Method and system for detecting objects in blind spots

Abstract

The invention relates to a method and system for detecting an object in a blind spot. The present disclosure relates to a method (200) for object detection in at least one blind spot (10) around a vehicle (20), comprising: capturing (202) image data of at least one area around the vehicle (20) at least partially covering the blind spot (10); at least partially processing (206) the obtained image data to locate at least one object in the obtained image data; at least partially converting (208) the processed image data into corresponding circumferential coordinates, the corresponding circumferential coordinates being defined with respect to a viewpoint at which the image data is obtained; transforming (210) the transformed circumferential coordinates at least partially into corresponding orthogonal coordinates, the corresponding orthogonal coordinates relating to an orthogonal viewpoint with respect to the blind spot (10); movement of at least a portion of the object relative to the blind spot (10) is detected (212) based at least on at least a portion of the transformed orthogonal coordinates. The disclosure also relates to a system (100) for object detection in a blind spot (10) around a vehicle (20).

Description

Technical Field

[0001] This disclosure relates to a method for detecting objects in a blind spot in front of a vehicle using a camera and a control unit. Furthermore, this disclosure relates to a system for detecting objects in a blind spot in front of a vehicle, the system including a camera and a control unit adapted to implement this method. Background Technology

[0002] As the global population increases, the number of vehicles on the road also increases. Furthermore, with this population growth, road congestion has increased due to traffic. As global traffic increases, so too has the number of road accidents. Various technologies are emerging to reduce the number of accidents. Significant advancements have been made in materials for bumpers and dashboards to reduce collision impact. Chassis materials have also been developed to absorb impact. Similarly, advancements have been made in the field of airbags. In addition, various developments have been undertaken in lighting technology to reduce the number of accidents.

[0003] Furthermore, there is development in near-field communication to avoid collisions by notifying the driver of the vehicle's presence. Similarly, sensors are included in vehicles to detect any other vehicles nearby or at a distance. Additionally, collision avoidance managed through inter-vehicle GPS data has been developed and investigated. Driver monitoring systems have also been developed to avoid road collisions. Furthermore, driving behavior prediction and estimation systems have been developed to avoid road collisions.

[0004] For example, US11054835B2 discloses a collision avoidance method that uses LiDAR (Light Detection and Ranging) data including one or more points and determines velocity constraints for each of those points. Similarly, CN105405320A discloses a car collision warning system using 3D reconstruction of polarized light images based on feature point extraction using the Harris operator. KR102605696B1 relates to a method for accurately estimating map-based CCTV camera pose and target coordinates, wherein the method includes: estimating the camera pose by accurately matching a map to road objects in an image captured by the CCTV camera; selecting a mapped object in the image; and estimating the coordinates of the selected mapped object in the image based on the estimated camera pose, wherein the camera pose includes information about the CCTV camera's position, pan, and tilt.

[0005] However, there is a lack of development in the field of using cameras to detect objects in front of a vehicle. Objects can be, but are not limited to, people, children, bicycles, and / or animals. More specifically, there is a lack of development in the field of detecting objects in the blind spot in front of a vehicle. The term blind spot can refer to a rectangular area that is not directly visible to the driver from their seated position. Furthermore, blind spots in front of a vehicle with a specific geometry can be fixed depending on the geometry of the vehicle. Moreover, the corresponding blind spot may also change as the geometry changes. Summary of the Invention

[0006] Therefore, there is a need to develop a cost-effective, safe, and reliable system and method for detecting objects in the blind spot in front of a vehicle.

[0007] Therefore, one object of this disclosure is to provide a method for detecting objects in the blind spot in front of a vehicle, thereby at least partially overcoming the known disadvantages of the prior art. Furthermore, an object is to develop a cost-effective and reliable system for detecting objects in the blind spot in front of a vehicle. Yet another object of this disclosure is to provide a cost-effective and safe method for detecting objects in the blind spot in front of a vehicle.

