Workshop distance measurement method, workshop distance measurement device, electronic device, computer program, and computer-readable recording medium

Abstract

Disclosed is a method for measuring a distance between vehicles, a device for measuring a distance between vehicles, an electronic device, a computer program, and a computer-readable recording medium. A method for measuring a distance between vehicles is disclosed. The present invention includes: a step of acquiring a travel image captured by a camera of a first vehicle; a step of detecting a second vehicle from the acquired travel image; a step of, if the second vehicle is not detected from the travel image, detecting a first feature point of a second vehicle region from a first frame corresponding to a frame in which the second vehicle is detected before a frame in which the second vehicle is not detected, among a plurality of frames constituting the travel image; a step of tracking the detected first feature point and detecting a second feature point in a second frame corresponding to a current frame; a step of calculating a feature point change value between the first feature point and the second feature point; and a step of calculating a distance from the camera of the first vehicle to the second vehicle based on the calculated feature point change value.

Description

[0001] This application is a divisional application of the invention patent application filed on January 20, 2021, with application number "202110076340.2" and entitled "Workshop Distance Measurement Method, Workshop Distance Measurement Device, Electronic Equipment, Computer Program and Computer-Readable Recording Medium". Technical Field

[0002] This invention relates to a method, apparatus, electronic device, computer program, and computer-readable recording medium for measuring inter-vehicle distances based on vehicle images. More specifically, it relates to a method, apparatus, electronic device, computer program, and computer-readable recording medium for measuring the distance between vehicles located close to each other by tracking feature points in vehicle images. Background Technology

[0003] The most important aspects of vehicle operation are safe driving and prevention of traffic accidents. To this end, vehicles are equipped with various auxiliary devices for controlling vehicle posture and the functions of vehicle structural components, as well as safety devices such as seat belts and airbags.

[0004] Furthermore, there is a recent trend towards installing devices such as black boxes in vehicles to store driving images and data transmitted from various sensors, enabling the identification of the cause of accidents. Black boxes or navigation applications can also be installed on portable devices such as smartphones and tablets, thus enabling their use as vehicle-mounted devices as described above.

[0005] Therefore, advanced driver assistance systems (ADAS) that utilize driving images captured while the vehicle is in motion have recently been developed and are being widely adopted to assist drivers in driving, thereby seeking to improve both safe driving and driver convenience.

[0006] Among the functions provided by the ADAS, the Forward Collision Warning System (FCWS) detects vehicles ahead of the vehicle in the captured driving images, measures the distance between the vehicle and the detected vehicles, and then issues a warning to the driver that there is a risk of collision based on the distance.

[0007] In other words, for FCWS (Forward-Looking Wire) detection, it is necessary to detect vehicles ahead. Existing technologies employ image processing methods that utilize the shadows of vehicles ahead within captured driving images, or machine learning methods that learn from numerous vehicle images for detection. Both of these methods achieve higher detection performance when the driving image contains a vehicle region including the lower portion of the vehicle.

[0008] However, vehicles encounter various driving environments while in motion, and there may be situations where the lower part of the vehicle is not captured in the driving image. For example, when a vehicle detects a vehicle ahead and slows down, the distance between the two vehicles will decrease. Figure 1 As shown, when the distance between vehicle 1 and the vehicle 2 in front is large (2-1), the image of the vehicle, including the lower part of the vehicle in front, may be included in the driving image captured by vehicle 1. However, when the distance between vehicle 1 and the vehicle 2 in front becomes smaller (for example, within 10m) (2-2), the lower part of the vehicle in front will not be included in the driving image.

[0009] As mentioned above, when a vehicle is in motion and the distance to the vehicle in front decreases while the lower part of the vehicle in front is not included in the driving image, there is a problem that the vehicle in front cannot be accurately detected by existing shadow-based or learning-based methods.

[0010] On the one hand, the vehicle-to-vehicle distance measurement technology is a core technology for autonomous driving that has been the subject of recent active discussion. If the vehicle in front is not accurately detected and the vehicle-to-vehicle distance cannot be measured in autonomous driving, it may be directly related to accidents. Therefore, the importance of vehicle-to-vehicle distance measurement technology is gradually increasing. Summary of the Invention

[0011] This invention is proposed to solve the above-mentioned problems. The purpose of this invention is to provide a vehicle image-based inter-vehicle distance measurement method, inter-vehicle distance measurement device, electronic device, computer program, and computer-readable recording medium that can track the object vehicle by tracking feature points and measure the distance between the object vehicle and the vehicle itself even when the lower part of the object vehicle is not captured as the distance between the vehicle and the object vehicle (the vehicle in front or behind) approaches, resulting in the object vehicle not being detected by driving images.

[0012] Furthermore, the present invention aims to provide a vehicle image-based vehicle distance measurement method, vehicle distance measurement device, electronic device, computer program, and computer-readable recording medium that provides prompts based on vehicle distance using measured distances.

[0013] In addition, the present invention aims to provide a vehicle image-based inter-vehicle distance measurement method, inter-vehicle distance measurement device, electronic device, computer program, and computer-readable recording medium that generate autonomous driving control signals for their own vehicles using measured distances.

[0014] To achieve the above objective, a processor-based inter-vehicle distance measurement method according to an embodiment of the present invention includes the following steps: acquiring a driving image captured by a camera of a first vehicle in motion; detecting a second vehicle from the acquired driving image; if the second vehicle is not detected from the driving image, detecting a first feature point of a region of the second vehicle from a first frame corresponding to a frame in which the second vehicle was detected before the frame in which the second vehicle was not detected, among a plurality of frames constituting the driving image; tracking the detected first feature point and detecting a second feature point in a second frame corresponding to the current frame; calculating a feature point change value between the first feature point and the second feature point; and calculating the inter-vehicle distance from the camera of the first vehicle to the second vehicle based on the calculated feature point change value.

[0015] Then, the step of detecting the second vehicle can be done by using a learning model built from vehicle images through machine learning or deep learning.

[0016] Alternatively, the step of detecting the first feature point can be performed if the second vehicle is not detected by the constructed learning model when the distance between the first vehicle and the second vehicle is close, in which case the step of detecting the first feature point in the region of the second vehicle is executed.

[0017] Then, the step of detecting the first feature point may be to set the middle area of ​​the vehicle in the second vehicle region of the frame as the region of interest, and detect the first feature point in the set region of interest.

[0018] Alternatively, the step of detecting the second feature point may involve using the optical flow of the detected first feature point to track the second feature point, thereby detecting the second feature point within the second frame.

[0019] Then, it may also include the step of filtering out second feature points that are not displayed in the second frame and first feature points corresponding to the second feature points that are not displayed in the second frame when using the optical flow to trace the second feature points.

[0020] Additionally, the step of calculating the feature point change value may include: calculating the average pixel position of the first feature point; calculating a first average pixel distance that averages the pixel distances from the calculated average pixel position of the first feature point to each of the first feature points; calculating the average pixel position of the second feature point; calculating a second average pixel distance that averages the pixel distances from the calculated average pixel position of the second feature point to each of the second feature points; and calculating the average pixel distance ratio between the first average pixel distance and the second average pixel distance.

[0021] Then, the step of calculating the vehicle distance may include: multiplying the image width of the second vehicle in the first frame by the calculated average pixel distance ratio, and then calculating the image width of the second vehicle in the second frame.

[0022] In addition, the step of calculating the vehicle-to-vehicle distance may also include: calculating the vehicle-to-vehicle distance from the camera of the first vehicle to the second vehicle based on the image width of the second vehicle in the calculated second frame, the focal distance of the first camera device, and the predicted width of the second vehicle.

[0023] Then, the step of calculating the vehicle-to-vehicle distance may further include: calculating the ratio of the image width of the detected second vehicle to the image width of the lane in which the second vehicle is located; determining the size class of the second vehicle based on the calculated ratio; and calculating the predicted width of the second vehicle based on the determined size class of the second vehicle.

[0024] Additionally, it may include the step of generating alert data to indicate the collision hazard level corresponding to the distance difference between the first vehicle and the second vehicle when the calculated workshop distance is less than a preset distance.

[0025] Then, it may also include the step of generating control signals based on the calculated inter-vehicle distance to control the autonomous driving of the first vehicle.

[0026] On one hand, in order to achieve the above-mentioned objective, a vehicle-to-vehicle distance measuring device according to an embodiment of the present invention includes: an image acquisition unit for acquiring a driving image captured by a camera of a first vehicle in motion; a vehicle detection unit for detecting a second vehicle from the acquired driving image; a feature point detection unit for detecting a first feature point of a region of the second vehicle from a first frame corresponding to a frame in which the second vehicle was detected before the frame in which the second vehicle was not detected, tracking the detected first feature point, and detecting a second feature point in a second frame corresponding to the current frame, among a plurality of frames constituting the driving image; a feature point change value calculation unit for calculating a feature point change value between the first feature point and the second feature point; and a vehicle-to-vehicle distance calculation unit for calculating the vehicle-to-vehicle distance from the camera of the first vehicle to the second vehicle based on the calculated feature point change value.

[0027] Then, the vehicle detection unit can detect the second vehicle using a learning model built from machine learning or deep learning of vehicle images.

[0028] Additionally, when the second vehicle is not detected by the constructed learning model as the distance between the first vehicle and the second vehicle approaches, the feature point detection unit can perform the detection of the first feature point in the region of the second vehicle.

[0029] Then, the feature point detection unit can set the middle area of ​​the vehicle as the region of interest in the second vehicle region of the frame, and detect the first feature point in the set region of interest.

[0030] In addition, the feature point detection unit can use the optical flow of the detected first feature point to track the second feature point, thereby detecting the second feature point in the second frame.

[0031] Then, when the feature point detection unit uses the optical flow to trace the second feature point, it can filter out the second feature points that are not displayed in the second frame and the first feature points corresponding to the second feature points that are not displayed in the second frame.

[0032] Additionally, the feature point change value calculation unit may include: an average pixel distance calculation unit, which calculates the average pixel position of the first feature point, calculates a first average pixel distance averaging the pixel distances from the calculated average pixel position of the first feature point to each of the first feature points, calculates the average pixel position of the second feature point, and calculates a second average pixel distance averaging the pixel distances from the calculated average pixel position of the second feature point to each of the second feature points; and a ratio calculation unit, which calculates the average pixel distance ratio between the first average pixel distance and the second average pixel distance.

[0033] Then, the vehicle distance calculation unit can multiply the image width of the second vehicle in the first frame by the calculated average pixel distance ratio to calculate the image width of the second vehicle in the second frame.

[0034] In addition, the workshop distance calculation unit can calculate the distance from the camera of the first vehicle to the second vehicle based on the image width of the second vehicle in the calculated second frame, the focal distance of the first shooting device, and the predicted width of the second vehicle.

[0035] Then, the vehicle distance calculation unit can calculate the image width ratio between the image width of the detected second vehicle and the image width of the lane in which the second vehicle is located, determine the size class of the second vehicle based on the calculated ratio, and calculate the predicted width of the second vehicle based on the determined size class of the second vehicle.

[0036] Additionally, it may include: a prompt data generation unit, which generates prompt data to indicate the collision hazard level corresponding to the distance difference between the first vehicle and the second vehicle when the calculated inter-vehicle distance is less than a preset distance.

[0037] It may further include: an autonomous driving control signal generation unit, which generates control signals based on the calculated inter-vehicle distance to control the autonomous driving of the first vehicle.

[0038] On the one hand, in order to achieve the above-mentioned objective, an electronic device for providing prompts for assisting the driver based on inter-vehicle distance according to an embodiment of the present invention includes: an output unit for outputting prompt information that the driver can confirm; an image acquisition unit for acquiring a driving image captured by a shooting device; a vehicle detection unit for detecting a second vehicle from the acquired driving image; a feature point detection unit for detecting a first feature point in the second vehicle region from a first frame corresponding to a frame before the frame in which the second vehicle was detected, tracking the detected first feature point, and detecting a second feature point in a second frame corresponding to the current frame, if the second vehicle is not detected from the driving image; a feature point change value calculation unit for calculating a feature point change value between the first feature point and the second feature point; an inter-vehicle distance calculation unit for calculating the inter-vehicle distance from the shooting device of the first vehicle to the second vehicle based on the calculated feature point change value; and a control unit for controlling the output unit to output a forward vehicle collision warning or a forward vehicle departure warning according to the calculated distance.

[0039] Then, the output unit further includes: a display unit that combines the captured driving image with a prompt object to output an augmented reality image; the control unit can generate a prompt object for the forward vehicle collision warning through the display unit, and overlay the generated prompt object for the forward vehicle collision warning with the forward vehicle display area of ​​the augmented reality image.

[0040] On the one hand, in order to achieve the above objectives, according to an embodiment of the present invention, a computer-readable recording medium containing a program for performing the above-described workshop distance measurement method can be provided.

[0041] In addition, to achieve the above objectives, according to an embodiment of the present invention, a program for performing the above-described workshop distance measurement method can be provided.

[0042] According to the various embodiments of the present invention described above, even if the lower part of the target vehicle is not captured as the distance between the vehicle and the target vehicle being measured approaches, resulting in the inability to detect the target vehicle through driving images, the distance between the target vehicle and the vehicle can still be measured by tracking feature points.

[0043] Furthermore, according to various embodiments of the present invention, even at close range, low-parameter terminals can achieve fast processing speeds through feature point-based tracking, enabling real-time processing.

[0044] Furthermore, according to various embodiments of the present invention, even when the area of ​​the target vehicle is obscured as the distance between the vehicle and the target vehicle approaches, the collision warning function and the departure warning function can be executed accurately.

