Method and system for identifying an object

Abstract

Each example of the present disclosure describes a method and apparatus for identifying an object. The method includes: at a first moment, acquiring raw data output by a camera mounted on a vehicle, the raw data output having one of a continuous data output format for a continuous model or a discrete data output format for a discrete model; feeding the raw data to a discrete statistical model to form discrete statistical model format data for the first moment, the discrete statistical model format data including a continuous part having a continuous data output format and a discrete part having a discrete data output format; fusing the discrete statistical model format data for the first moment with historical discrete statistical model format data to update the discrete statistical model format data for the first moment; and identifying an object from the updated discrete statistical model format data for the first moment, wherein the object includes a lane line characterized by a curve.

Description

Technical Field

[0001] The present disclosure relates generally to object identification and, more particularly, to identifying objects using discrete statistical models. Background of the Invention

[0003] An autonomous vehicle (also known as a driverless car, autonomous car, robot car) is a vehicle that can sense its environment and navigate without human input. An autonomous vehicle (hereinafter referred to as an "ADV") uses various technologies to detect its surroundings, such as radar, laser, GPS, odometry, and computer vision. An advanced control system interprets the sensed information to identify an appropriate navigation path, as well as obstacles and related landmarks.

[0004] More specifically, ADV collects sensor data from various on-board sensors, such as visual sensors (e.g., cameras), radar-based ranging sensors (e.g., laser radar, millimeter-wave radar, ultrasonic radar), etc. Cameras are generally used to capture real-time images of the road. For example, by processing real-time images, lane line edge curves can be extracted from image sequences to identify lane lines. Lane line recognition is the basis for judging lane departure, and its recognition quality will directly affect the timeliness and accuracy of vehicle warnings.

[0005] Cameras made by different manufacturers may use different formats of data to represent objects. For example, some cameras may use discrete points to represent objects, while other cameras may use polynomials to represent objects. Currently, separate data models need to be established for storage and processing of different data formats. There is no model that can support multiple different data formats at the same time.

[0006] In addition, the single-frame camera data obtained for a single moment is unstable and unreliable. It is ideal to fuse the camera data for this moment with the multi-frame camera data for the previous multiple moments to fit the data to identify the lane line.

[0007] Therefore, it is hoped to provide a solution that can simultaneously support a model of multiple data formats to mark lane lines in real time, so as to overcome the above-mentioned defects. Summary of the invention

[0008] The present disclosure aims to provide a method and apparatus for identifying an object.

[0009] According to a first exemplary embodiment of the present disclosure, a method for identifying an object is provided. The method includes: at a first moment, acquiring raw data output by a camera mounted on a vehicle, the raw data output having one of a continuous data output format for a continuous model or a discrete data output format for a discrete model; feeding the raw data to a discrete statistical model to form discrete statistical model format data for the first moment, the discrete statistical model format data including a continuous part having a continuous data output format and a discrete part having a discrete data output format; fusing the discrete statistical model format data for the first moment with historical discrete statistical model format data to update the discrete statistical model format data for the first moment; and identifying an object from the updated discrete statistical model format data for the first moment, wherein the object includes a lane line characterized by a curve.

[0010] According to a second exemplary embodiment of the present disclosure, a device for identifying an object is provided. The device includes: a data acquisition module, which is configured to acquire raw data output by a camera mounted on a vehicle at a first moment, the output raw data having one of a continuous data output format for a continuous model or a discrete data output format for a discrete model; a data feeding module, which is configured to feed the raw data to a discrete statistical model to form discrete statistical model format data for the first moment, the discrete statistical model format data including a continuous part having a continuous data output format and a discrete part having a discrete data output format; a data fusion module, which is configured to fuse the discrete statistical model format data for the first moment with the historical discrete statistical model format data to update the discrete statistical model format data for the first moment; and an object identification module, which is configured to identify an object from the updated discrete statistical model format data for the first moment, wherein the object includes a lane line characterized by a curve.

[0011] According to a third exemplary embodiment of the present disclosure, a vehicle is provided, comprising one or more cameras, which use a continuous model or a discrete model to represent an object, wherein the object includes a lane line represented by a curve; and an apparatus as in the second exemplary embodiment above.