[0008] This objective is achieved by the features of claim 1. Embodiments of the method according to this disclosure are described in claims 2 to 12.

[0009] Furthermore, this disclosure provides a system comprising a camera configured to capture image data of the front of a vehicle; and a control unit disposed on the vehicle and adapted to implement the method as described above. Embodiments of the system according to this disclosure are described in claims 13 and 14.

[0010] Furthermore, this disclosure relates to a computer-implemented method for object detection in a blind spot in front of a vehicle, the method being implemented by a control unit located in a base station, wherein the control unit is adapted to implement the method as described above.

[0011] Therefore, one aspect of this disclosure relates to a method for detecting an object in a blind spot in front of a vehicle, the method comprising: capturing image data by a camera; obtaining the captured image data from the camera by a control unit configured on the vehicle; processing the obtained image data by the control unit to locate the object in the obtained image data; converting the processed image data into corresponding circumferential coordinates by the control unit, wherein the corresponding circumferential coordinates relate to a viewpoint with respect to the position of the camera; transforming the transformed circumferential coordinates into corresponding orthogonal coordinates by the control unit, wherein the corresponding orthogonal coordinates relate to an orthogonal viewpoint with respect to the blind spot; and detecting the object in the blind spot by the control unit based on the transformed orthogonal coordinates.

[0012] In one respect, the object may be at least one of a nearby vehicle, an animal, or a vulnerable road user.

[0013] On the other hand, the processing steps may also include classifying objects in the acquired image data.

[0014] According to an embodiment, the camera can be placed on top of the vehicle's windshield to capture blind spots.

[0015] Converting the processed image data into corresponding circular coordinates may correspond to converting pixel coordinates into real-world coordinates, and / or transforming the converted circular coordinates into corresponding orthogonal coordinates may correspond to perspective transformation, which may be advantageous.

[0016] On the other hand, the detection of objects in blind spots can be based on matching the transformed orthogonal coordinates with the coordinates of a predetermined region of interest in the blind spot.

[0017] In another aspect of this disclosure, a system for detecting objects in a blind spot in front of a vehicle may have a camera configured to capture image data in front of the vehicle; and a control unit configured on the vehicle for: - acquiring the captured image data from the camera; - processing the acquired image data to locate an object in the acquired image data; - converting the processed image data into corresponding circumferential coordinates; wherein optionally, the corresponding circumferential coordinates relate to a viewpoint with respect to the position of the camera; - transforming the transformed circumferential coordinates into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates relate to an orthogonal viewpoint with respect to the blind spot; and - detecting the object entering the blind spot based on the transformed orthogonal coordinates.

[0018] The object can be at least one of a nearby vehicle, an animal, or a vulnerable road user.

[0019] The control unit can also be configured to classify objects in the acquired image data.

[0020] The camera can be placed on top of the vehicle's windshield to capture blind spots.

[0021] Converting the processed image data into corresponding circular coordinates corresponds to the conversion from pixel coordinates to real-world coordinates, and / or transforming the converted circular coordinates into corresponding orthogonal coordinates corresponds to perspective transformation.

[0022] In one aspect, the detection of an object in a blind spot can be based on matching at least one of the transformed orthogonal coordinates with the coordinates of a predetermined region of interest in the blind spot.

[0023] Another aspect of this disclosure relates to a computer-implemented method for detecting objects in a blind spot in front of a vehicle, the method being implemented by a control unit located at a base station, the method steps including: - acquiring image data from a camera; - processing the acquired image data to locate the object in the acquired image data; - converting the processed image data into corresponding circular coordinates; wherein the corresponding circular coordinates relate to a viewpoint with respect to the position of the camera; - transforming the transformed circular coordinates into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates relate to an orthogonal viewpoint with respect to the blind spot; and - detecting the object in the blind spot based on the transformed orthogonal coordinates.

[0024] This disclosure also relates to a computer-implemented method for detecting objects in a blind spot in front of a vehicle, wherein the detection of objects in the blind spot can be based on matching at least one of the transformed orthogonal coordinates with the coordinates of a predetermined region of interest in the blind spot.

