a target coordinate system to identify the target object and the objects around the target object.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of computer recognition, and particularly relates to a recognition method and system based on a coordinate system and an electronic device. BACKGROUND

[0002] At present, the recognition of objects is generally based on shooting images of the regions where the objects are located, and then extracting the objects from the shot images. Figure 1 An image obtained by shooting is exemplarily shown. The existing image recognition technology can recognize a vehicle running on an elevated bridge, a road under the elevated bridge, trees on both sides of the road, and a factory in the distance from the shot image. This image recognition technology is relatively simple and cannot recognize the relative positions and distances between the objects, for example, cannot determine the actual relative positions and distances between the vehicle running on the elevated bridge and the trees on the roadside under the bridge.

[0003] In order to accurately recognize the actual relative positions and distances between the objects, some existing technologies (for example, the patent application with the publication number CN109492639A) use a binocular camera, a structure camera, or a TOF (Time of flight) depth camera to shoot and obtain the original data (for example, the RGB information of two lenses, the RGB information of a camera, and the time difference of light return) of a scene where a target object is located, calculate the distance between the target object and the camera, form a two-dimensional bounding box based on the position contour, filter a target two-dimensional bounding box of a target position from the two-dimensional bounding box, and obtain the image coordinates of the target position from the target two-dimensional bounding box. This method only recognizes the coordinate position of the target object in the image. In a scene where the recognition result of the image needs to be used for corresponding control, this object recognition in the entire environment within the image range does not meet the control requirements, because this recognition is not accurate, and based on this inaccurate recognition, an incorrect control result will be generated.

[0004] For example, in the highway transportation industry, the vehicle needs to be controlled according to the red and green lights, green belts, and zebra crossings in front; in the rail transportation industry, the train needs to be controlled according to the signal lights, transponders, etc. in front, which all need accurate recognition.

[0005] Therefore, an accurate recognition technology is needed to achieve correct control. SUMMARY

[0006] To solve at least one of the above problems, the present application provides a recognition method and system based on a coordinate system and an electronic device.

[0007] A recognition method based on a coordinate system, the recognition method comprising:

[0008] recognizing objects in a target range;

[0009] establishing an identification coordinate system with a target object in the target range as the origin, the identification coordinate system including objects around the target object;

[0010] comparing the established identification coordinate system with a target coordinate system to identify the target object and the objects around the target object.

[0011] In some embodiments,

[0012] After identifying the objects in the target range, defining the size of the objects according to the use environment;

[0013] forming the objects in the coordinate system according to the defined size.

[0014] In some embodiments, the size includes the length, width and height of the objects, and the coordinate system is a three-dimensional coordinate system.

[0015] In some embodiments, the identifying the objects in the target range includes,

[0016] identifying the relative positions and distances of the objects in the target range before identifying the objects.

[0017] In some embodiments, the comparing the established identification coordinate system with a target coordinate system to identify the target object and the objects around the target object includes,

[0018] determining whether the identification coordinate system is the same as the target coordinate system, and if so, considering the identified objects in the target range to be correct; otherwise, considering the identified objects in the target range to be incorrect.

[0019] The present disclosure also provides a coordinate system-based identification system, which includes an identification device and a processing device, wherein,

[0020] The identification device is configured to identify objects in a target range.

[0021] The processing device is configured to establish an identification coordinate system with a target object in the target range as the origin, the identification coordinate system including objects around the target object; and compare the established identification coordinate system with a target coordinate system to identify the target object and the objects around the target object.

[0022] In some embodiments, after the identification device identifies the objects in the target range, the processing device defines the size of the objects according to the use environment; and forms the objects in the coordinate system according to the defined size.

[0023] In some embodiments, the processing device compares the established identification coordinate system with a target coordinate system to identify the target object and objects around the target object, including determining whether the identification coordinate system is the same as the target coordinate system, and if so, considering the identified objects in the target range to be correct; otherwise, considering the identified objects in the target range to be incorrect.

[0024] The present disclosure also provides an electronic device, including one or more processors, one or more memories electrically connected with the processors, wherein,

[0025] The one or more memories are used to store computer programs;

[0026] The one or more processors are used to execute the computer programs stored in the one or more memories to perform the coordinate system-based identification method according to any one of the above.

[0027] Compared with the prior art, the present application has the following advantages:

[0028] In the embodiments of the present disclosure, the identified front object is abstracted to form a coordinate system and compared with an internally stored coordinate system, and through the comparison of the coordinate systems, the positional relationship between each of the front settings can be accurately determined, and accurate control instructions can be generated.

