Vehicle searching method and device and storage medium

Abstract

The invention provides a vehicle searching method and device. The invention provides a method for searching a vehicle. The method comprises the following steps: reading video data of an automobile data recorder of the vehicle in response to a vehicle searching request of a user; extracting feature point data based on the video data; and matching the feature point data with a pre-collected map of a parking environment to obtain coordinates of the vehicle in a vector map.

Description

Technical Field

[0001] The present disclosure relates to the field of vehicles, and more particularly, to a method and device for finding a vehicle. Background Art

[0002] With the development of economy and society, vehicles have become an important tool in people's daily life.

[0003] In order to cope with the growing number of vehicles, the scale of parking lots is constantly expanding, and the paths in parking lots are becoming increasingly complex. When users return to the parking lot to look for their cars after completing their business, it is often difficult for them to find their own parked cars from among the many vehicles parked in the parking lot.

[0004] Furthermore, when a vehicle is parked in an indoor parking lot, it is difficult for a user to locate the vehicle through the GPS signal, since the Global Positioning System (GPS) signal quality is poor or even there is almost no GPS signal in the parking lot. Summary of the invention

[0005] According to a first aspect of the present disclosure, a method for finding a vehicle is provided, comprising: reading video data from a driving recorder of a vehicle in response to a user's request to find the vehicle; extracting feature point data based on the video data; and matching the feature point data with a pre-collected map of a parking environment to obtain the coordinates of the vehicle in a vector map.

[0006] According to an embodiment of the present disclosure, the matching is performed in the cloud, and the method further includes sending the coordinates of the vehicle in the vector map from the cloud to the user.

[0007] According to an embodiment of the present disclosure, the pre-collected map of the parking environment is a fused map pre-stored in the cloud including a point cloud base map and a vector map.

[0008] According to an embodiment of the present disclosure, extracting feature point data based on the video data includes: performing real-time localization and mapping (SLAM) modeling on the video data within a period of time before the vehicle stops to extract feature point data; filtering the feature point data to filter out feature point data corresponding to video frames with confidence levels lower than a confidence threshold; and sending the filtered feature point data to the cloud for the matching.

[0009] According to an embodiment of the present disclosure, when it is determined that the number of extracted feature points in the current video frame exceeds a quantity threshold, feature point extraction of the current video frame is stopped, the extracted feature points are sent to the cloud, and feature point extraction of the next video frame is started; and when it is determined that the number of all feature points extracted in the current video frame does not exceed the quantity threshold, the video points extracted in the current video frame are screened out, and feature point extraction of the next video frame is started.

[0010] According to an embodiment of the present disclosure, the video data is surround view video data of the vehicle surroundings acquired by a plurality of cameras having different shooting orientations in the driving recorder.

[0011] According to a second aspect of the present disclosure, there is provided a device for finding a vehicle, comprising: a video data reading module, configured to read video data from a driving recorder of a vehicle in response to a user's request to find the vehicle; a feature point extraction module, configured to extract feature point data based on the video data; and a feature point matching module, configured to match the feature point data with a pre-collected map of a parking environment to obtain the coordinates of the vehicle in a vector map.

[0012] According to an embodiment of the present disclosure, the feature point matching module is located in the cloud, and is further configured to send the coordinates of the vehicle in the vector map from the cloud to the user.

[0013] According to an embodiment of the present disclosure, the pre-collected map of the parking environment is a fused map pre-stored in the cloud including a point cloud base map and a vector map.

[0014] According to an embodiment of the present disclosure, extracting feature point data based on the video data includes: performing real-time localization and mapping (SLAM) modeling on the video data within a period of time before the vehicle stops to extract feature point data; filtering the feature point data to filter out feature point data corresponding to video frames with confidence levels lower than a confidence threshold; and sending the filtered feature point data to the cloud for the matching.

[0015] According to an embodiment of the present disclosure, extracting feature point data based on the video data includes: extracting feature points based on SLAM modeling for video frames in the video data within the period of time; when it is determined that the number of extracted feature points in the current video frame exceeds a quantity threshold, stopping the feature point extraction of the current video frame, sending the extracted feature points to the cloud, and starting the feature point extraction of the next video frame; and when it is determined that the number of all feature points extracted in the current video frame does not exceed the quantity threshold, screening out the feature points extracted from the current video frame, and starting the feature point extraction of the next video frame.