[0012] This summary is provided to introduce in simplified form a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. Additional aspects, features, and / or advantages of the examples will be set forth in part in the description that follows, and in part will be apparent from the description, or may be learned through practice of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other aspects and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments, taken in conjunction with the accompanying drawings, which illustrate, by way of example, the principles of the present disclosure. Note that the drawings are not necessarily drawn to scale.

[0014] Figure 1 A schematic diagram showing an autonomous driving vehicle 100 traveling on a road according to one or more embodiments of the present invention is shown.

[0015] Figure 2 An object 201 represented by a camera using a continuous model and an object 201 ′ represented by a camera using a discrete model are shown according to one or more embodiments of the present invention.

[0016] Figure 3 is a flow chart of an exemplary method 300 for identifying an object in accordance with one or more embodiments of the present invention.

[0017] Figure 4 Schematic diagrams 401 and 402 are shown for discrete statistical model format data fusion according to one or more embodiments of the present invention.

[0018] Figure 5 Shown according to Figure 3 Flow chart of method 500 for fusing discrete statistical model format data at a first moment with historical discrete statistical model format data in an embodiment.

[0019] Figure 6 is a block diagram of an apparatus 600 for identifying an object according to an embodiment of the present invention.

[0020] Figure 7 A general hardware environment in which the present disclosure is applicable is illustrated according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0021] In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the described exemplary embodiments. However, it will be apparent to one skilled in the art that the described embodiments may be practiced without some or all of these specific details. In other exemplary embodiments, well-known structures or processing steps are not described in detail to avoid unnecessarily obscuring the concepts of the present disclosure.

[0022] The term "vehicle" used in the specification refers to a car, an airplane, a helicopter, a ship, etc. For simplicity, the present invention is described in conjunction with a "car", but the embodiments described herein are not limited to "cars" but are applicable to other types of vehicles. The term "A or B" used in the specification refers to "A and B" and "A or B", and does not mean that A and B are exclusive unless otherwise specified.

[0023] Discrete statistical models

[0024] When autonomous vehicles are driving on the road, they need to know the real-time road conditions. To this end, various types of sensors are installed on the vehicle to act as the "eyes" of the vehicle. Currently, widely used sensors include visual sensors (e.g., cameras) and radar ranging sensors (such as lidar, millimeter wave radar, and ultrasonic radar). Among them, cameras are low-cost and can identify different objects. They have advantages in object height and width measurement accuracy, lane line recognition, and pedestrian recognition accuracy. They are indispensable sensors for lane departure warning and traffic sign recognition.

[0025] refer to Figure 1 , which shows a schematic diagram of an autonomous driving vehicle 100 traveling on a road. Although, for the purpose of illustration, Figure 1 The vehicle 100 in the figure uses a camera 101, a millimeter wave radar 102 and a laser radar 103 to identify an object, but those skilled in the art will fully appreciate that the technical solution of the present invention may use more or fewer sensors. According to one embodiment of the present invention, the object includes a lane line 104. However, those skilled in the art will fully appreciate that other objects represented by curves are also within the scope of the present application.

[0026] Generally speaking, the formats of camera data provided by various camera manufacturers are different. Usually, for each format of sensor data, one or more models are used to record the sensor data in that format. For example, if a vehicle is equipped with cameras manufactured by different manufacturers with different camera data formats, it is necessary to pre-define multiple models for different camera data formats. As a result, when processing a large amount of data, the data processing speed may be delayed due to the mutual conversion between data formats. In addition, in practice, it is very likely that there will be a situation where the pre-installed camera needs to be replaced with a camera with a different output format (for example, the originally installed camera is damaged, the user has a preference for a certain brand of camera, etc.). In this case, it is necessary to redefine the model for the camera data. This is usually inconvenient for the user and may cause program problems in subsequent use.

