TWI702577B - A method for generating a driving assistance image utilizing in a vehicle and a system thereof - Google Patents

Abstract

A method for generating a driving assistance image utilizing in a vehicle, comprising : (a) determining whether an object on the road is existing based on a detected result; (b) determining whether a distance between the object and the vehicle is more than a threshold value based on the detected result, when the object is existed; (c) selecting a first 3-dimention projection model when the distance between the object and the vehicle is equal or less than the threshold value; (d) selecting a second 3-dimention projection model when the distance between the object and the vehicle is more than the threshold value; (e) projecting an image surrounding the vehicle to the selected first 3-dimention projection model or the second 3-dimention projection model; and (f) generating a first field of view driving assistance image or a second field of view driving assistance image based on the result of step (e).

Description

Method and system for generating driving auxiliary image

The invention relates to an image generation method, in particular to a driving assistance image generation method for vehicles.

The car around-view system is a kind of advanced driver assistance system (ADAS) technology, which can instantly show the driver a bird's-eye 360-degree panoramic image of a vehicle and its surroundings to ensure parking or other low-speed driving conditions Driving safety. The existing car surround view display system obtains the image of the surrounding environment of the vehicle by using several wide-angle cameras installed at the front and rear and on both sides of the vehicle, and then converts these images through an inverse perspective projection based on a fixed three-dimensional projection model. The image obtained by the wide-angle camera is converted into a three-dimensional projection model, and then a perspective projection conversion is performed on the three-dimensional projection model from different viewing angles through a virtual camera to obtain two-dimensional images with different viewing angles.

However, the shape and size of the three-dimensional projection model used in the existing automotive surround view display system are fixed, and cannot be dynamically adjusted according to changes in the driving situation, so it cannot provide a good user experience in use.

Therefore, the object of the present invention is to provide a driving assistance image generation method and system that can dynamically adjust the functions and parameters of the three-dimensional projection model according to changes in the driving situation to provide a better user experience.

Therefore, the driving assistance image generation method of the present invention is suitable for a vehicle and is implemented by a processing module that is electrically connected to a photographing unit and an obstacle detection module. The photographing unit is used for photographing. And transmitting at least a surrounding image of the vehicle to the processing module, and the obstacle detection module is used to generate and transmit an obstacle detection result related to an obstacle corresponding to the vehicle to the processing module; The method includes the following steps:

(A) The processing module determines whether the obstacle detection result indicates the existence of the obstacle;

(B) When the processing module determines that the obstacle exists, wherein the obstacle detection result includes an obstacle distance between the obstacle and the vehicle, the processing module determines whether the obstacle distance is greater than a distance Threshold

(C) When the processing module determines that the obstacle distance is less than or equal to the distance threshold, the processing module selects a first three-dimensional projection model with an elliptical arc;

(D) When the processing module determines that the obstacle distance is greater than the distance threshold, the processing module selects a second three-dimensional projection model with a circular arc;

(E) The processing module uses an inverse perspective projection conversion method according to the selected first three-dimensional projection model or the second three-dimensional projection model to map the at least surrounding image to the first three-dimensional projection model or the second three-dimensional projection model Two and three-dimensional projection model; and

(F) The processing module uses a perspective projection conversion method to convert the result of step (E) into a corresponding first-view driving assistance image or a second-view driving assistance image.

Therefore, the driving assistance image generating system of the present invention can be used to implement the above driving assistance image generating method.

The effect of the present invention is that the processing module adopts different three-dimensional projection models according to the obstacle distance between the obstacle and the vehicle or the speed of the vehicle, thereby achieving a function of dynamically adjusting the three-dimensional projection model in response to changes in the driving situation Formulas and parameters, respectively corresponding to the situation to provide driving images of different fields of view, thereby providing the driver with a better experience. In addition, it can predict the next driving position of the vehicle based on the vehicle information and provide predictive driving assistance images.