[0045] Furthermore, according to various embodiments of the present invention, even if the area of ​​the target vehicle is obscured as the distance between the vehicle and the target vehicle approaches, the distance between the vehicle and the target vehicle can be accurately calculated, and the automatic driving control of the vehicle can be accurately executed. Attached Figure Description

[0046] Figure 1 This is a diagram used to illustrate driving images taken based on the distance between one's own vehicle and the vehicle in front, as well as problems with existing technology.

[0047] Figure 2 This is a block diagram illustrating a workshop distance measuring device according to an embodiment of the present invention.

[0048] Figure 3 This is a block diagram that more specifically illustrates a workshop distance measuring device according to an embodiment of the present invention.

[0049] Figure 4 This is a schematic diagram illustrating the feature point detection and tracking process according to an embodiment of the present invention.

[0050] Figure 5 This is a schematic diagram illustrating the construction of a learning dataset for vehicle detection according to an embodiment of the present invention, and the process of detecting vehicles using the constructed learning dataset.

[0051] Figure 6 This is a schematic diagram illustrating the feature point change value calculation unit according to an embodiment of the present invention.

[0052] Figure 7 This is a schematic diagram illustrating a method for calculating the average position of feature points according to an embodiment of the present invention.

[0053] Figure 8 This is a block diagram showing more specifically a workshop distance calculation unit according to an embodiment of the present invention.

[0054] Figure 9 This is a schematic diagram illustrating a workshop distance measurement method according to an embodiment of the present invention.

[0055] Figure 10 This is a schematic diagram illustrating the ratio between the image width of the second vehicle and the image width of the lane in which the second vehicle is located, according to an embodiment of the present invention.

[0056] Figure 11 This is a schematic diagram illustrating the process of fitting a vehicle region to the detected left and right boundary positions of a vehicle according to an embodiment of the present invention.

[0057] Figure 12This is a conceptual diagram illustrating the process of determining the size class of a second vehicle according to an embodiment of the present invention.

[0058] Figure 13 This is a flowchart illustrating a workshop distance measurement method according to an embodiment of the present invention.

[0059] Figure 14 This is a flowchart that more specifically illustrates the step (S150) of calculating the change value of feature points according to an embodiment of the present invention.

[0060] Figure 15 The flowchart illustrates more specifically the workshop distance calculation step (S160) according to an embodiment of the present invention.

[0061] Figure 16 This is a flowchart illustrating a workshop distance calculation method according to another embodiment of the present invention.

[0062] Figure 17 This is a block diagram illustrating an electronic device according to an embodiment of the present invention.

[0063] Figure 18 This is a schematic diagram illustrating a system network connected to an electronic device according to an embodiment of the present invention.

[0064] Figure 19 as well as Figure 20 This is a schematic diagram illustrating a forward vehicle collision prevention warning screen of an electronic device according to an embodiment of the present invention.

[0065] Figure 21 This is a schematic diagram illustrating the form of an electronic device according to an embodiment of the present invention when it does not have a camera unit.

[0066] Figure 22 This is a schematic diagram illustrating the form of an electronic device having a camera unit according to an embodiment of the present invention.

[0067] Figure 23 This is a schematic diagram illustrating an embodiment of a HUD (Head-Up Display) according to an embodiment of the present invention.

[0068] Figure 24 This is a block diagram illustrating an autonomous driving system according to an embodiment of the present invention.

[0069] Figure 25 This is a block diagram illustrating the structure of an autonomous vehicle according to an embodiment of the present invention.

[0070] Figure label:

[0071] 10: Workshop distance measuring device; 11: Image acquisition unit; 12: Vehicle inspection unit; 13: Feature point detection unit; 14: Feature point change value calculation unit; 15: Workshop distance calculation unit. Detailed Implementation

[0072] The following description is merely an example of the principles of the invention. Therefore, although not explicitly stated or shown in this specification, those skilled in the art can grasp the principles of the invention and invent various devices encompassed within the concept and scope of the invention. Furthermore, it should be understood that, in principle, all terms and embodiments set forth in this specification are intended to provide a clear understanding of the concept of the invention and are not intended to limit one to the embodiments and states specifically described above.

[0073] Furthermore, it should be understood that not only the principles, viewpoints, and embodiments of the present invention, but also all detailed descriptions of specific embodiments are intended to include structural and functional equivalents of the matters described. Additionally, it should be understood that these equivalents include not only those currently known, but also those to be developed in the future, i.e., all elements invented regardless of structure to perform the same function.

[0074] Therefore, for example, the block diagrams of the present invention should be understood as conceptual views illustrating exemplary circuits that embody the principles of the present invention. Similarly, it should be understood that all flowcharts, state transition diagrams, pseudocode, etc., illustrate various processes that can actually be displayed in a computer-readable medium and are executed by a computer or processor regardless of whether a computer or processor is explicitly shown.

[0075] The functionality of the various elements shown in the accompanying drawings, which include a processor or functional blocks represented as similar concepts, can be provided not only by dedicated hardware but also, for the appropriate software, as hardware capable of executing software. When provided by a processor, the functionality can be provided by a single dedicated processor, a single shared processor, or multiple individual processors, some of which can be shared.

[0076] Furthermore, the explicit use of the term as a processor, control, or similar concept should not be construed as excluding references to hardware capable of executing software, but is unrestricted and should be understood to include, implicitly, digital signal processor (DSP) hardware, read-only memory (ROM), random access memory (RAM), and non-volatile memory for storing software. Other well-known hardware may also be included.

[0077] Within the scope of the claims of this specification, the constituent elements shown as means for performing the functions described in the specific embodiments are intended to include all methods for performing the functions of all forms of software, and to be combined with appropriate circuitry for performing said software to perform said functions, such as combinations of circuit elements for performing said functions or firmware / microcode, etc. The invention as defined by the scope of the claims combines the functions provided by the various means described, and combines them with the manner claimed in the claims; therefore, any means capable of providing said functions is equivalent to those mastered from this specification.

[0078] The aforementioned objectives, features, and advantages will become more apparent from the accompanying drawings and the following detailed description thereof, thereby enabling those skilled in the art to readily implement the technical concept of the invention. Furthermore, in describing the invention, detailed descriptions of well-known techniques related to the invention are omitted where it is determined that such detailed descriptions would obscure the essence of the invention.

[0079] Before providing a detailed description of various embodiments of the present invention, the names used in the present invention are defined as follows.

[0080] In this specification, the inter-vehicle distance can refer to the distance in real-world coordinates. Specifically, the inter-vehicle distance can refer to the distance between the first vehicle and the second vehicle, or more precisely, the distance from the camera mounted on the first vehicle to the second vehicle.

[0081] Additionally, in this specification, the width of a vehicle can refer to the width of the vehicle in real-world coordinates, or the radius of the vehicle in real-world coordinates.

[0082] Additionally, in this specification, image width may refer to the pixel width of the image on the imaging surface of the imaging element of the imaging device.

[0083] Additionally, in this specification, pixel distance may refer to the distance between pixels on the imaging surface of the imaging element of the imaging device.

[0084] Hereinafter, various embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0085] Figure 2 This is a block diagram illustrating a workshop distance measuring device according to an embodiment of the present invention. Figure 3 This is a block diagram illustrating more specifically a workshop distance measuring device according to an embodiment of the present invention. (See reference) Figure 2 as well as Figure 3According to an embodiment of the present invention, the vehicle distance measuring device 10 may include all or part of an image acquisition unit 11, a vehicle detection unit 12, a feature point detection unit 13, a feature point change value calculation unit 14, a vehicle distance calculation unit 15, a prompt data generation unit 17, a driving control data generation unit 18, and a control unit 19. Furthermore, the feature point change value calculation unit 14 may include an average pixel distance calculation unit 14-1 and an average pixel distance ratio calculation unit 14-2.

[0086] The vehicle-to-vehicle distance measuring device 10 can measure the distance between a first vehicle, which serves as the reference for distance measurement, and a second vehicle, which serves as the object of distance measurement. The first vehicle is the reference vehicle, which may also be referred to as the "reference vehicle" or "own vehicle." The second vehicle is the object of distance measurement, which may also be referred to as the "object vehicle." Furthermore, the second vehicle is a vehicle located near the first vehicle, and may include a vehicle located in front of the first vehicle and a vehicle located behind the first vehicle.

[0087] The vehicle-to-vehicle distance measuring device 10 can calculate the distance between the first vehicle and the second vehicle by controlling the activation of feature point detection and tracking functions based on whether the second vehicle is detected in the driving image captured by the camera of the first vehicle.

[0088] Specifically, a first vehicle may be traveling in a lane, and a second vehicle may first appear in front of or behind the first vehicle while the first vehicle is traveling. In this case, the vehicle-to-vehicle distance measuring device 10 can detect the second vehicle from the driving image using machine learning or deep learning. Then, the vehicle-to-vehicle distance measuring device 10 can calculate the vehicle-to-vehicle distance between the second vehicle and the first vehicle detected by machine learning or deep learning.

[0089] However, if the distance between the first and second vehicles approaches and the lower part of the second vehicle is not captured, resulting in the second vehicle not being detected from the driving image using machine learning or deep learning, the vehicle-to-vehicle distance measuring device 10 can track detected feature points from the driving image and calculate the distance between the first and second vehicles. Specifically, the vehicle-to-vehicle distance measuring device 10 can select a first frame from the multiple frames constituting the driving image that corresponds to the frame before the frame where the second vehicle was detected, detect and track the first feature point in the second vehicle area within the selected first frame, detect a second feature point in the second frame corresponding to the current frame, calculate the feature point change value between the first and second feature points, and calculate the distance from the camera capturing the first vehicle to the second vehicle based on the calculated feature point change value.

[0090] That is, if the operation of the workshop distance measuring device 10 according to the present invention is described in the order of steps, it can be as follows: the first step is that the second vehicle first appears in front of the first vehicle; the second step is that the workshop distance measuring device 10 detects the second vehicle from the driving image through machine learning or deep learning and calculates the workshop distance between the detected second vehicle and the first vehicle; the third step is that if the workshop distance measuring device 10 does not detect the second vehicle from the driving image through machine learning or deep learning, it calculates the workshop distance between the second vehicle and the first vehicle from the driving image through feature point detection and tracking.

[0091] The workshop distance measuring device 10 can be embodied in software, hardware, or a combination thereof. As an example, depending on the hardware implementation, it can be embodied by at least one of ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), processors, controllers, microcontrollers, microprocessors, or other electronic units for performing functions.

[0092] For ease of explanation, the following example will be used to illustrate the various components of the vehicle that constitutes the vehicle-to-vehicle distance measuring device 10 in more detail.

[0093] The image acquisition unit 11 can acquire driving images captured by the camera device on the first vehicle. Specifically, the image acquisition unit 11 can acquire driving images captured by the camera device installed on the first vehicle in real time during the first vehicle's movement. The acquired driving images may include multiple lanes distinguished by vehicle lanes, a road composed of multiple lanes, and multiple vehicles traveling on the road.

[0094] Lane markings can refer to the lines on both sides of the lane in which a vehicle is positioned. Additionally, a lane can refer to a road formed by lane markings, such as one lane, two lanes, ... N lanes, for vehicles to travel on.

[0095] The vehicle detection unit 12 can detect a second vehicle from driving images acquired by the image acquisition unit 11. Specifically, the vehicle detection unit 12 can perform learning on the vehicle images using machine learning or deep learning to construct a learning model for vehicle detection, and then detect the second vehicle using the constructed learning model. The constructed model is an algorithm or program for detecting vehicles from images.

[0096] Then, the learning model used for vehicle detection can learn to become a more advanced model by utilizing the output values ​​that display the vehicle detection results. As an example, when the output result is incorrect, the user can input a response to the output result, and the vehicle detection unit 12 can use the learning model for vehicle detection to learn based on the driver's response.

[0097] That is, according to the present invention, machine learning or deep learning can be performed to generate a learning model for vehicle detection, and the generated model can be used to detect vehicles from driving images. Deep learning can apply the CNN (Convolutional Neural Network) algorithm, which is one type of neural network model. In this case, deep learning can perform learning by assuming various conditions to augment the driving images. These conditions are defined as those that cause changes in the images collected for the learning of the neural network model. Specifically, various images can be displayed due to factors such as image shift, rotation, brightness changes, and blurring; therefore, these can be taken into account to augment the data.

[0098] In addition, when the vehicle detection unit 12 detects multiple vehicles in the driving image, it can select a second vehicle as the distance measurement object from the multiple detected vehicles based on the driving status information, wherein the driving status information indicates whether the first vehicle is accurately driving in a specific lane or leaving a specific lane.

[0099] As an example, when the first vehicle is traveling in a specific lane, the vehicle detection unit 12 can select a second vehicle that is in the same lane as the first vehicle from among the multiple vehicles included in the driving image, and then detect the selected second vehicle.

[0100] As another example, when the first vehicle leaves a specific lane, the vehicle detection unit 12 can select a second vehicle in the lane that the first vehicle in the lane it is facing from among the multiple vehicles included in the driving image, and then detect the selected second vehicle.

[0101] On one hand, if the vehicle detection unit 12 detects a second vehicle, the inter-vehicle distance calculation unit 15 can calculate the distance between the detected second vehicle and the first vehicle. That is, if the vehicle detection unit 12 detects a second vehicle from the driving image using a learning model, it can activate the inter-vehicle distance calculation unit 15 to calculate the distance between the detected second vehicle and the first vehicle without activating the functions of the feature point detection unit 13 and the feature point change value calculation unit 14. The inter-vehicle distance calculation unit 15 can use the inter-vehicle distance calculation algorithm described later to calculate the distance between the first vehicle and the detected second vehicle.

