JP7229032B2 - External object detection device - Google Patents

Description

The present disclosure relates to an exterior object detection device that detects an exterior object.

In recent years, technologies have been developed for capturing images of the environment outside the vehicle using an onboard camera mounted on the vehicle, and detecting objects such as vehicles and pedestrians outside the vehicle based on the captured images (see Patent Documents 1 and 2). ). There is also a technique of setting a region of interest (ROI) for detecting a detection target in an image and performing detection processing of the detection target on the target region. At this time, for example, there is a method in which the size of the attention area is determined in advance, the attention area is moved on the image, and the detection processing is sequentially performed on the attention area after movement (window sliding method).

JP-A-2005-62910 JP 2017-207279 A

In the general Window Sliding method, as the resolution of the camera improves, more attention areas are generated, and as a result, the processing time required for detection increases.

It is desirable to provide an exterior object detection system that allows for efficient detection of exterior objects.

A vehicle exterior object detection apparatus according to an embodiment of the present disclosure sets a left end position and a right end position of a region of interest for detecting a detection target outside a vehicle in an image obtained by photographing the exterior of the vehicle. a left and right area setting unit , a distance calculation unit that calculates the distance of the attention area in the real space as the distance of the detection target in the real space, and the left end position and the right end position of the attention area in the image of the detection target. A left/right region moving unit that moves an object so that it is equal to or greater than a predetermined threshold value according to the distance of the object in the real space , and a detection process that performs detection processing for the target area after movement by the left/right region moving unit. and a part.

According to the object detection device outside the vehicle according to the embodiment of the present disclosure, since the setting of the attention area for detecting the detection target is optimized, the object outside the vehicle can be efficiently detected. It becomes possible.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a configuration diagram schematically showing one configuration example of an exterior environment recognition system including an exterior object detection device according to a first embodiment of the present disclosure; 1 is a block diagram schematically showing a configuration example of an external environment recognition device; FIG. FIG. 10 is an explanatory diagram showing an example of an image when a detection target is relatively far away from the host vehicle; FIG. 4 is an explanatory diagram showing an example of an image when a detection target is relatively close to the host vehicle; 7 is a flow chart showing an example of a process for determining an attention area by an external object detection unit; FIG. 10 is an explanatory diagram showing an example of thresholds at the right end of an attention area; FIG. 10 is an explanatory diagram showing an example of a left end threshold value of a region of interest; FIG. 10 is an explanatory diagram schematically showing an example of a process of determining upper end positions and lower end positions of a region of interest by an upper and lower region setting unit; FIG. 10 is an explanatory diagram schematically showing an example of a process of determining upper end positions and lower end positions of a region of interest by an upper and lower region setting unit; FIG. 10 is an explanatory diagram schematically showing an example of a process of determining upper end positions and lower end positions of a region of interest by an upper and lower region setting unit; FIG. 7 is an explanatory diagram schematically showing an example of attention area setting processing by a left and right area setting unit and movement processing of a right end position of the attention area by a left and right area moving unit; FIG. 7 is an explanatory diagram schematically showing an example of attention area setting processing by a left/right area setting unit and movement processing of a left end position of the attention area by a left/right area moving unit; FIG. 5 is an explanatory diagram showing an example of the relationship between the distance of a detection target and predetermined thresholds (thresholds at the left end and right end); FIG. 11 is an explanatory diagram showing a first modification of the relationship between the distance of the detection target and the predetermined thresholds (thresholds at the left end and right end); FIG. 12 is an explanatory diagram showing a second modification of the relationship between the distance of the detection target and the predetermined thresholds (thresholds at the left end and right end); It is an explanatory view showing an outline of Window Sliding method.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. The description will be given in the following order.
1. First Embodiment 1.0 Overview 1.1 Configuration 1.2 Operation 1.3 Effect 2. Other embodiments

<1. First Embodiment>
[1.0 Overview]
2. Description of the Related Art In recent years, a technology has been developed in which an on-vehicle camera captures an image of an environment outside the vehicle and, based on the captured image, detects a vehicle, a pedestrian, or the like outside the vehicle as a detection target. Techniques have been developed to detect vehicles, for example, by detecting tires. As a result, vehicles equipped with a collision prevention function that avoids a collision with a detected vehicle or the like, and ACC ( Adaptive Cruise Control) that maintains a predetermined distance from a preceding vehicle are becoming popular. Object detection devices used in such techniques are limited in processing time due to hardware performance requirements. In addition, the resolution of cameras used for detection has been improving more and more in recent years.