[0027] The present invention defines a discrete statistical model that can support both continuous and discrete models. The continuous model is used for data in polynomial format. For example, data in polynomial format is generally represented by the polynomial function y=c 0 +c 1 x+c 2 x 2 +c 3 x 3 +...+c n x n to describe objects (such as lane line 104), and the camera uses a continuous model {[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end]} to output data in polynomial format, and the output data is considered to have a continuous data output format. Among them, "start" represents the starting point of the curve described by the polynomial equation, and "end" represents the ending point of the curve described by the polynomial equation. Figure 2 An object 201 is shown represented by a camera using a continuous model.

[0028] The discrete model is used for data in a discrete format. For example, the discrete format data generally uses multiple discrete points to describe an object (such as lane line 104), and the camera uses the discrete model {(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )} to output data in discrete format. The output data is considered to have a discrete data output format. Figure 2 An object 201 ′ is shown that is represented by a camera using a discrete model.

[0029] The discrete statistical model of the present invention integrates the above continuous model and discrete model, and its format is as follows: {t, [c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]}, which represents the camera data output by the camera at time t. 0 ,c 1 ,c 2,c 3 ,...,c n ],[start,end]" is the continuous part, "[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]" are discrete parts, and the data in these two parts describe the same object.

[0030] By adopting the above discrete statistical model, it becomes possible to support two different formats of camera data with only one model, thereby making data processing more efficient and alleviating possible programming pressure.

[0031] Implementation

[0032] Figure 3 is a flow chart of an exemplary method 300 for identifying an object. For example, the method 300 may be executed on at least one processor (e.g., Figure 7 The method 300 may be implemented in a processor 704 of the vehicle, which may be located in a vehicle-mounted computer system, a remote server, or a combination thereof. Of course, in various aspects of the present invention, the method 300 may also be implemented by any suitable device capable of performing related operations.

[0033] The method 300 starts at step 310. At step 310, at a first moment (in the following description, "first moment" is understood as "real time"), first data output by a camera mounted on a vehicle is obtained at the first moment. The output first data has one of a continuous data format for the above-mentioned continuous model or a discrete data format for the above-mentioned discrete model. That is, the first data uses a polynomial to describe an object or uses discrete points to describe an object.

[0034] In step 320, the first data acquired is fed into the discrete statistical model to form discrete statistical model format data for the first moment. That is, by feeding the time information (such as a timestamp, e.g., t FTI(第一时刻) ) and the first data is fed into the discrete statistical model {t,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} to instantiate the discrete statistical model.

[0035] According to one embodiment of the present invention, if the first data has a continuous data output format (ie, {[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end]} format), the first data is placed in the [c0,c1,c2,c3,…,cn],[start,end] part of the discrete statistical model to use the first data as a continuous part of the discrete statistical model format data. Then, based on the polynomial corresponding to the first data, multiple discrete points on the curve described by the first data are calculated. The number of calculated discrete points can be arbitrary. Those skilled in the art will appreciate that the number of calculated discrete points can be selected depending on the processor capability of the vehicle. Of course, the number of calculated discrete points can also be determined based on other criteria. Then, the calculated discrete points are placed in the “[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]" part as the discrete part of the discrete statistical model format data.

[0036] According to one embodiment of the present invention, if the first data has a discrete data output format (ie, {(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )} format), then by putting the first data into the discrete statistical model "[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]” part to take the first data as the discrete part of the discrete statistical model format data. Then, a polynomial corresponding to the first data is calculated based on the first data. It should be understood by those skilled in the art that the discrete points can be mathematically processed by interpolation or curve fitting to obtain a curve represented by a polynomial. Then, the calculated polynomial is placed in the “[c 0 ,c 1 ,c 2 ,c 3 ,...,c n],[start,end]” part as the continuous part of the discrete statistical model format data.

[0037] It can be seen that after step 320, the continuous part and the discrete part of the discrete statistical model format data contain data for describing the same object.

[0038] In step 330, the discrete statistical model format data for the first moment is fused with the historical discrete statistical model format data to update the discrete statistical model format data for the first moment. In practice, the real-time sensor data acquired at a single moment cannot accurately describe the object. In particular, for objects with continuity, such as lane lines, the fusion of sensor data at multiple consecutive moments is required to describe the lane line.