Referring to FIG. 1, an embodiment of the driving assistance image generation system of the present invention is applicable to a vehicle. The driving assistance image generation system includes a shooting unit 11, an obstacle detection module 12, a sensing module 13, a lane line detection module 14 electrically connected to the shooting unit 11, and an electrical connection to the shooting The processing module 15 of the unit 11, the obstacle detection module 12, the sensing module 13 and the lane line detection module 14.

The photographing unit 11 includes four photographing modules (not shown in the figure) respectively arranged on a front bumper, a rear bumper, a left rear mirror, and a right rear mirror of the vehicle. For example, it is a fisheye camera. The shooting modules are used to photograph the surrounding environment of the vehicle to generate and transmit four surrounding images related to the surrounding environment of the vehicle to the processing module 15 respectively.

The obstacle detection module 12 is used to generate and transmit an obstacle detection result related to an obstacle corresponding to the vehicle to the processing module 15. Wherein, the obstacle detection module 12 uses a conventional obstacle detection technology to determine whether there is an obstacle corresponding to the vehicle, and when it is determined that the obstacle exists, it also detects the obstacle An obstacle distance from the vehicle and an obstacle azimuth angle of the obstacle relative to the vehicle, and the obstacle indicating the existence of the obstacle and including the obstacle distance and the obstacle azimuth angle is generated Detection result; and when it is determined that the obstacle does not exist, the obstacle detection result indicating that the obstacle does not exist is generated. In this embodiment, the obstacle detection module 12 includes a radar detector (not shown in the figure), and the distance between the obstacle and the vehicle is detected by the radar detector. And the obstacle azimuth angle of the obstacle relative to the vehicle; however, in other embodiments, the obstacle detection module 12 may also be based on the surrounding surroundings corresponding to the vehicle's previous environment captured by the photographing unit 11 Image to obtain the obstacle distance between the obstacle and the vehicle, and the obstacle azimuth angle of the obstacle relative to the vehicle, but not limited to this.

The sensing module 13 is used for sensing and transmitting a vehicle information related to the vehicle to the processing module 15, wherein the vehicle information includes a speed of the vehicle, indicating whether the vehicle's direction lights are turned on A turn signal parameter group, and a corner signal of the vehicle. Since the feature of the present invention does not lie in the manner in which the sensor module 13 obtains the vehicle information known to those skilled in the art, for the sake of brevity, their details are omitted here.

The lane line detection module 14 is used to generate and transmit a lane line detection result related to the lane on which the vehicle is traveling to the processing based on the surrounding image corresponding to the environment in front of the vehicle captured by the shooting unit 11 Module 15. Wherein, the lane line detection module 14 uses a conventional lane line recognition technology to determine whether there is at least one lane line corresponding to the lane in the surrounding image corresponding to the environment in front of the vehicle. When there is at least one lane line, a detection result of the lane line indicating the presence of the at least one lane line and including the identified at least one lane line is generated; and when it is determined that the obstacle does not exist, an indication is generated indicating the absence The lane line detection result of the at least one lane line.

The operation details of the components in the driving assistance image generation system will be described below in conjunction with the embodiment of the driving assistance image generation method of the present invention. This embodiment sequentially includes a projection model determination program and a row assistance image generation program.

1 and 2, the driving assistance image generation system implements the projection model determination procedure and the driving assistance image generation procedure of the driving assistance image generation method of the present invention, and includes the following steps. The projection model determination program explains how to choose different three-dimensional projection models according to different driving situations, and the driving assistance image generation program explains how to obtain driving assistance images with different fields of view.

In step 201, the obstacle detection module 12 continuously generates and transmits the obstacle detection result related to the obstacle corresponding to the vehicle to the processing module 15.

In step 202, the sensing module 13 continuously generates and transmits the vehicle information related to the vehicle to the processing module 15.

In step 203, the lane line detection module 14 continuously generates and transmits the lane line detection results related to the lane in which the vehicle is traveling to the processing module 15.

In step 204, the processing module 15 determines whether the received obstacle detection result indicates the presence of the obstacle. When the processing module 15 determines that the obstacle detection result indicates that the obstacle exists, the process proceeds to step 205; when the processing module 15 determines that the obstacle detection result indicates that the obstacle does not exist, The process proceeds to step 208.