As a technique for object detection, there is a method in which the position and size of an attention area for detecting an object to be detected in an image are determined in advance, and detection processing is performed on the attention area. In addition, there is a technique called a window sliding method in which an attention area of a specific size is slid (moved) on a screen and the attention area after movement is sequentially subjected to detection processing.

FIG. 16 shows an outline of the Window Sliding method. FIG. 16 shows an example in which the object to be detected is a vehicle tire. As shown in FIG. 16, for example, a region of interest 800 having a size suitable for tire detection is moved by one pixel, and detection processing is sequentially performed on the region of interest 800 after movement. Note that FIG. 16 shows an example in which the attention area 800 is simply moved in the horizontal direction. In this case, the attention area 800 is generated such that the left end 800L of the attention area and the right end 800R of the attention area move in units of one pixel. In such a method, as the resolution of the camera improves, more regions of interest 800 are generated, resulting in an increase in the processing time required for detection.

In order to solve this problem, a method of skipping the movement amount by a predetermined number of pixels (for example, one pixel) to generate the attention area 800 can be considered. However, since the detection accuracy of the detection target in the image varies depending on the distance of the detection target in the real space, the method of simply moving the attention area 800 by a predetermined number of pixels cannot be detected. This leads to a decrease in accuracy and detection efficiency.

For example, FIG. 3 shows an example of an image when the object to be detected is relatively far away from the own vehicle (40 m ahead in real space). FIG. 4 shows an example of an image when the object to be detected is relatively close to the own vehicle (20 m ahead in real space).

FIG. 3 shows an example of a detection object located 40 m ahead in the real space, for example, in the image. In FIG. 3, for example, 8 pix (pixels) in the image correspond to 20 cm in real space. On the other hand, FIG. 4 shows an example of a detection object located 20 m ahead in the real space, for example, in the image. In FIG. 4, for example, 8 pixels in the image correspond to 10 cm in real space.

As can be seen from FIGS. 3 and 4, for example, a detection target at a long distance can be seen very differently even if the region of interest is shifted by one pixel, whereas a detection target at a short distance has a region of interest of one pixel. Even if it shifts, the appearance does not change much compared to the case of the long distance. Therefore, when the Window Sliding method is used, it is possible to detect an object at a short distance even if the amount of movement of the attention area is increased. For example, when detecting a tire of a vehicle by the Window Sliding method, if there is a performance that can detect a tire 40 m ahead while shifting the attention area by 1 pixel, finding a tire 20 m ahead requires shifting 2 pixels by 2 pixels. detectable.

Therefore, in the present embodiment, when object detection is performed using the Window Sliding method, the movement amount of the attention area is optimized according to the distance of the detection target. This enables efficient detection of objects outside the vehicle.

[1.1 Configuration]
(External Environment Recognition System 100)
FIG. 1 schematically shows a configuration example of an exterior environment recognition system 100 including an exterior object detection device according to the first embodiment of the present disclosure.

The host vehicle 1 includes an external environment recognition system 100 . The vehicle exterior environment recognition system 100 includes two imaging devices 110 , an exterior environment recognition device 120 , and a vehicle control device (ECU: Engine Control Unit) 130 . The own vehicle 1 also includes a steering wheel 132 , an accelerator pedal 134 , a brake pedal 136 , a steering mechanism 142 , a drive mechanism 144 and a braking mechanism 146 .

The vehicle control device 130 receives operation inputs from the driver through the steering wheel 132, the accelerator pedal 134, and the brake pedal 136, and transmits them to the steering mechanism 142, the driving mechanism 144, and the braking mechanism 146, thereby controlling the own vehicle 1. . Vehicle control device 130 also controls steering mechanism 142 , drive mechanism 144 , and braking mechanism 146 in accordance with instructions from external environment recognition device 120 .

