A method and device for map detection

Abstract

An embodiment of the present invention provides a map detection method and apparatus, which relates to the technical field of data processing. The method includes: obtaining a first image collected by an unmanned device during driving. Identifying a first area where the device is located in the first image. Obtaining pixel points characterizing the features of the first area as first feature points. Determining the actual positions corresponding to the respective first feature points in the actual environment in which the unmanned device travels according to the pixel positions of the obtained first feature points. Determining a first position of the device according to the actual positions corresponding to the respective first feature points. Comparing the first position with a second position of the device recorded in the device map to obtain a map detection result. Applying the solution provided by the embodiment of the present invention can improve the efficiency of map detection.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a map detection method and apparatus. Background Art

[0002] In the prior art, a map can be used to record the positions of devices, so that functions such as device positioning and navigation can be realized based on the map. For example, a map can be used to record the positions of devices indoors, and an unmanned vehicle can travel indoors based on the positions of the devices recorded in the above map. However, if the position of the device changes, or new devices appear in the map area represented by the map, the positions of the devices recorded in the map will no longer be accurate. In order to determine the accuracy of the positions of the devices recorded in the map, it is necessary to detect the map.

[0003] In the prior art, it is often necessary to manually confirm the current positions of the devices in the map area and compare them with the positions of the devices recorded in the above map, so as to detect the map. However, when there are many objects in the above map area, the efficiency of map detection based on manual operation is low. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a map detection method and apparatus to improve the efficiency of map detection. The specific technical solutions are as follows:

[0005] In a first aspect, the embodiments of the present invention provide a map detection method, which includes:

[0006] Obtain a first image collected by an unmanned vehicle during driving;

[0007] Identify a first area where the device is located in the first image;

[0008] Obtain pixel points representing the characteristics of the first area as first feature points;

[0009] According to the pixel positions of the obtained first feature points, determine the actual positions corresponding to the respective first feature points in the actual environment in which the unmanned vehicle travels;

[0010] According to the actual positions corresponding to the respective first feature points, determine the first position of the device;

[0011] Compare the first position with the second position of the device recorded in the device map to obtain a map detection result.

[0012] In an embodiment of the present invention, the comparing the first position with the second position of the device recorded in the device map to obtain a map detection result includes:

[0013] Calculate the distance between the first position and the second position of the device recorded in the device map;

[0014] When the minimum distance among the calculated distances is greater than or equal to the first preset distance, generate a map detection result indicating that the device position is incorrect.

[0015] In one embodiment of the present invention, obtaining the pixel points representing the characteristics of the first region as the first feature points includes:

[0016] Obtain the pixel points representing the characteristics of the first region and having depth values within a preset depth range as the first feature points.

[0017] In one embodiment of the present invention, obtaining the pixel points representing the characteristics of the first region as the first feature points includes:

[0018] Obtain the pixel points representing the characteristics of the first region with corner response values greater than the preset response value as the first feature points.

[0019] In one embodiment of the present invention, the second position of the device recorded in the device map is determined by the following method:

[0020] Acquire a second image collected during the driving of the driverless device;

[0021] Identify the second region where the device is located in the second image;

[0022] Obtain the pixel points representing the characteristics of the second region as the second feature points;

[0023] According to the pixel positions of the obtained second feature points, determine the actual positions corresponding to the second feature points in the actual environment where the driverless device travels;

[0024] According to the actual positions corresponding to the second feature points, determine the second position of the device.

[0025] In one embodiment of the present invention, determining the second position of the device according to the actual positions corresponding to the second feature points includes:

[0026] According to the actual positions corresponding to the second feature points, determine the third position of the device;

[0027] Calculate the distance between the third position and the second position of the device currently recorded in the device map;

[0028] When the minimum distance among the calculated distances is greater than or equal to the second preset distance, add the third position as the new second position of the device to the device map;

[0029] When the minimum distance is less than a second preset distance, update a second position of a target device in the device map according to the third position, where the target device is a device whose distance between the second position before update and the third position is the minimum distance.

[0030] In one embodiment of the present invention, the method further includes:

[0031] Obtain a target image collected by an unmanned device, where the target image is an image collected when the unmanned device determines that the distance between its own position and a second position of a device recorded in the device map is less than a third preset distance;

[0032] Identify a third area where a display panel of an instrument is located in the target image;

[0033] Identify a fourth area for displaying information in the third area;

[0034] Perform character recognition on the fourth area to obtain instrument information.

[0035] In a second aspect, an embodiment of the present invention provides a map detection device, where the device includes:

[0036] An image acquisition module, configured to acquire a first image collected by an unmanned device during driving;

[0037] An area recognition module, configured to identify a first area where a device is located in the first image;

[0038] A feature point acquisition module, configured to acquire pixel points characterizing the features of the first area as first feature points;

[0039] An actual position determination module, configured to determine actual positions corresponding to the respective first feature points in the actual environment in which the unmanned device travels according to the pixel point positions of the respective first feature points obtained;

[0040] A first position determination module, configured to determine a first position of the device according to the actual positions corresponding to the respective first feature points;

[0041] A result acquisition module, configured to compare the first position with a second position of the device recorded in the device map to obtain a map detection result.

[0042] In one embodiment of the present invention, the result acquisition module is specifically configured to:

[0043] Calculate the distance between the first position and the second position of the device recorded in the device map;

[0044] When the minimum distance among the calculated distances is greater than or equal to the first preset distance, a map detection result indicating a device position error is generated.

[0045] In one embodiment of the present invention, the feature point obtaining module is specifically configured to:

[0046] Obtain pixel points that characterize the features of the first region and whose depth values belong to a preset depth interval as first feature points.

[0047] In one embodiment of the present invention, the feature point obtaining module is specifically configured to:

[0048] Obtain pixel points that characterize the features of the first region and whose corner point response values are greater than a preset response value as first feature points.

[0049] In one embodiment of the present invention, the device further includes a second position determination module for determining a second position of the device recorded in the device map. The second position determination module includes:

[0050] An image acquisition sub-module for acquiring a second image collected by the driverless device during driving;

[0051] A region recognition sub-module for recognizing a second region where the device is located in the second image;

[0052] A feature point obtaining sub-module for obtaining pixel points that characterize the features of the second region as second feature points;

[0053] An actual position determination sub-module for determining the actual positions corresponding to the respective second feature points in the actual environment in which the driverless device travels according to the pixel point positions of the obtained second feature points;

[0054] A second position determination sub-module for determining the second position of the device according to the actual positions corresponding to the respective second feature points.

