Method and apparatus for determining object position, electronic device, and storage medium

Abstract

The present disclosure provides a method and device for determining the position of an object, electronic equipment and storage medium, relating to the technical field of computer, and particularly to the technical field of artificial intelligence such as image processing, automatic driving, intelligent transportation, and intelligent logistics. The specific scheme is as follows: first, a fisheye camera acquires a first image at a current time, and identifies the first image to determine target boundary points of a target object contained in the first image; then, a historical size of the target object is acquired; and finally, based on the historical size and the target boundary points, a current position of the target object is determined.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to the fields of artificial intelligence technology such as image processing, autonomous driving, intelligent transportation, and smart logistics. Specifically, it relates to a method, apparatus, electronic device, and storage medium for determining the position of an object. Background Technology

[0002] Currently, in roadside perception multi-camera fusion technology, due to the influence of the fisheye camera shooting mode, the images captured by the fisheye camera may contain truncated objects, which will lead to inaccurate object information after fusion. Summary of the Invention

[0003] This disclosure aims to at least partially address one of the technical problems in the related art.

[0004] The first aspect of this disclosure provides a method for determining the position of an object, including:

[0005] Acquire the first image captured by the fisheye camera at the current moment;

[0006] The first image is identified to determine the target boundary points of the target objects contained in the first image.

[0007] Obtain the historical dimensions of the target object;

[0008] Based on the historical dimensions and the target boundary points, the current position of the target object is determined.

[0009] A second aspect of this disclosure provides an apparatus for determining the position of an object, comprising:

[0010] The first acquisition module is used to acquire the first image captured by the fisheye camera at the current moment;

[0011] The recognition module is used to recognize the first image and determine the target boundary points of the target objects contained in the first image.

[0012] The second acquisition module is used to acquire the historical dimensions of the target object;

[0013] The determination module is used to determine the current position of the target object based on the historical dimensions and the target boundary points.

[0014] A third aspect of this disclosure provides a computer device including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the method for determining the position of an object as provided in the first aspect of this disclosure.

[0015] A fourth aspect of this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for determining the position of an object as described in a first aspect of this disclosure.

[0016] A fifth aspect of this disclosure provides a computer program product including a computer program that, when executed by a processor, implements the method for determining the position of an object as described in a first aspect of this disclosure.

[0017] The method, apparatus, electronic device, and storage medium for determining the position of an object provided in this disclosure have the following beneficial effects:

[0018] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, the historical dimensions of the target object are acquired. Finally, based on the historical dimensions and the target boundary points, the current position of the target object is determined. Therefore, by identifying the target boundary points of the target object in the image captured by the fisheye camera at the current moment, and determining the current position of the target object based on the target boundary points and historical dimensions, the efficiency and accuracy of determining the object's position are improved.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0020] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0021] Figure 1 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0022] Figure 2 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0023] Figure 3 A schematic diagram of the target boundary points of the target object provided in the embodiments of this disclosure;

[0024] Figure 4 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0025] Figure 5 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0026] Figure 6 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0027] Figure 7 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0028] Figure 8 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure;

[0029] Figure 9 A schematic diagram illustrating the determination of the position and bounding box of a target object according to an embodiment of this disclosure;

[0030] Figure 10 This is a schematic diagram of the structure of the device for determining the position of an object provided in an embodiment of this disclosure;

[0031] Figure 11 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0032] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0033] This disclosure relates to the fields of artificial intelligence technology, such as autonomous driving and intelligent transportation.

[0034] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence.

[0035] Image processing refers to the techniques of analyzing, processing, and manipulating images to meet visual, psychological, or other requirements.

[0036] Autonomous driving, also known as driverless driving, computer-controlled driving, or wheeled mobile robots, is a cutting-edge technology that relies on computer and artificial intelligence technologies to achieve complete, safe, and efficient driving without human intervention.

[0037] Intelligent Traffic (IT) is the effective integration of advanced science and technology (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operations research, artificial intelligence, etc.) into transportation, service control, and vehicle manufacturing. It strengthens the connection between vehicles, roads, and users, thereby forming a comprehensive transportation system that ensures safety, improves efficiency, improves the environment, and saves energy.

[0038] Smart logistics refers to a modern logistics model that uses intelligent hardware and software, the Internet of Things, big data and other intelligent technologies to achieve refined, dynamic and visual management of all aspects of logistics, improve the intelligent analysis and decision-making and automated operation execution capabilities of the logistics system, and enhance the efficiency of logistics operations.

[0039] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0040] The following description, with reference to the accompanying drawings, describes a method, apparatus, electronic device, and storage medium for determining the position of an object according to embodiments of the present disclosure.

[0041] Figure 1 This is a flowchart illustrating a method for determining the position of an object provided in an embodiment of this disclosure.

[0042] like Figure 1 As shown, the method for determining the position of an object may include the following steps:

[0043] Step 101: Obtain the first image captured by the fisheye camera at the current moment.

[0044] The first image can be the image data captured by the fisheye camera at the current moment.

[0045] It should be noted that the first image may contain any type and number of objects, and this disclosure does not impose any restrictions on this.

[0046] Step 102: Identify the first image and determine the target boundary points of the target objects contained in the first image.

[0047] The target object can be any object in the first image. In some possible cases, the target object may be a truncated object in the first image, that is, the target object may be an object that was not fully captured by the fisheye camera. It can be any type of object, and this disclosure does not limit it.