Each of the two imaging devices 110 includes an imaging device such as a CCD (Charge-Coupled Device) or a CMOS (Complementary Metal-Oxide Semiconductor), captures an image of the environment outside the vehicle 1 in front of the vehicle 1, and obtains at least luminance information. generates a luminance image containing As the luminance image, for example, a color image represented by color values is generated. A color value is a numerical group consisting of, for example, one luminance (Y) and two chrominance (UV). Alternatively, the color value may be a numerical group consisting of three hues (R (red), G (green), B (blue)).

The imaging devices 110 are arranged substantially horizontally apart from each other on the traveling direction side of the own vehicle 1, for example, so that the optical axes of the two imaging devices 110 are substantially parallel to each other. Each of the two imaging devices 110 continuously generates a luminance image of a three-dimensional object existing in a detection area in front of the own vehicle 1, for example, every 1/60 second frame (60 fps). Here, three-dimensional objects recognized by the imaging device 110 include not only independent objects such as bicycles, pedestrians, vehicles, traffic lights, roads (courses), road signs, guardrails, and buildings, but also wheels of bicycles and vehicles. It also includes things that can be specified as part of it, such as tires.

The vehicle exterior environment recognition device 120 acquires luminance images from each of the two imaging devices 110, and arbitrarily extracts a block (for example, an array of horizontal 4 pixels×vertical 4 pixels) from one of the luminance images. Parallax information including the screen position indicating the position of an arbitrary block within the screen is derived using so-called pattern matching, which is searched from the luminance image of the . Here, horizontal indicates the screen horizontal direction of the captured image, and vertical indicates the screen vertical direction of the captured image. As this pattern matching, it is conceivable to compare the brightness (Y) between a pair of images in arbitrary block units. For example, SAD (Sum of Absolute Difference) that takes the difference in luminance, SSD (Sum of Squared Intensity Difference) that uses the square of the difference, and similarity of the variance value obtained by subtracting the average value from the luminance of each pixel There is a technique such as ZNCC (Zero-mean Normalized Cross Correlation) that takes a degree of correlation. The vehicle exterior environment recognition device 120 performs such block-based parallax derivation processing for all blocks displayed in the detection area (for example, 600 pixels×200 pixels). Here, the block has 4 pixels×4 pixels, but the number of pixels in the block can be set arbitrarily.

However, the external environment recognition device 120 can derive the parallax for each block, which is the detection resolution unit, but cannot recognize what kind of object the block is part of. Therefore, the parallax information is independently derived in units of detection resolution (for example, in units of blocks) in the detection area, not in units of objects. An image associated with parallax information derived in this way is called a distance image to distinguish it from the luminance image described above.

In addition, the vehicle exterior environment recognition device 120 uses three-dimensional position information in the real space including the relative distance to the own vehicle 1 calculated based on the luminance value (color value) based on the luminance image and the distance image. , blocks with the same color value and close three-dimensional position information are grouped as objects, and which specific object (for example, preceding vehicle or bicycle) corresponds to the object in the detection area in front of the own vehicle 1 is determined. identify. Further, when the external environment recognition device 120 identifies the three-dimensional object in this way, it avoids a collision with the three-dimensional object (collision avoidance control), and controls the host vehicle 1 so as to maintain a safe inter-vehicle distance from the preceding vehicle. control (cruise control). The above relative distance can be obtained by converting parallax information for each block in the distance image into three-dimensional position information using a so-called stereo method. Here, the stereo method is a method of deriving the relative distance of the object to the imaging device 110 from the parallax of the object by using the triangulation method.

(External environment recognition device 120)
Next, the configuration of the vehicle exterior environment recognition device 120 will be described in detail. Here, a detailed description will be given mainly of the part that performs processing related to the detection of an object outside the vehicle, which is characteristic of the present embodiment.

FIG. 2 schematically shows a configuration example of the vehicle exterior environment recognition device 120. As shown in FIG.