[0055] In one embodiment of the present invention, the second position determination sub-module is specifically configured to:

[0056] Determine a third position of the device according to the actual positions corresponding to the respective second feature points;

[0057] Calculate the distance between the third position and the second position of the device currently recorded in the device map;

[0058] When the minimum distance among the calculated distances is greater than or equal to the second preset distance, add the third position as the new second position of the device to the device map;

[0059] When the minimum distance is less than a second preset distance, update a second position of a target device in the device map according to the third position, where the target device is a device whose distance between the second position before update and the third position is the minimum distance.

[0060] In one embodiment of the present invention, the apparatus further includes:

[0061] A target image acquisition module, configured to acquire a target image collected by an unmanned device, where the target image is an image collected when the unmanned device determines that the distance between its own position and a second position of a device recorded in the device map is less than a third preset distance;

[0062] A third area acquisition module, configured to identify a third area where a display panel of an instrument is located in the target image;

[0063] A fourth area identification module, configured to identify a fourth area for displaying information in the third area;

[0064] An information identification module, configured to perform character recognition on the fourth area to obtain instrument information.

[0065] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;

[0066] The memory is used to store a computer program;

[0067] The processor, when executing the program stored on the memory, implements the method steps of any one of the first aspect.

[0068] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method steps of any one of the first aspect are implemented.

[0069] In a fifth aspect, an embodiment of the present invention further provides a computer program product including instructions, which when running on a computer, causes the computer to execute the method steps of any one of the first aspect.

[0070] Advantageous effects of the embodiments of the present invention:

[0071] During the process of map detection through the embodiments of the present invention, a first image collected by the driverless device during driving can be obtained. Identify the first area where the device is located in the first image. And obtain the pixel points characterizing the features of the first area as the first feature points. Determine the actual positions corresponding to the respective first feature points in the actual environment in which the driverless device travels according to the pixel positions of the obtained first feature points. Determine the first position of the device according to the actual positions corresponding to the respective first feature points. Compare the first position with the second position of the device recorded in the device map to obtain the map detection result.

[0072] As can be seen from the above, since the above first feature points can characterize the features of the first area where the device is located, it can be considered that the above first feature points can represent the first area where the device is located. Based on the actual positions corresponding to the first feature points, the first position of the device can be determined, and it can be considered that the above first position is the current position where the device is located. By comparing the above first position with the second position, it can be determined whether the second position recorded in the device map is accurate, thereby obtaining the map detection result. Moreover, the first position where the device is currently located during the above map detection process is obtained from the first image collected by the driverless device during driving, and the above map detection process does not require manual operation, which can improve the efficiency of map detection. Brief Description of the Drawings

[0073] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0074] Figure 1 It is a schematic flowchart of a map detection method provided in the embodiments of the present invention;

[0075] Figure 2 It is a schematic diagram of a first area provided in the embodiments of the present invention;

[0076] Figure 3 It is a schematic diagram of the first feature points in a first image provided in the embodiments of the present invention;

[0077] Figure 4 It is a schematic flowchart of the first method for determining the second position provided in the embodiments of the present invention;

[0078] Figure 5 It is a schematic flowchart of the second method for determining the second position provided in the embodiments of the present invention;

[0079] Figure 6 Schematic flowchart of a method for collecting instrument information provided in an embodiment of the present invention;

[0080] Figure 7A Schematic diagram of a third region provided in an embodiment of the present invention;

[0081] Figure 7B Schematic diagram of an edge detection result provided in an embodiment of the present invention;

[0082] Figure 7C Schematic diagram of a fourth region provided in an embodiment of the present invention;

[0083] Figure 7D Schematic diagram of a character region provided in an embodiment of the present invention;

[0084] Figure 8 Schematic diagram of the structure of a map detection device provided in an embodiment of the present invention;

[0085] Figure 9 Schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed implementation manners

[0086] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art based on this application belong to the scope of protection of the present invention.

[0087] Since the efficiency of map detection in the prior art is relatively low, to solve this problem, an embodiment of the present invention provides a map detection method and device.

[0088] In an embodiment of the present invention, a map detection method is provided. The above method includes:

[0089] Obtain a first image collected by the driverless device during driving.

[0090] Identify the first region where the device is located in the above first image.

[0091] Obtain the pixel points representing the characteristics of the above first region as the first feature points.

[0092] According to the pixel point positions of the obtained first feature points, determine the actual positions corresponding to the respective first feature points in the actual environment in which the driverless device travels.

[0093] Determine the first position of the device according to the actual positions corresponding to the respective first feature points.

[0094] Compare the first position with the second position of the device recorded in the device map to obtain a map detection result.

[0095] As can be seen from the above, since the first feature point can characterize the features of the first area where the device is located, it can be considered that the first feature point can represent the first area where the device is located. Based on the actual position corresponding to the first feature point, the first position of the device can be determined, and it can be considered that the first position is the current position of the device. By comparing the first position with the second position, it can be determined whether the second position recorded in the device map is accurate, so that a map detection result can be obtained. Moreover, the first position where the device is currently located during the above map detection process is obtained from the first image collected during the driving of the unmanned device, and the above map detection process does not require manual operation, which can improve the efficiency of map detection.

[0096] See Figure 1 , which is a schematic flowchart of a map detection method provided by an embodiment of the present invention. The above method includes the following steps S101-S106.

[0097] Specifically, the execution subject of the embodiment of the present invention can be a processor installed in an unmanned device. For example, the above unmanned device can be a device such as a robot or an unmanned vehicle. ROS (Robot Operating System) etc. can be installed in the processor of the above unmanned device. However, since there is a large amount of data to be processed and a large workload during the data processing process of the embodiment of the present invention, the execution subject of the embodiment of the present invention can also be a server communicatively connected to the unmanned device.

[0098] S101: Obtain a first image collected by the unmanned device during driving.

[0099] Among them, the above first image can be an image collected by an image acquisition device installed on the unmanned device.