[0102] However, when the distance between the first vehicle and the second vehicle approaches a point below the predetermined distance and the lower part of the second vehicle is not captured, causing the vehicle detection unit 12 to fail to detect the second vehicle from the driving image through the learning model, the functions of the feature point detection unit 13 and the feature point change value calculation unit 14 can be activated to track the second vehicle by tracking the feature points, and the vehicle-to-vehicle distance calculation unit 15 can be activated to calculate the distance between the first vehicle and the second vehicle.

[0103] In response, refer to Figure 4 Please provide more specific details.

[0104] Figure 4 This is a schematic diagram illustrating the feature point detection and tracking process according to an embodiment of the present invention. (Reference) Figure 4 When the distance between the first vehicle 21 and the second vehicle 22 is more than a predetermined distance (22-1), the driving image 23-1 captured by the camera of the first vehicle 21 includes the lower part of the second vehicle. Therefore, the vehicle detection unit 12 can detect the second vehicle image 23-2 in the driving image 23-1 by using a learning model built by machine learning or deep learning.

[0105] At this point, the dataset required for vehicle detection training can be constructed by classifying a dataset of vehicle rear-end images collected according to vehicle type (Sedan, SUV, Truck, Large car) based on detection distance (near, medium, and far), thereby building training data. Then, the vehicle detection unit 12 can generate a classifier, which is created using learning-based methods (Machine Learning, Deep Learning, etc.) to learn from the constructed training data. The vehicle detection unit 12 can then use the generated classifier to detect the second vehicle image 23-2 in the driving image 23-1.

[0106] Regarding the vehicle detection action of the vehicle detection unit 12, refer to Figure 5 Please provide more specific details.

[0107] Figure 5 This is a schematic diagram illustrating the process of constructing a learning dataset for vehicle detection and using the constructed learning dataset to detect vehicles.

[0108] refer to Figure 5 The process can begin by constructing a vehicle learning dataset (S1000), selectively learning from the constructed dataset (S1500), and then generating a classifier for vehicle classification (S1700). Then, by inputting images through camera 1700 (S1950), the vehicle detection learning system according to an embodiment of the present invention can detect vehicles using the generated classifier (S1900).

[0109] The specific steps to be performed are shown on the right side of the attached diagram.

[0110] First, the specific steps for constructing the vehicle learning dataset (S10000) are explained below. According to an embodiment of the present invention, the learning system for vehicle detection acquires the image to be learned (S1010), harvests the vehicle regions to be learned from the learned image (S1030), and annotates the attributes of the vehicles contained in the harvested vehicle regions. At this time, the annotated vehicle attributes can be vehicle type, distance of vehicles within the image, etc. Then, according to an embodiment of the present invention, the learning system for vehicle detection can use the harvested vehicle images and their attributes to generate a learning dataset (S1070).

[0111] The specific steps of the selective dataset learning step (S1500) are explained as follows. The learning system for vehicle detection according to an embodiment of the present invention can extract features from the constructed dataset (S1510). At this time, the feature extraction method can use (i) grayscale intensity, (ii) RGB color information (Red, Green, Blue (RGB) color information), (iii) HSV color information (Hue, Saturation, Value (HSV) color information), (iv) YIQ color information, and (v) edge information (grayscale, binary, eroded binary) etc. The learning system for vehicle detection according to an embodiment of the present invention classifies vehicles using the extracted features (S1530), and after strengthening the vehicle classification process through learning (S1550), a classifier for classifying vehicles can be generated (S1700).

[0112] Finally, the vehicle detection step (S1900) is described in detail below. According to an embodiment of the present invention, the learning system for vehicle detection can extract features (S1920) from the image input via camera 1700 (S1950), use a classifier on the extracted features to detect vehicles (S1930), and output the detection result (S1970).

[0113] On the one hand, if the second vehicle image 23-2 is detected in the vehicle detection unit 12, the vehicle distance calculation unit 15 can calculate the distance between the detected second vehicle 22 and the first vehicle 21.

[0114] However, vehicles encounter various driving environments while driving. When the distance between the first vehicle 21 and the second vehicle 22 gets closer (for example, when they get within 10m), the driving image captured by the camera of the first vehicle 21 does not include the lower part of the second vehicle. Therefore, the vehicle detection unit 12 may not be able to detect the image of the second vehicle through the learning model built by machine learning or deep learning.

[0115] As described above, when the vehicle detection unit 12 does not detect a second vehicle image from the driving image, the feature point detection unit 13 can select a first frame 24-1 from the multiple frames constituting the driving image that corresponds to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected. Then, the feature point detection unit 13 can detect feature points using the selected first frame 24-1. In this case, the feature point detection unit 13 can detect feature points using the first frame 24-1 where the region of interest has not been processed, or it can detect feature points using the first frame 24-2 where the region of interest has been processed. As an example, Figure 4 As shown, the detection unit 13 can set the second vehicle region as the region of interest in the first frame 24-1, generate a first frame 24-2 for processing the region of interest, and detect a first feature point 24-3 in the first frame 24-1 for processing the region of interest. Then, although recorded as... Figure 4 The reference numeral 24-3 in the attached drawing refers to only one point; however, the first feature point can refer to all points displayed in the first frame 24-2 that distinguishes the region of interest. In this case, the feature point detection unit 13 can set the middle area of ​​the vehicle as the region of interest in the second vehicle region of the first frame 24-2 that distinguishes the region of interest, and detect the first feature point 24-3 within the set region of interest. The middle area of ​​the vehicle may include a license plate area formed at the rear of the vehicle, a rear bumper area separated from the license plate area by a predetermined distance, and a trunk area.

[0116] The feature point detection unit 13 can use the Harris corner detection method or the FAST (features-from-accelerated-segment test) corner detection method to detect the first feature point.

[0117] Then, the feature point detection unit 13 can track the detected first feature point 24-3 and detect a second feature point in the second frame corresponding to the current frame. In this case, the feature point detection unit 13 can detect the second feature point using the second frame 25-1 where the region of interest has not been processed, or the feature point detection unit 13 can detect the feature point using the second frame 25-2 where the region of interest has been processed. As an example, Figure 4 As shown, the feature point detection unit 13 can set the second vehicle region as the region of interest, generate a second frame 25-2 for processing the region of interest, and detect the second feature point 25-3 in the second frame 25-2 for processing the region of interest. Then, although recorded as... Figure 4 The reference numeral 25-3 in the attached figure refers to only one point; however, the second feature point can refer to all the points displayed in the second frame 25-2 that distinguishes the region of interest.

[0118] At this time, the feature point detection unit 13 can use the optical flow of the detected first feature point 24-3 to track the second feature point 25-3, thereby detecting the second feature point 25-3 in the second frame 25-2 of the region of interest.

[0119] On one hand, when the feature point detection unit 13 tracks the second feature point 25-3 using optical flow, it can filter out the second feature point 25-3 not displayed in the second frame 25-2 of the region of interest, as well as the first feature point 24-3 corresponding to the second feature point not displayed in the second frame 25-2 of the region of interest. That is, when the distance between the first vehicle 21 and the second vehicle 22 is closer, the second feature point 25-3 corresponding to a portion of the first feature points 24-3 detected in the first frame 24-2 of the region of interest (e.g., the feature point located at the lower end of the vehicle in the detected first feature points 24-3) may not be displayed in the second frame 25-2 of the region of interest. Thus, when the feature point detection unit 13 tracks the second feature point 25-3 using optical flow, it can filter out the second feature point 25-3 not displayed in the second frame 25-2 of the region of interest, as well as the first feature point 24-3 corresponding to the second feature point not displayed in the second frame 25-2 of the region of interest, thereby improving the calculation execution speed.

[0120] On one hand, the feature point change value calculation unit 14 can calculate the feature point change value between the first feature point and the second feature point. The feature point change value calculation unit 14 may include an average pixel distance calculation unit 14-1 and an average pixel distance ratio calculation unit 14-2. Regarding the operation of the feature point change value calculation unit 14, refer to... Figure 6 Please provide more specific details.

[0121] Figure 6 This is a schematic diagram illustrating a feature point change value calculation unit according to an embodiment of the present invention. (Reference) Figure 6 The average pixel distance calculation unit 14-1 can calculate the average pixel position 24-5 of the first feature points 24-3 and the first average pixel distance, which is the pixel distance from the average position to each of the first feature points 24-3. Specifically, the average pixel distance calculation unit 14-1 can average the pixel position coordinates of the first feature points 24-3 in the region of interest 24-4 of the first frame 24-1 to calculate the coordinates of the average pixel position 24-5. The region of interest 24-4 can be the middle area of ​​the vehicle. For example, the region of interest may include a license plate area formed at the rear of the vehicle, a rear bumper area separated from the license plate area by a predetermined distance, and a trunk area. Then, although described as... Figure 6The reference numeral 24-3 in the attached figure refers to only one point; however, the first feature point can refer to all points displayed in the region of interest 24-4, excluding reference numeral 24-5.

[0122] Then, the average pixel distance calculation unit 14-1 can average the pixel position coordinates of the first feature point 24-3 within the region of interest 24-4 and calculate the coordinates of the average pixel position 24-5.

[0123] refer to Figure 7 To illustrate the average pixel position calculation process, as an example, when the feature point is composed of a first point with (x,y) as the pixel position coordinate value, a second point with (x`,y`) as the pixel position coordinate value, and a third point with (x``,y``) as the pixel position coordinate value, the average pixel distance calculation unit 14-1 can calculate the average pixel position coordinate value (mx,my) by arithmetically averaging the pixel position coordinate values ​​of the first point, the second point, and the third point.

[0124] On the one hand, return to Figure 6 The average pixel distance calculation unit 14-1 can calculate a first average pixel distance, which is an average of the pixel distances from the calculated average pixel position 24-5 to each first feature point 24-3. The first frame 24-1 can be the frame that detected the second vehicle among the multiple frames constituting the driving image, preceding the frame in which the second vehicle was not detected.

[0125] Furthermore, the average pixel distance calculation unit 14-1 can calculate the average pixel position 25-5 of the second feature points 25-3 tracked by optical flow and a second average pixel distance, which is the average pixel distance from the average pixel position to each of the second feature points 25-3. Specifically, the average pixel distance calculation unit 14-1 can average the pixel position coordinates of the second feature points 25-3 in the region of interest 25-4 of the second frame 25-1 to calculate the coordinates of the average pixel position 25-5. Figure 6 In the first frame 24-1, the feature point represented by the dashed line indicates the position of the first feature point 24-3 within the region of interest 24-4. Then, the point represented by the solid line indicates the second feature point 25-3, although it is recorded as... Figure 6 The reference numeral 25-3 in the attached figure refers to only one point; however, the second feature point can refer to all points displayed in the region of interest 25-4, excluding reference numeral 25-5.

[0126] Then, the average pixel distance calculation unit 14-1 can set a region of interest 25-4 including the second feature point 25-3, average the pixel position coordinates of the feature points 25-3 within the region of interest 25-4, and calculate the coordinates of the average pixel position 25-5. Then, a second average pixel distance, which is the average of the pixel distances from the calculated average pixel position 25-5 to each feature point 25-3, can be calculated. The second frame 25-1 can be the current frame.

[0127] On the one hand, according to the above embodiments, an example is given of setting the middle area of ​​the vehicle as the region of interest 24-4 and 25-4 in frames 24-1 and 25-1, thereby performing feature point detection, tracking, and average pixel distance calculation. However, it is not limited to this. According to other embodiments of the present invention, the entire frames 24-1 and 25-1 can also be set as the region of interest to perform the aforementioned feature point detection, tracking, and average distance calculation.

[0128] On one hand, based on the above actions, the first average pixel distance and the second average pixel distance are calculated, and the average pixel distance ratio calculation unit 14-2 can calculate the average pixel distance ratio between the first average pixel distance and the second average pixel distance. Specifically, the average pixel distance ratio calculation unit 14-2 can calculate the average pixel distance ratio by dividing the second average pixel distance by the first average pixel distance, as shown in the following mathematical formula 1.

[0129]

Mathematical Formula 1

[0130]

[0131] Here, Ratio1 can refer to the average pixel distance ratio, curAvgDist can refer to the second average pixel distance, and preAvgDist can refer to the first average pixel distance.

[0132] On one hand, the vehicle-to-vehicle distance calculation unit 15 can calculate the vehicle-to-vehicle distance between the first vehicle and the second vehicle. Specifically, the vehicle-to-vehicle distance calculation unit 15 can calculate the distance from the camera on the first vehicle to the second vehicle based on the image width of the second vehicle, the focal distance of the camera device on the first vehicle, and the predicted width of the second vehicle. Regarding the vehicle-to-vehicle distance calculation unit 15, refer to... Figures 8 to 12 Please provide more specific details.

[0133] on the one hand, Figure 6 The example illustrates the use of a first frame 24-1 (without processed regions of interest) and a second frame 25-1 (without processed regions of interest) to calculate feature point change values; however, the invention is not limited thereto. According to other embodiments of the invention, a first frame 24-2 (with processed regions of interest) and a second frame 25-2 (with processed regions of interest) can also be used to calculate feature point change values.

[0134] Figure 8 This is a block diagram showing more specifically the workshop distance calculation unit 15 according to an embodiment of the present invention. Figure 9 This is a schematic diagram illustrating a workshop distance measurement method according to an embodiment of the present invention. (Reference) Figure 8 as well as Figure 9 The workshop distance calculation unit 15 may include an image width ratio calculation unit 15-1, a vehicle size level calculation unit 15-2, a vehicle width calculation unit 15-3, and a distance calculation unit 15-4.

[0135] A camera device 50 may be installed in a first vehicle (not shown) to capture images of the vehicle in motion. The camera device 50 may be mounted on the first vehicle and may consist of a car dash cam or a car video recorder that captures images of the surroundings while the vehicle is in motion or parked. Alternatively, the camera device 50 may also consist of a camera integrated into a navigation system or a camera built into the driver's mobile phone, wherein the navigation system provides route guidance to the driver of the first vehicle.