[0029] In the embodiments of the present disclosure, not only the visual recognition from the image itself is considered, but also the features of multiple objects are considered, and a coordinate system with the target object as the origin is combined to perform identification through the coordinate system.

[0030] Other features and advantages of the present application will be described in the following description, and some will become apparent from the description, or will be understood through implementation of the present application. The purposes and other advantages of the present application can be achieved and obtained through the structures indicated in the specification, claims, and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0032] Figure 1 A schematic diagram of taking a picture is shown;

[0033] Figure 2 A flowchart of a coordinate system-based identification method according to an embodiment of the present disclosure is shown;

[0034] Figure 3 A schematic diagram of a rail vehicle identification structure is shown according to an embodiment of the present disclosure;

[0035] Figure 4 A schematic diagram of a different type of signal is shown according to an embodiment of the present disclosure;

[0036] Figure 5 A schematic diagram of a two-dimensional coordinate system is shown according to an embodiment of the present disclosure;

[0037] Figure 6 A structural block diagram of an identification system is shown according to an embodiment of the present disclosure;

[0038] Figure 7 An electronic device structural block diagram is shown according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0039] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in detail with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0040] Figure 2 A flowchart of a coordinate system-based identification method is shown according to an embodiment of the present disclosure, as shown in Figure 2 The coordinate system-based identification method of the present embodiment includes: identifying objects in a target range; establishing an identification coordinate system with a target object in the target range as the origin, the identification coordinate system including objects around the target object; and comparing the established identification coordinate system with a target coordinate system to identify the target object and the objects around the target object.

[0041] In the present embodiment, the rail vehicle is taken as an example to illustrate and explain the technical solutions of the present disclosure, but it should be noted that the present disclosure is not limited to rail vehicles. Figure 3 A schematic diagram of a rail vehicle identification structure is shown according to an embodiment of the present disclosure. As shown in Figure 3As shown, the left track has a train running, and the right track has no train running. The train on the left track runs towards the signal post marked as S6, and when the train runs to a certain distance from the signal post S6, the recognition system on the train recognizes the equipment in the three-dimensional modeling recognition area of the geographical feature in front, which can include a laser radar, a long-focus camera, a near-focus camera, a microwave radar, etc. The recognition of the recognition area can include signal boxes, signal posts, and trackside equipment, etc. As known, the trackside equipment includes switch machines, signal posts, track circuits, transponders, or electrical boxes, etc. In order to more safely and accurately recognize the trackside equipment, the above-mentioned recognition system can also use a 2-by-2-by-2 safety calculation method.

[0042] In the embodiment of the present disclosure, the recognition system of the train can compare the physical features of the longitudinal section of the track line obtained by scanning with the stored map to determine the position of the train, so as to realize the autonomous positioning of the train on the line. The train control system (TCMS) calculates the allowable running speed of the train according to the electronic map, and monitors the running of the train through the allowable running speed; when it is determined that the train is approaching a certain distance from the signal post in the electronic map (for example, approaching a predetermined distance from the S6 signal post), the recognition system automatically searches for the front signal post and recognizes the signboard and display color of the signal post. According to the principle of detecting how far and setting the speed as high, the running speed of the current position of the train is calculated according to the distance between the current position of the train and the position of the signal post, the braking performance of the train, and the protection distance. Exemplarily, if the recognition system now detects within an effective distance of 800 meters, the current allowable speed of the train is calculated according to 800 meters; if it detects 20 meters, the current allowable speed of the train is calculated according to 20 meters.

[0043] The recognition system recognizes the signal post by scanning the mechanism shape, light position sequence, and signal post signboard (unique identification number) of the signal post, and checks according to the stored three-dimensional geographical information at the signal post to determine the name of the signal post. According to the determined name of the signal post, the absolute position of the train is corrected by querying the signal post in the stored map.

[0044] 1) Signal post itself feature recognition

[0045] The recognition system can recognize the signal post by the shape of the signal post and the signal post signboard (unique identification number, such as Figure 3The "S6" in the image identification process identifies the object ahead as a signal: using a telephoto and a short-focus camera to capture the outline of the signal on the track and the sequence of the illuminated lights in the signal positions, the image recognition process uses the outline and the position of the illuminated lights in the signal mechanism as key identification points for identification. This is compared with the object outline data (which can be supplemented with the scanning conclusions of the object outline using LiDAR, such as outline shape and size) and the light position sequence data stored in the database to identify the object as a signal and the type of signal.