[0048] Among them, the target boundary points can be the visible boundary points of the target object in the image captured by the fisheye camera.

[0049] In this disclosure, after obtaining the first image captured by the fisheye camera at the current moment, since the fisheye camera shoots vertically downwards, only objects within the field of view captured by the fisheye camera can be completely captured. When a part of the target object is not within the field of view, the fisheye camera will only capture the other part of the target object within the field of view. At this time, the image of the target object in the first image is truncated and incomplete. In order to determine the current position of the target object, the target boundary points of the target object contained in the first image can be determined first.

[0050] It should be noted that in order to accurately determine the position of the target object, it is usually necessary to determine at least two target boundary points, but this disclosure does not limit this.

[0051] Step 103: Obtain the historical dimensions of the target object.

[0052] In this disclosure, after determining the target boundary points of the target object contained in the first image, the historical dimensions of the target object can be obtained in order to improve the accuracy of determining the current position of the target object.

[0053] It should be noted that the historical size of the target object can be the historical size of the target object captured by the fisheye camera before the current moment, or it can be the historical size of the target object captured by other cameras. This disclosure does not limit it in this way.

[0054] It should be noted that other camera types may be the same as or different from fisheye cameras. For example, other cameras may be fisheye cameras or bullet cameras, etc. This disclosure does not limit them.

[0055] Step 104: Determine the current position of the target object based on historical dimensions and target boundary points.

[0056] In this disclosure, after obtaining the target visible point and historical size of the target object, the current position of the target object can be determined based on the target visible point and historical size, thereby improving the accuracy of determining the position of the target object.

[0057] Optionally, in this disclosure, the current bounding box corresponding to the target object can also be determined based on the historical location and target boundary points.

[0058] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, the historical dimensions of the target object are acquired. Finally, based on the historical dimensions and the target boundary points, the current position and bounding box of the target object are determined. Thus, by identifying the image captured by the fisheye camera at the current moment and determining the target boundary points of the target object in the image, and by determining the current position of the target object based on the target boundary points and the historical dimensions, the efficiency and accuracy of determining the object's position are improved.

[0059] Figure 2 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure.

[0060] like Figure 2 As shown, the method for determining the position of an object may include the following steps:

[0061] Step 201: Obtain the first image captured by the fisheye camera at the current moment.

[0062] The specific implementation of step 201 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0063] Step 202: Identify the first image and determine multiple boundary points of the target object contained in the first image.

[0064] In this disclosure, after obtaining the first image captured by the fisheye camera at the current moment, the first image can be identified to determine multiple boundary points of the target object contained in the first image, thereby providing conditions for improving the determination of the current position of the target object.

[0065] Step 203: Among the multiple boundary points, the boundary points that meet the requirements of the fisheye camera's setting position are determined as the target boundary points of the target object.

[0066] The requirements for the setting positions of the boundary points and the fisheye camera can be any pre-set requirements, or they can be requirements determined according to needs. For example, the requirement for the setting positions of the boundary points and the fisheye camera can be that they are closest to the setting position of the fisheye camera, etc. This disclosure does not limit this.

[0067] In this disclosure, after determining multiple boundary points of the target object contained in the first image, the boundary point closest to the setting position of the fisheye camera among the multiple boundary points can be determined as the target boundary point of the target object.

[0068] In this disclosure, after determining multiple boundary points of the target object contained in the first image, in order to improve the accuracy of the determined position of the target object, at least two target boundary points can be determined from these multiple boundary points.

[0069] The following is combined Figure 3 Taking the determination of two target boundary points as an example, this embodiment of the present disclosure provides an example of determining the target boundary points of a target object. Figure 3 This is a schematic diagram of the target boundary points of the target object provided in the embodiments of this disclosure, wherein the vertical shooting range of the fisheye camera is as follows: Figure 3 As shown, the image captured by the fisheye camera at the current moment only contains a portion of the target object. The boundary points of this portion of the target object in the captured image are boundary point A, boundary point B, boundary point C, and boundary point D. Among these boundary points, the two boundary points closest to the fisheye camera's setting position can be identified as the target boundary points. Figure 3 In this context, boundary points A and B are the closest boundary points to the fisheye camera's location compared to boundary points C and D. Therefore, boundary points A and B can be identified as the target boundary points of the target object, thereby improving the accuracy and reliability of the identified target boundary points. This disclosure does not limit this aspect.

[0070] Step 204: Obtain the historical dimensions of the target object.

[0071] Step 205: Determine the current position of the target object based on historical dimensions and target boundary points.

[0072] The specific implementation of steps 204 to 205 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0073] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine multiple boundary points of the target object contained in the first image. Then, the boundary points among the multiple boundary points that meet the requirements of the fisheye camera's setting position are determined as the target boundary points of the target object. Next, the historical dimensions of the target object are acquired. Finally, based on the historical dimensions and the target boundary points, the current position of the target object is determined. Thus, by identifying the multiple boundary points of the target object contained in the image captured by the fisheye camera, determining the boundary points among the multiple boundary points of the target object contained in the image that meet the requirements of the fisheye camera's setting position as the target boundary points of the target object, and determining the current position of the target object based on the target boundary points and the historical dimensions, the accuracy of determining the position of the target object is improved.

[0074] Figure 4 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure.