[0136] The imaging device 50 may include a lens 51 and an imaging element 52, although not in the... Figure 6 As shown, it may further include all or part of a lens unit, an aperture, an aperture drive unit, an image sensor control unit, and an image processor. The lens unit 51 can perform the function of collecting optical signals, and the optical signals transmitted through the lens unit 51 reach the imaging area of ​​the image sensor 52 for optical imaging. The image sensor 52 can be a CCD (Charge Coupled Device), a CIS (Complementary Metal Oxide Semiconductor Image Sensor), or a high-speed image sensor, etc., that converts optical signals into electrical signals.

[0137] On one hand, the workshop distance calculation unit 15 can use the driving images captured by the camera device 50 of the first vehicle to calculate the distance between the camera device 50 installed on the first vehicle and the second vehicle 30 based on the following mathematical formula 2.

[0138]

Mathematical Formula 2

[0139] D≤W×(f÷w)

[0140] Where D can be the distance from the camera device mounted on the first vehicle to the second vehicle, W can be the width of the second vehicle, f can be the focal length of the camera device, and w can be the image width of the second vehicle.

[0141] That is, the distance D from the camera device installed on the first vehicle to the second vehicle can refer to the distance from the camera device installed on the first vehicle to the second vehicle in real-world coordinates.

[0142] Then, the width W of the second vehicle can refer to the width of the second vehicle in real-world coordinates.

[0143] Then, the image width can refer to the pixel width of the second vehicle on the imaging surface of the imaging element 52 of the imaging device 50. The image width w of the second vehicle can be the same value as Vehicle W in Equation 3 described later.

[0144] On one hand, the vehicle-to-vehicle distance calculation unit 15 first calculates the ratio between the image width of the second vehicle 30 and the image width of the lane where the second vehicle 30 is located from the driving image acquired by the camera device 50 of the first vehicle. Based on the calculated ratio, it determines the size class of the second vehicle 30 among multiple size classes, and then calculates the width W of the second vehicle 30 based on the determined size class. (Reference) Figure 10 The operation of the workshop distance calculation unit 15 will be explained in more detail.

[0145] Figure 10 This is a schematic diagram illustrating the ratio between the image width of the second vehicle and the image width of the lane in which the second vehicle is located, according to an embodiment of the present invention. Figure 10 As shown, the driving image 45 captured by the camera device 50 of the first vehicle may include a second vehicle 30 traveling in front of the first vehicle, a lane 40 in which the second vehicle is traveling, a left lane 41 and a right lane 42 that distinguish the lane 40 from other lanes.

[0146] In this case, the image width ratio calculation unit 15-1 can calculate the image width VehicleW of the second vehicle 30. Specifically, when the vehicle detection unit 12 detects a vehicle from the driving image using the already constructed learning model, the image width ratio calculation unit 15-1 can identify the left boundary 31 and right boundary 32 of the second vehicle 30 in the detected image of the second vehicle 30. (Reference) Figure 11 To explain the boundary recognition in more detail.

[0147] refer to Figure 11When a vehicle is detected using a learning model in an image frame (WXH) 81 acquired by the camera, a detected vehicle region 82 can be harvested. For the harvested region (w`X h`) 83, a vertical edge 84 is detected through Sobel calculus.

[0148] Then, the cumulative value of the vertical histogram can be calculated on the detected vertical edges, and the point where the maximum histogram value is located can be detected as the left and right boundary positions of the vehicle (85).

[0149] Then, the vehicle area can be fitted to the detected left and right boundary positions of the vehicle.

[0150] On one hand, the image width ratio calculation unit 15-1 can determine the image width between the identified left boundary 31 and the identified right boundary 32 as the image width of the second vehicle, VehicleW.

[0151] Furthermore, the image width ratio calculation unit 15-1 can identify the left lane 41 and right lane 42 of the lane 40 in which the second vehicle 30 is traveling in the acquired driving image 45. Then, the image width ratio calculation unit 15-1 can set a line 33 indicating the position of the second vehicle 30 within the lane. The line 33 indicating the position of the second vehicle 30 within the lane can be formed by extending the bottom edge of the second vehicle 30 in the driving image 45. As an example, it can be formed by extending the bottom edge of the left wheel and the bottom edge of the right wheel of the second vehicle 30.

[0152] On the one hand, the first fulcrum 43 and the second fulcrum 44 can be determined, and the image width between the first fulcrum 43 and the second fulcrum 44 is determined as the image width LaneW of the lane where the second vehicle 30 is located. The first fulcrum 43 is the fulcrum where the line 33 representing the position of the second vehicle 30 in the lane intersects with the left lane line 41, and the second fulcrum 44 is the fulcrum where the line 33 representing the position of the second vehicle 30 in the lane intersects with the right lane line 42.

[0153] On the one hand, if the image width of the second vehicle (VehicleW) and the image width of the lane where the second vehicle 30 is located (LaneW) are calculated, the image width ratio calculation unit 15-1 can calculate the ratio between the image width of the second vehicle in front and the image width of the lane where the second vehicle is located based on the following mathematical formula 3.

[0154]

Mathematical Expression 3

[0155] Ratio2=(VehiclcW / LaneW)×100

[0156] Wherein, Vehicle W can refer to the image width of the second vehicle, Lane W can refer to the image width of the lane in which the second vehicle is located, and Ratio 2 can refer to the ratio between the image width of the second vehicle in front and the image width of the lane in which the second vehicle is located.

[0157] As described above, if the distance between the first vehicle and the second vehicle becomes closer, the image width of the second vehicle and the image width of the lane in which the second vehicle is located will increase; conversely, if the distance between the first vehicle and the second vehicle becomes farther, the image width of the second vehicle and the image width of the lane in which the second vehicle is located will decrease. However, this ratio is not affected by the distance between the first vehicle and the second vehicle, but is proportional to the size of the second vehicle. Therefore, according to the present invention, it can be used as an indicator for calculating the size of the second vehicle.

[0158] On the one hand, based on the above example, if the ratio between the image width of the second vehicle and the image width of the lane in which the second vehicle is located is calculated, the vehicle-to-vehicle distance calculation unit 15 can determine the size class of the second vehicle among multiple size classes. For this purpose, refer to... Figure 12 To explain in more detail.

[0159] Figure 12 This is a conceptual diagram illustrating the size classification process of a second vehicle according to an embodiment of the present invention. Figure 12 As shown, the vehicle size class calculation unit 15-2 divides the ratio value into multiple intervals. For each interval, the vehicle size class can be calculated based on the critical value table that matches the size class of the second vehicle.

[0160] As an example, the threshold table can be divided into three intervals based on the first and second values. When the value is less than the first value, it can be matched with the first size class corresponding to small cars. When the calculated proportion is greater than the first value but less than the second value, it can be matched with the second size class corresponding to medium-sized cars. When the calculated proportion is greater than the second value, it can be matched with the third size class corresponding to large cars.

[0161] In this case, when the ratio calculated by the ratio calculation unit 15-1 is less than the first value, the vehicle size classification calculation unit 15-2 can determine the size classification of the second vehicle as the first size classification. Then, when the ratio calculated by the image width ratio calculation unit 15-1 is greater than the first value but less than the second value, the vehicle size classification calculation unit 15-2 can determine the size classification of the second vehicle as the second size classification. Then, when the ratio calculated by the image width ratio calculation unit 15-1 is greater than the second value, the vehicle size classification calculation unit 15-2 can determine the size classification of the second vehicle as the third size classification. As an example, the first value can be 48%, and the second value can be 60%.

[0162] The vehicle width calculation unit 15-3 can determine the width of the second vehicle based on the size class of the second vehicle. Specifically, as shown in Table 1 below, the storage unit can store vehicle widths for multiple size classes. In this case, the vehicle width calculation unit 15-3 can determine the width of the second vehicle, VehicleW, by detecting the vehicle width corresponding to the determined size class from the vehicle widths stored in the storage unit in advance.

[0163] Table 1

[0164] Actual width of the vehicle 1500mm 1900mm 2500mm

[0165] Then, as shown in mathematical formula 2 above, the distance calculation unit 15-4 can divide the focal length f of the shooting device 50 by the image width w of the second vehicle 30, and then multiply it by the width W of the second vehicle 30 calculated by the vehicle width calculation unit 15-3, thereby being able to calculate the distance between the shooting device 50 and the second vehicle 30.

[0166] On one hand, if the distance between the shooting device 50 and the second vehicle 30 is calculated, the distance calculation unit 15-4, in order to accurately calculate the distance between the vehicles, appropriately corrects the distance value between the shooting device 50 and the second vehicle 30, thereby enabling the calculation of the distance value between the first vehicle and the second vehicle 30 on which the shooting device 50 is installed. According to the present invention, the error in the inter-vehicle distance between the first vehicle and the second vehicle can be reduced, thereby enabling more accurate measurement of the inter-vehicle distance.

[0167] That is, for small cars, medium cars, and large cars that are separated from the first vehicle by the same distance but have different widths, if we want the distance values ​​calculated based on the above mathematical formula 2 to be the same, we need to know the accurate width of each vehicle. However, in existing image recognition and detection, it is impossible to confirm all the data related to all vehicle types. Therefore, the existing technology does not consider the actual width of many vehicles (e.g., small cars, medium cars, and large cars) with different widths, and measures the vehicle width by processing the vehicle width with a preset specific constant value. Therefore, there is a problem of inaccurate measured vehicle distance values.

[0168] However, in order to solve the aforementioned problem, the present invention distinguishes vehicles ahead into small cars, medium cars, and large cars by utilizing the ratio between the image width of the vehicle ahead and the image width of the lane. Then, based on the distinguished results, the inter-vehicle distance is measured according to the average width that matches each of the small, medium, and large cars, thereby reducing errors and measuring the inter-vehicle distance more accurately.

[0169] On the one hand, as described above, if the expected width of the second vehicle is calculated, the vehicle detection unit 12 continuously measures the image width of the second vehicle in the environment of detecting the second vehicle, thereby calculating the distance between the camera of the first vehicle and the second vehicle.

[0170] However, if the distance between the first vehicle and the second vehicle approaches to a point where the lower part of the second vehicle is not captured, and the vehicle detection unit 12 fails to detect the second vehicle from the driving image using the learning model, it may be unable to measure the image width of the second vehicle.

[0171] Therefore, according to the present invention, when the distance between the first vehicle and the second vehicle approaches to a point where the lower part of the second vehicle is not captured, causing the vehicle detection unit 12 to fail to detect the second vehicle from the driving image using the learning model, the image width of the second vehicle can be predicted based on the average distance ratio calculated by the average pixel distance ratio calculation unit 14-2. In this regard, refer again... Figure 6 Please provide a detailed explanation.

[0172] Refer to the above Figure 6 The first frame 24-1 is the frame that corresponds to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected, which constitutes the driving image. The vehicle detection unit 12 can detect the second vehicle 30 in the first frame 24-1, and the image width ratio calculation unit 15-1 can calculate the image width 24-6 of the second vehicle.

[0173] However, since the second frame 25-1 is the current frame, it is a frame in which the second vehicle has not been detected. Therefore, the image width ratio calculation unit 15-1 can apply the image width 24-6 of the second vehicle and the average pixel distance ratio calculated by the average pixel distance ratio calculation unit 14-2 to the following mathematical formula 4 to calculate the predicted value of the image width of the second vehicle in the second frame 25-1.

[0174]

Mathematical Expression 4

[0175] curVehicleW=Ratiol×preVehicleW

[0176] Here, curVehicleW can refer to the image width of the second vehicle in the second frame 25-1, Ratio 1 can refer to the average pixel distance ratio calculated according to mathematical formula 1, and preVehicleW can refer to the image width of the second vehicle in the first frame 24-1.

[0177] Then, as shown in the above mathematical formula 2, the distance calculation unit 15-4 can divide the focal length f of the shooting device 50 by the image width (curVehicleW) of the second vehicle 30 in the second frame 25-1, and then multiply it by the width W of the second vehicle 30 calculated by the vehicle width calculation unit 15-3, thereby being able to calculate the distance between the shooting device 50 and the second vehicle 30.

[0178] Therefore, according to the present invention, even if the lower part of the second vehicle is not captured as the distance between the first vehicle and the second vehicle approaches, resulting in the inability to detect the target vehicle through driving images, the distance between one's own vehicle and the target vehicle can be accurately measured through feature point tracking.

[0179] In other embodiments of the present invention, the vehicle-to-vehicle distance calculation unit 15 monitors the distance between the second vehicle and the first vehicle by calculating the ratio of the image width 24-6(w) of the second vehicle detected in the first frame 24-1 to the image width 25-6(w) of the second vehicle detected in the second frame 25-1. The control unit 19 can use the distance monitored by the vehicle-to-vehicle distance calculation unit 15 to provide the driver with various functions related to vehicle movement, such as collision warning and adaptive cruise control.

[0180] On the one hand, when the distance calculated by the workshop distance calculation unit 15 is less than the preset distance, the prompt data generation unit 17 can generate prompt data to indicate the collision hazard level corresponding to the distance difference between the first vehicle and the second vehicle.

[0181] In addition, the driving control data generation unit 18 can generate control data based on the distance calculated by the inter-vehicle distance calculation unit 15 to control the automatic driving of the first vehicle.

[0182] The operation of the prompt data generation unit 17 and the driving control data generation unit 18 will be explained below based on the control unit 19.