[0016] According to an embodiment of the present disclosure, the video data is surround view video data of the vehicle surroundings acquired by a plurality of cameras having different shooting orientations in the driving recorder.

[0017] According to a third aspect of the present disclosure, there is provided a device for finding a vehicle, comprising: a memory having computer instructions stored thereon; and a processor, wherein the instructions, when executed by the processor, cause the processor to execute the method according to the first aspect.

[0018] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing instructions is provided, wherein the instructions enable a processor to execute the method according to the first aspect.

[0019] By utilizing the method and device provided by the present disclosure, the defects of the prior art can be overcome, and an intelligent reverse vehicle search function with low cost and high versatility can be realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] These and / or other aspects and advantages of the present disclosure will become more clear and easier to understand from the following detailed description of the embodiments of the present disclosure in conjunction with the accompanying drawings, in which:

[0021] Figure 1 A flow chart of a method for finding a vehicle according to an embodiment of the present disclosure is shown;

[0022] Figure 2 A schematic diagram of a system for finding a vehicle according to an embodiment of the present disclosure is shown;

[0023] Figure 3 shows a non-limiting example of image data of a vehicle's surroundings according to an embodiment of the present disclosure;

[0024] Figure 4 An example feature extraction process according to an embodiment of the present disclosure is shown;

[0025] Figure 5 A flow chart of a method for finding a vehicle in collaboration between a vehicle side, a cloud side, and a user side according to an embodiment of the present disclosure is shown;

[0026] Figure 6 A schematic diagram of a device for finding a vehicle according to an embodiment of the present disclosure is shown;

[0027] Figure 7 A schematic diagram of a device for finding a vehicle according to an embodiment of the present disclosure is shown.

[0028] It should be understood that these drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification, and are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation of the present invention. In addition, in the drawings, the same reference numerals generally represent the same parts or steps. DETAILED DESCRIPTION

[0029] In order to better illustrate the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should be understood that based on the embodiments described in the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present invention, and the embodiments described herein are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, which are merely illustrative and exemplary, and therefore should not be interpreted as limiting the scope of the present invention.

[0030] In the prior art, the reverse vehicle search system used in parking lots where there may be no GPS signal or the signal is poor usually uses the infrastructure of the parking lot (such as multiple Bluetooth base stations, cameras, etc.) arranged inside the parking lot to cooperate with the vehicle to locate and record the vehicle, and send the location information to the user, thereby providing the user with a reverse vehicle search function.

[0031] There are some problems with such existing technologies. Specifically, such technical solutions need to rely on the infrastructure of the parking lot, such as Bluetooth base stations and cameras, which requires a lot of software, hardware infrastructure costs and maintenance costs. On the other hand, parking lots that are not equipped with such infrastructure cannot provide users with car search services. In addition, since there are currently no unified regulations or standards for the technical solutions for the parking lot infrastructure, most parking lots use their own customized technical solutions. Therefore, from the user's perspective, it may be difficult to achieve a unified car search solution that can adapt to each parking lot, causing great inconvenience.

[0032] In view of the above problems in the prior art, the present disclosure provides a method and device for finding a vehicle.

[0033] Figure 1 A flow chart of a method 100 for finding a vehicle according to an embodiment of the present disclosure is shown. The method 100 may be executed on the vehicle side.

[0034] like Figure 1 As shown, at step S101, the video data of the vehicle's driving recorder can be read in response to the user's vehicle search request.

[0035] The driving recorder of a vehicle may be any device for recording the driving process of the vehicle on which the driving recorder is installed. Generally, the driving recorder includes a camera to capture images or videos of the surrounding environment including the vehicle driving, thereby recording the driving process of the vehicle. In the present disclosure, the device on the vehicle side of the execution method 100 may be coupled to the driving recorder installed on the vehicle (such as by a cable), so that the video data of the driving recorder can be read for further processing. In an embodiment according to the present disclosure, the driving recorder may include a plurality of cameras with different shooting orientations, so that video data of the surrounding environment of the vehicle at different angles can be obtained. As a non-limiting example, the driving recorder may include at least 4 cameras with different shooting orientations, such as a front-view camera, a left camera, a right camera, and a rear-view camera, to form a surround-view camera subsystem, so that the video in any surrounding environment up to 360° around the vehicle can be obtained and the vehicle positioning is performed based on the feature points existing in the 360° environment to complete the vehicle search.