[0075] like Figure 4 As shown, the method for determining the position of an object may include the following steps:

[0076] Step 401: Obtain the first image captured by the fisheye camera at the current moment.

[0077] Step 402: Identify the first image and determine the target boundary points of the target objects contained in the first image.

[0078] The specific implementation of steps 401 to 402 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0079] Step 403: Obtain the image sequence acquired during the first consecutive time period before the current time.

[0080] It should be noted that the length of the first continuous time period can be any preset length, or it can be determined according to actual needs. For example, the length of the first continuous time period can be 10 seconds, etc., and this disclosure does not limit it in this way.

[0081] The image sequence can be a sequence of images acquired within a first consecutive time period.

[0082] It should be noted that the method for obtaining the image sequence within the first continuous time period can be arbitrary. For example, the image sequence can be obtained by tracking and capturing objects within the first continuous time period through a video stream, and this disclosure does not limit this method.

[0083] Step 404: Recognize the image sequence to determine the reference size of the target object and the number of times the reference size appears.

[0084] The reference size can be the size of the target object captured in the image sequence.

[0085] It should be noted that the number of reference dimensions corresponding to the target object can be arbitrary. For example, the target object can have one reference dimension, or it can have multiple reference dimensions; this disclosure does not limit this.

[0086] In this disclosure, after obtaining the image sequence acquired within the first continuous time period, the image sequence can be identified to determine the reference size of the target object contained in the image sequence and the number of times the reference size appears, thereby providing a data basis for determining the accuracy of the target object size.

[0087] It should be noted that when there are multiple reference dimensions, the number of times each reference dimension appears can be any number of times, and the number of times each reference dimension appears may be different from or the same as the number of times other reference dimensions appear. This disclosure does not limit this.

[0088] In this disclosure, after determining the reference dimensions corresponding to the target object, in the case of multiple reference dimensions, in order to improve the reliability of the determined target object's position and bounding box, the occurrence frequency of each reference dimension can also be determined.

[0089] Step 405: Determine the historical dimensions of the target object based on the reference dimensions and the number of times the reference dimensions appear.

[0090] In this disclosure, after determining the reference size of the target object and the number of times the reference size occurs, in order to improve the accuracy of the determined historical size of the target object, the historical size of the target object can be determined based on the reference size of the target object and the number of times the reference size occurs.

[0091] It should be noted that when there are multiple reference dimensions, in order to improve the accuracy of the historical dimensions of the target object, after determining the number of occurrences of each reference dimension, the historical dimensions of the target object can also be determined based on the number of occurrences of each reference dimension. This disclosure does not limit this.

[0092] Step 406: Determine the current position of the target object based on historical dimensions and target boundary points.

[0093] The specific implementation of step 406 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0094] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, an image sequence captured in a first continuous time period before the current moment is acquired, and the image sequence is identified to determine the reference size of the target object and the frequency of occurrence of the reference size. Subsequently, based on the reference size and the frequency of occurrence of the reference size, the historical size of the target object is determined. Finally, based on the historical size and the target boundary points, the current position of the target object is determined. Thus, by determining the target boundary points of the target object contained in the image captured by the fisheye camera at the current moment, determining the reference size of the target object and the frequency of occurrence of the reference size based on the image sequence in the continuous time period before the current moment, determining the historical size of the target object based on the reference size and the frequency of occurrence of the reference size, and determining the current position of the target object based on the historical size and the target boundary points, the reliability of determining the position of the target object is improved.

[0095] Figure 5 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure.

[0096] like Figure 5 As shown, the method for determining the position of an object may include the following steps:

[0097] Step 501: Obtain the first image captured by the fisheye camera at the current moment.

[0098] Step 502: Identify the first image and determine the target boundary points of the target objects contained in the first image.

[0099] Step 503: Obtain the image sequence acquired during the first consecutive time period before the current time.

[0100] The specific implementation of steps 501 to 503 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0101] Step 504: In the case that the image sequence contains a first sequence captured by a fisheye camera and a second sequence captured by another camera, the first sequence and the second sequence are identified respectively to obtain a first object and first information of the first object contained in the first sequence, and a second object and second information of the second object contained in the second sequence.

[0102] It should be noted that the type of the other camera may be different from, or may be the same as, the type of fisheye camera. For example, the other camera could also be a fisheye camera, etc., and this disclosure does not limit it in this respect.

[0103] It should be noted that the first sequence may be different from the second sequence, but this disclosure does not limit this.

[0104] The first information can be used to characterize the first object, and it can be any information. For example, the first information can be the shape, type, position, etc. of the first object, and this disclosure does not limit it.

[0105] It should be noted that the second object may be the same as the first object, or it may be different from the first object; this disclosure does not limit this.

[0106] The second information can be used to characterize the second object, and it can be any information. For example, the second information can be the shape, type, position, etc. of the second object, and this disclosure does not limit it.

[0107] It should be noted that the second piece of information may be different from the first piece of information, or it may be the same as the first piece of information; this disclosure does not limit this.

[0108] In this disclosure, after obtaining the image sequence within a first continuous time period, if the image sequence includes a first sequence captured by a fisheye camera and a second sequence captured by another camera, object recognition can be performed on the first sequence and the second sequence respectively to obtain the first object and the first information of the first object contained in the first sequence, and the second object and the second information of the second object contained in the second sequence, thereby providing a data basis for determining whether the first object and the second object are both target objects.