[0183] The control unit 19 controls the overall operation of the inter-vehicle distance measuring device 10. Specifically, the control unit 19 can control all or part of the image acquisition unit 11, vehicle detection unit 12, feature point detection unit 13, feature point change value calculation unit 14, inter-vehicle distance calculation unit 15, prompt data generation unit 17, and driving control data generation unit 18.

[0184] In particular, the control unit 19 can control the vehicle detection unit 12 to detect the second vehicle from the driving image captured by the camera of the first vehicle in motion. If the second vehicle is not detected from the driving image, the control unit 13 can select the first frame corresponding to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected from the multiple frames constituting the driving image. The control unit 13 can detect the first feature point of the first feature point tracking detection in the second vehicle area in the selected first frame, detect the second feature point in the second frame corresponding to the current frame, and control the feature point change value calculation unit 14 to calculate the feature point change value between the first feature point and the second feature point. The control unit 15 can calculate the distance from the camera of the first vehicle to the second vehicle based on the calculated feature point change value.

[0185] Furthermore, if the distance information between the first vehicle and the second vehicle is obtained, the control unit 19 can control the prompt data generation unit 17 to generate prompt data based on this information to assist the driver of the first vehicle in safe driving. Specifically, when the distance calculated by the distance calculation unit 15 is less than a preset distance, the prompt data generation unit 17 can generate prompt data to indicate the distance difference between the first vehicle and the second vehicle. As an example, the prompt data generated by the prompt data generation unit 17 can be data that warns of the need to pay attention to the distance via voice or data that provides visual prompts.

[0186] As another example, when the distance calculated by the vehicle distance calculation unit 15 is less than a preset distance, the warning data generation unit 17 can generate data to indicate the collision hazard level corresponding to the distance difference between the first vehicle and the second vehicle. For example, the distance difference between the first vehicle and the second vehicle is divided into multiple levels. When the vehicle distance is less than a first value, data to indicate a first hazard level can be generated. When the vehicle distance is greater than the first value but less than a second value, data to indicate a second hazard level with a higher degree of hazard than the first hazard level can be generated. When the vehicle distance is greater than the second value, data to indicate a third hazard level with a higher degree of hazard than the second hazard level can be generated.

[0187] On one hand, if the inter-vehicle distance information between the first vehicle and the second vehicle is obtained, the control unit 19 can control the driving control data generation unit 18 to generate driving control data based on this information to control the automatic driving of the first vehicle. Specifically, when the first vehicle is driving in automatic driving mode, and the inter-vehicle distance calculated by the inter-vehicle distance calculation unit 15 is less than a preset distance, the control unit 19 can control the driving control data generation unit 18 to generate driving control data for controlling the automatic driving of the first vehicle (for example, controlling the speed of the first vehicle to decelerate from the current speed to a predetermined speed or controlling it to stop, etc.). The driving control data generated by the driving control data generation unit 18 can be transmitted to the automatic driving control unit that uniformly controls the automatic driving of the first vehicle, and the first vehicle automatic driving control unit can then control various units (brakes, steering wheel, electric motor, engine, etc.) installed in the first vehicle based on this information to control the automatic driving of the first vehicle.

[0188] The following is for reference Figures 13 to 16 A workshop distance measurement method according to an embodiment of the present invention will be described in more detail.

[0189] Figure 13 This is a flowchart illustrating a workshop distance measurement method according to an embodiment of the present invention. Figure 13 As shown, firstly, the camera can capture a driving image of the first vehicle in motion (S100).

[0190] Then, it can be determined whether a second vehicle has been detected from the acquired driving images (S120). The detection of the second vehicle from the acquired driving images can be performed by a learning model built by machine learning or deep learning.

[0191] Assuming a second vehicle is detected from the acquired driving image (S120:Y), the distance from the first vehicle's camera to the detected second vehicle can be calculated (S170). The distance calculation step (S170) calculates the image width ratio between the detected second vehicle's image width and the lane's image width. Based on the calculated image width ratio, the size class of the second vehicle is determined. Based on the determined size class of the second vehicle, the predicted width of the second vehicle is calculated. The image width of the second vehicle, the focal distance of the first camera, and the predicted width of the second vehicle are applied to the aforementioned mathematical formula 2 to calculate the distance from the first vehicle's camera to the second vehicle.

[0192] However, if no second vehicle is detected from the acquired driving image (S120:N), a first frame corresponding to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected can be selected from the multiple frames constituting the driving image. A first feature point is then detected in the second vehicle region within the selected first frame (S130). That is, the step of detecting the first feature point (S130) is performed when the second vehicle is not detected by the constructed learning model as the distance between the first and second vehicles approaches. The step of detecting the first feature point (S130) can be performed by setting the middle area of ​​the vehicle as the region of interest within the second vehicle region of the first frame, and detecting the first feature point within the set region of interest.

[0193] Then, the detected first feature point can be tracked to detect a second feature point in the second frame corresponding to the current frame (S140). Specifically, the step of detecting the second feature point (S140) can use the optical flow of the detected first feature point to track the second feature point, thereby detecting the second feature point in the second frame.

[0194] In addition, according to an embodiment of the present invention, it may further include the step of filtering out second feature points not displayed in the second frame and first feature points corresponding to the second feature points not displayed in the second frame when using optical flow to track the second feature points.

[0195] Then, the feature point change value between the first feature point and the second feature point can be calculated (S150). This will be discussed later. Figure 14 Explain the steps for calculating the change value of feature points (S150).

[0196] Then, based on the calculated feature point change values, the distance from the camera device of the first vehicle to the second vehicle can be calculated (S160). Wherein, reference... Figure 15 The steps for calculating the distance are explained in more detail (S160).

[0197] Figure 14 This is a flowchart illustrating more specifically the step (S150) of calculating the feature point change value according to an embodiment of the present invention. (See reference...) Figure 14The average pixel position of the first feature point can be calculated (S210). Then, a first average pixel distance, which is the average of the calculated average pixel position to the pixel distance between each first feature point, is calculated (S220). Specifically, the average pixel distance calculation unit 14-1 can average the pixel position coordinate values ​​of the first feature points, calculate the coordinate values ​​of the average pixel position, and calculate the first average pixel distance, which is the average of the calculated evaluation pixel position and the pixel distance between each first feature point. Here, the first frame is the frame corresponding to the frame in the plurality of frames constituting the driving image where the second vehicle was detected before the frame where the second vehicle was not detected, and can be a frame before the second frame.

[0198] Then, the average pixel position of the second feature points can be calculated (S230). Then, a second average pixel distance, which is the average of the calculated average pixel positions to the pixel distances between each of the second feature points, can be calculated (S240). Specifically, the average pixel distance calculation unit 14-1 can average the pixel position coordinate values ​​of the second feature points, calculate the coordinate values ​​of the average pixel positions, and calculate the second average pixel distance, which is the average of the calculated evaluation pixel positions and the pixel distances between each of the second feature points. The second frame can be a frame following the first frame.

[0199] Then, the average pixel distance ratio between the first average pixel distance and the second average pixel distance can be calculated (S250). Specifically, as shown in the above mathematical formula 1, the average pixel distance ratio calculation unit 14-2 can divide the second average pixel distance by the first average pixel distance to calculate the average pixel distance ratio.

[0200] Figure 15 This is a flowchart illustrating more specifically the workshop distance calculation step (S160) according to an embodiment of the present invention. (See reference...) Figure 15 The second vehicle can be detected from the first frame (S310), and the image width of the second vehicle can be calculated from the first frame (S320).

[0201] Then, based on the image width of the second vehicle in the first frame and the average pixel distance ratio calculated by the average pixel distance ratio calculation unit, the predicted value of the image width of the second vehicle in the second frame can be calculated (S330). That is, the second frame is a frame in which the second vehicle was not detected, so the image width ratio calculation unit 15-1 can apply the image width of the second vehicle and the average pixel distance ratio calculated by the average pixel distance ratio calculation unit 14-2 to the above mathematical formula 4 to calculate the predicted value of the image width of the second vehicle in the second frame.

[0202] Then, based on the calculated image width of the second vehicle in the second frame, the focal distance of the camera device of the first vehicle, and the predicted width of the second vehicle, the distance from the camera device of the first vehicle to the second vehicle can be calculated (S340). Specifically, as shown in the above mathematical formula 2, the distance calculation unit 15-4 can divide the focal length f of the camera device of the first vehicle by the image width (curVehicleW) of the second vehicle in the second frame, and then multiply it by the predicted width of the second vehicle 30 calculated by the vehicle width calculation unit 15-3, thereby calculating the distance between the camera device of the first vehicle and the second vehicle.

[0203] The process of calculating the predicted width of the second vehicle can consist of the following steps: calculating the ratio between the image width of the second vehicle detected by the vehicle detection unit 12 and the image width of the lane in which the second vehicle is located; determining the size class of the second vehicle based on the calculated ratio; and calculating the predicted width of the second vehicle based on the determined size class. The predicted width of the second vehicle can be calculated and stored in advance during the period when the vehicle detection unit 12 detects the second vehicle.

[0204] Figure 16 This is a flowchart illustrating a workshop distance calculation method according to another embodiment of the present invention.

[0205] refer to Figure 16 First, the vehicle distance measuring device 10 receives the current frame (frame i) via the image acquisition unit 11 (S1100). Then, the vehicle distance measuring device 10 uses a learning model to detect a second vehicle in the input frame i (S1105). Assuming that a second vehicle is detected in the frame i (S1110:N), the device calculates the distance between itself and the second vehicle detected by the learning model (S1115), and stores the frame i and the detected vehicle region (S1120). At this time, the detected vehicle region can be set to a rectangle, a circle, or a polygon, but is not limited to these.

[0206] Then, if the i-th frame and the detected vehicle area are stored, the vehicle distance measuring device 10 updates i to i+1 (S1125) and obtains the i+1-th frame as the current frame input.

[0207] However, if no second vehicle is detected in step S1110 (S1110:Y), the vehicle distance measuring device 10 checks whether a stored vehicle region exists in the previous frame (the (i-1)th frame) (S1130). Then, in S1130, if a stored vehicle region exists in the (i-1)th frame (S1130:Y), the vehicle distance measuring device 10 extracts a first feature point from the region of interest of the vehicle region in the (i-1)th frame (S1135), and after extracting a second feature point corresponding to the first feature point extracted in the (i-1)th frame in the i-th frame (S1140), the distance between the device and the second vehicle is calculated using the difference between the average pixel position of the first feature point and the average pixel position of the second feature point (S1145).

[0208] Conversely, in step S1130, if there is no stored vehicle area in the (i-1)th frame (S1130:N), the vehicle distance measuring device 10 determines that there was no second vehicle previously, and inputs a new current frame obtained from the image acquisition unit 11 (S1125).

[0209] On the one hand, the workshop distance measuring device 10 can be composed of a module of an electronic device that outputs various prompts to assist the driver, thereby enabling it to perform route prompting functions. In this regard, refer to... Figures 15 to 17 More detailed explanation.

[0210] Figure 17 This is a block diagram illustrating an electronic device according to an embodiment of the present invention. Figure 17 As shown, the electronic device 100 may include all or part of a storage unit 110, an input unit 120, an output unit 130, a distance measurement unit 140, an augmented reality providing unit 160, a control unit 170, a communication unit 180, a sensing unit 190, and a power supply unit 195.

[0211] The electronic device 100 can consist of various devices such as smartphones, tablets, laptops, PDAs (personal digital assistants), PMPs (portable multimedia players), smart glasses, Google Glass, navigators, car dashcams, or car video recorders that can provide driving-related prompts to the driver of the vehicle, and can be installed in the vehicle.

[0212] Driving-related prompts can include various prompts to assist drivers, such as route prompts, lane departure prompts, lane keeping prompts, prompts for vehicles ahead to depart, traffic light change prompts, prompts for vehicles ahead to avoid collisions, lane change prompts, lane prompts, and curve prompts.

[0213] The route suggestions can include augmented reality route suggestions that combine the user's location, direction, and other information from the image in front of the moving vehicle, or 2D or 3D route suggestions that combine the user's location, direction, and other information from 2D or 3D map data.

[0214] Furthermore, route suggestions can be derived from aviation map data by incorporating various information such as the user's location and direction, thus providing aviation map route suggestions. The concept of route suggestions can be interpreted as including not only suggestions when the user is driving, but also when the user is walking or running.

[0215] In addition, the lane departure warning can indicate whether a moving vehicle has deviated from the lane.

[0216] In addition, lane keeping assist can prompt the vehicle to return to its original lane.

[0217] Additionally, the vehicle ahead departure alert can indicate whether the vehicle in front of a parked vehicle has departed. This vehicle ahead departure alert can be performed using the inter-vehicle distance calculated by the inter-vehicle distance measurement unit 140.

[0218] Additionally, traffic light change indicators can alert vehicles that are stopped to whether the traffic light signal ahead has changed. For example, it can indicate when the red light (stop signal) changes to a green light (go signal).

[0219] In addition, the forward collision prevention warning is provided when the distance to a vehicle in front of a stopped or moving vehicle is within a certain distance to prevent a collision. This forward collision prevention warning can be implemented using the inter-vehicle distance calculated by the inter-vehicle distance measurement unit 140.

[0220] Additionally, lane change prompts can be used to indicate the route to the destination, prompting vehicles to change lanes from their current lane to other lanes.

[0221] In addition, lane indicators can show the vehicle the lane it is currently in.

[0222] In addition, curve indications can indicate that the road the vehicle will be traveling on in the future is a curve.

[0223] Similar to the forward-facing image of a vehicle that can provide the aforementioned various prompts, driving-related images can be captured by a camera installed on the vehicle or a smartphone camera. The camera can be integrated with the electronic device 100 installed on the vehicle to capture images of the front of the vehicle.