As shown in FIG. 2 , the vehicle external environment recognition device 120 includes an I/F (interface) section 150 , a data holding section 152 and a central control section 154 .

The I/F unit 150 is an interface for bidirectionally exchanging information with the imaging device 110 and the vehicle control device 130 . The data holding unit 152 is composed of a RAM (Random Access Memory), a flash memory, a HDD (Hard Disk Drive), etc., and holds various information required for processing of each unit in the central control unit 154 described below.

The central control unit 154 is composed of a semiconductor integrated circuit including a central processing unit (CPU), a ROM (Read Only Memory) storing programs and the like, and a RAM as a work area. The central control unit 154 controls the I/F unit 150, the data holding unit 152 and the like through the system bus 156. FIG.

The central control unit 154 has an external object detection unit 160 . The central control unit 154 also includes a road surface identification unit that generates a road surface model using a known technique and identifies a road surface (road surface) area on the distance image and the luminance image.

The vehicle exterior object detection unit 160 detects various vehicle exterior objects as objects to be detected. Objects to be detected include not only independent objects such as bicycles, pedestrians, vehicles, traffic lights, roads (courses), road signs, guardrails, and buildings, but also bicycle wheels, vehicle tires, etc. Includes identifiable items. The vehicle-outside object detection unit 160 detects a vehicle as a detection target by detecting tires, for example. Exterior object detection unit 160 corresponds to a specific example of an exterior object detection device according to an embodiment of the present disclosure.

The vehicle-outside object detection unit 160 includes a left/right region setting unit 161 , a left/right region moving unit 162 , a top/bottom region setting unit 163 , and a detection processing unit 164 .

The left and right region setting unit 161 sets the left end position and the right end position of the attention region for detecting the detection target in the image.

The left/right area moving unit 162 moves the left end position and the right end position of the attention area so that they are equal to or greater than a predetermined threshold value according to the distance of the detection object. Further, the left/right area moving unit 162 moves the left end position and the right end position so that the width W of the attention area is within a predetermined range. As shown in FIGS. 13 to 15, which will be described later, the left/right area moving unit 162 uses threshold information indicating the relationship between the distance of the detection object and a predetermined threshold to move the left end position to the right end position. determine the threshold for

Here, the threshold information may be information different for each type of detection target. For example, as shown in FIGS. 14 and 15 to be described later, the threshold information may be different information depending on whether the detection target is a vehicle or a person. Also, the threshold information may be information according to the weather or the like. For example, the threshold information may be different information according to daytime, nighttime, rain, fog, etc., as shown in FIGS. 14 and 15, which will be described later. The threshold information is held by the data holding unit 152, for example.

Based on the representative distance in the image, the upper and lower region setting unit 163 sets the upper end position and the lower end position of the attention region as shown in FIGS. 8 to 10 described later. At this time, the upper and lower area setting unit 163 refers to the position information in the three-dimensional space and adjusts the attention area on the luminance image so that it becomes a rectangular area.

The detection processing unit 164 performs a detection target object detection process on the attention area after being moved by the left/right area moving unit. A known technique can be used for a specific method of detection processing.

[1.2 Operation]
Next, the operation of the vehicle-external object detection unit 160 that performs processing related to the detection of the vehicle-external object, which is characteristic of the present embodiment, will be described in detail.

(Set the left and right edges of the attention area)
FIG. 5 shows an example of the flow of attention area determination processing by the vehicle-external object detection unit 160 . FIG. 6 shows an example of the right end threshold of the attention area. FIG. 7 shows an example of the left edge threshold of the attention area. FIG. 11 schematically shows an example of attention area setting processing by the left/right area setting unit 161 and movement processing of the right end position of the attention area by the left/right area moving unit 162 . FIG. 12 schematically shows an example of attention area setting processing by the left/right area setting unit 161 and movement processing of the left end position of the attention area by the left/right area moving unit 162 .