[0109] Step 505: Based on the first information and the second information, determine whether the first object and the second object are both target objects.

[0110] In this disclosure, after obtaining the first information of the first object and the second information of the second object, in order to improve the accuracy of determining the position of the object, it is possible to determine whether the first object and the second object are target objects based on the first information and the second information.

[0111] Step 506: If both the first object and the second object are target objects, determine the reference dimensions of the target objects in the first sequence and the second sequence, and the number of times the reference dimensions appear in the first sequence and the second sequence.

[0112] In this disclosure, when both the first object and the second object are target objects, since the reference size of the target object in the first sequence and the reference size in the second sequence may deviate, in order to improve the accuracy of the determined historical size of the target object, the reference size of the target object in the first sequence and the second sequence, and the number of times the reference size appears in the first sequence and the second sequence can be determined first.

[0113] It should be noted that when both the first object and the second object are target objects, and there are multiple reference dimensions for the target objects, the number of times each reference dimension of the target object appears in the first sequence and the second sequence can be determined, thereby improving the accuracy of determining the historical dimensions of the target objects. This disclosure does not limit this.

[0114] It should be noted that the reference dimensions of the target object in the first sequence and the second sequence may be different or the same, and this disclosure does not limit this.

[0115] It should be noted that the number of times the reference size of the target object appears in the first sequence and the second sequence may be different or the same, and this disclosure does not limit this.

[0116] Step 507: Determine the historical dimensions of the target object based on the reference dimensions and the number of times the reference dimensions appear.

[0117] Step 508: Determine the current position of the target object based on historical dimensions and target boundary points.

[0118] The specific implementation of steps 507 to 508 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0119] In this embodiment of the present disclosure, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, an image sequence captured in a first continuous time period before the current moment is acquired. If the image sequence contains a first sequence captured by the fisheye camera and a second sequence captured by another camera, the first sequence and the second sequence are identified respectively to obtain a first object and its first information contained in the first sequence, a second object and its second information contained in the second sequence, and whether the first object and the second object are both target objects based on the first information and the second information. If the first object and the second object are both target objects, the reference size of the target object in the first sequence and the second sequence, and the number of times the reference size appears in the first sequence and the second sequence are determined. Then, the historical size of the target object is determined based on the reference size and the number of times the reference size appears. Finally, the current position of the target object is determined based on the historical size and the target boundary points. Therefore, by using the image captured by the fisheye camera at the current moment, the target boundary points of the target object are determined. Based on the image sequences captured by the fisheye camera and another camera, the object, object information, and reference size in each sequence are determined. When the object in each sequence is the target object, the historical size of the target object is determined based on the reference size of the target object and the number of times the reference size appears. Based on the historical size and the target boundary points, the current position of the target object is determined, thereby improving the accuracy of determining the position of the target object.

[0120] Figure 6 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure.

[0121] like Figure 6 As shown, the method for determining the position of an object may include the following steps:

[0122] Step 601: Obtain the first image captured by the fisheye camera at the current moment.

[0123] Step 602: Identify the first image and determine the target boundary points of the target objects contained in the first image.

[0124] Step 603: Obtain the image sequence acquired during the first consecutive time period before the current time.

[0125] Step 604: In the case that the image sequence contains a first sequence captured by a fisheye camera and a second sequence captured by another camera, the first sequence and the second sequence are identified respectively to obtain a first object and first information of the first object contained in the first sequence, and a second object and second information of the second object contained in the second sequence.

[0126] The specific implementation of steps 601 to 604 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0127] Step 605: Determine the first probability ellipse associated with the first object based on the type of the first object in the first information, and determine the second probability ellipse associated with the second object based on the type of the second object in the second information.

[0128] It should be noted that the first probability ellipse associated with the first object may be different depending on the type of the first object, and this disclosure does not limit this.

[0129] It should be noted that the second probability ellipse associated with the determined second object may be different depending on the type of the second object, and this disclosure does not limit this.

[0130] It should be noted that the type of the first object may be different from or the same as the type of the second object; this disclosure does not limit this.

[0131] In this disclosure, after the fisheye camera and another camera are deployed, the probability ellipse of the first object and the probability ellipse of the second object can be determined based on historical data statistics and analysis, as well as the type of the first object and the type of the second object.

[0132] In this disclosure, after obtaining the first information of the first object contained in the first sequence and the second information of the second object contained in the second sequence, a first probability ellipse of the first object can be determined based on the type of the first object in the first information, and a second probability ellipse of the second object can be determined based on the type of the second object in the second information, thereby providing conditions for determining whether the first object and the second object are both target objects.

[0133] Step 606: Based on the first probability ellipse, the second probability ellipse, the first position in the first information, and the second position in the second information, determine whether the first object and the second object are both target objects.

[0134] It should be noted that the first position in the first information may be different from, or may be the same as, the second position in the second information, and this disclosure does not limit this.

[0135] In this disclosure, after determining the first probability ellipse of the first object and the second probability ellipse of the second object, in order to improve the accuracy of determining whether the first object and the second object are both target objects, it is possible to determine whether the first object and the second object are both target objects based on the first probability ellipse, the second probability ellipse, the first position of the first object in the first information, and the second position of the second object in the second information.