In step 205, the processing module 15 determines whether the obstacle distance in the obstacle detection result is greater than a distance threshold. When the processing module 15 determines that the obstacle distance is less than or equal to the distance threshold, the process proceeds to step 206; when the processing module 15 determines that the obstacle distance is greater than the distance threshold, the process proceeds to step 207. It is worth mentioning that the distance threshold is the product of the speed of the vehicle and a preset time, and the preset time may be the time to last point to brake. In this embodiment , The preset time is 2.5 seconds. The processing module 15 first obtains the distance threshold value according to the vehicle speed sensed by the sensing module 13 and the preset time, and then determines whether the obstacle distance is greater than the distance threshold value.

…(1) In step 206, the processing module 15 selects a first three-dimensional projection model with an elliptical arc (see FIG. 3), and the first three-dimensional projection model is on a first plane formed by an X axis and a Z axis The projection of the first three-dimensional projection model is shown in Figure 4, and the projection of the first three-dimensional projection model on a second plane formed by a Y axis and the Z axis is shown in Figure 5. The projection on the third plane formed by the Y axis is shown in FIG. 6. Wherein, the curve 501 (see FIG. 4) projected on the first plane by the curved surface of the first three-dimensional projection model and the curve 502 (see FIG. 5) projected on the second plane are both an elliptical arc. The function of the first three-dimensional projection model can be expressed as the following formula (1).

…(1)

為車輛之長度，H為對應於該第一三維投影模型之一虛擬相機701的高度，

，

，

為該車輛之寬度，

為一預設距離。在本實施方式中該預設距離

設定為2公尺，然而，

之值可視使用需求而調整，並不以此為限，

之值越大，該第一三維投影模型之曲面在該第一平面上投影出之橢圓弧線，與在該第二平面上投影出之橢圓弧線皆越扁平；

之值越小，該第一三維投影模型之曲面在該第一平面上投影出之橢圓弧線，與在該第二平面上投影出之橢圓弧線皆越接近圓弧線。當

之值設為0時，該第一三維投影模型之曲面在該第一平面上投影出之曲線，與在該第二平面上投影出之曲線即為圓弧線。 among them,

Is the length of the vehicle, H is the height of a virtual camera 701 corresponding to the first three-dimensional projection model,

,

,

Is the width of the vehicle,

Is a preset distance. In this embodiment, the preset distance

Set to 2 meters, however,

The value can be adjusted according to the needs of use, and is not limited to this.

The larger the value, the flatter the elliptical arc projected on the first plane by the curved surface of the first three-dimensional projection model and the elliptical arc projected on the second plane;

The smaller the value is, the elliptical arc projected by the curved surface of the first three-dimensional projection model on the first plane and the elliptical arc projected on the second plane are closer to the circular arc. when

When the value is set to 0, the curve projected by the curved surface of the first three-dimensional projection model on the first plane and the curve projected on the second plane are arc lines.

...(2) In step 207, the processing module 15 selects a second three-dimensional projection model with a circular arc line (see FIG. 7). The projection of the second three-dimensional projection model on the first plane is as shown in FIG. The projection of the second three-dimensional projection model on the second plane is shown in FIG. 9, and the projection of the second three-dimensional projection model on the third plane is shown in FIG. 10. Wherein, the curve 503 (see FIG. 8) projected on the first plane by the curved surface of the second three-dimensional projection model and the curve 504 (see FIG. 9) projected on the second plane are both a circular arc. . The function of the second three-dimensional projection model can be expressed as the following formula (2).

...(2)

，在本實施方式中，該第一三維投影模型之該虛擬相機701的高度與該第二三維投影模型之該虛擬相機701的高度相同，亦即H’= H。類似地，該第二三維投影模型中的

之值亦可視使用需求而調整，並不以此為限，

之值越大，該第二三維投影模型之曲面在該第一平面上投影出之圓弧線，與在該第二平面上投影出之圓弧線皆越接近直線。當

之值趨近無窮大時，該第二三維投影模型概呈圓柱型。 Wherein, H'corresponds to the height of a virtual camera 701 of the second three-dimensional projection model,

In this embodiment, the height of the virtual camera 701 of the first three-dimensional projection model is the same as the height of the virtual camera 701 of the second three-dimensional projection model, that is, H′=H. Similarly, the second three-dimensional projection model

The value can also be adjusted according to the needs of use, and is not limited to this.