[0036] Reference Figure 2 For further explanation, Figure 2 A schematic diagram of a system 200 for finding a vehicle according to an embodiment of the present disclosure is shown.

[0037] like Figure 2 As shown, the system 200 may include a user-side device 210 and a vehicle-mounted device 220 , wherein the vehicle-mounted device 220 may be communicatively coupled to the user-side device 210 .

[0038] The user device 210 may be a portable device of the user, such as a mobile phone, etc., through which the user can interact with the vehicle-mounted device 220 to perform a vehicle-finding process.

[0039] The vehicle-mounted device 220 may be configured to execute the method 100 for finding a vehicle described above. Specifically, the vehicle-mounted device 220 may include a driving recorder 221, a data processing device 222, and a data storage device 223, wherein the data processing device 222 may be configured to read and process the video data of the driving recorder 221, and the data storage device 223 may be configured to store the video data of the driving recorder 221. The data processing device 222 may be coupled to the driving recorder 221, for example, via a cable; and the data processing device 222 may be coupled to the data storage device 223, for example, via a USB interface.

[0040] In an embodiment of the present disclosure, a user may remotely send a request to find a vehicle to the vehicle-mounted device 220 through the user-side device 210. Upon receiving the user's request to find a vehicle, the vehicle-mounted device 220 may be awakened from a sleep state and start to power on. The data processing device 222 in the vehicle-mounted device 220 may read the video data of the driving recorder 221 within a period of time before parking (such as 10 minutes before parking) from the data storage device 223 in response to the user's request to find a vehicle.

[0041] In an embodiment according to the present disclosure, the system 200 may further include a cloud server 230, so that some data processing operations that could have been performed by the vehicle-mounted device 220 may be completed in the cloud. For example, some data processing operations may involve a large amount of computation, and the local data processing resources of the vehicle-mounted device 220 may not be sufficient to process them locally, or the cost of local processing is too high. For example, some data processing operations may require pre-storage of a large amount of data (such as a large amount of map data), and the local data storage resources of the vehicle-mounted device 220 may not be sufficient for local storage, or the cost of local storage is too high.

[0042] In the case where the system 200 includes a cloud server 230, the user-side device 210 and the vehicle-mounted device 220 can be respectively communicatively coupled to the cloud server 230, and can communicate via the cloud server 230 and directly communicate with each other. As a non-limiting example, a remote wake-up of the vehicle-mounted device 220 initiated by the user can be performed by the cloud server 230.

[0043] Specifically, if Figure 2 As shown, the cloud server 230 can receive a vehicle search request initiated by a user from the user-side device 210, and remotely wake up the vehicle-mounted device 220 in response to the vehicle search request, and then the vehicle-mounted device 220 starts to power on and run.

[0044] Return to reference Figure 1 At step S102, feature point data may be extracted based on the video data. In the present disclosure, feature points may refer to points that can represent visual information of objects in the surrounding environment, so that the vehicle can be positioned and found based on the feature points.

[0045] Reference Figure 3 For further explanation, Figure 3 The video frame 300 is an exemplary video of the vehicle surrounding environment captured in the method for finding a vehicle according to an embodiment of the present disclosure. The video frame 300 includes pillars (e.g., structural pillars or load-bearing pillars) and signboards (e.g., direction signs) in an indoor parking lot. These objects are representative objects that are helpful for matching visual feature points.

[0046] In step S102, feature points can be extracted from the video frame by various image processing algorithms. As non-limiting examples, algorithms such as scale-invariant feature transform (SIFT), fast robust features (SURF), gradient position orientation histogram (GLOH) and the like can be used to extract features such as corners and edges of the above-mentioned representative objects. As a non-limiting example, the extraction can be achieved by Simultaneous Localization and Mapping (SLAM) modeling. Figure 3 As shown, when processing the video frame 300 , the corner points P11 , P12 , P13 , and P14 of the sign 301 as a representative object can be extracted as feature points, so that the vehicle can be matched and located based on the feature points of the sign 301 .

[0047] It should be understood that the above feature extraction method is merely a non-limiting example, and feature point data may also be extracted based on video data in other ways, such as through deep learning.

[0048] In an embodiment according to the present disclosure, feature points may be extracted frame by frame from one or more channels of video data recorded by one or more cameras in a driving recorder within a period of time before parking (such as 10 minutes before parking).