[0039] According to one embodiment of the present invention, it is assumed that a camera mounted on a vehicle outputs camera data at a certain time interval. The time interval may vary according to requirements, such as 0.1 seconds, 0.2 seconds, 0.5 seconds, 1 second, and so on. Of course, other time intervals are also within the scope of the present invention. According to one or more embodiments of the present invention, historical discrete statistical model format data may be formed at several consecutive moments before a first moment, and stored in a memory of the vehicle or cached for fast reading. As should be understood, the historical discrete statistical model format data has the same data format as the discrete statistical model format data for the first moment, and is formed at one or more moments before the first moment in the same manner as the discrete statistical model format data for the first moment.

[0040] According to one embodiment of the present invention, after fusion, the discrete part of the fused discrete statistical model format data includes data of all discrete parts in multiple discrete statistical model format data. For example, during fusion, the discrete part of the updated discrete statistical model format data for the first moment and all discrete parts in the historical discrete statistical model format data are aggregated. Thus, after fusion, the discrete part of the updated discrete statistical model format data for the first moment includes data of the discrete part of the initial discrete statistical model format data for the first moment and data of all discrete parts in the historical discrete statistical model format data. According to one embodiment of the present invention, considering the computing pressure and memory capacity, during fusion, duplicate data or abnormal data in the aggregated discrete data is removed or filtered. For example, in the discrete part of the updated discrete statistical model format data for the first moment, duplicate or abnormal discrete points can be removed.

[0041] According to one embodiment of the present invention, based on the discrete part data of the fused discrete statistical model format data, the continuous data output format data of the curve corresponding to the discrete part data of the fused discrete statistical model format data is calculated as the continuous part of the fused discrete statistical model format data. For example, based on the above discrete part of the updated (i.e., fused) discrete statistical model format data for the first moment, the polynomial corresponding to the updated (i.e., fused) discrete statistical model format data for the first moment is used as the continuous part of the updated (i.e., fused) discrete statistical model format data for the first moment.

[0042] Figure 4 Schematic diagrams 401 and 402 for discrete statistical model format data fusion according to one or more embodiments of the present invention are shown. Specifically, diagram 401 shows a single fusion, while diagram 402 shows iterative fusion.

[0043] Diagram 401 is a schematic diagram showing the fusion of discrete statistical model format data for a first moment with historical discrete statistical model format data including a plurality of discrete statistical model format data for a plurality of previous moments within a threshold time period before the first moment. FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} and {t FTI-1 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]},{t FTI-2 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]}……{t FTI-n ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} are merged to update {t FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]}. Wherein, “FTI” means “first moment in time”. Wherein, the interval between two adjacent moments (for example, t FTI With t FTI-1 The time interval between t and t can be predetermined as described above, and t FTI-n to FTI The threshold time period between {t can also be selected as needed. For example, when the predetermined time interval is 0.1 seconds, the threshold time period can be selected as 1 second, thereby selecting 10 (that is, in this case, n is 10) historical general statistical model format data within 1 second before the first moment to be fused. FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} After fusion with multiple previous historical general statistical model format data, {t FTI,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} is updated to {t FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 )',(x 2 ,y 2 )',....,(x m ,y m )']}. Thus, the method shown in FIG401 is to convert {t FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} is merged with the historical statistical model format data once to update {t FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]}.

[0044] Illustration 402 is a schematic diagram showing iteratively fusing a plurality of statistical model format data for a plurality of previous moments within a threshold time period before a first moment. Continuing with the above example, assuming that the threshold time period is 1 second and the predetermined time interval between two adjacent moments is 0.1 second. For a plurality of previous moments within the threshold time period before a first moment, iteratively perform the following steps until the discrete statistical model format data for the first moment is updated: the discrete statistical model format data for each moment is fused with the discrete statistical model format data for a subsequent moment after a predetermined time interval to update the discrete statistical model format data for the subsequent moment.