[0025] This disclosure also relates to a computer-implemented method for detecting objects in the blind spot in front of a vehicle, wherein converting processed image data into corresponding circular coordinates may correspond to converting pixel coordinates into real-world coordinates, and / or wherein transforming the converted circular coordinates into corresponding orthogonal coordinates may correspond to perspective transformation.

[0026] The term "circular coordinates" must be understood in the sense of the subject matter to be protected as a coordinate system having at least one, optionally two and / or three non-orthogonal coordinates, coordinate lines and / or coordinate axes, optionally polar coordinates, cylindrical coordinates, spherical coordinates and / or at least one angular coordinate.

[0027] This disclosure provides a safe, reliable, and cost-effective method and system for detecting objects in front of a vehicle. In particular, it eliminates the need for cost-intensive systems such as lidar or radar systems. Furthermore, existing cameras can be used, or additional cameras can be added to existing camera surveillance systems. Therefore, existing vehicle equipment can be at least partially utilized and / or expanded using the methods and systems of this disclosure.

[0028] It should be noted that the features individually set forth in the following description may be combined with each other in any technically advantageous manner, and other forms of this disclosure are also described. However, it should be understood that this disclosure is not limited to the precise arrangements and means shown. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate implementations of systems, apparatus, and methods consistent with this specification and, together with this specification, serve to explain the advantages and principles consistent with this disclosure. The drawings are not necessarily drawn to scale. The same numerals used in the drawings denote the same parts. However, it should be understood that the use of numerals to refer to parts in a given drawing is not intended to limit the parts to those labeled with the same numerals in another drawing. This specification further characterizes and describes this disclosure in detail, particularly in conjunction with the accompanying drawings. Attached Figure Description

[0029] Other aspects, advantages, and salient features of this disclosure will become apparent to those skilled in the art from the following detailed description of exemplary embodiments disclosed in conjunction with the accompanying drawings, in which:

[0030] Figure 1 The blind spot in front of a vehicle according to an embodiment of the present disclosure is shown.

[0031] Figure 2 A block diagram of an object detection system according to an embodiment of the present disclosure is shown.

[0032] Figure 3 A camera positioned on a vehicle according to an embodiment of the present disclosure is shown.

[0033] Figure 4 A camera positioned on a vehicle and a blind spot in front of the vehicle are shown according to an embodiment of the present disclosure.

[0034] Figure 5 An orthogonal viewpoint for the blind spot in front of a vehicle is shown according to an embodiment of the present disclosure.

[0035] Figure 6 A flowchart of method steps for object detection according to an embodiment of the present disclosure is shown.

[0036] Figure 7 A flowchart of method steps for object detection according to another embodiment of the present disclosure is shown. Detailed Implementation

[0037] The foregoing objects, features, and advantages of this disclosure will become more apparent from the following detailed description in conjunction with the accompanying drawings. However, various modifications may be applied to this disclosure, and this disclosure may have various embodiments. Specific embodiments of this disclosure illustrated in the accompanying drawings will be described in detail below.

[0038] In the following description, numerous specific details are set forth for purposes of explanation in order to provide a thorough understanding of this disclosure. However, it will be apparent to those skilled in the art that this disclosure may be practiced without these specific details. Descriptions of well-known components and processing techniques have been omitted so as not to unnecessarily obscure the embodiments herein. The examples used herein are intended only to facilitate an understanding of how the embodiments herein may be practiced, and further to enable those skilled in the art to practice the embodiments herein. Therefore, these examples should not be construed as limiting the scope of the embodiments herein.

[0039] In this specification, references to "an embodiment" or "embodiment" mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment of this disclosure. The phrase "in an embodiment" appearing in various places throughout the specification does not necessarily refer to all of the same embodiment, nor is it a single or alternative embodiment mutually exclusive with other embodiments. Furthermore, various features that may be exhibited by some embodiments but not others are described. Similarly, various requirements are described that may be required by some embodiments but not by others.