[0136] Step 607: If both the first object and the second object are target objects, determine the reference dimensions of the target objects in the first sequence and the second sequence, and the number of times the reference dimensions appear in the first sequence and the second sequence.

[0137] Step 608: Determine the historical dimensions of the target object based on the reference dimensions and the number of times the reference dimensions appear.

[0138] Step 609: Determine the current position of the target object based on historical dimensions and target boundary points.

[0139] The specific implementation of steps 607 to 609 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0140] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target objects contained in the first image. Then, an image sequence captured in a first continuous time period before the current moment is acquired. If the image sequence contains a first sequence captured by the fisheye camera and a second sequence captured by another camera, the first sequence and the second sequence are identified respectively to obtain a first object and its first information contained in the first sequence, and a second object and its second information contained in the second sequence. Then, based on the type of the first object in the first information, a first probability ellipse associated with the first object is determined, and based on the type of the second object in the second information, a second probability ellipse associated with the second object is determined. Based on the first probability ellipse, the second probability ellipse, the first position in the first information, and the second position in the second information, it is determined whether the first object and the second object are both target objects. If both the first object and the second object are target objects, the reference size of the target object in the first sequence and the second sequence, and the number of times the reference size appears in the first sequence and the second sequence are determined. Finally, based on the reference size and the number of times the reference size appears, the historical size of the target object is determined, and based on the historical size and the target boundary points, the current position of the target object is determined. Therefore, by using the image captured by the fisheye camera at the current moment, the target boundary point of the target object is determined. Based on the image sequences captured by the fisheye camera and another camera in the continuous time period before the current moment, the objects contained in each image sequence and the information of the objects are determined. When the objects contained in each image sequence are the target object, the historical size of the target object is determined based on the reference size of the target object and the number of times the reference size appears. Based on the historical size and the target boundary point, the current position of the target object is determined, thereby improving the accuracy of the determined position of the target object.

[0141] Figure 7 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure.

[0142] like Figure 7 As shown, the method for determining the position of an object may include the following steps:

[0143] Step 701: Obtain the first image captured by the fisheye camera at the current moment.

[0144] Step 702: Identify the first image and determine the target boundary points of the target objects contained in the first image.

[0145] Step 703: Obtain the image sequence acquired during the first consecutive time period before the current time.

[0146] Step 704: Recognize the image sequence to determine the reference size of the target object and the number of times the reference size appears.

[0147] The specific implementation of steps 701 to 704 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0148] Step 705: When there are multiple reference dimensions, determine the total number of occurrences of the multiple reference dimensions.

[0149] In this disclosure, after determining the reference dimensions and the number of times the reference dimensions occur, in the case of multiple reference dimensions, in order to improve the accuracy of determining the historical dimensions of the target object based on the number of times the reference dimensions occur, the total number of times the multiple reference dimensions occur can be determined first.

[0150] It should be noted that when there is only one reference dimension, the reference dimension can be directly determined as the historical dimension of the target object, and this disclosure does not limit this.

[0151] Step 706: Determine the average number of occurrences based on the total number of occurrences and the number of reference dimensions.

[0152] It should be noted that the number of reference dimensions can be any number, and this disclosure does not limit this number.

[0153] In this disclosure, after determining the total number of occurrences of multiple reference dimensions, the average number of occurrences can also be determined based on the total number of occurrences and the number of reference dimensions.

[0154] It should be noted that the average number of occurrences may vary depending on the total number of occurrences and the number of reference dimensions, and this disclosure does not impose any limitations on this.

[0155] Step 707: Determine the standard deviation based on the mean frequency of occurrence and the number of reference dimensions.

[0156] In this disclosure, after determining the mean frequency of occurrence, the standard deviation can be determined based on the mean frequency of occurrence and the number of reference dimensions, thereby providing conditions for improving the accuracy of determining the historical dimensions of the target object.

[0157] Step 708: Determine the target occurrence frequency based on the mean and standard deviation of the occurrence frequency.

[0158] In this disclosure, after determining the mean and standard deviation of the occurrences, the frequency of target occurrences can be determined based on the 3σ criterion, along with the mean and standard deviation of the occurrences. For example, if the standard deviation is σ and the mean of occurrences is μ, then based on the 3σ criterion, the frequency of occurrences within the interval (μ-3σ, μ+3σ) can be determined as the frequency of target occurrences, thus providing a data basis for improving the determination of the historical dimensions of the target object. This disclosure does not limit this aspect.

[0159] It should be noted that the mean and standard deviation of the occurrence frequency may differ, and the number of occurrences of the target may vary. This disclosure does not impose any restrictions on this.

[0160] Step 709: Determine the historical size of the target object based on the reference size associated with the number of times the target appears.

[0161] In this disclosure, after determining the number of times a target appears, the historical size of the target object can be determined based on the reference size associated with the number of times the target appears, thereby improving the accuracy of the determined historical size of the target object.

[0162] In some possible implementations, after determining the number of times the target appears, and when there are multiple reference dimensions associated with the number of times the target appears, in order to improve the accuracy of the determined historical dimensions of the target object, the average of the reference dimensions associated with the number of times the target appears can be determined as the historical dimensions of the target object.