[0049] In an embodiment according to the present disclosure, the confidence of each video can be determined based on feature point extraction and the corresponding video frames can be screened based on the confidence of each video frame to filter out feature point data corresponding to video frames with confidence lower than a confidence threshold (such as 80%). In the present disclosure, the confidence of a video frame is a value that can characterize the quality of the video frame. The confidence of a calculated video frame can be determined in various appropriate ways. For example, the more feature points extracted from a video frame, the higher the confidence of the video frame can be determined.

[0050] In the case where the system includes a cloud server, the filtered feature point data can be sent to the cloud for subsequent processing. In this way, the feature points of relatively high-quality video frames can be sent to the cloud server, thereby improving the processing quality; and, since only the coordinates of the feature points in the relatively high-quality video frames need to be sent to the cloud server instead of sending the entire video frame to the cloud server for subsequent processing, the amount of data transmission required to be performed between the vehicle-mounted device and the cloud server is greatly reduced.

[0051] In the case where the system does not include a cloud server, subsequent processing can also be performed locally on the vehicle-mounted device.

[0052] In the embodiment according to the present disclosure, in the process of extracting feature points, it is not necessary to perform complete feature point extraction on each video frame of the read video data, thereby further improving the processing speed of feature point extraction and reducing the amount of calculation.

[0053] Figure 4 FIG. 2 shows an example feature extraction process according to an embodiment of the present disclosure. Figure 4 As shown, at S401, feature points of the current frame may be extracted; then, at S402, it may be determined whether the number of feature points currently extracted from the current frame exceeds a threshold number.

[0054] If it is determined at S402 that the number of feature points extracted from the current frame exceeds the threshold number, all feature points extracted from the current frame can be sent to the cloud at S404, and then the next video frame of the current frame is used as the current frame at S405, and the process returns to S401 to start the next round of feature point extraction.

[0055] In other words, in the process of extracting feature points from video data frame by frame, if it is determined that the number of feature points extracted from the current video frame for which feature point extraction is being performed exceeds the number threshold, feature point extraction for the current video frame can be stopped and feature point extraction for the next video frame can be started. Since the number of feature points extracted from the current video frame exceeds the number threshold, it indicates that the confidence of the current video frame is higher than the confidence threshold, and it can be considered that a sufficient number of high-quality feature points have been extracted from the current video frame, so there is no need to continue to extract feature points from the current video frame, but feature point extraction can be started for the next video frame.

[0056] In an embodiment according to the present disclosure, Figure 2 The system 200 shown includes a cloud server 230 and when feature point matching is performed at the cloud server 230, in the process of extracting feature points frame by frame, after stopping the feature point extraction of the current video frame and before starting the feature extraction of the next video frame, the extracted feature point data of the current video frame can be further sent to the cloud, so that subsequent feature point processing can be performed on the cloud as soon as possible.

[0057] If it is determined at S402 that the number of feature points extracted from the current frame does not exceed the threshold number, it is determined at S403 whether the current frame has been completely extracted. If it is determined at S403 that the current frame has been completely extracted, the next video frame of the current frame can be used as the current frame at S405, and the process returns to S401 to start the next round of feature point extraction. If it is determined at S403 that the current frame has not been completely extracted, the process returns to S401 to continue extracting feature points of the current frame.

[0058] In other words, in the process of extracting feature points from video data frame by frame, when it is determined that the number of all feature points extracted from the current video frame does not exceed the number threshold, the feature points extracted from the current video frame can be screened out and feature point extraction for the next video frame can be started. Since the number of all feature points extracted from the current video frame does not exceed the number threshold, it indicates that the confidence of the current video frame is lower than the confidence threshold, and it can be considered that the quality of the current video frame is low, and it is difficult to extract sufficiently rich feature information from it, so the feature data extracted from the video frame is no longer considered in subsequent processing.

[0059] Return to reference Figure 1 At step S103, the feature point data may be matched with a pre-collected map of the parking environment to obtain the coordinates of the vehicle in the vector map.

[0060] In an embodiment according to the present disclosure, the feature matching step may be performed at a vehicle-mounted device or at a cloud server.

[0061] When the feature point matching step is performed at the cloud server, the pre-collected map of the parking environment may be a fused map pre-stored in the cloud including a point cloud base map and a vector map. Figure 1 Step S103 may further include sending the feature point data to a cloud server, and receiving the coordinates of the position of the vehicle in the vector map obtained by matching the feature points from the cloud server. This can advantageously reduce the data storage and processing requirements for the vehicle-mounted device, and improve the versatility of the method and device proposed in this article in different parking lots.