[0046] The system uses photographs of the illuminated light colors taken by a short-throw camera as samples for color recognition. Color recognition technology is then used to identify the colors of the illuminated lights on the traffic signal. The system also uses a short-throw camera to photograph the traffic signal sign, and image recognition technology is used to perform text recognition on the photographed sign to identify the traffic signal's name. Based on the traffic signal's name, its outline, number of light positions, and color sequence can be found in a database. Then, the collected data on the traffic signal's outline, identification number, number of light positions, and the color sequence of the currently illuminated light (a total of four markers) are compared with the database to perform traffic signal identification.

[0047] Figure 4 Schematic diagrams of different types of signals according to embodiments of the present disclosure are shown. Figure 4 As shown, signal signs S7 and S11 indicate three-light signal controllers, while signal sign S9 indicates two-light signal controllers. The color sequences of the signal lights differ between S7 and S9 type signal controllers, meaning the different colors are arranged in different sequences at the signal light positions. In this embodiment, the outline of an object and the sequence of different colors at the signal light positions can be obtained through a recognition system, and compared with the stored signal outline and signal light position sequence to determine the signal controller and its type. In this embodiment, the route information prepared for the train by the ground signaling system can be determined based on the identified signal controller name and the displayed colors of the signal lights.

[0048] This embodiment of the disclosure can identify the geographical environmental features surrounding the signal. A three-dimensional lidar in the identification system is used to scan the train's direction of travel. The above example uses a signal to illustrate the signal identification method; other devices can also be identified using the same method, and this embodiment will not elaborate further.

[0049] In the embodiments of the present disclosure, the scanned objects are abstracted in terms of the objects themselves (contours), including two-dimensional abstraction, such as the length and width of the objects, or three-dimensional abstraction, such as the three-dimensional contours of the objects, including the length and width. All the scanned objects are abstracted in terms of coordinate positions with the signal machine as the coordinate origin, including two-dimensional coordinate position abstraction with the two-dimensional abstraction results or three-dimensional coordinate position abstraction with the three-dimensional abstraction results, and then the objects are defined according to the use environment, for example, the minimum size and the maximum size of the two-dimensional or three-dimensional contours are specified, and then the object data is processed again to form a coordinate graph of the positions of the objects of the specified size in the coordinate system of the two-dimensional or three-dimensional information abstraction.

[0050] Figure 5 A schematic diagram of a two-dimensional coordinate system according to an embodiment of the present disclosure is shown. As shown in the figure, a two-dimensional coordinate system with the signal machine as the origin is constructed, in which Figure 5 Figure 3 In the three-dimensional modeling and recognition area of the geographic features, the transponders, cable boxes, equipment boxes and other equipment are labeled with the coordinates of the equipment in the coordinate system, and different sizes of the equipment are displayed according to the abstraction results.

[0051] In the two-dimensional coordinate system, the main geographic information includes the objects corresponding to the signal machine, such as the contours, the number and the positions in the coordinate system of the trackside equipment with regular size (which can be adjusted according to the environment). In the three-dimensional coordinate system, the height of the objects is also included. In the embodiments of the present disclosure, the three-dimensional information database can be set according to the needs of the field environment, and if necessary, the necessary increase or decrease of the objects or the setting of the objects in different positions can be set to distinguish.

[0052] The formed coordinate system and the stored coordinate system are compared to achieve the recognition and verification of the target object (the signal machine) and the geographic environment around the target object. For example, if the formed coordinate system is compared with the stored coordinate system and it is found that they are completely the same, it is considered that the recognized target object (for example, the signal machine) and the objects (transponders, equipment boxes, cable boxes) around the target object are correct, and at this time, the train operation can be controlled according to the recognition result. Compared with the prior art in which only a certain target in the photographed image is analyzed, in the embodiments of the present disclosure, the recognition accuracy is improved through the recognition of multiple objects in the image, the establishment of the coordinate system and the comparison of the coordinate systems.

[0053] On the basis of the above method, the present disclosure also provides a system for implementing the above method, as shown in Figure 6 ​As shown, the coordinate system-based identification system of the embodiment of the present disclosure comprises an identification device and a processing device. The identification device is configured to execute the step of identifying the object in the target range as described above. The processing device is configured to establish an identification coordinate system with the target object in the target range as the origin, the identification coordinate system comprising the object around the target object, and compare the established identification coordinate system with a target coordinate system to identify the target object and the object around the target object.