[0163] In some possible implementations, after determining the number of times the target appears, if there are multiple reference dimensions associated with the number of times the target appears, the maximum and minimum values ​​among the reference dimensions associated with the number of times the target appears can be determined. Then, other reference dimensions associated with the number of times the target appears, excluding the maximum and minimum values, can be determined. The average of the other reference dimensions or the reference dimension with the highest number of occurrences can then be determined as the historical dimension of the target object, thereby improving the accuracy of determining the historical dimension of the target object.

[0164] Step 710: Determine the current position of the target object based on historical dimensions and target boundary points.

[0165] The specific implementation of step 710 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0166] In this embodiment of the present disclosure, firstly, a first image captured by a fisheye camera at the current moment is acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, an image sequence captured in a first continuous time period before the current moment is acquired, and the image sequence is identified to determine the reference size of the target object and the number of times the reference size appears. If there are multiple reference sizes, the total number of times the multiple reference sizes appear is determined. Then, based on the total number of times the appearances and the number of reference sizes, the mean number of appearances is determined. Then, based on the mean number of appearances and the number of reference sizes, the standard deviation is determined. Based on the mean number of appearances and the standard deviation, the number of times the target appears is determined. Finally, based on the reference size associated with the number of times the target appears, the historical size of the target object is determined. Based on the historical size and the target boundary points, the current position of the target object is determined. Therefore, by using images captured by a fisheye camera at the current moment, the target boundary points of the target object are determined. Based on the image sequence of continuous time periods before the current moment, the reference size of the target object is determined. When there are multiple reference sizes, the mean occurrence frequency is determined based on the total number of occurrences of multiple reference sizes and the number of reference sizes. Based on the mean occurrence frequency and standard deviation, the target occurrence frequency is determined. Based on the reference sizes associated with the target occurrence frequency, the historical size of the target object is determined. Finally, based on the historical size and the target boundary points, the position of the target object is determined, thereby improving the accuracy of determining the position of the target object.

[0167] Figure 8 This is a flowchart illustrating a method for determining the position of an object according to an embodiment of the present disclosure.

[0168] like Figure 8 As shown, the method for determining the position of an object may include the following steps:

[0169] Step 801: Obtain the first image captured by the fisheye camera at the current moment.

[0170] Step 802: Identify the first image and determine the target boundary points of the target objects contained in the first image.

[0171] Step 803: Obtain the historical dimensions of the target object.

[0172] The specific implementation of steps 801 to 803 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0173] Step 804: Determine the relative position of the target boundary point in the target object in the first image.

[0174] It should be noted that the relative position of the target boundary point within the target object can be any position, and this disclosure does not impose any restrictions on it.

[0175] In this disclosure, after determining the target boundary point of the target object in the first image, in order to improve the accuracy of determining the current position and bounding box of the target object, the relative position of the target boundary point in the target object can be determined first.

[0176] It should be noted that, in order to improve the accuracy of the current position of the determined target object, at least two target boundary points can be determined. The following example of determining two target boundary points further illustrates the solution provided in this disclosure. After determining the two target boundary points, the relative positions of these two target boundary points within the target object can be determined first. The relative positions of these two target boundary points within the target object can be arbitrary, and the relative positions of these two target boundary points within the target object are not identical; this disclosure does not impose any limitations on this.

[0177] Step 805: Starting from the target boundary point, extend the target object along the relative position by the length of the historical dimension to obtain the current position and bounding box of the target object.

[0178] In this disclosure, after determining the relative position of the target boundary point in the target object, the target object can be extended along the historical dimension from the target boundary point as the starting point to obtain the current position and bounding box of the target object, thereby improving the accuracy of determining the current position and bounding box of the target object.

[0179] It should be noted that after determining two target boundary points, the target object can be extended along the relative position of these two target boundary points to obtain the current position and bounding box of the target object, based on the relative position of these two target boundary points. This improves the accuracy of the determined position and bounding box of the target object. This disclosure does not limit this.

[0180] The following is combined Figure 9 Taking the determination of two target boundary points as an example, the method for determining the current position and bounding box of a target object provided in this embodiment will be further explained. Figure 9 This is a schematic diagram illustrating the determination of the position and bounding box of a target object according to an embodiment of this disclosure.

[0181] like Figure 9 As shown, A and B are two target boundary points of the target object, respectively; bounding box 1 is the bounding box formed by the historical dimensions of the target object; bounding box 2 is the bounding box formed by the dimensions of the target object captured by the fisheye camera; and bounding box 3 is the bounding box determined based on the two target boundary points and the historical dimensions.

[0182] Typically, in roadside perception multi-camera fusion technology, due to the influence of the fisheye camera shooting mode, the position and bounding box size of the target object in the image acquired by the fisheye camera may deviate from the actual position and bounding box size of the target object, resulting in low accuracy of the acquired target object position and bounding box.

[0183] like Figure 9 As shown, taking the example of a target object being truncated in an image captured by a fisheye camera, the truncation of the target object in the captured image causes a deviation between the determined current position and bounding box of the target object and its true position and bounding box, resulting in the following... Figure 9 The bounding box 2 shown is used to improve the accuracy of determining the position and bounding box of a target object. Based on the method for determining the object position provided in this embodiment, the current position and bounding box of the target object are determined by using two target boundary points A and B of the target object in the image acquired by the fisheye camera, as well as the historical dimensions of the target object, resulting in the following: Figure 9 The bounding box 3 shown in the figure.