[0040] Furthermore, although the following description contains numerous details for illustrative purposes, any person skilled in the art will understand that many variations and / or modifications to said details are within the scope of this disclosure. Similarly, although many features of this disclosure are described by way of each other or in combination with each other, those skilled in the art will understand that many of these features may be provided independently of the others. Therefore, this description of the disclosure is set forth without losing its generality or imposing any limitation on it.

[0041] In the accompanying drawings, the thickness of layers and regions may be exaggerated for clarity. When referring to an element or layer "on" or "above" another element or layer, this includes cases where the other layer or element is inserted between them and cases where the element or layer is directly above the other element or layer. In principle, throughout the specification, reference numerals denote elements. In the following description, the same reference numerals are used to denote elements having the same function within the same concept shown in the drawings of each embodiment of this disclosure.

[0042] Detailed descriptions of known functions or configurations relevant to this disclosure will be omitted where such descriptions unnecessarily obscure the essential points of this disclosure. Furthermore, the numbers used in the description herein (e.g., first, second, etc.) are merely identifiers used to distinguish one element from another.

[0043] In addition, for ease of writing instructions only, the terms “module” and “unit” used to refer to components in the following description are given or used in combination, and these terms do not have different meanings or functions in themselves.

[0044] Furthermore, the use of singular terms, such as “a,” should not be construed as limiting the number of components or the details of a particular component. Additionally, various terms and / or phrases describing or indicating positional or directional references, such as, but not limited to, “top,” “bottom,” “front,” “rear,” “forward,” “backward,” “end,” “external,” “internal,” “left,” “right,” “vertical,” “horizontal,” etc., may refer to one or more specific parts generally visible from a user’s advantageous position during use or operation, and such terms and / or phrases should not be construed as restrictive but merely as a representative basis for describing this disclosure to those skilled in the art. Furthermore, the suffixes “area,” “part,” and “unit” used for components in the following description are given or combined only for ease of drafting the specification and do not have a distinguishing meaning or function from each other.

[0045] As used below, the term vehicle 20 refers to any internal combustion engine or battery-powered motor vehicle, driven by a driver, with or without a trailer. For example, vehicle 20 can be a car, truck, tractor, bus, motorcycle, or trailer.

[0046] Figure 1 A top view of a vehicle with blind spot 10 according to an embodiment of the present disclosure is shown. Blind spot 10 may refer to a rectangular area in front of the vehicle 20 that cannot be directly seen by the driver in the passenger and / or driving position. Figure 2 As shown, according to an embodiment of this disclosure, a block diagram of a system 100 for detecting objects in a blind spot 10 in front of a vehicle 20 is illustrated. Objects may be, but are not limited to, nearby vehicles, animals, and vulnerable road users. Objects may be moving or stationary. Vulnerable road users may be defined according to at least one of the United Nations standards, the EU Intelligent Transport Systems Directive, or any similar government transport standards.

[0047] System 100 includes a camera 102 and a control unit 104. Camera 102 can be a conventionally known image capturing device, such as a fisheye camera. Camera 102 can be as follows: Figures 3 to 5 As shown, the camera 102 is located at the top of the windshield of vehicle 20 to at least partially or completely capture blind spot 10. In one embodiment, the camera 102 can be as follows: Figure 3 As shown, it is located at the top center of the windshield. In another embodiment, two cameras 102 are located at the top right and top left of the windshield, respectively. Furthermore, as... Figure 3 , 4 As shown in Figure 5, camera 102 can be aligned with or parallel to the chassis of vehicle 20. Figure 4 and 5 As shown, camera 102 can be configured to capture images of the blind spot 10 in front of vehicle 20. Figure 4 and 5 As shown, camera 102 can capture images of objects that intrude into blind spot 10. Camera 102 can capture images at predetermined time intervals or continuously. Camera 102 can wirelessly or via a wired connection known to those skilled in the art to transmit the captured image data to control unit 104.