In FIG. 5, first, the outside object detection unit 160 sets the left edge position (left edge coordinate) xL of the attention area to a by the left and right area setting unit 161 (step S11). Next, the outside object detection unit 160 sets the value of the parameter i to 1 and sets the value of the parameter i' to an invalid value (eg -∞) by the left and right area setting unit 161 (step S12). Next, the vehicle exterior object detection unit 160 sets the right end position (right end coordinate) xR of the attention region to a+i by the left and right region setting unit 161 (step S13). As a result, for example, a region of interest is set as shown in FIGS. 11A and 12A.

Next, the vehicle-outside object detection unit 160 determines whether or not the lateral width W (mm) of the attention area is less than the threshold value (minimum value) Wa (mm) using the left/right region moving unit 162 (step S14). When it is determined that the width W of the attention area is not less than the threshold Wa (step S14: N), next, the outside object detection unit 160 causes the left/right area moving unit 162 to adjust the width W of the attention area to the threshold ( Maximum value) Wb (mm) or more is determined (step S15). If it is determined that the width W of the attention area is not equal to or greater than the threshold value Wb (step S15: N), then the exterior object detection unit 160 causes the detection processing unit 164 to detect the detection target object in the attention area. A detection process is performed (step S16). After that, the outside object detection unit 160 sets the value of the parameter i' to i by the left/right area moving unit 162 (step S17). The width W and the thresholds Wa and Wb are values in real space (actual distances).

When it is determined that the width W of the attention area is less than the threshold Wa (step S14: Y), and after the process of step S17, the vehicle exterior object detection unit 160 causes the left/right area moving unit 162 to change the value of the parameter i to It is set to i+1 (step S18). Next, the outside object detection unit 160 uses the left/right area moving unit 162 to determine whether or not the value of the parameter i and the value of the parameter i' are separated from each other by a threshold value TR (mm) or more at the right end of the attention area ( step S19). If it is determined that the value of the parameter i and the value of the parameter i' are not separated from each other by the threshold TR at the right end of the attention area (step S19: N), the outside object detection unit 160 proceeds to the process of step S18. return. If it is determined that the value of the parameter i and the value of the parameter i' are separated from each other by the threshold value TR at the right end of the attention area (step S19: Y), the outside object detection unit 160 proceeds to the process of step S13. move on. Note that the value of the threshold value TR on the right end is a value in real space (actual distance). By the above processing, the attention area is set as shown in FIG. 11B, for example, with respect to the right end of the attention area. Also, as schematically shown in FIG. 6, the right end 210R of the attention area 210 is set to be equal to or greater than the threshold value TR. Incidentally, FIG. 11B schematically shows the judgment processing of steps S14, S15, and S19 in FIG.

Further, when it is determined that the width W of the attention area is equal to or greater than the threshold value Wb (step S15: Y), the outside object detection unit 160 causes the left/right area moving unit 162 to set the value of the parameter a' to a. (step S20). Next, the vehicle-outside object detection unit 160 sets the value of the parameter a to a+1 by the left/right region moving unit 162 (step S21). Next, the outside object detection unit 160 uses the left/right area moving unit 162 to determine whether or not the value of the parameter a′ and the value of the parameter a are separated by a threshold value TL (mm) or more at the left end of the attention area ( step S22). If it is determined that the value of the parameter a' and the value of the parameter a are not separated from each other by the left end threshold value TL of the attention area (step S22: N), the outside object detection unit 160 proceeds to the process of step S21. return. If it is determined that the value of the parameter a′ and the value of the parameter a are separated by the threshold value TL or more at the left end of the attention area (step S22: Y), the outside object detection unit 160 proceeds to the process of step S11. move on. Note that the value of the leftmost threshold TL is a value in real space (actual distance). The value of the threshold value TL at the left end and the value of the threshold value TR at the right end may be the same value. By the above processing, the attention area is set as shown in FIG. 12B, for example, with respect to the left end of the attention area. Also, as schematically shown in FIG. 7, the left end 210L of the attention area 210 is set to be equal to or greater than the threshold TL. Note that FIG. 12B schematically shows the judgment processing of steps S14, S15, and S22 in FIG.

(Setting the top and bottom edges of the region of interest)
8 to 10 schematically show an example of processing for determining the upper end position and the lower end position of the attention area by the upper and lower area setting section 163. FIG.