[0062] The pre-collected point cloud base map of the parking environment (such as a parking lot) may include point cloud data measured at roads and intersections in the parking lot for various objects in the parking lot (such as pillars, signboards, etc.), which are suitable for matching with images actually captured by the vehicle for visual repositioning; the vector map may include vector graphic elements (such as points, lines, rectangles, polygons, circles, arcs, etc.) describing the geometric characteristics of roads and intersections in the parking lot, which are suitable for being presented on the vehicle display for the driver to observe and understand his position in real time. It should be noted that the point cloud base map and the vector map contained in the fused map are associated with each other, that is, each location in the parking lot (such as a road or intersection) can be represented by a specific vector graphic element, and the point cloud pattern that can be observed or measured from the location is also unique to the location, that is, the vector graphic element and the point cloud pattern at the location have a unique mapping relationship. Since the point cloud base map in the fused map includes the point cloud data of the entire scene obtained by measuring each object in the scene in advance, the feature points extracted from the actually captured video frame are matched with it, and the specific location of the shooting location in the scene can be determined based on the similarity between the extracted feature points and the point cloud data.

[0063] Furthermore, in order to make the technical solution of the present disclosure applicable to different locations (such as different parking lots), the fused map of the location can be acquired in advance. When the feature point matching is performed locally on the vehicle-mounted device, it is only necessary to update and store the fused map of the location locally on the vehicle-mounted device; when the feature point matching is performed in the cloud, it is only necessary to store the fused map of the location in the fused map library of the cloud server.

[0064] Figure 5 A flowchart of a method 500 for finding a vehicle collaboratively performed on the vehicle side, the cloud side, and the user side is shown according to an embodiment of the present disclosure.

[0065] like Figure 5 As shown, on the user side, at S511, the user's portable device can send a vehicle search request, so that the vehicle-mounted device is powered on to perform subsequent steps for searching the vehicle. The vehicle search request can be sent directly to the vehicle-mounted device on the vehicle side; the vehicle search request can also be sent to a cloud server on the cloud side, and forwarded by the cloud server to the vehicle-mounted device on the vehicle side.

[0066] Next, on the vehicle side, at S522, in response to the vehicle search request, the vehicle-mounted device can be awakened and powered on; at S523, the vehicle-mounted device can read the video data of the driving recorder; and at S524, the vehicle-mounted device can extract feature points, filter the video frames based on the confidence obtained in the process of extracting feature points, and send the feature point data of the filtered video frames to the cloud side for feature point matching in the cloud.

[0067] Next, on the cloud side, at S535, the cloud server can receive feature point data from the vehicle side; at S536, the cloud server can match the received feature point data with the fused map to obtain the vector coordinates of the vehicle, and send the vector coordinates to the user's portable device for the user to use to find the vehicle. Specifically, according to the matching degree between the extracted feature points and the point cloud data and the mapping relationship between the point cloud base map and the vector map, the specific position coordinates of the vehicle in the vector map can be determined according to the visual positioning algorithm, and the position coordinates can be sent to the user's device.

[0068] Finally, on the user side, at S517, the user's portable device may receive the vector coordinates obtained by matching the feature points; and at S518, the user's portable device may instruct the user to search for the vehicle based on the vector coordinates. As a non-limiting example, at S518, a rendered parking lot plan may be displayed on the user's portable device based on the coordinates in the vector map, in which the location of the vehicle is indicated, and the path from the user's current location to the location of the vehicle may also be displayed on the user's portable device, so that when the user views the display in the portable device, the positioning result of the search for the vehicle is presented to the user, guiding the user to search for the vehicle. As a non-limiting example, at S518, as a non-limiting example, at S518, the user may be informed of the location indicator of the vehicle being sought based on the coordinates in the vector map, such as "B1-256", thereby informing the user that the vehicle is located in parking space No. 256 on the first underground floor.