[0048] Control unit 104 can be configured to acquire captured image data from camera 102. Control unit 104 can be a processor or an electronic control unit. In one embodiment, control unit 104 can be an electronic control unit 104 of vehicle 20. In one embodiment, control unit 104 can be located within vehicle 20. In another embodiment, control unit 104 can be located at a base station. The base station can be a centralized location configured to receive data from multiple vehicles 20, particularly receiving data wirelessly.

[0049] The control unit 104 can access pre-stored image data. In another embodiment, the control unit 104 may have a data storage unit to store the pre-stored image data. The pre-stored image data includes information about objects having various visual attributes (e.g., shape, size, and color) to identify patterns such as pedestrians, animals, vehicles, walls, trees, etc. The pre-stored image data can be used by artificial intelligence methods to identify patterns of pre-stored objects. The control unit 104 may also have stored coordinates of a predetermined region of interest in the blind spot 10. In one case, the predetermined region of interest may be part or all of the blind spot 10.

[0050] Furthermore, the control unit 104 can be configured to process the acquired image data. Processing the acquired image data may include classifying objects in the acquired image data. Classification may include categorizing objects such as people, animals, and vehicles, and / or may include categorizing them by hazard level, such as high hazard or low hazard, or may include categorizing them by the probability of a potential collision with an object, such as high or low probability.

[0051] Furthermore, processing of the acquired image data includes locating objects within the acquired image data. Object localization provides the pixel coordinates of the object within the acquired image data. Control unit 104 can be configured to perform object classification and localization in the acquired image data using artificial intelligence methods such as deep learning methods and / or neural network models, where the neural network model involves multiple convolutional layers. In one embodiment, the multiple convolutional layers may have a YOLO backbone, a RESNET head, segmentation, and semantic segmentation leading to classification and localization. Deep learning methods and neural networks are specifically used to find projection points of one or more objects, such as people, vehicles, cyclists, animals, walls, trees, etc., onto the ground. For example, the projection points of the left and right feet on the ground or the front and rear wheels of a bicycle are local. For example, by using projection points from a top-view or bird's-eye view, calibration can be used to assign pixels or groups of pixels to one or more locations on the ground. The location of an object can be obtained by using one or more, or even all, of the detected projection points of an object connected to the ground.

[0052] In addition to projection points, deep learning methods and / or neural networks use additional points (e.g., frames to create human body points or vehicle shape points) and / or features such as size, shape, color, etc., to classify objects as, for example, humans, vehicles, or trees, and compute the true and complete object shape and position relative to vehicle 20. Using projection points and / or additional features, objects can be tracked, optionally by using cosine similarity of feature vectors. Using the tracking history, trajectory extrapolation can be performed to predict possible collisions with the front of vehicle 20.

[0053] Furthermore, the control unit 104 can be configured to convert the acquired image data into corresponding circumferential coordinates, wherein the corresponding circumferential coordinates relate to the viewpoint with respect to the camera's position. In one embodiment, the circumferential coordinates may be the coordinates of the outer edge of an object (e.g., ...). Figure 4 (As shown). In one embodiment, the object is a person, and the circular coordinates can be the coordinates of a limb. Converting the processed image data into corresponding circular coordinates corresponds to a conversion from pixel coordinates to real-world coordinates.

[0054] Furthermore, the control unit 104 can be configured to transform the converted circumferential coordinates into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates relate to orthogonal viewpoints with respect to the blind spot 10, such as... Figure 5 As shown. For example, as Figure 5As shown, an object can be represented as a dot from an orthogonal viewpoint directly above blind spot 10. The transformation from the converted circular coordinates to the corresponding orthogonal coordinates can correspond to a perspective transformation, where the transformed coordinates can correspond to real-world 2D or 3D coordinates with an origin relating to, for example, the position of the camera or a specific point on vehicle 20 or a specific point in the surrounding environment of vehicle 20, such as a specific point on the ground. Furthermore, if at least one of the transformed orthogonal coordinates matches the coordinates of a predetermined region of interest in blind spot 10, control unit 104 can be configured to detect an object entering blind spot 10.