After setting the left end position and the right end position of the attention area as described above (before performing the detection processing in step S16 in FIG. 5), the vehicle exterior object detection unit 160 causes the upper and lower area setting unit 163 to set the upper end position of the attention area. and bottom position.

The upper and lower area setting unit 163 refers to the position information in the three-dimensional space and calculates representative distances at the left end position and right end position of the attention area. Also, the upper and lower area setting unit 163 refers to the position information in the three-dimensional space and adjusts the attention area on the luminance image so that it becomes a rectangular area as shown in FIG.

The vertical region setting unit 163 first calculates a representative distance for each horizontal position, for example, from the left end to the right end of the distance image or the luminance image. Specifically, the upper and lower area setting unit 163 first divides, for example, the distance image or the luminance image into a plurality of strip-shaped divided areas in the horizontal direction. Each divided area is a vertically extending area having a width of a predetermined number of pixels in the horizontal direction. Subsequently, the upper and lower region setting unit 163 generates a histogram of relative distances for all pixel blocks located above the road surface, for each divided region, based on the positional information in the three-dimensional space. Based on the histogram, upper and lower area setting section 163 specifies the relative distance corresponding to the peak with the highest frequency. Here, "equivalent to a peak" refers to a peak value or a value near the peak that satisfies an arbitrary condition. The vertical area setting unit 163 sets the relative distance corresponding to this peak as the representative distance for each divided area (for each horizontal position). As a result, a plurality of representative distances 220 indicated by thick black lines are calculated in FIGS. 8 and 9 .

First, as shown in FIG. 8, the upper/lower region setting unit 163 sets the intersection point between a straight line extending downward from the position of the representative distance at the left end of the region of interest and the plane indicated by the road surface model in the distance image and the luminance image. Ask for Then, the vertical position of the intersection is determined (point A in FIG. 8). Similarly, in the range image and the luminance image, the intersection of a straight line extending downward from the position of the representative distance at the right end of the attention area and the plane indicated by the road surface model is obtained. Then, the vertical position of the intersection is determined (point B in FIG. 8).

Next, the upper and lower region setting unit 163 obtains the distance d [mm] between the points A and B in the three-dimensional space, and in the range image and the luminance image, the point extended by d [mm] in the vertical direction from the points A and B Find C and D (Fig. 9). Thus, as shown in FIG. 10, a parallelogram 201 having points A, B, C, and D as vertices is obtained. The upper and lower area setting unit 163 sets the rectangular area 202 including the parallelogram 201 as the attention area. That is, the upper end of the rectangular area 202 becomes the upper end of the attention area, and the lower end of the rectangular area 202 becomes the lower end of the attention area.

(Relationship between distance of object to be detected and predetermined threshold)
FIG. 13 shows an example of the relationship between the distance of the detection object and the predetermined thresholds (left edge threshold TL and right edge threshold TR) used when setting the left edge and right edge of the attention area. The value of the threshold value TL at the left end and the value of the threshold value TR at the right end may be the same value. Hereinafter, the threshold value TL at the left end and the threshold value TR at the right end will be collectively described as predetermined threshold values. In FIG. 13, the value of the distance to the detection object and the value of the predetermined threshold are values in real space.

When there is an object to be detected in the distance, detection processing is performed by generating and identifying the attention area more densely depending on the atmospheric conditions (dust, rain, etc.) and hardware performance (lens MTF characteristics, etc.). It is possible to reduce the situation in which the detection is not actually made. Therefore, the left/right area moving unit 162 changes the predetermined threshold used when determining the left end and right end of the attention area according to the distance of the detection object. Specifically, as shown in FIG. 13, non-linear data representing the relationship between the distance of the object to be detected and a predetermined threshold is held as threshold information in the data holding unit 152 (FIG. 2). determines a predetermined threshold for the region of interest that is actually used by referring to this data according to the distance.

FIG. 14 shows a first modification of the relationship between the distance of the detection target and the predetermined threshold. FIG. 15 shows a second modification of the relationship between the distance of the detection target and the predetermined thresholds (thresholds at the left end and right end). In FIGS. 14 and 15, the value of the distance to the detection object and the value of the predetermined threshold are values in real space. FIG. 14 shows an example in which the object to be detected is a vehicle, and FIG. 15 shows an example in which the object to be detected is a person.