[0100] Specifically, during the movement of the driverless device, it can use image acquisition devices installed on itself, such as cameras, video cameras, RGBD cameras, etc., to continuously acquire environmental images, and then use SLAM (Simultaneous Localization and Mapping) technology to perform self-localization based on the acquired environmental images. Among them, in addition to collecting RGB images, the above-mentioned RGBD camera can also collect a depth map in which the pixel value of each pixel point is a depth value. The pixel points at the same position in the RGB image and the depth map collected at the same moment correspond to the same position in the actual environment. The above-mentioned RGBD camera can continuously collect an image stream of RGB images and depth maps at 30 FPS.

[0101] In addition, during the process of self-localization of the above-mentioned driverless device, it can also determine its own pose based on devices such as encoders and inertial sensors installed on itself for self-localization.

[0102] Furthermore, a lidar can be installed on the above-mentioned driverless device, and then laser SLAM can be performed based on the laser point cloud collected by the lidar to achieve the localization of the driverless device.

[0103] The above-mentioned SLAM technology is only one way to achieve the localization of the driverless device, and other existing technologies can also be used to achieve the localization of the driverless device. The embodiments of the present invention do not limit this.

[0104] In the case where the execution subject of the embodiment of the present invention is a server, the processor of the above-mentioned driverless device can compress the above-mentioned first image and then send it to the server to reduce the amount of data to be transmitted during the transmission of the first image and improve the transmission efficiency. For example, the above-mentioned first image can be compressed into a jpg format image with a size of 960 pixels x 540 pixels, or it can be an image of other sizes.

[0105] In addition, in the case where the above-mentioned first image is collected by an RGBD camera, the processor can mix the first image and the depth map and then send it to the server. After receiving the mixed first image and depth map, the server can determine each pair of first images and depth maps with the same acquisition time according to the timestamps included in the first image and the depth map.

[0106] In an embodiment of the present invention, various images collected by the above-mentioned driverless device during driving can be obtained, and the feature points included in each obtained image can be determined respectively, and the image with the number of included feature points greater than the preset number is used as the above-mentioned first image.

[0107] S102: Identify the first area where the device is located in the above-mentioned first image.

[0108] In one embodiment of the present invention, a sample image including the above device can be used to train a neural network model, and the trained neural network model is used to identify the first region where the device is located in the first image.

[0109] It is also possible to use algorithms in the prior art such as the YOLO (You Only Look Once: Unified, Real-Time Object Detection) algorithm and the R-CNN (Region-Convolutional Neural Networks) algorithm to identify the first region where the device is located in the first image, which will not be elaborated in the embodiments of the present invention.

[0110] See Figure 2 , which is a schematic diagram of a first region provided by an embodiment of the present invention.

[0111] Among them, the region framed by the black wireframe in the figure is the above-mentioned first region.

[0112] S103: Obtain the pixel points representing the features of the above first region as the first feature points.

[0113] In one embodiment of the present invention, each pixel point representing the image features in the first image can be identified, and then the pixel points located in the first region among the identified pixel points are determined as the first feature points.

[0114] In another embodiment of the present invention, it is also possible to directly identify the pixel points representing the features of the first region in the first region as the first feature points.

[0115] Since the above first region is the image region where the device is located, it can be considered that the obtained first feature points are the feature points representing the features of the device.

[0116] Specifically, the above first feature points can be collected by using the ORB (Oriented FAST and Rotated BRIEF) algorithm provided by the OpenCV platform. Then, the collected first feature points can be called ORB feature points, and the above ORB feature points can be represented by ORB descriptors in binary form. In addition, the above first feature points can also be collected by other algorithms in the prior art, which is not limited in the embodiments of the present invention.

[0117] See Figure 3 , which is a schematic diagram of the first feature points in the first image provided by an embodiment of the present invention.

[0118] Among them, the black dots in the figure are the first feature points in the above-mentioned first image.

[0119] In addition, the above-mentioned first feature points can also be obtained through the following step A.

[0120] Step A: Obtain the pixel points that characterize the features of the above-mentioned first region and whose depth values belong to a preset depth interval as the first feature points.

[0121] Specifically, if the depth value of the pixel point in the above-mentioned first image is too small, it means that when the above-mentioned first image is collected, the actual object corresponding to the pixel point with too small depth value is too close to the image acquisition device, and the above-mentioned actual object may be an occlusion covering the lens of the above-mentioned image acquisition device. The feature points contained in the image area corresponding to the occlusion in the collected first image will affect the accuracy of the subsequent determination of the first position of the device.

[0122] In addition, if the depth value of the pixel point in the above-mentioned first image is too large, it means that when the above-mentioned first image is collected, the actual object corresponding to the pixel point with too large depth value is too far from the image acquisition device. Due to the limitation of the image acquisition ability of the image acquisition device, the information of the above-mentioned pixel point collected may be inaccurate. The feature points contained in the image area corresponding to the object at a relatively far distance in the collected first image will also affect the accuracy of the subsequent determination of the first position of the device.

[0123] Therefore, the above-mentioned preset depth interval can be set, and the pixel points that characterize the features of the above-mentioned first region and whose depth values belong to the preset depth interval are used as the first feature points, so as to remove the pixel points with too large or too small depth values.

[0124] In an embodiment of the present invention, when the above-mentioned first image is collected by an RGBD camera, the RGBD camera will also collect the depth map corresponding to the above-mentioned first image while collecting the above-mentioned first image, and the pixel points at the same position in the depth map and the first image correspond to the same position in the actual environment. Therefore, the depth value of the above-mentioned pixel point can be obtained according to the depth map collected at the same time as the above-mentioned first image.

[0125] In another embodiment of the present invention, the depth value of the above-mentioned pixel point can also be calculated based on the pose of the image acquisition device when the above-mentioned first image is collected and the position of the pixel point in the above-mentioned first image. Specifically, the above-mentioned depth value can be calculated based on the depth value calculation method in the prior art, and the embodiments of the present invention do not limit this.

[0126] Furthermore, the above-mentioned first feature points can also be obtained through the following step B.

[0127] Step B: Obtain the pixel points that represent the features of the first region and whose corner response values are greater than a preset response value, as the first feature points.