[0055] Furthermore, after detection, the control unit 104 can be configured to notify the driver of the vehicle 20. The notification to the driver of the vehicle 20 can be, but is not limited to, visual, audio, and / or tactile notifications. In another embodiment, the control unit 104 can send signals and / or even control at least one driver assistance system and / or the function of the vehicle 20, such as braking or acceleration.

[0056] also, Figure 6 A flowchart of a method 200 for detecting objects in a blind spot in front of a vehicle 20 according to an embodiment is shown. Method 200 begins with a start step 201. In a step 202 of capturing image data, a camera 102 captures image data. In a step 204 of obtaining the captured image data, a control unit 104 configured on the vehicle 20 and / or at a base station receives the received image data from the camera 102. Furthermore, in a step 206 of processing the acquired image data, the control unit 104 processes the acquired image data to locate objects in the acquired image data. In a step 208 of converting the processed image data, the control unit 104 converts the processed image data into corresponding circumferential coordinates, wherein the corresponding circumferential coordinates may relate to a viewpoint with respect to the position of the camera 102. Furthermore, in a step 210 of transforming the converted circumferential coordinates, the control unit 104 transforms the converted circumferential coordinates into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates may relate to an orthogonal viewpoint with respect to the blind spot. Step 210 leads to the step of detecting an object in blind spot 212, wherein the control unit 104 detects an object in blind spot 10 when at least one of the transformed orthogonal coordinates matches the coordinates of a predetermined region of interest in blind spot 10. Method 200 ends at ending step 214.

[0057] Similarly, Figure 7A flowchart of a method 300 for detecting objects in a blind spot in front of a vehicle 20 according to another embodiment is shown. Method 300 begins with a start step 301. In the step of acquiring image data 302, a control unit 104 located at a base station acquires image data from a camera 102. The base station may be a centralized location configured to receive data from multiple vehicles 20. In an embodiment, the base station may be located in a cloud location. Furthermore, in the step of processing the acquired image data 304, the control unit 104 processes the acquired image data to locate objects in the acquired image data. In the step of converting the processed image data 306, the control unit 104 converts the processed image data into corresponding circular coordinates, wherein the corresponding circular coordinates may be related to a viewpoint about the position of the camera 102. Furthermore, in the step of transforming the converted circular coordinates 308, the control unit 104 transforms the converted circular coordinates into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates may be related to an orthogonal viewpoint about the blind spot 10. Step 308 leads to step 310, which detects an object in the blind spot, wherein the control unit 104 detects an object in the blind spot 10 when at least one of the transformed orthogonal coordinates matches the coordinates of a predetermined region of interest in the blind spot 10. Method 300 ends at step 312.

[0058] Therefore, this disclosure provides a system for detecting objects in the blind spot in front of a vehicle 20. Furthermore, this disclosure provides a method for detecting objects in the blind spot in front of a vehicle 20.

[0059] Although the subject matter of this disclosure has been described in language specific to structural features and / or actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are disclosed as examples of implementing the claims, and other equivalent features and actions are intended to fall within the scope of the claims; that is, the features disclosed in the foregoing description, claims, and drawings may be necessary individually or in any combination to implement this disclosure in various embodiments. The embodiments shown herein are merely examples of this disclosure and should not be construed as limiting. Alternative embodiments that may be considered by those skilled in the art are also covered by the scope of this disclosure.