[0045] For example, first, {t FTI-n ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} and {t FTI-n+1 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} are fused to obtain the updated {t FTI-n+1 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'}. Next, {t FTI-n+1 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,cn ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'} and {t FTI-n+2 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} are fused to obtain the updated {t FTI-n+2 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'}. Next, {t FTI-n+2 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'} and {t FTI-n+3 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m)]} are fused to obtain the updated {t FTI-n+3 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'}. Finally, {t FTI-1 ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'} and {t FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ],[start,end],[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]} are fused to obtain the updated {t FTI ,[c 0 ,c 1 ,c 2 ,c 3 ,...,c n ]',[start,end]',[(x 1 ,y 1 ),(x 2 ,y 2 ),....,(x m ,y m )]'}. Thus, this process can be considered as "iterative fusion".

[0046] In order to make the fused data more accurate, the following mathematical methods can be used in the fusion process. Figure 5 Shown according to Figure 3Flowchart of a method 500 for fusing discrete statistical model format data for a first moment with historical discrete statistical model format data. In step 510, historical discrete statistical model format data is obtained, the historical discrete statistical model format data comprising a plurality of discrete statistical model format data for a plurality of previous moments within a threshold time period before the first moment.

[0047] In step 520, the discrete statistical model format data for the first moment and the historical discrete statistical model format data are converted to the same coordinate system. For example, assuming that the vehicle is the origin of the local coordinate system, the direction of the vehicle is the x-axis of the local coordinate system, and the direction perpendicular to the direction of the vehicle is the y-axis of the local coordinate system. Then, as the vehicle moves from t FTI-1 to FTI Traveling a distance L in the direction of travel, it can be understood that at t FTI The origin of the local coordinate system at t FTI-1 The origin of the local coordinate system at t is moved by (Lx,Ly). Through coordinate transformation, all historical discrete statistical model format data are converted to t FTI The data used for fusion are in the same coordinate system.

[0048] According to another embodiment of the present invention, various types of coordinate systems used by the historical discrete statistical model format data and the discrete statistical model format data for the first moment can be converted into a world coordinate system, so that the data used for fusion are all in the same coordinate system. Various coordinate conversion methods include, but are not limited to, translation and rotation of coordinates in two-dimensional space, translation and rotation of coordinates in three-dimensional space, and the like.

[0049] In step 530, by using Figure 4 Any one of the two fusion modes 401 and 402 shown in the figure fuses the discrete statistical model format data for the first moment and the historical discrete statistical model format data in the same coordinate system to update the discrete statistical model format data for the first moment.

[0050] In one embodiment, a weighted average algorithm may also be used for fusion. For example, during fusion, the historical discrete statistical model format data recorded closer to the first moment in time is assigned a higher weight, while the historical discrete statistical model format data recorded farther from the first moment in time is assigned a lower weight. Of course, other weighting methods may also be conceived.

[0051] return Figure 3In step 340, the object is identified based on the updated discrete statistical model format data for the first moment. In one embodiment, a mathematical algorithm may be used to identify the lane curve represented from the data in the discrete data output format in the updated discrete statistical model format data for the first moment, the mathematical algorithm comprising one of the following: a regression algorithm, an interpolation algorithm, and a fitting algorithm.

[0052] Thus, by using the solution of the present invention, the data set can be processed more quickly by putting different types of data into a unified data model. By using a data model that supports both continuous and discrete models of labeling, the vehicle system can be compatible with both cameras. Moreover, historical information is taken into account, making the marking of lane lines more accurate.

[0053] Figure 6 6 is a block diagram of a device 600 for identifying an object according to an embodiment of the present invention. All functional blocks of the device 600 (including various units in the device 600) can be implemented by hardware, software, or a combination of hardware and software. Those skilled in the art should understand that Figure 6 The functional blocks described in the description may be combined into a single functional block or divided into multiple sub-functional blocks.

[0054] The device 600 may include a data acquisition module 610, which is configured to acquire raw data output by a camera mounted on the vehicle at a first moment, and the output raw data has one of a continuous data output format for a continuous model or a discrete data output format for a discrete model. The device 600 includes a data feeding module 620, which is configured to feed the raw data to a discrete statistical model to form discrete statistical model format data for the first moment, and the discrete statistical model format data includes a continuous part with a continuous data output format and a discrete part with a discrete data output format. The device 600 includes a data fusion module 630, which is configured to fuse the discrete statistical model format data for the first moment with the historical discrete statistical model format data to update the discrete statistical model format data for the first moment. In addition, the device 600 also includes an object identification module 640, which is configured to identify an object from the updated discrete statistical model format data for the first moment, wherein the object includes a lane line represented by a curve.