[0224] As another example, the camera may be installed independently of the electronic device 100 on the vehicle to capture images of the front of the vehicle. In this case, the camera may be a separate vehicle imaging device positioned facing the front of the vehicle, and the electronic device 100 receives the captured images via wired / wireless communication input with the separate vehicle imaging device, or the electronic device 100 may receive the captured images when a storage medium for storing images captured by the vehicle imaging device is inserted into the electronic device 100.

[0225] Hereinafter, based on the above, an electronic device 100 according to an embodiment of the present invention will be described in more detail.

[0226] The storage unit 110 performs functions for storing various data and applications required for the operation of the electronic device 100. In particular, the storage unit 110 can store data required for the operation of the electronic device 100, such as the operating system, route exploration applications, map data, etc. In addition, the storage unit 110 can store data generated by the operation of the electronic device 100, such as explored route data, received images, etc.

[0227] The storage unit 110 can be composed not only of memory-type storage elements such as RAM (Random Access Memory), flash memory, ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable and Programmable ROM), registers, hard disks, removable disks, memory cards, and USIM (Universal Subscriber Identity Module), but also of removable storage elements such as USB storage devices.

[0228] The input unit 120 performs the function of converting physical inputs from outside the electronic device 100 into specific electrical signals. The input unit 120 may include all or part of the user input unit 121 and the microphone unit 123.

[0229] The user input unit 121 can receive user input such as touch and swipe gestures. The user input unit 121 may be composed of at least one of various button shapes, a touch sensor that receives touch input, and a proximity sensor that receives the action of being approached.

[0230] The microphone unit 123 can receive the user's voice as well as sounds generated inside and outside the vehicle.

[0231] The output unit 130 is a device that outputs data from the electronic device 100 to a user via images and / or voice. The output unit 130 may include all or part of the display unit 131 and the audio output unit 133.

[0232] Display unit 131 is a device that outputs visually identifiable data to a user. Display unit 131 may be a display unit provided on the front of the housing of electronic device 100. Alternatively, display unit 131 may be integrated with electronic device 100 to output visually identifiable data, or it may be set independently of electronic device 100 to output visually identifiable data, similar to a HUD.

[0233] The audio output unit 133 is a device that outputs audibly recognizable data to a user. The audio output unit 133 may be composed of a speaker that uses sound to represent data to be communicated to the user of the electronic device 100.

[0234] The workshop distance measuring unit 140 can perform the functions of the workshop distance measuring device 10 described above.

[0235] Augmented Reality Provider 160 can provide an augmented reality field of view mode. Augmented reality can be a method of providing additional information (e.g., graphic elements representing points of interest (POIs), graphic elements indicating the danger of collision with vehicles ahead, graphic elements indicating the distance between vehicles, graphic elements indicating curves, and various additional information to help the driver drive safely) by visually overlaying them onto the screen that presents the real world that the user actually sees.

[0236] The augmented reality providing unit 160 may include all or part of a calibration unit, a 3D space generation unit, an object generation unit, and a mapping unit.

[0237] The calibration unit can perform calibration to infer camera parameters belonging to the camera from the images captured by the camera. These camera parameters, which constitute the camera matrix, can include extrinsic parameters and intrinsic parameters. The camera matrix represents information about the relationships between the captured space and the images.

[0238] The 3D space generation unit can generate a virtual 3D space based on images captured by the camera. Specifically, the 3D space generation unit can apply camera parameters estimated by the calibration unit to 2D captured images to generate a virtual 3D space.

[0239] The object generation unit can generate objects for prompts in augmented reality, such as objects for preventing collisions ahead, route prompts, lane change prompts, lane departure prompts, and curve prompts.

[0240] The mapping unit can map objects generated by the object generation unit to the virtual 3D space generated by the 3D space generation unit. Specifically, the mapping unit can determine the position of the object generated in the object generation unit in the virtual 3D space and map the object at the determined position.

[0241] On the one hand, the communication unit 180 may be provided to enable the electronic device 100 to communicate with other devices. The communication unit 180 may include all or part of the location data unit 181, the wireless network unit 183, the broadcast transceiver unit 185, the mobile communication unit 186, the short-range communication unit 187, and the wired communication unit 189.

[0242] The location data unit 181 is a device for acquiring location data via GNSS (Global Navigation Satellite System). GNSS refers to a navigation system that uses radio signals received from artificial satellites to calculate the location of a receiving terminal. Specific examples of GNSS, depending on their operating entity, include GPS (Global Positioning System), Galileo, GLONASS (Global Orbiting Navigational Satellite System), COMPASS, IRNSS (Indian Regional Navigation Satellite System), and QZSS (Quasi-Zenith Satellite System). According to an embodiment of the present invention, the location data unit 181 can receive GNSS signals serving the area where the electronic device 100 is used to acquire location data. Alternatively, in addition to GNSS, the location data unit 181 can also acquire location data through communication with a base station or access point (AP).

[0243] Wireless Network Unit 183 is a device that connects to a wireless network to acquire or transmit data. Wireless Network Unit 183 can connect to the network via various communication protocols, defined as transmitting and receiving wireless data using WLAN (Wireless Local Area Network), Wibro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), and HSDPA (High-Speed ​​Downlink Packet Access).

[0244] The Broadcast Transceiver Unit 185 is a device that transmits and receives broadcast signals through various broadcast systems. Broadcast systems that can be transmitted and received through the Broadcast Transceiver Unit 185 include DMBT (Digital Multimedia Broadcasting Terrestrial), DMBS (Digital Multimedia Broadcasting Satellite), MediaFLO (Media Forward Link Only), DVBH (Digital Video Broadcast Handheld), and ISDBT (Integrated Services Digital Broadcast Terrestrial), among others. Broadcast signals transmitted and received through the Broadcast Transceiver Unit 185 can include traffic data, lifestyle data, and more.

[0245] The Mobile Communications Department 186 can connect to the mobile communication network according to various mobile communication standards such as 3G (3rd Generation), 3GPP (3rd Generation Partnership Project), and LTE (Long Term Evolution) to conduct voice and data communication.

[0246] The short-range communication unit 187 is a device for performing short-range communication. As described above, the short-range communication unit 187 can communicate via Bluetooth, RFID (Radio Frequency Identification), IrDA (Infrared Data Association), UWB (Ultra Wide Band), ZigBee (Wireless Personal Area Network), NFC (Near Field Communication), Wi-Fi (Wireless Fidelity), etc.

[0247] The wired communication unit 189 is an interface device that enables wired connection between the electronic device 100 and other devices. The wired communication unit 189 may be a USB module that communicates via a USB interface (Port).

[0248] The communication unit 180 can communicate with other devices using at least one of the location data unit 181, wireless network unit 183, broadcast transceiver unit 185, mobile communication unit 186, short-range communication unit 187, and wired communication unit 189.

[0249] As an example, when the electronic device 100 does not include a camera function, it can use images captured by a vehicle image capturing device such as a transceiver vehicle recorder or a vehicle video recorder, which are at least one of the short-range communication unit 187 and the wired communication unit 189.

[0250] As another example, when communicating with multiple devices, one can communicate through the short-range communication unit 187, and the other through the wired communication unit 189.

[0251] The sensing unit 190 is a device capable of sensing the current state of the electronic device 100. The sensing unit 190 may include all or part of the motion sensing unit 191 and the light sensing unit 193.

[0252] The motion sensing unit 191 can sense the motion of the electronic device 100 in 3D space. The motion sensing unit 191 may include a 3-axis geomagnetic sensor and a 3-axis accelerometer. By combining the motion data acquired by the motion sensing unit 191 with the position data acquired by the position data unit 181, the trajectory of the vehicle equipped with the electronic device 100 can be calculated more accurately.

[0253] The light sensor 193 is a device for detecting the ambient illuminance of the electronic device 100. The brightness of the display unit 131 can be changed to correspond to the ambient brightness using the illuminance data acquired by the light sensor 193.

[0254] The power supply unit 195 is a device that provides the necessary power for the operation of the electronic device 100 or other devices connected to the electronic device 100. The power supply unit 195 may be a device that receives power from a battery built into the electronic device 100 or from an external power source such as a vehicle. In addition, the power supply unit 195 may be composed of a wired communication module 119 or a device that receives power wirelessly, depending on the form in which the power is supplied.

[0255] The control unit 170 controls the overall operation of the electronic device 100. Specifically, the control unit 170 can control all or part of the storage unit 110, input unit 120, output unit 130, inter-station distance measurement unit 140, augmented reality providing unit 160, communication unit 180, sensing unit 190, and power supply unit 195.

[0256] Specifically, the control unit 170 can control the output unit 130 to output a forward collision warning or a forward vehicle departure warning based on the inter-vehicle distance calculated by the inter-vehicle distance measurement unit 140. As an example, the output unit 130 may include a display unit 131 that combines a captured driving image with a warning object to output an augmented reality image. In this case, the control unit 170 can control the display unit 131 to generate a warning object for the forward collision warning or forward vehicle departure warning, and overlay the generated warning object with the forward vehicle display area of ​​the augmented reality image.

[0257] When a forward collision warning is issued, the displayed warning objects can represent different levels of collision hazard based on the difference in distance between the first and second vehicles. For example, the distance difference between the first and second vehicles can be divided into multiple levels. When the distance is less than a first value, it can represent a warning object indicating the first level of hazard. When the distance is greater than the first value but less than a second value, it can represent a warning object indicating the second level of hazard, which is higher than the first level. When the distance is greater than the second value, it can represent a warning object indicating the third level of hazard, which is higher than the second level.

[0258] Alternatively, when issuing a departure alert for a vehicle ahead, the alert objects can be differentiated based on the departure request level corresponding to the distance difference between the first and second vehicles. For example, the distance difference between the first and second vehicles can be divided into multiple levels. When the distance is less than a first value, it can represent an alert for a first departure request level. When the distance is greater than the first value but less than a second value, it can represent an alert for a second departure request level, indicating a departure that needs to be faster than the first. When the distance is greater than the second value, it can represent an alert for a third departure request level, indicating a departure that needs to be faster than the second.

[0259] Figure 18 This is a schematic diagram illustrating a system network connected to an electronic device according to an embodiment of the present invention. Figure 18 As shown, an electronic device 100 according to an embodiment of the present invention may be composed of various devices installed on a vehicle, such as a navigator, a vehicle image capturing device, a smartphone or other vehicle augmented reality interface providing device, and may be connected to various communication networks and other electronic devices 61 to 64.

[0260] In addition, the electronic device 100 can work in conjunction with the GPS module based on the radio signals received from the artificial satellite 70, thereby being able to calculate the current location and the current time.

[0261] Each artificial satellite 70 can transmit L-band frequencies with different frequency bands. Electronic device 100 can calculate the current position based on the time taken for the L-band frequencies transmitted from each artificial satellite 70 to reach electronic device 100.

[0262] On one hand, electronic device 100 can wirelessly connect to network 90 via communication unit 180, control station (ACR) 80, base station (RAS) 85, access point (AP), etc. When electronic device 100 is connected to network 90, it is also indirectly connected to other electronic devices 61 and 62 connected to network 90, thereby enabling data exchange.

[0263] On the one hand, the electronic device 100 can also be indirectly connected to the network 90 through other devices 63 with communication functions. For example, when the electronic device 100 does not have a module that can connect to the network 90, it can communicate with other devices 63 with communication functions through a short-range communication module or the like.

[0264] Figure 19 as well as Figure 20 This is a schematic diagram illustrating a forward vehicle collision prevention warning screen of an electronic device according to an embodiment of the present invention. Figure 19 as well as Figure 20As shown, the electronic device 100 can generate a warning object indicating the danger of a vehicle collision based on the distance between its own vehicle and the vehicle in front, and output the generated warning object through augmented reality.

[0265] As an example, such as Figure 19 As shown, when the distance between one's own vehicle and the vehicle in front is greater than a predetermined distance, the electronic device 100 can generate and display a attention prompt object 1501 to remind the user of a state that requires attention.

[0266] In addition, such as Figure 20 As shown, when the distance between one's own vehicle and the vehicle in front approaches within a predetermined distance and the risk of collision with the vehicle in front increases, the electronic device 100 can generate and display a hazard warning object 1601 to indicate the risk of collision.

[0267] The attention warning object 1501 and the danger warning object 1601 are distinguished by different colors and sizes, thereby improving the driver's recognition. Furthermore, as an example, the warning objects 1501 and 1601 are composed of texture images, thus enabling augmented reality representation.

[0268] Furthermore, the electronic device 100 can also quantify the distance between its own vehicle and the vehicle in front, calculated by the distance measurement unit 140, and display it on the screen to make it easier for the driver to recognize the distance to the vehicle in front. As an example, when the distance measurement unit 140 calculates the distance between its own vehicle and the vehicle in front, the electronic device 100 can generate prompt objects 1502 and 1602 indicating the distance and display them on the screen.

[0269] In addition, electronic device 100 can also output various prompts via voice.

[0270] Figure 21 This is a schematic diagram illustrating the form of an electronic device according to an embodiment of the present invention when it does not have a camera unit. Figure 21 As shown, the vehicle imaging device 200, which is independently configured from the electronic device 100, can be configured as a system according to an embodiment of the present invention using wired / wireless communication methods.

[0271] The electronic device 100 may include a display unit 131, a user input unit 121, and a microphone unit 123 disposed on the front of the housing 191.

[0272] The vehicle imaging device 200 may include a camera 222, a microphone 224, and an adhesive part 281.

[0273] Figure 22This is a schematic diagram illustrating the form in which an electronic device according to an embodiment of the present invention includes a camera unit. Figure 22 As shown, when the electronic device 100 includes a camera unit 150, it can be a device in which the camera unit 150 of the electronic device 100 captures images of the front of the vehicle, and the user identifies the display portion of the electronic device 100. Thus, a system according to an embodiment of the present invention can be constructed.