[0054] Based on the above method, the embodiment of the present disclosure further provides an electronic device, which comprises Figure 7 As shown, the electronic device of the embodiment of the present disclosure comprises one or more processors, one or more memories electrically connected to the processors, and the one or more memories are configured to store computer programs. The one or more processors are configured to execute the computer programs stored in the memories to execute the coordinate system-based identification method as described above.

[0055] It should be noted that in the embodiment of the present disclosure, a two-dimensional coordinate system is exemplarily shown, and a three-dimensional coordinate system (also including the height of the object) is also applicable to the present disclosure.

[0056] In the embodiment of the present disclosure, the correlation between the target object and the surrounding objects and the correlation between the objects are comprehensively considered, and finally the accurate identification is realized through the establishment of the coordinate system and the comparison of the coordinate system.

[0057] As known, the identification of a single object may have a risk of error. For example, there may be multiple signal machines along the track, and there are multiple sets of traffic lights at the two adjacent intersections of the road. If only the signal machines or the traffic lights are identified, there may be a risk of error, for example, the intention is to obtain the traffic light state of the first intersection, but the traffic light of the adjacent second intersection is photographed and identified, which will have extremely dangerous consequences. In the embodiment of the present disclosure, not only the analysis of the target object is considered, but also the analysis of the objects around the target object. Compared with the identification of a single object, the accuracy of the identification is effectively improved. Considering the joint identification of multiple objects, there is still a certain risk, for example, there are multiple signal machines, transponders and equipment boxes along the track. If only the multiple devices are identified, there may also be a problem of identification error. Therefore, in the embodiment of the present disclosure, the positional relationship between the multiple objects is also considered, the target object and the surrounding environment are expressed and stored in the form of the coordinate system, and the target object and the surrounding objects are expressed and stored in the form of the coordinate system. The expression and storage of the environment are more concise, and the data calling is more rapid. In summary, in the embodiment of the present disclosure, the features of the surrounding environment and the positional relationship between the objects are comprehensively considered, and the comparison and analysis are performed in the form of the coordinate system, thereby realizing the fast and accurate identification. For the issuance of the control instruction based on the identification result, the risk is effectively reduced.

[0058] Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood that modifications can be made to the foregoing embodiments, or additional implementations can be implemented, without departing from the spirit and scope of the inventive subject matter. Accordingly, the present application is not limited to the implementations described herein, but is intended to be defined by the claims set forth below, and equivalents thereof.

Claims

1. A coordinate system-based recognition method, the recognition method comprising: Identifying objects within a target range includes identifying the relative positions and distances between each object within the target range; A recognition coordinate system is established with the target object within the target range as the origin, and the recognition coordinate system includes the objects surrounding the target object; The established recognition coordinate system is compared with the target coordinate system to identify the target object and the objects around the target object; wherein, if the recognition coordinate system is the same as the target coordinate system, the identified objects within the target range are considered to be correct; otherwise, the identified objects within the target range are considered to be incorrect.

2. The identification method according to claim 1, wherein, After identifying objects within the target range, the size of the objects is defined according to the usage environment; The object is formed in the coordinate system according to the defined dimensions.

3. The identification method according to claim 1, wherein, The dimensions include the length, width, and height of the object, and the coordinate system is a three-dimensional coordinate system.

4. A coordinate-based recognition system, the recognition system comprising a recognition device and a processing device, wherein, The identification device is used to identify objects within a target range, including identifying the relative positions and distances between various objects within the target range; The processing device is used to establish a recognition coordinate system with the target object within the target range as the origin, and the recognition coordinate system includes objects surrounding the target object; The established recognition coordinate system is compared with the target coordinate system to identify the target object and the objects around the target object; if the recognition coordinate system is the same as the target coordinate system, the identified objects within the target range are considered correct; otherwise, the identified objects within the target range are considered incorrect.

5. The identification system according to claim 4, wherein, After the identification device identifies an object within the target range, the processing device defines the size of the object according to the usage environment; and forms the object in the coordinate system according to the defined size.

6. An electronic device, the electronic device comprising one or more processors and one or more memories electrically connected to the processors, wherein, The one or more memories are used to store computer programs; The one or more processors are configured to execute a computer program stored in the one or more memories to perform the coordinate-based identification method as described in any one of claims 1-3.

Images