[0055] Reference Figure 7, the computing device 700 will now be described, which is an example of a hardware device applicable to various aspects of the present disclosure. The computing device 700 can be any machine configured to perform processing and / or computing, and can be, but is not limited to, a workstation, a server, a desktop computer, a laptop computer, a tablet computer, a personal digital assistant, a smart phone, a vehicle-mounted computer, or any combination thereof. The above-mentioned system can be implemented in whole or in part by the computing device 700 or a similar device or system.

[0056] The computing device 700 may include an element that is connected to or in communication with the bus 702, possibly via one or more interfaces. For example, the computing device 700 may include a bus 702, and one or more processors 704, one or more input devices 706, and one or more output devices 708. The one or more processors 704 may be any type of processor, and may include, but are not limited to, one or more general-purpose processors and / or one or more special-purpose processors (such as special processing chips). The input device 706 may be any type of device that can input information into the computing device, and may include, but are not limited to, a mouse, a keyboard, a touch screen, a microphone, and / or a remote controller. The output device 708 may be any type of device that can present information, and may include, but are not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The computing device 700 may also include or be connected to a non-transitory storage device 710, which may be any storage device that is non-transitory and can achieve data storage, and may include but is not limited to a disk drive, an optical storage device, a solid-state storage, a floppy disk, a floppy disk, a hard disk, a tape or any other magnetic medium, an optical disk or any other optical medium, a ROM (read-only memory), a RAM (random access memory), a cache memory and / or any other memory chip or memory cartridge, and / or any other medium from which a computer can read data, instructions and / or code. The non-transitory storage device 710 may be separable from the interface. The non-transitory storage device 710 may have data / instructions / code for implementing the above methods and steps. The computing device 700 may also include a communication device 712. The communication device 712 may be any type of device or system that can achieve communication with an external device and / or network, and may include but is not limited to a modem, a network card, an infrared communication device, a Bluetooth communication device, a wireless communication device, a wireless communication device such as ... TM devices, 802.11 devices, WiFi devices, WiMax devices, wireless communication devices and / or chipsets such as cellular communication facilities, and the like.

[0057] When the computing device 700 is used as a vehicle-mounted device, the computing device 700 may also be connected to external devices, such as a GPS receiver, sensors for sensing different environmental data (such as an accelerometer, a wheel speed sensor, a gyroscope), etc. In this way, the computing device 700 may, for example, receive location data and sensor data indicating the driving conditions of the vehicle. When the computing device 700 is used as a vehicle-mounted device, the computing device 700 may also be connected to other facilities for controlling the driving and operation of the vehicle (such as an engine system, wipers, an anti-lock braking system, etc.).

[0058] In addition, the non-transitory storage device 710 may have map information and software elements so that the processor 704 can perform route guidance processing. In addition, the output device 708 may include a display for displaying a map, a location marker of the vehicle, and an image indicating the driving situation of the vehicle. The output device 708 may also include a speaker or interface with headphones for audio guidance.

[0059] The bus 702 may include, but is not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus. Specifically, for in-vehicle devices, the bus 702 may include a Controller Area Network (CAN) bus or other architectures designed for applications in automobiles.

[0060] The computing device 700 may also include a working memory 714 , which may be any type of working memory that may store instructions and / or data useful for operation of the processor 704 , and may include, but is not limited to, random access memory and / or read-only memory devices.

[0061] Software elements may be located in the working memory 714, including but not limited to an operating system 716, one or more applications 718, drivers, and / or other data and code. Instructions for executing the above methods and steps may be included in one or more applications 718, and the above-mentioned units of the apparatus 800 may be implemented by the processor 704 reading and executing the instructions of one or more applications 718. The executable code or source code of the instructions of the software elements may be stored in a non-transitory computer-readable storage medium (such as the above-mentioned storage device 710), and may be read into the working memory 714, possibly by compilation and / or installation. The executable code or source code of the instructions of the software elements may also be downloaded from a remote location.