[0274] Figure 23 This is a schematic diagram illustrating an embodiment of a HUD (Head-Up Display) according to an embodiment of the present invention. Figure 23 As shown, the HUD can communicate with other devices via wired / wireless communication to display augmented reality cues on the HUD.

[0275] As an example, augmented reality can be provided through image synthesis using a HUD on the vehicle's windshield or another image output device. As described above, the augmented reality providing unit 160 can generate realistic images or interface images overlaid on the glass. Thus, an augmented reality navigation system or a vehicle infotainment system can be constructed.

[0276] On the one hand, the inter-vehicle distance measuring device 10 is composed of a module of a system for autonomous driving, thereby enabling it to perform route suggestion functions. In this regard, refer to... Figure 24 as well as Figure 25 Please provide more specific details.

[0277] Figure 24 This is a block diagram illustrating the structure of an autonomous vehicle according to an embodiment of the present invention. Figure 24 As shown, the autonomous vehicle 2000 according to this embodiment may include a control device 2100, sensing modules 2004a, 2004b, 2004c, 2004d, an engine 2006, and a user interface 2008.

[0278] The autonomous vehicle 2000 can have either an autonomous driving mode or a manual mode. For example, it can switch from manual mode to autonomous driving mode, or vice versa, based on user input received through the user interface 2008.

[0279] When the vehicle 2000 is running in automatic driving mode, the automatic vehicle 2000 can operate under the control of the control device 2100.

[0280] In this embodiment, the control device 2100 may include a controller 2120, a sensor 2110, a wireless communication device 2130, and a target detection device 2140. The controller 2120 includes a memory 2122 and a processor 2124.

[0281] In this embodiment, the target detection device 2140 is used to detect targets located outside the vehicle 2000. The target detection device 2140 can detect targets located outside the vehicle 2000 and generate target information based on the detection results.

[0282] Target information may include information related to the presence of a target, the target's location, the distance between the vehicle and the target, and the relative speed between the vehicle and the target.

[0283] The targets can include various objects located outside the vehicle 2000, such as lane markings, other vehicles, pedestrians, traffic signals, light, roads, structures, speed bumps, ground features, and animals. Traffic signals can include traffic lights, traffic signs, patterns or text drawn on the road surface. Light can be light generated from lights on other vehicles, light generated from streetlights, or sunlight.

[0284] Then, structures can be objects located around roads and fixed to the ground. For example, structures can include streetlights, trees, buildings, utility poles, traffic lights, and bridges. Ground features can include hills, slopes, etc.

[0285] The target detection device 2140 may include a camera module. The controller 2120 can extract object information from external images captured by the camera module and process the information related thereto.

[0286] Additionally, the target detection device 2140 may also include an imaging device for identifying the external environment. Besides LIDAR (Light Detection and Ranging), RADAR (Radar), GPS (Global Positioning System) devices, distance measurement devices (Odometry), and other computer vision devices, ultrasonic sensors, and infrared sensors can be used. These devices can operate selectively or simultaneously as needed to achieve more precise sensing.

[0287] Additionally, sensor 2110 can be connected to sensing modules 2004a, 2004b, 2004c, and 2004d for sensing the vehicle's internal / external environment to acquire various sensing information. Sensor 2110 may include posture sensors (e.g., yaw, roll, and pitch sensors), impact sensors, wheel sensors, speed sensors, tilt sensors, weight sensors, heading sensors, gyroscope sensors, position modules, vehicle forward / reverse sensors, battery sensors, fuel sensors, tire sensors, steering sensors rotated by the steering wheel, vehicle interior temperature sensors, vehicle interior humidity sensors, ultrasonic sensors, illuminance sensors, accelerator pedal positioning sensors, brake pedal positioning sensors, etc.

[0288] Therefore, sensor 2110 can acquire sensing signals related to vehicle posture information, vehicle collision information, vehicle direction information, vehicle position information (GPS information), vehicle angle information, vehicle speed information, vehicle acceleration information, vehicle tilt information, vehicle forward / reverse information, battery information, fuel information, tire information, headlight information, vehicle interior temperature information, vehicle interior humidity information, steering wheel rotation angle, vehicle exterior illuminance, pressure applied to the accelerator pedal, and pressure applied to the brake pedal.

[0289] In addition, sensor 2110 may also include accelerator pedal sensor, pressure sensor, engine speed sensor, air flow sensor (AFS), intake air temperature sensor (ATS), coolant temperature sensor (WTS), throttle position sensor (TPS), TDC sensor, crank angle (CAS) sensor, etc.

[0290] As described above, sensor 2110 can generate vehicle status information based on the sensed data.

[0291] The wireless communication device 2130 is designed to enable wireless communication with the autonomous vehicle 2000. For example, the autonomous vehicle 2000 can communicate with a user's smartphone or other wireless communication device 2130, other vehicles, a central control unit (traffic control unit), a server, etc. The wireless communication device 2130 can transmit and receive wireless signals according to a connection wireless protocol. The wireless communication protocol can be Wi-Fi, Bluetooth, Long-Term Evolution (LTE), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), or Global Systems for Mobile Communications (GSM), but the specific communication protocol is not limited to these.

[0292] In addition, in this embodiment, the autonomous vehicle 2000 can also conduct vehicle-to-vehicle communication via the wireless communication device 2130. That is, the wireless communication device 2130 can communicate with other vehicles on the road and other vehicles via vehicle-to-vehicle communication. The autonomous vehicle 2000 can send and receive driving warnings, traffic information, and other information through vehicle-to-vehicle communication, and can also send and receive invitations to other vehicles. For example, the wireless communication device 2130 can conduct V2V communication via a dedicated short-range communication (DSRC) device or a C-V2V (Cellular-V2V) device. In addition to vehicle-to-vehicle communication, the wireless communication device 2130 can also conduct vehicle-to-everything (V2X) communication between vehicles and other things (e.g., electronic devices carried by pedestrians).

[0293] In this embodiment, the controller 2120, as a unit controlling the overall operation of various units within the vehicle 2000, can be set up by the vehicle manufacturer during vehicle manufacturing, or further set up after vehicle manufacturing to perform autonomous driving functions. Alternatively, it can include a structure where the controller 2120 set during manufacturing is upgraded to continuously perform additional functions. The controller 2120 can also be named ECU (Electronic Control Unit).

[0294] The controller 2120 can collect various data from connected sensors 2110, target detection device 2140, wireless communication device 2130, etc., and based on the collected data, transmit control signals to sensors 2110, engine 2006, user interface 2008, wireless communication device 2130, and target detection device 2140, including other structures within the vehicle. Additionally, although not shown, control signals can also be transmitted to acceleration devices, braking systems, steering devices, or navigation systems related to vehicle operation.

[0295] In this embodiment, the controller 2120 can control the engine 2006. For example, it can sense the speed limit of the road on which the autonomous vehicle 2000 is traveling and control the engine 2006 to ensure that the driving speed does not exceed the speed limit, or control the engine 2006 to accelerate the autonomous vehicle 2000 within the speed limit range.

[0296] Additionally, the controller 2120 can sense the distance between the autonomous vehicle 2000 and the vehicle in front while the vehicle is driving, and control the engine 2006 to control the vehicle speed based on the inter-vehicle distance. Specifically, the autonomous vehicle 2000 may be equipped with an inter-vehicle distance measuring device 10 according to an embodiment of the present invention. The inter-vehicle distance measuring device 10 can measure the distance between the vehicle 1000 and the target vehicle, and can transmit the measured inter-vehicle distance value to the controller 2120.

[0297] In this situation, the controller 2120 can control the deceleration, acceleration, and speed maintenance of the vehicle 2000 based on the vehicle distance information acquired by the vehicle distance measuring device 10, thereby controlling the vehicle's automatic driving. Specifically, when the acquired vehicle distance is less than a preset distance, the controller 2120 can control various units on the vehicle 2000 (brakes, steering wheel, etc.) to control the vehicle 2000's speed to decelerate from the current speed to a predetermined speed or to stop the vehicle 2000. That is, the controller 2120 can control the automatic driving of the vehicle 2000 based on the vehicle distance acquired by the vehicle distance measuring device 10.

[0298] In another embodiment of the present invention, the controller 2120 may also send instructions to the drive unit of the vehicle 2000 to control the driving speed so that the inter-vehicle distance obtained by the inter-vehicle distance measuring device 10 is kept at a preset certain distance.

[0299] Furthermore, according to another embodiment of the present invention, when the vehicle-to-vehicle distance obtained by the vehicle-to-vehicle distance measuring device 10 is greater than a preset distance, the controller 2120 can control various units (brakes, steering wheel, etc.) on the vehicle 2000 to accelerate the speed of the vehicle 2000 from the current speed to a predetermined speed. That is, the controller 2120 can control the automatic driving of the vehicle 2000 based on the vehicle-to-vehicle distance obtained by the vehicle-to-vehicle distance measuring device 10.

[0300] The workshop distance measuring device 10 can be constituted by a module within the control device 2100 of the autonomous vehicle 2000. That is, the memory 2122 and processor 2124 of the control device 2100 can be configured to implement the workshop distance measuring method according to the present invention through software.

[0301] When other vehicles or obstacles are in front of the vehicle, the engine 2006 or braking system can be controlled to slow the vehicle down. In addition to speed, the trajectory, driving route, and steering angle can also be controlled. Furthermore, the controller 2120 can generate the necessary control signals based on the vehicle's driving lanes, driving signals, and other external environmental information to control the vehicle's movement.

[0302] In addition to generating its own control signals, the controller 2120 can also communicate with surrounding vehicles or a central server, and transmit instructions for controlling surrounding devices through the received information, thereby controlling the vehicle's movement.

[0303] Furthermore, when the position or shooting angle of the camera module 2150 changes, it may be difficult to accurately identify vehicles or lane lines according to this embodiment. Therefore, to prevent this, the controller 2120 can also generate a control signal to control the calibration of the camera module 2150. Thus, in this embodiment, the controller 2120 generates a calibration control signal to the camera module 2150, so that even if the installation position of the camera module 2150 changes due to vibration or impact caused by the movement of the autonomous vehicle 2000, the camera module 2150 can maintain its normal installation position, orientation, and shooting angle. When the pre-stored initial installation position, orientation, and shooting angle information of the camera module 2150 and the initial installation position, orientation, and shooting angle information of the camera module 2150 measured during the driving of the autonomous vehicle 2000 change and exceed a critical value, the controller 2120 can generate a control signal to perform the calibration of the camera module 2150.

[0304] In this embodiment, the controller 2120 may include a memory 2122 and a processor 2124. The processor 2124 can run software stored in the memory 2122 according to control signals from the controller 2120. Specifically, the controller 2120 may store data and instructions for performing the workshop distance measurement method according to the present invention in the memory 2122, and the processor 2124 may execute the instructions to implement one or more methods disclosed herein.

[0305] At this time, memory 2122 can be stored in a recording medium executable by non-volatile processor 2124. Memory 2122 can store software and data through appropriate internal and external devices. Memory 2122 can be composed of RAM (random access memory), ROM (read only memory), hard disk, or a memory 2122 device connected to an adapter.

[0306] The memory 2122 can store at least the operating system (OS), user applications, and executable instructions. The memory 2122 can also store application data and arranged data structures.

[0307] The processor 2124, as a microprocessor or a suitable electronic processor, can be a controller, microcontroller, or state machine.

[0308] The processor 2124 may be embodied in a combination of computing devices, which may consist of a digital signal processor, a microprocessor, or a combination thereof.

[0309] On one hand, the autonomous vehicle 2000 may further include a user interface 2008, which is used to input information from the user in relation to the aforementioned control device 2100. The user interface 2008 can provide information input to the user through appropriate interaction. For example, it may consist of a touchscreen, a keyboard, operation buttons, etc. The user interface 2008 can transmit input or instructions to the controller 2120, which, in response to the input or instructions, can control the vehicle's movements.

[0310] In addition, the user interface 2008 enables external devices of the autonomous vehicle 2000 to communicate with the autonomous vehicle 2000 via the wireless communication device 2130. For example, the user interface 2008 can be connected to a smartphone, tablet, or other computer device.

[0311] Furthermore, while this embodiment describes the autonomous vehicle 2000 as including an engine 2006, it may also include other types of propulsion systems. For example, the vehicle may operate on electric power, hydrogen power, or a hybrid system combining both. Therefore, the controller 2120 may include propulsion mechanisms based on the propulsion system of the autonomous vehicle 2000, and then provide control signals based thereon to the respective propulsion mechanism structures.

[0312] The following is for reference. Figure 25 The specific structure of the control device 2100 for performing the workshop distance measurement method according to the present embodiment will be described in more detail.

[0313] The control device 2100 includes a processor 2124. The processor 2124 can be a common single or multiple chip microprocessor, a dedicated microprocessor, a microcontroller, a programmable gate array, etc. The processor can also be referred to as a central processing unit (CPU). Furthermore, in this embodiment, the processor 2124 can also be used in combination of multiple processors.

[0314] The control device 2100 also includes a memory 2122. The memory 2122 can also be any electronic component capable of storing electronic information. The memory 2122 can also include combinations of memories other than a single memory.

[0315] Data and instructions 2122a used to execute the workshop distance measurement method according to the present invention can also be stored in memory 2122. When processor 2124 executes instruction 2122a, all or part of instruction 2122a and data 2122b required for executing the instruction can also be loaded onto processor 2124a, 2124b.

[0316] The control device 2100 may also include a transmitter 2130a, a receiver 2130b, or a transceiver 2130c that allows signal transmission and reception. One or more antennas 2132a, 2132b may also be electrically connected to the transmitter 2130a, the receiver 2130b, or each of the transceivers 2130c, and may further include wires.