[0069] It should be understood that Figure 5 The steps shown in the embodiment do not necessarily have to be performed sequentially, and some of the steps may be performed in parallel or in a repeated cycle. Figure 4 As described above, feature point extraction based on video data can be performed frame by frame, so Figure 5 At S524 in the above, the vehicle-mounted device may send the extracted feature points from one or more video frames to the cloud for feature point matching immediately after successfully extracting the feature points, instead of sending the extracted feature points to the cloud together after completing the feature point extraction for all video frames of the video data in the time period. In addition, in the case of extracting feature points frame by frame and sending the extracted feature points (such as in the reference Figure 4 In the described embodiment), when the cloud server successfully matches the feature point to the point in the fused map at S536, it can proceed directly from step S536 to S517; and when the feature point matching is unsuccessful, it can return from step S536 to S524 to extract the feature point, that is, try to extract the feature point from the next video frame.

[0070] And it should be understood that Figure 5 The steps shown in the figure do not necessarily have to be performed on the user side, vehicle side or cloud side as shown, and some of the steps may also be performed on other sides. As a non-limiting example, when the number of feature points to be matched and the data size do not exceed the processing capacity of the vehicle-mounted device, Figure 5 Steps S535 and S536 shown as being executed on the cloud side may also be executed at the vehicle-mounted device.

[0071] Figure 6 FIG. 6 is a schematic diagram of a system 600 for finding a vehicle according to an embodiment of the present disclosure. Figure 6 As shown, the system 600 may include a device 610 arranged at the vehicle end and a device 620 arranged at the cloud end. The device 610 at the vehicle end may include a video data reading module 611, a feature point extraction module 612, and a feature point matching module 613. The device 620 at the cloud end may include a feature point matching module 621, so that the feature point matching module may be arranged at at least one of the cloud end or the vehicle end.

[0072] Specifically, on the vehicle side, the video data reading module 611 can be configured to read the video data of the vehicle's driving recorder in response to the user's vehicle search request. On the vehicle side, the feature point extraction module 612 can be configured to extract feature point data based on the video data. The feature point matching module 613 or 621 on the vehicle side or the cloud can be configured to match the feature point data with the pre-collected map of the parking environment to obtain the coordinates of the vehicle in the vector map.

[0073] It should be understood that Figure 6 The device 610 disposed on the vehicle side and the device 620 disposed on the cloud are shown as non-limiting examples. The devices 610 and 620 may also include one or more devices not disposed on the cloud side. Figure 6 Modules shown in, such as transceiver modules, etc.

[0074] Figure 7 FIG. 7 is a schematic diagram of a vehicle-finding device 700 according to an embodiment of the present disclosure. Figure 7 As shown, the device 700 may include a processor 701 and a memory 702 .

[0075] The processor 701 may be any device with processing capabilities capable of implementing the functions of the embodiments of the present disclosure, for example, it may be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field programmable gate array signal (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components, or any combination thereof, designed to perform the functions described herein.

[0076] Memory 702 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may also include other removable / non-removable, volatile / non-volatile computer system memory, such as a hard drive, floppy disk, CD-ROM, DVD-ROM, or other optical storage media.

[0077] In this embodiment, the memory 702 stores computer program instructions, and the processor 701 can execute the instructions stored in the memory 702. When the computer program instructions are executed by the processor, the processor executes the method for finding a vehicle in the embodiment of the present disclosure. Figures 1 to 5 The description is basically the same, so in order to avoid repetition, it will not be repeated.

[0078] The method / device for finding a vehicle according to the present disclosure may also be implemented by providing a computer program product containing a program code for implementing the method or device, or by any storage medium storing such a computer program product.

[0079] Advantages of the method and device for finding a vehicle according to the present disclosure include but are not limited to:

[0080] The recorded video of the driving recorder is used as the video source for extracting feature points. The driving recorder is a common vehicle-mounted device, so there is no need to add video acquisition equipment, which reduces the cost of finding the vehicle;

[0081] Use multi-visual source fusion and filter video frames based on confidence, so as to efficiently and accurately obtain the best visual data and reduce computing power consumption;

[0082] In combination with cloud computing, only the coordinates of feature points in relatively high-quality video frames need to be sent to the cloud server instead of all video frames, thereby greatly reducing the amount of communication data between the vehicle-mounted device and the cloud server;

[0083] The use of visual positioning and pre-collected fusion maps reduces the hardware dependence of parking infrastructure compared to the reverse car search solution in the existing technology, thereby reducing costs and improving versatility.

[0084] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the embodiments of the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0085] The block diagrams of the devices, equipment, devices, and systems involved in the embodiments of the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by operators in the art, these devices, equipment, devices, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The word "such as" used here refers to the phrase "such as but not limited to", and can be used interchangeably with it.