[0062] From the above embodiments, it will be clear to those skilled in the art that the present disclosure can be implemented by software with necessary hardware, or by hardware, firmware, etc. Based on such understanding, the embodiments of the present disclosure can be implemented in part in the form of software. Computer software can be stored in a readable storage medium such as a floppy disk, hard disk, optical disk, or flash memory of a computer. The computer software includes a series of instructions to enable a computer (e.g., a personal computer, a service station, or a network terminal) to perform a method or a part thereof according to a corresponding embodiment of the present disclosure.

[0063] Throughout the specification, reference has been made to "one example" or "an example", which means that the specifically described features, structures, or characteristics are included in at least one example. Therefore, the use of such phrases may refer to more than one example. Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more examples.

[0064] However, one skilled in the relevant art may recognize that the examples may be practiced without one or more of the specific details, or with other methods, resources, materials, etc. In other instances, well-known structures, resources, or operations are not shown or described in detail to avoid obscuring aspects of the examples.

[0065] Although examples and applications have been illustrated and described, it should be understood that these examples are not limited to the precise configuration and resources described above. Various modifications, changes and variations that are obvious to those skilled in the art can be made to the arrangement, operation and details of the methods and systems disclosed herein without departing from the scope of the examples claimed.

Claims

1. A method for identifying an object, comprising: At a first moment, raw data output by a camera mounted on a vehicle is acquired, wherein the output raw data has one of a continuous data output format for a continuous model or a discrete data output format for a discrete model; Feeding the raw data to a discrete statistical model to form discrete statistical model format data for the first moment, the discrete statistical model format data comprising a continuous part having the continuous data output format and a discrete part having the discrete data output format; Merging the discrete statistical model format data for the first moment with the historical discrete statistical model format data to update the discrete statistical model format data for the first moment; as well as identifying an object from the updated discrete statistical model format data for the first time instant, wherein the object comprises a lane line represented by a curve; The step of feeding the raw data to a discrete statistical model to form discrete statistical model format data for the first moment further comprises: If the original data has a continuous data output format, the original data is used as a continuous part of the discrete statistical model format data, and a plurality of discrete points on a curve described by the original data is calculated as a discrete part of the discrete statistical model format data; If the original data has a discrete data output format, continuous data output format data of a curve corresponding to the original data is calculated based on the original data as a continuous part of the discrete statistical model format data, and the original data is used as a discrete part of the discrete statistical model format data.

2. The method according to claim 1, characterized in that The continuous model uses a polynomial to describe the curve, and the discrete model uses a plurality of discrete points to describe the curve.

3. The method according to claim 1, characterized in that The camera includes one or more of the following: a monocular camera, a binocular stereo vision camera, a panoramic vision camera, and an infrared camera.

4. The method according to claim 1, characterized in that The step of fusing the discrete statistical model format data for the first moment with the historical discrete statistical model format data further includes: The discrete statistical model format data for the first moment is fused with historical discrete statistical model format data including multiple discrete statistical model format data for multiple previous moments within a threshold time period before the first moment, so that the discrete part of the fused discrete statistical model format data includes data of all discrete parts in the multiple discrete statistical model format data, and based on the discrete part data of the fused discrete statistical model format data, the continuous data output format data of the curve corresponding to the discrete part data of the fused discrete statistical model format data is calculated as the continuous part of the fused discrete statistical model format data.

5. The method according to claim 1, characterized in that The step of fusing the discrete statistical model format data for the first moment with the historical discrete statistical model format data further includes: For a plurality of discrete statistical model format data at a plurality of previous time instants within a threshold time period before the first time instant, iteratively performing: fusing the discrete statistical model format data for each moment with the discrete statistical model format data for a subsequent moment after a predetermined time interval to update the discrete statistical model format data for the subsequent moment until the discrete statistical model format data for the first moment is updated, so that the discrete parts of the fused discrete statistical model format data include data of all discrete parts in the plurality of discrete statistical model format data, and Based on the discrete part data of the fused discrete statistical model format data, the continuous data output format data of the curve corresponding to the discrete part data of the fused discrete statistical model format data is calculated as the continuous part of the fused discrete statistical model format data.