[0317] The control unit 2100 may also include a digital signal processor (DSP) 2170. The DSP 2170 enables the vehicle to process digital signals quickly.

[0318] The control device 2100 may also include a communication interface 2180. The communication interface 2180 may further include one or more interfaces and / or communication modules for connecting other devices to the control device 2100. The communication interface 2180 allows a user to interact with the control device 2100.

[0319] The various structures of the control device 2100 can also be connected into one unit via one or more buses 2190. The bus 2190 may also include a power bus, a control signal bus, a status signal bus, a data bus, etc. Under the control of the processor 2124, the structures can transmit mutual information through the bus 2190 and perform the expected functions.

[0320] On the one hand, for ease of explanation, the above embodiments have been described using the example of calculating the distance between a reference vehicle and the vehicle in front, but this is not a limitation. The inter-vehicle distance measurement method according to the present invention can also be applied in the same way to the case of calculating the distance between a reference vehicle and the vehicle behind.

[0321] On the one hand, the terms "first," "second," "third," and "fourth," as used in the specification and claims, are used, if described herein, to distinguish between similar constituent elements and are not fixed; they are used to indicate a specific order or sequence of occurrence. Terms used interchangeably with these terms should be understood to be interchangeable in appropriate contexts so that embodiments of the invention described herein can be performed, for example, in other orders not shown or described herein. Similarly, when a method is described herein as comprising a series of steps, the order of these steps disclosed herein is not necessarily the order in which they are performed; any described steps may be omitted and / or any other steps not described herein may be added to the method. For example, without departing from the scope of the invention, a first constituent element may be named a second constituent element, and similarly, a second constituent element may be named a first constituent element.

[0322] Furthermore, the terms "left side," "right side," "front," "rear," "upper," "bottom," "above," and "below" used in the specification and claims are for illustrative purposes and do not necessarily indicate fixed relative positions. Terms used interchangeably with these terms should be understood as being interchangeable in suitable contexts so that embodiments of the invention described herein can operate in directions other than those shown or described herein. The term "connection" as used herein is defined as a direct or indirect connection by electrical or non-electrical means. Objects described herein as "adjacent" may, depending on the context in which the term is used, be objects that are physically in contact with each other, adjacent to each other, or within the same general scope or area. The phrase "in one embodiment" as used herein refers to the same embodiment, but is not always the case.

[0323] In addition, various variations of the terms "connected," "linked," "linked," "connected," "combined," and "combined" used in the specification and the scope of the claims are used to mean that they are directly connected to or indirectly connected to other constituent elements.

[0324] Conversely, when it comes to a constituent element being "directly connected" or "directly linked" to other constituent elements, it should be understood that there are no other constituent elements in between.

[0325] Furthermore, the suffixes “module” and “section” used for constituent elements in this specification are assigned or used interchangeably only for the convenience of writing the specification, and do not have the meaning or function of distinguishing each other.

[0326] Furthermore, the terminology used in this specification is for illustrative purposes and not for limiting the invention. Unless otherwise explicitly stated in the text, the singular forms used herein include the plural forms. In this application, terms such as "constituting" or "comprising" should not be construed as necessarily including all of the multiple constituent elements or steps described in the specification, but should be interpreted as possibly excluding some of the constituent elements or steps, or possibly including further constituent elements or steps.

[0327] The foregoing description focuses on preferred embodiments of the present invention. All embodiments and conditional examples disclosed in this specification are provided to facilitate understanding of the principles and concepts of the invention by those skilled in the art. It should be understood that the invention can be embodied in modified forms without departing from its essential characteristics.

[0328] Therefore, the disclosed embodiments should be considered in the context of illustration rather than limitation. The scope of the invention is set forth in the claims rather than in the description, and all differences within the same scope should be interpreted as being included in the invention.

[0329] On the one hand, the workshop distance measurement methods described above according to various embodiments of the present invention can be embodied in a program and provided to a server or device. Thus, each device connects to a server or device storing the program, thereby enabling the download of the program.

[0330] Furthermore, the methods described above according to various embodiments of the present invention can be embodied in a program and provided by storing them in various non-transitory computer-readable media. A non-transitory computer-readable medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short period of time, such as registers, caches, or memory. Specifically, the various applications or programs described above can be provided by storing them in non-transitory computer-readable media such as CDs, DVDs, hard disks, Blu-ray discs, USB drives, memory cards, and ROMs.

[0331] Furthermore, the preferred embodiments of the present invention have been shown and described above. However, the present invention is not limited to the specific embodiments described above. Those skilled in the art can make various modifications without departing from the spirit of the present invention as claimed in the claims. Moreover, such modifications should not be understood separately from the technical concept or prospects of the present invention.

Claims

1. A workshop distance measurement method, which is a workshop distance measurement method utilizing a processor, characterized in that, include: The steps for acquiring driving images captured by a camera on a moving vehicle; The steps of detecting a second vehicle from the acquired driving images using machine learning or deep learning; as well as The step of calculating the inter-vehicle distance between the detected second vehicle and the first vehicle. The calculations mentioned above include: When the second vehicle is detected through the machine learning or deep learning, the distance to the detected second vehicle is calculated; and When the second vehicle is not detected by the machine learning or deep learning, the distance to the second vehicle is calculated based on the change value of the feature points detected in the second vehicle region in the first frame of the driving image, which corresponds to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected. The variation values ​​of the feature points include: Detect feature points in the first frame and feature points in the current frame that correspond to the feature points in the first frame; For the first frame and the current frame, respectively calculate the average pixel distance, which is the average pixel distance from the average pixel position of the feature points to the pixel distance between each feature point; and Calculate the ratio between the average pixel distance of the first frame and the average pixel distance of the current frame.

2. The workshop distance measurement method according to claim 1, characterized in that, The steps for detecting the second vehicle include: The steps to construct a vehicle learning dataset; Selective dataset learning steps; and Vehicle inspection procedures.

3. The workshop distance measurement method according to claim 2, characterized in that, The steps for constructing the vehicle learning dataset include: The steps to acquire the images needed for learning; The steps to extract the vehicle region to be learned from the acquired learning images; and The steps for annotating the attributes of the vehicles contained in the harvested vehicle area. The vehicle attributes annotated include the type of vehicle and the distance of vehicles within the image.

4. The workshop distance measurement method according to claim 3, characterized in that, The selective dataset learning steps include: The steps to extract features from the constructed dataset; The steps for classifying vehicles using the extracted features; and The steps involved in learning and enhancing the process of classifying vehicles are to generate a classifier for classifying vehicles.

5. The workshop distance measurement method according to claim 4, characterized in that, The feature extraction step utilizes at least one of the following to extract the features: (i) grayscale intensity, (ii) red, green, and blue RGB color information, (iii) chroma, saturation, and purity HSV color information, (iv) YIQ color information, and (v) edge information (grayscale, binary, and eroded binary).

6. The workshop distance measurement method according to claim 4, characterized in that, The vehicle inspection steps include: The steps are: extracting features from the input image, using the classifier to detect vehicles based on the extracted features, and outputting the detected results.

7. The workshop distance measurement method according to claim 1, characterized in that, The steps for calculating the workshop distance include: The steps of identifying the left and right boundaries of the second vehicle from the image of the detected second vehicle; and The step of determining the image width between the identified left boundary and the identified right boundary as the image width of the second vehicle.

8. The workshop distance measurement method according to claim 7, characterized in that, The step of calculating the workshop distance further includes: The steps include determining a first pivot point where the line representing the position of the second vehicle in the lane intersects with the left lane line and a second pivot point where the line representing the position of the second vehicle in the lane intersects with the right lane line, and determining the image width between the first pivot point and the second pivot point as the image width LaneW of the lane where the second vehicle is located.

9. The workshop distance measurement method according to claim 8, characterized in that, The step of calculating the workshop distance further includes: The step of calculating the image width ratio between the calculated image width of the second vehicle and the calculated image width of the lane in which the second vehicle is located; The step of determining the size class of the second vehicle based on the calculated proportions; and The step of calculating the predicted width of the second vehicle based on the determined size class of the second vehicle.

10. The workshop distance measurement method according to claim 7, characterized in that, The step of determining the image width of the second vehicle is to harvest the detected vehicle region. For the harvested region, vertical edges are detected by Sobel calculus. The cumulative value of the vertical histogram is calculated on the detected vertical edges, and the point where the maximum histogram value is located is detected as the left and right boundary positions of the vehicle.

11. The workshop distance measurement method according to claim 10, characterized in that, The step of determining the image width of the second vehicle further includes the step of fitting the vehicle region to the detected left and right boundary positions of the vehicle.

12. The workshop distance measurement method according to claim 1, characterized in that, Further includes: If the second vehicle is not detected from the driving image, then in the plurality of frames constituting the driving image, the step of detecting a first feature point of the second vehicle region from the first frame corresponding to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected is performed. The steps of tracking the detected first feature point and detecting the second feature point in the second frame corresponding to the current frame; The step of calculating the feature point change value between the first feature point and the second feature point; as well as The step of calculating the inter-vehicle distance from the camera device of the first vehicle to the second vehicle based on the calculated feature point change values.

13. A workshop distance measuring device, characterized in that, include: The image acquisition unit acquires driving images captured by the camera of the first vehicle in motion; The vehicle inspection department uses machine learning or deep learning to detect a second vehicle from the acquired driving images; as well as The vehicle-to-vehicle distance calculation unit calculates the vehicle-to-vehicle distance between the detected second vehicle and the first vehicle. The workshop distance calculation unit is configured as follows: When the second vehicle is detected through the machine learning or deep learning, the distance to the detected second vehicle is calculated; and When the second vehicle is not detected by the machine learning or deep learning, the distance to the second vehicle is calculated based on the change value of the feature points detected in the second vehicle region in the first frame of the driving image, which corresponds to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected. The variation values ​​of the feature points include: Detect feature points in the first frame and feature points in the current frame that correspond to the feature points in the first frame; For the first frame and the current frame, respectively calculate the average pixel distance, which is the average pixel distance from the average pixel position of the feature points to the pixel distance between each feature point; and Calculate the ratio between the average pixel distance of the first frame and the average pixel distance of the current frame.

14. The workshop distance measuring device according to claim 13, characterized in that, The vehicle inspection department: Construct a vehicle learning dataset; Learning from selective datasets; and Perform vehicle inspection.

15. The workshop distance measuring device according to claim 14, characterized in that, To construct the vehicle learning dataset, the vehicle detection department: Obtain the images needed for learning; The acquired learning images reveal the vehicle regions that need to be learned; and Annotate the attributes of the vehicles included in the harvested vehicle area. The vehicle attributes annotated include the type of vehicle and the distance of vehicles within the image.

16. The workshop distance measuring device according to claim 15, characterized in that, In order to learn the selected dataset, the vehicle detection department: Extract features from the constructed dataset; Vehicles are classified using the extracted features; as well as By learning and enhancing the process of classifying vehicles, a classifier for classifying vehicles is generated.

17. The workshop distance measuring device according to claim 16, characterized in that, The vehicle detection unit uses at least one of the following to extract the features: (i) grayscale intensity, (ii) red, green, and blue RGB color information, (iii) chromaticity, saturation, and purity HSV color information, (iv) YIQ color information, and (v) edge information (grayscale, binary, and eroded binary).

18. The workshop distance measuring device according to claim 16, characterized in that, The vehicle detection unit extracts features from the input image, uses the classifier to detect vehicles based on the extracted features, and outputs the detected results.

19. The workshop distance measuring device according to claim 13, characterized in that, The inter-vehicle distance calculation unit identifies the left and right boundaries of the second vehicle from the detected image of the second vehicle; and The image width between the identified left boundary and the identified right boundary is determined as the image width of the second vehicle.

20. The workshop distance measuring device according to claim 19, characterized in that, The vehicle distance calculation unit determines a first pivot point where the line representing the position of the second vehicle in the lane intersects with the left lane line and a second pivot point where the line representing the position of the second vehicle in the lane intersects with the right lane line, and determines the image width between the first pivot point and the second pivot point as the image width LaneW of the lane where the second vehicle is located.

21. The workshop distance measuring device according to claim 20, characterized in that, The workshop distance calculation unit calculates the image width ratio between the calculated image width of the second vehicle and the calculated image width of the lane where the second vehicle is located. The size class of the second vehicle is determined based on the calculated proportions; and The predicted width of the second vehicle is calculated based on the determined size class of the second vehicle.

22. The workshop distance measuring device according to claim 19, characterized in that, The workshop distance calculation unit detects the vehicle area and uses Sobel's algorithm to detect the vertical edges of the detected area. It calculates the cumulative value of the vertical histogram at the detected vertical edges and detects the point where the maximum histogram value is located as the left and right boundary positions of the vehicle.

23. The workshop distance measuring device according to claim 22, characterized in that, The workshop distance calculation unit fits the vehicle area to the left and right boundary positions of the detected vehicle.

24. The workshop distance measuring device according to claim 13, characterized in that, Further includes: If the second vehicle is not detected in the driving image, the feature point detection unit detects a first feature point of the second vehicle region in the first frame corresponding to the frame in which the second vehicle was detected before the frame in which the second vehicle was not detected, and tracks the detected first feature point, and detects a second feature point in the second frame corresponding to the current frame. The feature point change value calculation unit calculates the feature point change value between the first feature point and the second feature point; The workshop distance calculation unit calculates the workshop distance from the camera device of the first vehicle to the second vehicle based on the calculated feature point change value.

25. A computer-readable recording medium, characterized in that, It contains a computer program for performing the workshop distance measurement method according to any one of claims 1 to 12.

Images