[0128] Since in most cases, the pixel points located at the edges of the image can better reflect the features of the image region, the above corner response value can reflect the possibility that the pixel point is located at the edge of the image. The greater the above corner response value, the greater the possibility that the pixel point is located at the edge of the image. Therefore, pixel points with corner response values greater than the preset response value can be screened as the above first feature points.

[0129] Specifically, the above corner response value can be the Harris corner response value.

[0130] S104: According to the pixel positions of the obtained first feature points, determine the actual positions corresponding to each first feature point in the actual environment where the unmanned device travels.

[0131] In an embodiment of the present invention, the three-dimensional coordinates of the actual position corresponding to the above first feature point can be calculated through the following formula.

[0132]

[0133] Among them, the above Z is the depth value of the first feature point, P uv is the pixel coordinate value of the first feature point in the above first image, u is the abscissa value of the first feature point in the above first image, v is the ordinate value of the first feature point in the above first image, K is the camera internal parameter matrix of the image acquisition device that acquires the above first image, R is the camera pose of the above image acquisition device, P w is the three-dimensional coordinate of the actual position corresponding to the above first feature point, t is the translation vector of the above camera pose, and the above T is the transformation matrix representing the above camera pose.

[0134] In another embodiment of the present invention, the above actual position can also be calculated by other methods in the prior art, and the embodiments of the present invention do not limit this.

[0135] S105: According to the actual positions corresponding to each first feature point, determine the first position of the device.

[0136] Since the above first feature points are the feature points located in the above first region, and the above first region is the region where the device is located in the above first image, it can be considered that the above first feature points can represent the above device.

[0137] In an embodiment of the present invention, the average value, weighted average value, etc. of the coordinate values of the actual positions corresponding to each first feature point can be calculated as the coordinate value of the first position of the above device.

[0138] Specifically, during the calculation of the weighted average, the weights can be set manually. If the area of the upper part of the device included in the above first image is large, most of the obtained first feature points correspond to the upper part of the device. Therefore, the weights corresponding to the first feature points closer to the lower part of the above device can be larger. And so on, if the area of the lower part of the device included in the above first image is large, the weights corresponding to the first feature points closer to the upper part of the above device can be larger. If the area of the left part of the device included in the above first image is large, the weights corresponding to the first feature points closer to the right part of the above device can be larger. If the area of the right part of the device included in the first image is large, the weights corresponding to the first feature points closer to the left part of the above device can be larger.

[0139] S106: Compare the above first position with the second position of the device recorded in the device map to obtain a map detection result.

[0140] In an embodiment of the present invention, if the calculated first position is close to the second position of the device recorded in the above device map, it can indicate that the above device has not undergone a large position change, and then the above map detection result can be that the map data is accurate.

[0141] In addition, the above first feature points can be added to the feature point library corresponding to the above device, and the second position of the above device can be recalculated according to the actual positions corresponding to the feature points stored in the feature point library. Each feature point corresponding to the above device is stored in the above feature point library.

[0142] It is also possible to calculate the average value, weighted average value, etc. between the coordinates of the above first position and the second position, and update the second position of the above device recorded in the above device map.

[0143] In another embodiment of the present invention, if the gap between the calculated first position and the second position recorded in the above device map is large, it indicates that the calculation of the first position of the above device is incorrect, or the above device is a newly emerged device, or it may also be that the position of the above device has changed. Therefore, the above map detection result can be that the map data is inaccurate, and an alarm message can be sent to notify the staff that the above map data is inaccurate.

[0144] In an embodiment of the present invention, the above step S106 can be implemented through the following steps C - step D.

[0145] Step C: Calculate the distance between the above first position and the second position of the device recorded in the device map.

[0146] In one embodiment of the present invention, the distance between the first position and the second position can be calculated based on the coordinate values of the above-mentioned first position and the second position.

[0147] Since the number of devices recorded in the above device map may be greater than 1, the distances between the above first position and the second positions of each device recorded in the above device map can be calculated respectively.

[0148] Step D: When the minimum distance among the calculated distances is greater than or equal to the first preset distance, generate a map detection result indicating that the device position is incorrect.

[0149] Specifically, since the installation positions of most devices do not change significantly in most cases, it can be considered that the device with the smallest distance between the corresponding second position and the first position has the greatest possibility of being the same device as the device corresponding to the first position. That is, if the calculated minimum distance is greater than or equal to the first preset distance, it means that the distances between the obtained first position and the second positions of each device recorded in the map are all large, and the first position of the calculated device does not match the second positions of each device recorded in the map. Therefore, it can be considered that the device position recorded in the map is incorrect, and a map detection result indicating that the device position is incorrect is generated.

[0150] As can be seen from the above, since the above first feature point can characterize the features of the first area where the device is located, it can be considered that the above first feature point can represent the first area where the above device is located. Based on the actual position corresponding to the first feature point, the first position of the above device can be determined, and it can be considered that the above first position is the current position of the above device. By comparing the above first position with the second position, it can be determined whether the second position recorded in the above device map is accurate, so as to obtain a map detection result. Moreover, the first position where the device is currently located in the above map detection process is obtained from the first image collected during the driving process of the driverless device, and the above map detection process does not require manual operation, which can improve the efficiency of map detection.

[0151] See Figure 4 , which is a schematic flowchart of the first method for determining the second position provided by the embodiment of the present invention.

[0152] Specifically, the second position of the device recorded in the above device map can be determined through the following steps S401 - S405.

[0153] S401: Obtain a second image collected by the driverless device during the driving process.

[0154] S402: Identify the second area where the device is located in the above second image.

[0155] S403: Obtain the pixel points representing the characteristics of the second region as the second feature points.

[0156] S404: Determine the actual positions corresponding to each of the second feature points in the actual environment where the driverless device travels according to the pixel positions of the obtained second feature points.

[0157] S405: Determine the second position of the device according to the actual positions corresponding to each second feature point.

[0158] In one embodiment of the present invention, the above steps S401 - S405 are similar to the foregoing steps S101 - S105, and the only difference is that the first image in the foregoing steps S101 - S105 is replaced by a second image, and the obtained first position is replaced by a second position. This embodiment of the present invention will not be elaborated herein.