The larger the value, the closer the arc line projected by the curved surface of the second three-dimensional projection model on the first plane to the arc line projected on the second plane. when

When the value of is approaching infinity, the second three-dimensional projection model is almost cylindrical.

In step 208, the processing module 15 determines whether the vehicle speed in the received vehicle information is greater than a speed threshold. When the processing module 15 determines that the vehicle speed is not greater than the speed threshold, the process proceeds to step 209; when the processing module 15 determines that the vehicle speed is greater than the speed threshold, the process proceeds to step 210.

In step 209, the processing module 15 selects the first three-dimensional projection model (see FIG. 3).

In step 210, the processing module 15 selects the second three-dimensional projection model (see FIG. 7).

In step 211, the processing module 15 uses an inverse perspective projection transformation method according to the selected first three-dimensional projection model or the second three-dimensional projection model to map the surrounding images to the first three-dimensional projection model or The second three-dimensional projection model.

When the obstacle distance is less than or equal to the distance threshold, the advantage of selecting the first three-dimensional projection model is that the first-view driving assistance image (see Figure 14) obtained according to the first three-dimensional projection model has a relatively large proximity Visual range; when the obstacle distance is greater than the distance threshold, the advantage of selecting the second three-dimensional projection model is that the second-view driving assistance image (see Figure 15) obtained according to the second and third projection model The effect of magnifying distant objects. If the obstacle does not exist, since the first-view auxiliary driving image obtained based on the first three-dimensional projection model has a larger near-view range, when the vehicle speed is low, the first three-dimensional projection model will be more Good visual experience; because the second-view driving assistance image obtained based on the second three-dimensional projection model has the effect of magnifying distant objects, when the vehicle speed is high, the second three-dimensional projection model will have better vision Experience. Therefore, it can be seen from Figure 11, Figure 14 and Figure 15 that the imaging using the first three-dimensional projection model (see the solid line 102 in Figure 11) will have a larger near visual range, and the second three-dimensional projection model (see Figure 11) The imaging at the dotted line 101) will have the effect of magnifying distant objects. Therefore, switching between different 3D projection models according to different obstacle distances can provide a better visual experience for the driver.

In step 212, the processing module 15 converts the result of mapping the surrounding images to the first three-dimensional projection model or the second three-dimensional projection model into a corresponding first-view driving assistance image or A second-view driving assistance image.

The following will describe the details of the driving assistance image generation program with the attached illustration. The driving assistance image generation program can project the generated three-dimensional projection model into driving assistance images with different viewing angles according to different driving situations. Whether the obstacle detection result indicates the existence of the obstacle, the driving assistance image generation process can be further subdivided into a first generation process and a second generation process.

1 and 12, when the obstacle detection result indicates that the obstacle exists, the driving assistance image generation system uses the first generation process to obtain driving assistance images of different fields of view, and includes the following steps.

In step 301, the processing module 15 calculates a relationship related to the height of the virtual camera 701 (see FIG. 3 and FIG. 7) based on the obstacle distance of the obstacle detection result and the height of the virtual camera 701. The first tilt angle.

In step 302, the processing module 15 uses perspective projection conversion based on the results of the surrounding images being mapped to the first three-dimensional projection model or the second three-dimensional projection model, the first pan angle and the obstacle azimuth To convert a corresponding two-dimensional first-view driving assistance image or a second-view driving assistance image.

…(3) It is worth mentioning that the calculation formula of the perspective projection conversion is as the following formula (3).