[0184] Depend on Figure 9 It can be seen that by using the method for determining the position of an object based on the embodiments of this disclosure, the error between the current position and bounding box of the target object determined based on the image acquired by the fisheye camera and the current true position and true bounding box of the target object can be effectively reduced, thereby improving the accuracy of determining the position and bounding box of the target object.

[0185] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, the historical dimensions of the target object are acquired. Next, the relative positions of the target boundary points in the first image within the target object are determined. Finally, starting from the target boundary points, the target object is extended along the length of the historical dimensions along the relative positions to obtain the current position and bounding box of the target object. Therefore, by identifying the image captured by the fisheye camera, determining the target boundary points of the target object, and determining the current position of the target object based on the relative positions of the target boundary points within the target object and the length of the historical dimensions of the target object, the accuracy of the determined current position and bounding box of the target object is improved.

[0186] To implement the above embodiments, this disclosure also proposes a device for determining the position of an object.

[0187] Figure 10 This is a schematic diagram of the structure of the device for determining the position of an object provided in an embodiment of this disclosure.

[0188] like Figure 10As shown, the device 1000 for determining the position of an object includes: a first acquisition module 1001, an identification module 1002, a second acquisition module 1003, and a determination module 1004.

[0189] The first acquisition module 1001 is used to acquire the first image captured by the fisheye camera at the current moment;

[0190] The recognition module 1002 is used to recognize the first image and determine the target boundary points of the target object contained in the first image;

[0191] The second acquisition module 1003 is used to acquire the historical dimensions of the target object;

[0192] The determination module 1004 is used to determine the current position of the target object based on historical dimensions and target boundary points.

[0193] In one possible implementation of this disclosure, the identification module 1002 is specifically used for:

[0194] The first image is identified to determine multiple boundary points of the target object contained in the first image;

[0195] Among multiple boundary points, the boundary point that meets the requirements of the fisheye camera's setting position is determined as the target boundary point of the target object.

[0196] In one possible implementation of this disclosure, the second acquisition module 1003 is specifically used for:

[0197] Acquire the image sequence acquired during the first consecutive time period prior to the current moment;

[0198] The image sequence is identified to determine the reference size of the target object and the number of times the reference size appears;

[0199] The historical dimensions of the target object are determined based on the reference dimensions and the frequency of their occurrence.

[0200] In one possible implementation of this disclosure, the second acquisition module 1003 is further configured to:

[0201] In the case where the image sequence contains a first sequence captured by a fisheye camera and a second sequence captured by another camera, the first sequence and the second sequence are identified respectively to obtain a first object and first information of the first object contained in the first sequence, and a second object and second information of the second object contained in the second sequence.

[0202] Based on the first information and the second information, determine whether the first object and the second object are both target objects;

[0203] When both the first object and the second object are target objects, determine the reference dimensions of the target objects in the first sequence and the second sequence, and the number of times the reference dimensions appear in the first sequence and the second sequence.

[0204] In one possible implementation of this disclosure, the second acquisition module 1003 is further configured to:

[0205] Based on the type of the first object in the first information, determine the first probability ellipse associated with the first object; based on the type of the second object in the second information, determine the second probability ellipse associated with the second object.

[0206] Based on the first probability ellipse, the second probability ellipse, the first position in the first information, and the second position in the second information, determine whether the first object and the second object are both target objects.

[0207] In one possible implementation of this disclosure, the second acquisition module 1003 is further configured to:

[0208] When there are multiple reference dimensions, determine the total number of occurrences of the multiple reference dimensions;

[0209] Determine the average number of occurrences based on the total number of occurrences and the number of reference dimensions;

[0210] Determine the standard deviation based on the mean frequency of occurrence and the number of reference dimensions;

[0211] The frequency of the target occurrence is determined based on the mean and standard deviation of the occurrences.

[0212] The historical dimensions of the target object are determined based on the reference dimensions associated with the number of times the target appears.

[0213] In one possible implementation of this disclosure, the second acquisition module 1003 is further configured to:

[0214] The average of the reference dimensions associated with the number of times the target appears is used to determine the historical dimensions of the target object.

[0215] In one possible implementation of this disclosure, the second acquisition module 1003 is further configured to:

[0216] Determine the maximum and minimum values ​​in the reference dimensions associated with the number of times the target occurs;

[0217] Determine reference dimensions other than the maximum and minimum values ​​that are associated with the frequency of the target occurrence;

[0218] The average of other reference dimensions or the reference dimension that appears most frequently is determined as the historical dimension of the target object.

[0219] In one possible implementation of this disclosure, the determining module 1004 is further configured to:

[0220] Determine the relative positions of the target boundary points within the target object in the first image;

[0221] Starting from the target boundary point, extend the target object along the relative position by the length of its historical dimensions to obtain the current position and bounding box of the target object.

[0222] The functions and specific implementation principles of the modules described in this embodiment can be found in the above method embodiments, and will not be repeated here.

[0223] In this embodiment, a first image captured by a fisheye camera at the current moment is first acquired, and the first image is identified to determine the target boundary points of the target object contained in the first image. Then, the historical dimensions of the target object are acquired. Finally, based on the historical dimensions and the target boundary points, the current position of the target object is determined. Therefore, by identifying the target boundary points of the target object in the image captured by the fisheye camera at the current moment, and determining the current position of the target object based on the target boundary points and historical dimensions, the efficiency and accuracy of determining the object's position are improved.