[0159] As can be seen from the above, since the above second feature points can represent the characteristics of the second region where the device is located, it can be considered that the above second feature points can represent the second region where the device is located. Based on the actual positions corresponding to the second feature points, the second position of the device can be determined, and it can be considered that the second position is the current position where the device is located. And the process of determining the second position is obtained from the second image collected during the driving of the driverless device, and the process of determining the second position does not require manual operation, which can improve the efficiency of determining the second position.

[0160] See Figure 5 , which is a schematic flowchart of the second method for determining the second position provided by the embodiment of the present invention. Compared with the embodiment shown in the foregoing Figure 4 , the above step S405 can be implemented by the following steps S405A - S405D.

[0161] S405A: Determine the third position of the device according to the actual positions corresponding to each second feature point.

[0162] Specifically, the above step S405A is similar to the foregoing step S105, and this embodiment of the present invention will not be elaborated herein.

[0163] S405B: Calculate the distance between the third position and the second position of the device currently recorded in the device map.

[0164] Specifically, the above step S405B is similar to the foregoing step C, and this embodiment of the present invention will not be elaborated herein.

[0165] S405C: In the case where the minimum distance among the calculated distances is greater than or equal to the second preset distance, add the third position as the new second position of the device to the device map.

[0166] If the above minimum distance is greater than or equal to the above second preset distance, it can be considered that the distance between the device corresponding to the above third position and the device corresponding to the minimum distance recorded in the device map is relatively far, and they are not the same device. The device corresponding to the above third position may be a newly appeared device in the actual environment, so the above third position can be added to the device map as the position of the new device.

[0167] S405D: When the above minimum distance is less than the second preset distance, update the second position of the target device in the above device map according to the above third position.

[0168] Wherein, the above target device is: the device whose distance between the second position before update and the above third position is the above minimum distance.

[0169] If the above minimum distance is less than the above second preset distance, it can be considered that the distance between the device corresponding to the above third position and the device corresponding to the minimum distance in the device map is relatively close, and they are the same device. Therefore, the second position of the above target device in the above device map can be updated based on the above third position.

[0170] Specifically, the average value or weighted average value, etc. of the above third position and the second position of the target device originally recorded in the above device map can be calculated, and the calculation result is used as the new second position of the target device recorded in the above device map.

[0171] In addition, the above second feature point can also be added to the feature point library corresponding to the above target device, and the second position of the above target device is recalculated according to the actual positions corresponding to the feature points stored in the feature point library.

[0172] As can be seen from the above, when the distance between the device corresponding to the third position and the target device is relatively far, it is considered that the device corresponding to the third position and the target device are not the same device, so the device corresponding to the above third position can be determined as the newly discovered device in the above actual environment and added to the above device map to make the above device map more accurate. Otherwise, it is considered that the device corresponding to the above third position and the target device are the same device, and the second position of the target device recorded in the device map can be updated based on the third position, so that the second position of the target device recorded in the device map is more accurate, and further the above device map is more accurate.

[0173] See Figure 6 , which is a schematic flowchart of a method for collecting instrument information provided by an embodiment of the present invention. The above method includes the following steps S601 - S604.

[0174] S601: Obtain a target image collected by an unmanned device.

[0175] Wherein, the above-mentioned target image is an image collected when the distance between the position of the above-mentioned driverless device itself and the second position of the device recorded in the above-mentioned device map is less than a third preset distance.

[0176] When the execution subject in the embodiment of the present invention is the processor of the driverless device, the above-mentioned processor can calculate the distance between the position of the driverless device itself and the second position. When the above-mentioned distance is less than the third preset distance, it controls the image acquisition device installed on itself to acquire the target image of the above-mentioned device, and then the above-mentioned processor can receive the target image sent by the image acquisition device. Among them, the image acquisition device for acquiring the target image and the image acquisition device for acquiring the above-mentioned first image can be the same or different. For example, the image acquisition device for acquiring the target image can be a monocular camera, and the above-mentioned image acquisition device for acquiring the first image can be an RGBD camera. The above two cameras can be installed at different positions of the above-mentioned driverless device and have different orientations.

[0177] When the execution subject in the embodiment of the present invention is the server, the above-mentioned server can calculate the distance between the position of the above-mentioned driverless device itself and the second position. When the above-mentioned distance is less than the third preset distance, it sends an image acquisition instruction to the processor of the above-mentioned driverless device, so that the above-mentioned processor controls the image acquisition device to perform image acquisition after receiving the above-mentioned image acquisition instruction. Then the above-mentioned server can receive the target image sent by the above-mentioned processor.

[0178] S602: Identify the third area where the display panel of the instrument is located in the above-mentioned target image.

[0179] Wherein, in addition to the area for displaying information, the above-mentioned display panel may also include other parts such as buttons and indicator lights.

[0180] Specifically, in one embodiment of the present invention, the area where the device is located in the above-mentioned target image can be identified first, and then the third area can be identified from the area where the device is located, or the third area can be directly identified from the above-mentioned target image.

[0181] In one embodiment of the present invention, since the position and size of the display panel on the device are relatively fixed, after the area where the device is located is identified, the above-mentioned third area can also be determined according to the preset position and size of the instrument on the above-mentioned device.

[0182] See Figure 7A , which is a schematic diagram of a third area provided in the embodiment of the present invention.

[0183] As can be seen from the figure, the display panel of the above-mentioned device includes an area for displaying information, a button area, etc., and the information displayed on the above-mentioned display panel is "40.0".

[0184] Specifically, the above-mentioned step S602 is similar to the previous step S102, and the only difference is that the first region is recognized in the above-mentioned step S102, while the third region is recognized in the step S602. The present invention will not elaborate on this.

[0185] S603: Identify the fourth region for displaying information in the above-mentioned third region.

[0186] Specifically, the above-mentioned displayed information can be represented in the form of numbers, characters, etc.

[0187] In an embodiment of the present invention, since the region for displaying information on the device is often at a fixed position on the display panel and the size of the region for displaying information is often relatively fixed, the above-mentioned fourth region can be determined from the above-mentioned third region according to the position of the preset region for displaying information on the display panel and the size of the region for displaying information.

[0188] In another embodiment of the present invention, the fourth region for displaying information can be identified from the above-mentioned third region through the following steps E-step F.

[0189] Step E: Perform edge detection on the above-mentioned third region.