6. The method according to claim 4 or 5, characterized in that The method further includes removing duplicate data from all discrete parts of the plurality of discrete statistical model format data.

7. The method according to claim 1, characterized in that The fusion further includes: taking the position of the vehicle at the first moment as the origin of the local coordinate system to convert the historical discrete statistical model format data into the local coordinate system, so that the historical discrete statistical model format data and the discrete statistical model format data for the first moment are both represented by the local coordinate system.

8. The method according to claim 1, characterized in that The fusion further includes: converting different types of coordinate systems used in the historical discrete statistical model format data and in the discrete statistical model format data for the first moment into a world coordinate system, so that both the historical discrete statistical model format data and the discrete statistical model format data for the first moment are represented by the same coordinate system.

9. The method according to claim 1, characterized in that Identifying the object from the updated discrete statistical model format data for the first moment further includes: using a mathematical algorithm to identify the object from the data with a discrete data output format in the updated discrete statistical model format data for the first moment, the mathematical algorithm including one of the following: a regression algorithm, an interpolation algorithm, and a fitting algorithm.

10. A device for identifying an object, comprising: A data acquisition module, the data acquisition module being configured to acquire raw data output by a camera mounted on the vehicle at a first moment, the output raw data having one of a continuous data output format for a continuous model or a discrete data output format for a discrete model; a data feeding module, the data feeding module being configured to feed the raw data to a discrete statistical model to form discrete statistical model format data for the first moment, the discrete statistical model format data comprising a continuous part having a continuous data output format and a discrete part having a discrete data output format; A data fusion module, wherein the data fusion module is configured to fuse the discrete statistical model format data for the first moment with the historical discrete statistical model format data to update the discrete statistical model format data for the first moment; as well as an object identification module configured to identify the object from the updated discrete statistical model format data for the first moment, wherein the object includes a lane line represented by a curve; The data fusion module is further configured to: If the original data has a continuous data output format, the original data is used as a continuous part of the discrete statistical model format data, and a plurality of discrete points on a curve described by the original data is calculated as a discrete part of the discrete statistical model format data; If the original data has a discrete data output format, continuous data output format data of a curve corresponding to the original data is calculated based on the original data as a continuous part of the discrete statistical model format data, and the original data is used as a discrete part of the discrete statistical model format data.

11. The device according to claim 10, characterized in that The data fusion module is further configured to: The discrete statistical model format data for the first moment is fused with historical discrete statistical model format data including multiple discrete statistical model format data for multiple previous moments within a threshold time period before the first moment, so that the discrete part of the fused discrete statistical model format data includes data of all discrete parts in the multiple discrete statistical model format data, and based on the discrete part data of the fused discrete statistical model format data, the continuous data output format data of the curve corresponding to the discrete part data of the fused discrete statistical model format data is calculated as the continuous part of the fused discrete statistical model format data.

12. The device according to claim 10, characterized in that The data fusion module is further configured to: For a plurality of discrete statistical model format data at a plurality of previous time instants within a threshold time period before the first time instant, iteratively performing the following: fusing the discrete statistical model format data for each moment with the discrete statistical model format data for a subsequent moment after a predetermined time interval to update the discrete statistical model format data for the subsequent moment, until the discrete statistical model format data for the first moment is updated, so that the discrete parts of the fused discrete statistical model format data include data of all discrete parts in the plurality of discrete statistical model format data, and Based on the discrete portion data of the fused discrete statistical model format data, continuous data output format data of a curve corresponding to the discrete portion data of the fused discrete statistical model format data is calculated as a continuous portion of the fused discrete statistical model format data.

13. A vehicle comprising: One or more cameras that use one of a continuous model or a discrete model to represent objects, wherein the objects include lane lines represented by curves; as well as The device as claimed in claim 10.

Images