[0224] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0225] Figure 11 A schematic block diagram of an example electronic device 1100 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0226] like Figure 11As shown, device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1102 or a computer program loaded from storage unit 1108 into random access memory (RAM) 1103. The RAM 1103 may also store various programs and data required for the operation of device 1100. The computing unit 1101, ROM 1102, and RAM 1103 are interconnected via bus 1104. Input / output (I / O) interface 1105 is also connected to bus 1104.

[0227] Multiple components in device 1100 are connected to I / O interface 1105, including: input unit 1106, such as keyboard, mouse, etc.; output unit 1107, such as various types of monitors, speakers, etc.; storage unit 1108, such as disk, optical disk, etc.; and communication unit 1109, such as network card, modem, wireless transceiver, etc. Communication unit 1109 allows device 1100 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0228] The computing unit 1101 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the various methods and processes described above, such as methods for determining the position of an object. For example, in some embodiments, the method for determining the position of an object may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1100 via ROM 1102 and / or communication unit 1109. When the computer program is loaded into RAM 1103 and executed by the computing unit 1101, one or more steps of the methods for determining the position of an object described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform a method for determining the position of an object by any other suitable means (e.g., by means of firmware).

[0229] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0230] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable device for determining the position of an object, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0231] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0232] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0233] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0234] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0235] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0236] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least, for example, three, unless otherwise explicitly specified. In the description of this disclosure, the words "if" and "suppose" as used may be interpreted as "when," "in response to a determination," or "in the circumstances."

[0237] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining the position of an object, comprising: Acquire the first image captured by the fisheye camera at the current moment; The first image is identified to determine the target boundary points of the target objects contained in the first image. Acquire the image sequence acquired during a first consecutive time period prior to the current time; The image sequence is identified to determine the reference size of the target object and the number of times the reference size appears; When there are multiple reference dimensions, determine the total number of occurrences of the multiple reference dimensions; The average number of occurrences is determined based on the total number of occurrences and the number of reference dimensions. The standard deviation is determined based on the mean frequency of occurrence and the number of reference dimensions. The target occurrence frequency is determined based on the mean and standard deviation of the occurrence frequency. The historical size of the target object is determined based on the reference size associated with the number of times the target appears; Determine the relative position of the target boundary point within the target object in the first image; Starting from the target boundary point, the target object is extended along the relative position by the length of the historical dimension to obtain the current position and bounding box of the target object.

2. The method as described in claim 1, wherein, The step of identifying the first image and determining the target boundary points of the target objects contained in the first image includes: The first image is identified to determine multiple boundary points of the target object contained in the first image; Among the plurality of boundary points, the boundary point that meets the requirements of the fisheye camera's setting position is determined as the target boundary point of the target object.

3. The method as described in claim 1, wherein, The step of identifying the image sequence and determining the reference size of the target object and the frequency of occurrence of the reference size includes: When the image sequence contains a first sequence captured by the fisheye camera and a second sequence captured by another camera, the first sequence and the second sequence are identified respectively to obtain a first object and first information of the first object contained in the first sequence, and a second object and second information of the second object contained in the second sequence. Based on the first information and the second information, determine whether the first object and the second object are both the target object; When both the first object and the second object are the target objects, determine the reference size of the target objects in the first sequence and the second sequence, and the number of times the reference size appears in the first sequence and the second sequence.

4. The method of claim 3, wherein, The step of determining whether the first object and the second object are both the target object based on the first information and the second information includes: Based on the type of the first object in the first information, determine the first probability ellipse associated with the first object; based on the type of the second object in the second information, determine the second probability ellipse associated with the second object. Based on the first probability ellipse, the second probability ellipse, the first position in the first information, and the second position in the second information, it is determined whether the first object and the second object are both the target object.

5. The method of claim 1, wherein, Determining the historical size of the target object based on a reference size associated with the number of times the target appears includes: The average of the reference dimensions associated with the number of times the target appears is determined as the historical size of the target object.

6. The method of claim 1, wherein, Determining the historical size of the target object based on a reference size associated with the number of times the target appears includes: Determine the maximum and minimum values ​​in the reference dimensions associated with the number of times the target occurs; Determine other reference dimensions besides the maximum and minimum values ​​associated with the number of times the target occurs; The average of the other reference dimensions or the reference dimension that appears most frequently is determined as the historical dimension of the target object.

7. A device for determining the position of an object, wherein, The device includes: The first acquisition module is used to acquire the first image captured by the fisheye camera at the current moment; The recognition module is used to recognize the first image and determine the target boundary points of the target objects contained in the first image. The second acquisition module is configured to acquire an image sequence collected within a first continuous time period prior to the current time; identify the image sequence to determine the reference size of the target object and the number of times the reference size appears; if there are multiple reference sizes, determine the total number of times the multiple reference sizes appear; determine the mean number of occurrences based on the total number of occurrences and the number of reference sizes; determine the standard deviation based on the mean number of occurrences and the number of reference sizes; determine the number of times the target appears based on the mean number of occurrences and the standard deviation; and determine the historical size of the target object based on the reference size associated with the number of times the target appears. The determination module is used to determine the relative position of the target boundary point in the target object in the first image; starting from the target boundary point, the target object is extended along the relative position by the length of the historical dimension to obtain the current position and bounding box of the target object.

8. An electronic device, characterized in that, include: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

9. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.

Images