[0190] In an embodiment of the present invention, edge detection can be performed on the above-mentioned third region based on the Robert operator, Sobel operator, Laplace operator, or other image edge detection algorithms in the prior art. The embodiments of the present invention will not elaborate on this.

[0191] See Figure 7B , which is a schematic diagram of an edge detection result provided by an embodiment of the present invention.

[0192] Among them, the above-mentioned Figure 7B is the edge detection result obtained by performing edge detection on the above-mentioned Figure 7A shown third region.

[0193] Step F: Determine the fourth region for displaying information in the above-mentioned third region according to the detected edge.

[0194] Specifically, the region enclosed by the edge with the largest enclosed area can be used as the above-mentioned fourth region.

[0195] In addition, since the shape of the region for displaying information on the above-mentioned device is often fixed, the region enclosed by the edge with the same shape as the region for displaying information can be selected as the above-mentioned fourth region.

[0196] See Figure 7C , which is a schematic diagram of a fourth region provided by an embodiment of the present invention.

[0197] Among them, the above-mentioned Figure 7C is the fourth area included in the third area shown above. Figure 7A shown above.

[0198] S604: Perform character recognition on the above-mentioned fourth area to obtain instrument information.

[0199] In one embodiment of the present invention, the above-mentioned instrument information can be recognized by using the character recognition algorithm in the prior art, and the embodiments of the present invention will not elaborate on this.

[0200] In another embodiment of the present invention, the above-mentioned instrument information can also be obtained through the following steps G - step I.

[0201] Step G: Determine the color values of each pixel point included in the above-mentioned fourth area.

[0202] Specifically, the above-mentioned color values can be color values in the RGB color space or color values in the HSV color space.

[0203] Step H: For each determined color value, count the number of pixel points corresponding to this color value.

[0204] Specifically, the color values of each pixel point in the above-mentioned fourth area can be traversed to determine the number of pixel points corresponding to each color value.

[0205] Step I: According to the target color value, identify the character area included in the above-mentioned fourth area, perform character recognition on the above-mentioned character area, and obtain instrument information.

[0206] Among them, the above-mentioned target color value is: the color value among the determined color values that corresponds to the largest number of pixel points.

[0207] Specifically, since most areas in the area for displaying information are often used to display information during the process of the above-mentioned device displaying information in the instrument, it can be considered that the color represented by the target color value corresponding to the largest number of pixel points among the determined color values is the color of the displayed information. Therefore, the smallest rectangular area including all pixel points with the above-mentioned target color value can be used as the above-mentioned character area.

[0208] After determining the above-mentioned character area, image segmentation can be performed on the above-mentioned character area to separately obtain the image areas where each character is located, and then character recognition is performed on the image areas where each character is located to obtain instrument information.

[0209] Specifically, the above process can be implemented by using the image segmentation algorithm and character recognition algorithm in the prior art, and the embodiments of the present invention will not elaborate on this.

[0210] In addition, after the above-mentioned fourth region is recognized, binarization processing can be performed on the fourth region. The pixel values of the pixel points with the target color value are set to the first pixel value, and the pixel values of other pixel points are set to the second pixel value, so as to perform binarization processing on the above-mentioned second region, which can highlight the information displayed in the second region. Furthermore, noise removal processing can be performed on the fourth region, and the existing noise removal methods in the prior art can be used to process the above-mentioned fourth region, which will not be elaborated in the embodiments of the present invention.

[0211] In addition, since the color of the information displayed by the above instrument is often relatively fixed during the information display process, the above target color value can also be a preset color value.

[0212] See Figure 7D , which is a schematic diagram of a character region provided in an embodiment of the present invention.

[0213] Among them, the above Figure 7D The shown region is the character region included in the above-mentioned Figure 7C shown fourth region.

[0214] As can be seen from the above, when the above-mentioned driverless device moves and the distance to the device is less than the preset distance, the target image of the device can be collected. Since the distance between the driverless device and the device is relatively close, the collected target image can clearly show the instrument of the above device. And by identifying the information region included in the target image and determining the content represented by the information region, the instrument information can be obtained. Therefore, the instrument information can be collected during the movement of the above-mentioned driverless device, and the process of collecting the instrument information does not require manual operation, which can improve the efficiency of collecting the instrument information.

[0215] In an embodiment of the present invention, during the movement of the above-mentioned driverless device in the actual environment, it is necessary to continuously determine its own position based on the feature points included in the environmental images collected by the image acquisition device, so as to achieve self-positioning. However, if the images collected by the above-mentioned driverless device contain fewer features or do not contain any features, it is difficult for the above-mentioned driverless device to determine its own position, resulting in the driverless device being difficult to continue to determine the driving route.

[0216] In the above case, based on the tracking thread of the open-source code of the ORB-SLAM2 system framework, the environmental image in the historical environmental images that matches the current environmental image can be determined through methods such as the bag-of-words model, projection matching, and pose optimization, and the relocalization of the driverless device can be performed based on the determined historical environmental image. Specifically, the method of performing relocalization based on the tracking thread belongs to the prior art, and the embodiments of the present invention will not elaborate on this. Of course, other relocalization methods can also be used for relocalization, and the embodiments of the present invention do not limit this.

[0217] In another embodiment of the present invention, after obtaining the environmental image, some pixel points in the environmental image can be selected, and based on the pixel point positions of the selected pixel points in the above environmental image and the camera pose of the image acquisition device when the above environmental image is acquired, the three-dimensional coordinates of the actual positions corresponding to the selected pixel points are calculated, and the three-dimensional point cloud blocks of the actual positions represented by each environmental image are obtained based on the above three-dimensional coordinates. After splicing the three-dimensional point cloud blocks corresponding to different environmental images, the three-dimensional point cloud of the above actual environment can be obtained as the point cloud map of the above actual environment.

[0218] Specifically, the above calculation of three-dimensional coordinates, obtaining three-dimensional point cloud blocks, and splicing the three-dimensional point cloud blocks can all be implemented by common methods in the prior art, and the embodiments of the present invention do not limit this.

[0219] Corresponding to the foregoing map detection method, an embodiment of the present invention further provides a map detection device.