…(3)

)為該第一三維投影模型或該第二三維投影模型中待投影至該第一視野行車輔助影像或該第二視野行車輔助影像的點，(u,v)為(

)經投影後的二維點，[R]=

，

=0，

為該第一俯攝角度，

為該障礙物方位角，[T]=

，(

)為該虛擬相機701的位置，

、

表示該虛擬相機701在對應二維點(u,v)之U、V軸的焦距，

、

表示第一視野行車輔助影像或該第二視野行車輔助影像的圖像中心點。 among them,(

) Is a point in the first three-dimensional projection model or the second three-dimensional projection model to be projected to the first-view driving auxiliary image or the second-view driving auxiliary image, (u, v) is (

) Two-dimensional point after projection, [R]=

,

=0,

Is the first tilt angle,

Is the azimuth angle of the obstacle, [T]=

, (

) Is the position of the virtual camera 701,

,

Indicates the focal length of the virtual camera 701 on the U and V axes corresponding to the two-dimensional point (u, v),

,

Represents the image center point of the first-view driving assistance image or the second-view driving assistance image.

1 and 13, when the obstacle detection result indicates that there is no obstacle, the driving assistance image generation system uses the second generation process to obtain driving assistance images of different fields of view, and includes the following steps.

In step 304, the processing module 15 determines whether one of the turn signals of the vehicle is turned on according to the turn signal parameter group of the vehicle information. When the processing module 15 determines that the direction lights are turned on, the process proceeds to step 305; when the processing module 15 determines that the direction lights are not turned on, the process proceeds to step 307.

In step 305, the processing module 15 calculates a steering azimuth corresponding to the turned-on turn-on light according to the turned-on turn-on light indicated by the turn-on light parameter group.

=0，

為該第二俯攝角度，

為該轉向方位角。 In step 306, the processing module 15 uses the result of mapping the surrounding images to the first three-dimensional projection model or the second three-dimensional projection model, a preset second panning angle, and the steering azimuth angle. The perspective projection conversion is used to convert the corresponding two-dimensional first-view driving assistance image or the second-view driving assistance image. Among them, the calculation formula of the perspective projection conversion is the same as formula (3), and

=0,

Is the second pan angle,

Is the steering azimuth.

In step 307, the processing module 15 determines whether the lane line detection result indicates at least one lane line. When the processing module 15 determines that the lane line detection result indicates the at least one lane line, the process proceeds to step 308; when the processing module 15 determines that the lane line detection result does not indicate any lane line, the process Go to step 314.

In step 308, the processing module 15 determines whether the lane is an intersection according to the at least one lane line. When the processing module 15 determines that the lane is the intersection, the process proceeds to step 309; when the processing module 15 determines that the lane is not the intersection, the process proceeds to step 311.

In step 309, the processing module 15 calculates an Alley View corresponding to the intersection based on the at least one lane line of the lane, using the conventional way of obtaining the lane view angle.

In step 310, the processing module 15 uses the perspective projection to convert the results of the surrounding images to the first three-dimensional projection model or the second three-dimensional projection model to convert the corresponding two The first visual field driving assistance image or the second visual field driving assistance image.

In step 311, the processing module 15 predicts the next driving position of the vehicle based on the vehicle speed of the vehicle information and the at least one lane line.

In step 312, the processing module 15 calculates a third pan angle related to the virtual camera 701's tilt shooting based on the next driving position and the height of the virtual camera 701 (see Figures 3 and 7) , And calculate a first driving azimuth angle of the next driving position relative to the vehicle according to the next driving position.

=0，

為該第三俯攝角度，

為該第一行駛方位角。 In step 313, the processing module 15 uses the result of mapping the surrounding images to the first three-dimensional projection model or the second three-dimensional projection model, and the third pan angle and the first driving azimuth angle. The perspective projection conversion is used to convert the corresponding two-dimensional first-view driving assistance image or the second-view driving assistance image. Among them, the calculation formula of the perspective projection conversion is the same as formula (3), and

=0,

Is the third pan angle,

Is the first driving azimuth.

In step 314, the processing module 15 predicts the next driving position of the vehicle based on the vehicle speed and the corner signal. It is worth mentioning that this embodiment uses, for example, a Kalman Filter to track and predict the next driving position of the vehicle.