[0060] Reference tag list

[0061] 10 blind spots

[0062] 20 vehicles

[0063] 100 System

[0064] 102 camera

[0065] 104 control unit

[0066] Methods for detecting 200 objects

[0067] The beginning of the 201 method

[0068] Steps to capture image data (202)

[0069] 204 Steps to obtain captured image data

[0070] 206 Steps for processing the acquired image data

[0071] Steps for converting image data after 208 transformation

[0072] Steps for transforming circumferential coordinates using the 210 transformation

[0073] 212 Steps for detecting objects in blind spots

[0074] End of Method 214

[0075] Object detection method implemented by 300 computers

[0076] The start of the 301 method

[0077] 302 Steps to obtain image data

[0078] 304 Steps for processing the obtained image data

[0079] Steps for converting image data after 306 transformation

[0080] Steps for transforming circumferential coordinates using the 308 transformation

[0081] 310 Steps for detecting objects in blind spots

[0082] 312 This method ends

Claims

1. A method (200) for detecting objects in at least one blind spot (10) around a vehicle (20), the method comprising: Capture (202) image data of at least one area in the surrounding environment of the vehicle (20) that at least partially covers the blind spot (10); The obtained image data is processed at least partially (206) to locate at least one object in the obtained image data; The processed image data is at least partially converted (208) into corresponding circumferential coordinates, wherein the corresponding circumferential coordinates are defined relative to the viewpoint from which the image data is obtained; The transformed circumferential coordinates are at least partially transformed (210) into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates relate to orthogonal viewpoints with respect to the blind spot (10); and (212) The movement of at least a portion of the object relative to the blind spot (10) is detected based at least a portion of the transformed orthogonal coordinates.

2. The method (200) according to claim 1, wherein (i) The blind spot may optionally be located in front of, to the side and / or behind the vehicle relative to the main direction of travel and / or the position and / or orientation of the cab and / or the driver's seat; (ii) The image data is captured by at least one camera (102); (iii) The processing, conversion, transformation and / or detection are performed at least in part by at least one control unit (104), which is optionally arranged at least in part in the vehicle (20) or at at least one base station, wherein optionally, the image data is obtained by the control unit (104) from the camera (102); (iv) Define circumferential coordinates relative to the viewpoint of the position of the camera (102), and / or define the origin of the coordinates by the viewpoint, in particular the viewpoint of the camera (102); (v) The object is at least one of the following: a nearby vehicle, person, animal, cyclist, pedestrian, wall, building, obstacle, plant, tree, and / or vulnerable road user; and / or (vi) Movement of at least the portion of the object includes at least part of entering and / or leaving the blind spot (10).

3. The method (200) according to claim 1 or 2, wherein, The processing (206) step further includes classifying the objects in the acquired image data, wherein, optionally, the classification includes categorizing the objects, optionally people, animals, cyclists, walls, obstacles, plants, trees, vulnerable road users, buildings, vehicles, pedestrians, etc., and / or includes categorizing the level of danger, optionally high danger level or low danger level, and / or includes categorizing the probability of potential collision with the object, optionally high probability or low probability.

4. The method (200) according to claim 2 or 3, wherein The control unit (104) is configured to employ artificial intelligence methods, particularly deep learning methods and / or neural network models, optionally to perform classification and / or localization of objects in the acquired image data, wherein, Optionally, the neural network model involves multiple convolutional layers, which may have a YOLO backbone, a RESNET head, segmentation, and semantic segmentation that lead to classification and localization.

5. The method (200) according to any one of claims 2 to 4, wherein The control unit (104), particularly when using artificial intelligence methods, is configured to locate at least one projection point on the ground of one or more elements in the captured image data, said one or more elements optionally being said objects, said objects optionally being at least one person, at least one vehicle, at least one cyclist, at least one animal, at least one wall, at least one tree, at least one obstacle, at least one vulnerable road user, at least one plant, at least one building and / or the like, wherein optionally at least one calibration is used to optionally assign pixels or groups of pixels of the captured image data to one or more locations on the ground by means of a projection point from a top view or bird's-eye view.

6. The method (200) according to claim 5, wherein In addition to the projection point, particularly after the element is identified as an object, the element and / or object are specifically classified by the control unit (104) and / or artificial intelligence methods using at least one additional point characterizing the object and / or object features such as size, shape, color, etc., optionally classifying it as a person, animal, cyclist, wall, vehicle, plant, building, obstacle, vulnerable road user, plant or tree, etc., and optionally calculating the true and complete object shape and position relative to the vehicle (20).