[0220] See Figure 8 , which is a schematic structural diagram of a map detection device provided by an embodiment of the present invention. The above device includes:

[0221] An image acquisition module 801, configured to acquire a first image collected by the driverless device during driving;

[0222] A region recognition module 802, configured to recognize a first region where the device is located in the first image;

[0223] A feature point acquisition module 803, configured to acquire pixel points characterizing the features of the first region as first feature points;

[0224] An actual position determination module 804, configured to determine the actual positions corresponding to each first feature point in the actual environment in which the driverless device travels according to the pixel point positions of the acquired first feature points;

[0225] A first position determination module 805, configured to determine the first position of the device according to the actual positions corresponding to each first feature point;

[0226] A result obtaining module 806 is configured to compare the first position with a second position of a device recorded in a device map to obtain a map detection result.

[0227] As can be seen from the above, since the first feature point can characterize the features of the first area where the device is located, it can be considered that the first feature point can represent the first area where the device is located. Based on the actual position corresponding to the first feature point, the first position of the device can be determined, and it can be considered that the first position is the current position where the device is located. By comparing the first position with the second position, it can be determined whether the second position recorded in the device map is accurate, thereby obtaining a map detection result. Moreover, the first position where the device is currently located during the above map detection process is obtained from the first image collected during the driving of the driverless device, and the above map detection process does not require manual operation, which can improve the efficiency of map detection.

[0228] In an embodiment of the present invention, the result obtaining module 806 is specifically configured to:

[0229] Calculate the distance between the first position and the second position of the device recorded in the device map;

[0230] In the case where the minimum distance among the calculated distances is greater than or equal to a first preset distance, generate a map detection result indicating that the device position is incorrect.

[0231] In an embodiment of the present invention, the feature point obtaining module 803 is specifically configured to:

[0232] Obtain pixel points that characterize the features of the first area and whose depth values belong to a preset depth interval as the first feature points.

[0233] In an embodiment of the present invention, the feature point obtaining module 803 is specifically configured to:

[0234] Obtain pixel points that have a corner response value greater than a preset response value and characterize the features of the first area as the first feature points.

[0235] In an embodiment of the present invention, the device further includes a second position determining module for determining the second position of the device recorded in the device map. The second position determining module includes:

[0236] An image acquisition sub-module for acquiring a second image collected during the driving of the driverless device;

[0237] A region recognition sub-module for recognizing the second region where the device is located in the second image;

[0238] A feature point acquisition sub-module, configured to acquire pixel points characterizing the features of the second region as second feature points;

[0239] An actual position determination sub-module, configured to determine the actual positions corresponding to the respective second feature points in the actual environment in which the driverless device travels according to the pixel positions of the acquired second feature points;

[0240] A second position determination sub-module, configured to determine the second position of the device according to the actual positions corresponding to the respective second feature points.

[0241] As can be seen from the above, since the above-mentioned second feature points can characterize the features of the second region where the device is located, it can be considered that the above-mentioned second feature points can represent the second region where the above-mentioned device is located. Based on the actual positions corresponding to the second feature points, the second position of the above-mentioned device can be determined, and it can be considered that the above-mentioned second position is the current position where the above-mentioned device is located. And the process of determining the second position is obtained from the second image collected during the driving of the driverless device, and the process of determining the second position does not require manual operation, which can improve the efficiency of determining the second position.

[0242] In an embodiment of the present invention, the second position determination sub-module is specifically configured to:

[0243] Determine the third position of the device according to the actual positions corresponding to the respective second feature points;

[0244] Calculate the distance between the third position and the second position of the device currently recorded in the device map;

[0245] In the case where the minimum distance among the calculated distances is greater than or equal to the second preset distance, add the third position as the new second position of the device to the device map;

[0246] In the case where the minimum distance is less than the second preset distance, update the second position of the target device in the device map according to the third position, where the target device is: the device with the minimum distance between the second position before update and the third position.

[0247] In an embodiment of the present invention, the device further includes:

[0248] A target image acquisition module, configured to acquire a target image collected by the driverless device, where the target image is an image collected when the driverless device determines that the distance between its own position and the second position of the device recorded in the device map is less than the third preset distance;

[0249] A third region acquisition module, configured to identify the third region where the display panel of the instrument is located in the target image;

[0250] A fourth area recognition module, configured to recognize a fourth area for displaying information in the third area;

[0251] An information recognition module, configured to perform character recognition on the fourth area to obtain instrument information.

[0252] As can be seen from the above, when the above-mentioned driverless device is moving and the distance between the device and the equipment is less than a preset distance, the target image of the equipment can be collected. Since the distance between the driverless device and the equipment is relatively close, the collected target image can clearly show the instrument of the above-mentioned equipment. And by identifying the information area included in the target image and determining the content represented by the information area, the instrument information can be obtained. Therefore, the instrument information can be collected during the movement of the above-mentioned driverless device, and the process of collecting the instrument information does not require manual operation, which can improve the efficiency of collecting the instrument information.

[0253] An embodiment of the present invention further provides an electronic device, as Figure 9 shown, including a processor 901, a communication interface 902, a memory 903, and a communication bus 904. Among them, the processor 901, the communication interface 902, and the memory 903 complete mutual communication through the communication bus 904.

[0254] The memory 903 is used to store a computer program;

[0255] The processor 901 is configured to implement any of the method steps of the above-mentioned map detection method when executing the program stored in the memory 903.

[0256] When using the electronic device provided by the embodiment of the present invention to detect a map, since the above-mentioned first feature point can characterize the features of the first area where the device is located, it can be considered that the above-mentioned first feature point can represent the first area where the above-mentioned device is located. Based on the actual position corresponding to the first feature point, the first position of the above-mentioned device can be determined, and it can be considered that the above-mentioned first position is the current position where the above-mentioned device is located. By comparing the above-mentioned first position with the second position, it can be determined whether the second position recorded in the map of the above-mentioned device is accurate, so as to obtain a map detection result. And, the first position where the device is currently located during the above-mentioned map detection process is obtained from the first image collected during the driving process of the driverless device, and the process of the above-mentioned map detection does not require manual operation, which can improve the efficiency of map detection.

[0257] The communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity, only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0258] The communication interface is used for communication between the above electronic device and other devices.