As shown in FIGS. 14 and 15, the predetermined threshold used when setting the left end and right end of the attention area is different information depending on the type of the detection object, for example, when the detection object is a vehicle and when the detection object is a person. The information may be different depending on the case. Also, as shown in FIGS. 14 and 15, the threshold information may be different information according to day, night, rain, fog, and the like.

[1.3 Effect]
As described above, according to the object detection apparatus outside the vehicle according to the first embodiment, the setting of the attention area for detecting the detection target is optimized, so that the object outside the vehicle can be detected efficiently. can be detected.

<2. Other Embodiments>
The technology according to the present disclosure is not limited to the description of the above embodiments, and various modifications are possible.

In the above embodiment, the program that causes the computer to function as the vehicle exterior environment recognition device 120 is provided by a computer-readable storage medium such as a flexible disk, a magneto-optical disk, a ROM, a CD, a DVD, or a BD that records the program. may be Here, the program means data processing means written in any language or writing method.

In the above-described embodiment, the processing by the vehicle-external object detection unit 160 does not necessarily have to be time-sequentially processed in the order shown in the flowchart of FIG. 5, and may include parallel or subroutine processing.

In addition, in the above-described embodiment, the central control unit 154 is not limited to being configured by a semiconductor integrated circuit including a central processing unit (CPU), ROM, RAM, etc., but may be FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit) or the like may be used. Alternatively, it may be composed of one or a plurality of central processing units, FPGAs, and ASICs.

Further, in the above-described embodiment, the distance information of the detection object is acquired by the stereo method using the two imaging devices 110, but the method of acquiring the distance information is not limited to this. For example, radar may be used to acquire distance information.

Reference Signs List 1 Own vehicle 100 Vehicle environment recognition system 110 Imaging device 120 Vehicle environment recognition device 130 Vehicle control device 132 Steering wheel 134 Accelerator pedal 136 Brake pedal 142 Steering mechanism 144 drive mechanism 146 braking mechanism 150 I/F section 152 data holding section 154 central control section 156 system bus 160 external object detection section 161 left and right area setting section 162 Left and right area moving unit 163 Upper and lower area setting unit 164 Detection processing unit 201 Parallelogram 202 Rectangular area 210 Region of interest (ROI) 210L Left end of region of interest 210R Right end of region of interest , 220... Representative distance TL... Left edge threshold TR... Right edge threshold xL... Left edge position (left edge coordinates) xR... Right edge position (right edge coordinates) W... Width of attention area Wa... Width of attention area Threshold (minimum value), Wb .

Claims (6)

a left and right region setting unit for setting the left end position and the right end position of a region of interest for detecting a detection target outside the vehicle in an image obtained by photographing the outside of the vehicle ;
a distance calculation unit that calculates the distance on the real space of the attention area as the distance on the real space of the detection target;
a left-right area moving unit that moves the left end position and the right end position of the attention area in the image so as to be equal to or greater than a predetermined threshold value according to the distance of the detection target in real space ;
and a detection processing unit that performs detection processing of the detection object on the attention area after movement by the left/right area moving unit. The left-right region moving unit moves the predetermined threshold value used for moving between the left end position and the right end position based on threshold information indicating the relationship between the distance of the detection object in the real space and the predetermined threshold value. The external object detection device according to claim 1, wherein the object detection device determines: The object detection device outside the vehicle according to claim 2 , wherein the threshold information is information different for each type of the detection target. 4. The outside-vehicle object detection device according to claim 2 , wherein the threshold information is information according to the weather. 5. The left/right area moving unit moves the left end position and the right end position in the image so that the horizontal width of the attention area in real space is within a predetermined range. 3. The external object detection device according to 1. 6. The object outside the vehicle according to any one of claims 1 to 5, further comprising an upper and lower area setting unit that sets an upper end position and a lower end position of the attention area in the image based on the representative distance in the image. detection device.

Images