In step 315, the processing module 15 calculates a fourth panning angle related to the virtual camera 701's tilt shooting based on the next driving position and the height of the virtual camera 701, and according to the next driving position Calculate the second driving azimuth angle of the next driving position relative to one of the vehicles.

=0，

為該第四俯攝角度，

為該第二行駛方位角。 In step 316, the processing module 15 uses the result of mapping the surrounding images to the first three-dimensional projection model or the second three-dimensional projection model, and the fourth panning angle and the second driving azimuth angle. The perspective projection conversion is used to convert the corresponding two-dimensional first-view driving assistance image or the second-view driving assistance image. Among them, the calculation formula of the perspective projection conversion is the same as formula (3), and

=0,

Is the fourth tilt angle,

Is the second driving azimuth.

The driving assistance image generation program of this embodiment determines the viewing angle of the virtual camera 701 according to different driving situations when generating the driving assistance image. When there is an obstacle, at least the corresponding one is converted according to the azimuth of the obstacle. The visual field driving assistance image is used to remind the driver to pay attention to obstacles to avoid collision. When the driver is turning, the corresponding visual field driving assistance image is converted at least according to the steering direction to assist the driver to pay attention to turning. Road conditions. In addition, the processing module 15 can predict the next driving position of the vehicle based on the driving trajectory or body signal, and set the observation angle to the next driving position in advance, so that the driver can know the road conditions in advance. Respond in advance. However, in other embodiments of the present invention, the priority of steps 304 and 307 can be adjusted according to driving needs, and is not limited thereto.

In summary, the driving assistance image generation method of the present invention uses the processing module 15 to select different three-dimensional projection models according to the obstacle position and the vehicle speed, so that the imaging is more realistically presented in response to the difference in driving situations. In addition, The processing module 15 determines the viewing angle of the virtual camera 701 in response to different driving situations to generate driving assistance images that are more suitable for the driving situation. The processing module 15 can also predict the next driving position of the vehicle in advance. Setting the viewing angle to the next driving position allows the driver to know the road conditions in advance and make a response, thereby providing the driver with a better user experience, so the objective of the invention can indeed be achieved.

However, the above are only examples of the present invention. When the scope of implementation of the present invention cannot be limited by this, all simple equivalent changes and modifications made in accordance with the scope of the patent application of the present invention and the content of the patent specification still belong to Within the scope of the patent for the present invention.

:車輛之長度

:車輛之寬度 H、H’:虛擬相機的高度

:預設距離 15:處理模組

:橢圓弧線所對應之橢圓的長軸

:橢圓弧線所對應之橢圓的短軸

:圓弧線所對應之圓的半徑 101:第二三維投影模型 102:第一三維投影模型 201~212:步驟 301~302:步驟 304~316:步驟11: Shooting unit 12: Obstacle detection module 13: Sensing module 14: Lane line detection module 15: Processing module 501~502: Elliptical arc 503~504: Circular arc 701: Virtual camera

: Length of vehicle

: The width of the vehicle H, H': the height of the virtual camera

: Preset distance 15: Processing module

: The major axis of the ellipse corresponding to the arc of the ellipse

: The minor axis of the ellipse corresponding to the arc of the ellipse

: The radius of the circle corresponding to the arc 101: the second three-dimensional projection model 102: the first three-dimensional projection model 201~212: steps 301~302: steps 304~316: steps

Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, in which: Figure 1 is a block diagram illustrating an embodiment of the auxiliary image generation system for a one-line vehicle of the present invention; Figure 2 is a flowchart illustrating an embodiment of the driving assistance image generation method of the present invention; Figure 3 is a three-dimensional schematic diagram illustrating a first three-dimensional projection model; 4 is a schematic front view illustrating the projection of the first three-dimensional projection model on a first plane; 5 is a schematic side view illustrating the projection of the first three-dimensional projection model on a second plane; 6 is a schematic top view illustrating the projection of the first three-dimensional projection model on a third plane; Figure 7 is a three-dimensional schematic diagram illustrating a second three-dimensional projection model; Figure 8 is a schematic front view illustrating the projection of the second three-dimensional projection model on the first plane; Figure 9 is a schematic side view illustrating the projection of the second three-dimensional projection model on the second plane; 10 is a schematic top view illustrating the projection of the second three-dimensional projection model on the third plane; Figure 11 is a schematic diagram illustrating the relationship between different obstacle distances and the first and second three-dimensional projection models; 12 is a flowchart illustrating a first generation process of one of the driving assistance image generating procedures of the driving assistance image generating method of the present invention; FIG. 13 is a flowchart illustrating a second generation process of the driving assistance image generating program in the embodiment of the driving assistance image generating method of the present invention; FIG. 14 is a schematic diagram of an experiment showing the first-view driving assistance image presented after applying the first three-dimensional projection model of the present invention; and FIG. 15 is a schematic diagram of an experiment showing the second-view driving assistance image presented after applying the second three-dimensional projection model of the present invention.