7. The method (200) according to any one of the preceding claims, wherein, The camera (102) is placed on top of the windshield or rear windshield of the vehicle (20) and / or on the A-pillar and / or the rearview device of the vehicle (20), and / or optionally adapted to capture the blind spot (10).

8. The method (200) according to any one of the preceding claims, wherein The step of converting the processed image data (208) to the corresponding circumferential coordinates corresponds to the conversion from pixel coordinates to real-world coordinates, and / or The transformation (210) step corresponds to a perspective transformation, which optionally provides 2D or 3D real-world coordinates of the recording area, which optionally has a coordinate origin related to the position of the camera or a specific point of the vehicle (20) or a specific point around the vehicle (20), optionally the specific point around the vehicle (20) is a specific point on the ground.

9. The method (200) according to any one of the preceding claims, wherein, The control unit (104) has pre-stored coordinates of a predetermined region of interest, particularly the coordinates of the predetermined region of interest in the blind spot (10), wherein, optionally, the predetermined region of interest is part or all of the blind spot (10), wherein, optionally, the detection (212) step is based on matching at least one of the transformed orthogonal coordinates with the coordinates of the predetermined region of interest in the blind spot (10) to detect the object entering the blind spot (10).

10. The method (200) according to any one of the preceding claims, wherein After an object is detected, the control unit (104) is configured to notify the driver or passenger of the vehicle (20), wherein optionally, the notification is a visual notification, an audio notification and / or a tactile notification, and / or the control unit (104) sends a signal and / or even controls at least one driving assistance system and / or a function of the vehicle (20) such as braking or acceleration.

11. The method (200) according to any one of the preceding claims, wherein The control unit (104) has access to pre-stored image data, wherein optionally, the control unit (104) has a data storage unit to store the pre-stored image data, wherein optionally, the pre-stored image data includes information about objects having varying visual attributes such as shape, size, and color, particularly object features, wherein attributes are used to identify patterns of elements and / or objects, wherein optionally, patterns of pedestrians, plants, vulnerable road users, people, buildings, obstacles, animals, vehicles, cyclists, walls, trees, etc., wherein optionally, the pre-stored image data is used by the control unit (104) and / or artificial intelligence methods to identify patterns of pre-stored objects in captured image data, wherein optionally, the objects are tracked by using projection points and / or object features, optionally by using cosine similarity of feature vectors, wherein, By utilizing the tracking history, the trajectory extrapolation is derived to predict possible collisions with the vehicle (20), particularly with the front of the vehicle (20).

12. The method (200) according to any one of the preceding claims, wherein Method (200) is a computer-implemented method having a control unit (104) arranged at a base station, wherein optionally, the base station is a centralized location configured to specifically wirelessly receive data from multiple vehicles (20) and / or the base station is located in a cloud location.

13. A system (100) for object detection in a blind spot (10) around a vehicle (20), the system (100) comprising: A camera (102) is configured to capture image data of the environment surrounding the vehicle (20); as well as The control unit (104) is disposed on and / or included in the vehicle (20) and is adapted to implement the method (200) according to any one of the preceding claims.

14. The system (100) according to claim 11, wherein, The control unit (104) is configured for - Obtain the captured image data from the camera (102), - At least partially process the acquired image data to locate at least one object in the acquired image data. - The processed image data is at least partially converted into corresponding circular coordinates; wherein the corresponding circular coordinates relate to a viewpoint with respect to the position of the camera (102). - Transform the transformed circumferential coordinates at least partially into corresponding orthogonal coordinates, wherein the corresponding orthogonal coordinates relate to orthogonal viewpoints with respect to the blind spot (10), and / or - Detect at least a portion of the object entering the blind spot (10) based at least a portion of the transformed orthogonal coordinates.

Images