[0259] The memory may include a Random Access Memory (RAM), and may also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0260] The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0261] In another embodiment provided by the present invention, a computer-readable storage medium is also provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements any of the method steps of the above map detection method.

[0262] When the computer program stored in the computer-readable storage medium provided by the embodiment of the present invention is executed to detect a map, since the above-mentioned first feature point can characterize the features of the first area where the device is located, it can be considered that the above-mentioned first feature point can represent the first area where the device is located. Based on the actual position corresponding to the first feature point, the first position of the device can be determined, and it can be considered that the above-mentioned first position is the current position where the device is located. By comparing the above-mentioned first position with the second position, it can be determined whether the second position recorded in the device map is accurate, so that a map detection result can be obtained. Moreover, the first position where the device is currently located during the above-mentioned map detection process is obtained from the first image collected during the driving process of the driverless device, and the above-mentioned map detection process does not require manual operation, which can improve the efficiency of map detection.

[0263] In another embodiment provided by the present invention, a computer program product containing instructions is further provided. When it runs on a computer, it causes the computer to execute any of the method steps of the above-mentioned map detection method.

[0264] When the computer program stored in the computer-readable storage medium provided by the embodiment of the present invention is executed to detect a map, since the above-mentioned first feature point can characterize the features of the first area where the device is located, it can be considered that the above-mentioned first feature point can represent the first area where the device is located. Based on the actual position corresponding to the first feature point, the first position of the device can be determined, and it can be considered that the above-mentioned first position is the current position where the device is located. By comparing the above-mentioned first position with the second position, it can be determined whether the second position recorded in the device map is accurate, so that a map detection result can be obtained. Moreover, the first position where the device is currently located during the above-mentioned map detection process is obtained from the first image collected during the driving process of the driverless device, and the above-mentioned map detection process does not require manual operation, which can improve the efficiency of map detection.

[0265] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0266] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article, or device including the element.

[0267] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for devices, electronic devices, computer-readable storage media, and computer program products, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0268] The above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are all included within the protection scope of the present invention.

Claims

1. A map detection method, characterized in that, The method includes: Obtaining a first image collected by the driverless device during driving; Identifying a first area where the device is located in the first image; Obtaining pixel points that characterize the features of the first area and whose depth values belong to a preset depth interval, or obtaining pixel points that characterize the features of the first area and whose corner response values are greater than a preset response value, as first feature points; Determining the actual positions corresponding to the respective first feature points in the actual environment in which the driverless device travels according to the pixel positions of the obtained first feature points; Determining a first position of the device according to the actual positions corresponding to the respective first feature points; Comparing the first position with a second position of the device recorded in the device map to obtain a map detection result; the map detection result is that the device map is accurate or the device map is inaccurate; Wherein, the second position of the device recorded in the device map is determined by the following method: Obtaining a second image collected by the driverless device during driving; Identifying a second area where the device is located in the second image; Obtaining pixel points that characterize the features of the second area as second feature points; Determining the actual positions corresponding to the respective second feature points in the actual environment in which the driverless device travels according to the pixel positions of the obtained second feature points; Determining a third position of the device according to the actual positions corresponding to the respective second feature points; Calculating the distance between the third position and the second position of the device currently recorded in the device map; In the case where the minimum distance among the calculated distances is greater than or equal to a second preset distance, adding the third position as the new second position of the device to the device map; In the case where the minimum distance is less than the second preset distance, updating the second position of the target device in the device map according to the third position, where the target device is: the device whose distance between the second position before updating and the third position is the minimum distance; 2. The method according to claim 1, wherein The comparing the first position with the second position of the device recorded in the device map to obtain a map detection result includes: Calculating the distance between the first position and the second position of the device recorded in the device map; In the case where the minimum distance among the calculated distances is greater than or equal to a first preset distance, generating a map detection result indicating that the device position is incorrect; 3. The method according to any one of claims 1-2, characterized in that, The method further includes: Obtaining a target image collected by the driverless device, where the target image is an image collected when the driverless device determines that the distance between its own position and the second position of the device recorded in the device map is less than a third preset distance; Identifying a third area where the display panel of the instrument is located in the target image; Identifying a fourth area for displaying information in the third area; Performing character recognition on the fourth area to obtain instrument information; 4. A map detection device, characterized in that, The device includes: An image acquisition module for obtaining a first image collected by the driverless device during driving; An area identification module for identifying a first area where the device is located in the first image; A feature point acquisition module, configured to acquire pixel points that characterize the features of the first region and whose depth values belong to a preset depth interval, or acquire pixel points that characterize the features of the first region and whose corner response values are greater than a preset response value, as first feature points; An actual position determination module, configured to determine the actual positions corresponding to the respective first feature points in the actual environment in which the driverless device travels according to the pixel positions of the acquired first feature points; A first position determination module, configured to determine the first position of the device according to the actual positions corresponding to the respective first feature points; A result acquisition module, configured to compare the first position with the second position of the device recorded in the device map to obtain a map detection result; the map detection result is that the device map is accurate or the device map is inaccurate; The device further includes a second position determination module for determining the second position of the device recorded in the device map, and the second position determination module includes: An image acquisition sub-module, configured to acquire a second image collected by the driverless device during travel; A region recognition sub-module, configured to recognize a second region where the device is located in the second image; A feature point acquisition sub-module, configured to acquire pixel points that characterize the features of the second region, as second feature points; An actual position determination sub-module, configured to determine the actual positions corresponding to the respective second feature points in the actual environment in which the driverless device travels according to the pixel positions of the acquired second feature points; A second position determination sub-module, configured to determine the second position of the device according to the actual positions corresponding to the respective second feature points; The second position determination sub-module is specifically configured to: Determine a third position of the device according to the actual positions corresponding to the respective second feature points; Calculate the distance between the third position and the second position of the device currently recorded in the device map; In the case where the minimum distance among the calculated distances is greater than or equal to a second preset distance, add the third position as the new second position of the device to the device map; In the case where the minimum distance is less than the second preset distance, update the second position of the target device in the device map according to the third position, where the target device is: the device whose distance between the second position before update and the third position is the minimum distance.

5. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus; The memory is used for storing a computer program; The processor is configured to implement the method steps described in any one of claims 1-3 when executing the program stored on the memory.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the method steps described in any one of claims 1-3.

Images