201~212: Steps

Claims (11)

A driving assistance image generation method is suitable for a vehicle and is implemented by a processing module that is electrically connected to a shooting unit and an obstacle detection module. The shooting unit is used to generate and transmit related Images of at least one perimeter of the surrounding environment of the vehicle are sent to the processing module, and the obstacle detection module is used to generate and transmit an obstacle detection result related to an obstacle corresponding to the vehicle to the processing module Group, the driving assistance image generation method includes the following steps: (A) The processing module determines whether the obstacle detection result indicates the existence of the obstacle; (B) When the processing module determines that the obstacle exists, wherein the obstacle detection result includes an obstacle distance between the obstacle and the vehicle, the processing module determines whether the obstacle distance is greater than a distance Threshold (C) When the processing module determines that the obstacle distance is less than or equal to the distance threshold, the processing module selects a first three-dimensional projection model with an elliptical arc; (D) When the processing module determines that the obstacle distance is greater than the distance threshold, the processing module selects a second three-dimensional projection model with a circular arc; (E) The processing module uses an inverse perspective projection conversion method according to the selected first three-dimensional projection model or the second three-dimensional projection model to map the at least surrounding image to the first three-dimensional projection model or the second three-dimensional projection model Two and three-dimensional projection model; and (F) The processing module uses a perspective projection conversion method to convert the result of step (E) into a corresponding first-view driving assistance image or a second-view driving assistance image. According to the driving assistance image generation method of claim 1, the processing module is also electrically connected to a sensing module, and the sensing module is used for sensing and transmitting a vehicle information related to the vehicle to the processing module , Wherein, in the step (B), the vehicle information includes a vehicle speed of the vehicle, and the distance threshold value is the product of the vehicle speed and a predetermined time. The driving assistance image generation method of claim 1, wherein, in the step (C), the function of the first three-dimensional projection model is expressed as follows:

, among them,

Is the length of the vehicle, H is the height of a virtual camera corresponding to the first three-dimensional projection model,

,

,

Is the width of the vehicle,

Is a preset distance. The driving assistance image generation method of claim 1, wherein, in the step (D), the function of the second three-dimensional projection model is expressed as follows:

, among them,

Is the length of the vehicle, H is the height of a virtual camera corresponding to the second three-dimensional projection model,