[0086] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0087] It should also be noted that in the device and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0088] For ordinary operators in the art, it is understood that all or any part of the methods and devices disclosed herein can be implemented in hardware, firmware, software or a combination thereof in any computing device (including processors, storage media, etc.) or a network of computing devices. The hardware can be a general-purpose processor, digital signal processor (DSP), ASIC, field programmable gate array signal (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components or any combination thereof designed to perform the functions described herein. The general-purpose processor can be a microprocessor, but as an alternative, the processor can be any commercially available processor, controller, microcontroller or state machine. The processor can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors cooperating with a DSP core, or any other such configuration. The software can be present in any form of computer-readable tangible storage medium. By way of example and not limitation, such computer-readable tangible storage media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage or other magnetic storage devices or any other tangible media that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer. As used herein, disc includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc.

[0089] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0090] The above description of the disclosed aspects is provided to enable any operator in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to operators in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0091] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for finding a vehicle, comprising: Reading video data from a driving recorder of a vehicle in response to a vehicle search request from a user; Extracting feature point data based on the video data; as well as The feature point data is matched with a pre-collected map of the parking environment to obtain the coordinates of the vehicle in the vector map.

2. The method according to claim 1, wherein: The matching is performed in a cloud, and the method further comprises sending the coordinates of the vehicle in a vector map from the cloud to the user.

3. The method according to claim 2, wherein: The pre-collected parking environment map is a fused map pre-stored in the cloud including a point cloud base map and a vector map.

4. The method according to claim 1, wherein: Extracting feature point data based on the video data includes: Performing real-time localization and mapping (SLAM) modeling on video data of a period of time before the vehicle stops to extract feature point data; Screening the feature point data to remove feature point data corresponding to video frames having a confidence level lower than a confidence level threshold; and The selected feature point data is sent to the cloud for the matching.

5. The method according to claim 4, wherein: When it is determined that the number of feature points extracted from the current video frame exceeds the number threshold, the feature point extraction of the current video frame is stopped, the extracted feature points are sent to the cloud, and the feature point extraction of the next video frame is started; as well as When it is determined that the number of all feature points extracted from the current video frame does not exceed the number threshold, the feature points extracted from the current video frame are screened out, and feature point extraction for the next video frame is started.

6. The method according to claim 1, wherein: The video data is surround view video data of the vehicle's surroundings acquired by a plurality of cameras having different shooting orientations in the driving recorder.

7. A device for finding a vehicle, comprising: A video data reading module, configured to read the video data of the vehicle's driving recorder in response to a user's vehicle search request; A feature point extraction module, configured to extract feature point data based on the video data; as well as The feature point matching module is configured to match the feature point data with a pre-collected map of the parking environment to obtain the coordinates of the vehicle in the vector map.

8. The device according to claim 7, wherein: The feature point matching module is located in the cloud, and is also configured to send the coordinates of the vehicle in the vector map from the cloud to the user.

9. The device according to claim 8, wherein: The pre-collected parking environment map is a fused map pre-stored in the cloud including a point cloud base map and a vector map.

10. The device according to claim 7, wherein: Extracting feature point data based on the video data includes: Performing real-time localization and mapping (SLAM) modeling on video data of a period of time before the vehicle stops to extract feature point data; Screening the feature point data to remove feature point data corresponding to video frames having a confidence level lower than a confidence level threshold; and The selected feature point data is sent to the cloud for the matching.

11. The device according to claim 10, wherein: When it is determined that the number of feature points extracted from the current video frame exceeds the number threshold, the feature point extraction of the current video frame is stopped, the extracted feature points are sent to the cloud, and the feature point extraction of the next video frame is started; as well as When it is determined that the number of all feature points extracted from the current video frame does not exceed the number threshold, the feature points extracted from the current video frame are screened out, and feature point extraction for the next video frame is started.

12. The device according to claim 7, wherein: The video data is surround view video data of the vehicle's surroundings acquired by a plurality of cameras having different shooting orientations in the driving recorder.

13. A device for finding a vehicle, comprising: a memory having computer instructions stored thereon; and processor, Wherein the instructions, when executed by the processor, cause the processor to perform the method according to any one of claims 1-6.

14. A non-transitory computer-readable storage medium storing instructions, the instructions causing a processor to perform the method according to any one of claims 1-6.

Images