,

Is the width of the vehicle,

Is a preset distance. The driving assistance image generation method of claim 1, wherein the step (F) includes the following sub-steps: (F-1) The processing module calculates the correlation between the obstacle distance of the obstacle detection result and the height of a virtual camera corresponding to the selected first three-dimensional projection model or the second three-dimensional projection model A panoramic angle taken by the virtual camera, wherein the obstacle detection result also includes an obstacle azimuth angle of the obstacle relative to the vehicle; and (F-2) The processing module obtains the corresponding first three-dimensional projection model or the second three-dimensional projection model based on the result of mapping the at least surrounding image to the first three-dimensional projection model or the second three-dimensional projection model, the pan angle, and the obstacle azimuth angle. The visual field driving assistance image or the second visual field driving assistance image. According to the driving assistance image generation method of claim 1, the processing module is also electrically connected to a sensing module, and the sensing module is used for sensing and transmitting a vehicle information related to the vehicle to the processing module , The driving assistance image generation method further includes the following steps after step (A): (G) When the processing module determines that there is no obstacle, wherein the vehicle information includes a speed of the vehicle, the processing module determines whether the speed is greater than a speed threshold; (H) When the processing module determines that the vehicle speed is less than or equal to the speed threshold, the processing module selects the first three-dimensional projection model; and (I) When the processing module determines that the vehicle speed is greater than the speed threshold, the processing module selects the second three-dimensional projection model. According to the driving assistance image generation method of claim 1, the processing module is also electrically connected to a sensing module, and the sensing module is used for sensing and transmitting a vehicle information related to the vehicle to the processing module , Where this step (F) includes the following sub-steps: (F-1) The vehicle information includes a turn signal parameter group indicating whether multiple turn signals of the vehicle are turned on, and the processing module determines whether one of the turn signals of the vehicle is turned on according to the turn signal parameter group; (F-2) When the processing module determines that the direction light is turned on, calculate a steering azimuth angle corresponding to the direction light; and (F-3) The processing module obtains the corresponding one according to the result of mapping the at least surrounding image to the first three-dimensional projection model or the second three-dimensional projection model, a preset panoramic angle and the steering azimuth The first-view driving assistance image or the second-view driving assistance image. According to the driving assistance image generation method of claim 1, the processing module is also electrically connected to a lane line detection module, and the lane line detection module is used to generate and transmit information related to a lane of the vehicle. The detection result of a lane line is sent to the processing module, wherein the step (F) includes the following sub-steps: (F-1) The processing module determines whether the lane line detection result indicates at least one lane line; (F-2) When the processing module determines that the lane line detection result indicates the at least one lane line, the processing module determines whether the lane is an intersection according to the at least one lane line; (F-3) When the determination result of step (F-2) is yes, the processing module calculates a view angle of an alley corresponding to the intersection according to the at least one lane line; and (F-4) The processing module obtains the corresponding first-view driving assistance image or the result of mapping the at least surrounding image to the first three-dimensional projection model or the second three-dimensional projection model, and the lane angle of view. The second-view driving assistance image. According to the driving assistance image generation method of claim 8, the processing module is also electrically connected to a sensing module, and the sensing module is used for sensing and transmitting a vehicle information related to the vehicle to the processing module , Where, after step (F-2), it also includes the following sub-steps: (F-5) When the result of step (F-2) is no, the vehicle information includes a speed of the vehicle, and the processing module predicts the next one of the vehicle based on the at least one lane line and the speed of the vehicle. Driving position (F-6) The processing module calculates the height of a virtual camera corresponding to the selected first three-dimensional projection model or the second three-dimensional projection model according to the next driving position Take a panoramic angle, and calculate the driving azimuth angle of the next driving position relative to the vehicle according to the next driving position; and (F-7) The processing module obtains the corresponding first field of view according to the result of mapping the at least surrounding image to the first three-dimensional projection model or the second three-dimensional projection model, the pan angle and the driving azimuth angle The driving assistance image or the second view driving assistance image. According to the driving assistance image generation method of claim 8, the processing module is also electrically connected to a sensing module, and the sensing module is used for sensing and transmitting a vehicle information related to the vehicle to the processing module , Where, after step (F-1), it also includes the following sub-steps: (F-8) When the processing module determines that the lane line detection result does not indicate any lane line, the vehicle information includes a speed of the vehicle, and a corner signal of the vehicle, and the processing module is based on the The vehicle speed and the corner signal predict the next driving position of the vehicle; (F-9) The processing module calculates the height of a virtual camera corresponding to the selected first three-dimensional projection model or the second three-dimensional projection model based on the next driving position and the elevation of the virtual camera Take a panoramic angle, and calculate the driving azimuth angle of the next driving position relative to the vehicle according to the next driving position; and (F-10) The processing module obtains the corresponding first field of view according to the result of the at least surrounding image being mapped to the first three-dimensional projection model or the second three-dimensional projection model, the pan angle and the driving azimuth The driving assistance image or the second view driving assistance image. A driving assistance image generation system for executing the driving assistance image generation method described in claim items 1-10.

Images