JP2021140723A - Vehicle exterior environment recognition device - Google Patents

Abstract

To improve determination accuracy of a specific object.SOLUTION: A vehicle exterior environment recognition device includes: a first width derivation unit for deriving a first width of a first solid object based on a luminance image capable of specifying luminance of an imaging object; a second width derivation unit for deriving a second width of a second solid object based on a distance image capable of specifying a distance of the imaging object; an overlap degree derivation unit for obtaining an overlap width in which the first width and the second width overlap in a horizontal direction to determine a larger one of an overlap width/the first width and an overlap width/the second width as an overlap degree; and a specific object determination unit for determining the first solid object and the second solid object as the same specific object if the overlap degree is a predetermined threshold value of 196 or more.SELECTED DRAWING: Figure 6

Description

The present invention relates to an external environment recognition device that identifies a specific object existing in the traveling direction of the own vehicle.

Conventionally, as in Patent Document 1, the preceding vehicle located in front of the own vehicle is detected to reduce the damage caused by the collision with the preceding vehicle, and the follow-up control is performed so as to keep the distance between the preceding vehicle and the vehicle at a safe distance. The technology to do is known.

Japanese Patent No. 3349060

In order to reduce the damage caused by the collision between the own vehicle and the preceding vehicle and to realize the follow-up control that causes the own vehicle to follow the preceding vehicle, first of all, is the three-dimensional object located around the own vehicle a specific object such as a vehicle? You have to decide whether or not. For example, the shape of the three-dimensional object in the luminance image captured by the imaging unit is recognized and determined as the preceding vehicle, or the relative distance and speed of the three-dimensional object in the distance image are recognized and determined as the preceding vehicle.

However, when the three-dimensional object is located far from the own vehicle, the area occupied by the three-dimensional object itself in the image becomes small, so that the shape in the luminance image and the relative distance in the distance image fluctuate, and the determination accuracy of the specific object is improved. It caused a decline.

In view of such a problem, an object of the present invention is to provide an external environment recognition device capable of improving the determination accuracy of a specific object.

In order to solve the above problems, the vehicle exterior environment recognition device of the present invention has a first width derivation unit that derives the first width of the first three-dimensional object based on a brightness image that can specify the brightness of the imaged object, and an imaged object. The overlap width is obtained by finding the overlap width in which the second width derivation unit for deriving the second width of the second three-dimensional object and the first width and the second width overlap in the horizontal direction based on the distance image in which the distance can be specified. / The first width and the overlap width / the second width, whichever is larger, is the overlap degree derivation part, and if the overlap degree is equal to or more than a predetermined threshold, the first three-dimensional object and the second three-dimensional object are combined. It is provided with a specific object determination unit for determining the same specific object.

The specific object determination unit may lower the threshold value for the degree of overlap determined to be the same specific object.

The specific object determination unit may lower the threshold value for the degree of duplication and then increase the threshold value for the degree of duplication determined that they are not the same specific object.

The specific object determination unit may change the threshold value for the degree of duplication stepwise according to the number of times that the same specific object is determined.

The specific object determination unit may change the threshold value depending on whether or not the wiper is operated.

The specific object determination unit may change the threshold value according to the speed of the own vehicle.

According to the present invention, it is possible to improve the determination accuracy of a specific object.

FIG. 1 is a block diagram showing the connection relationship of the vehicle exterior environment recognition system. FIG. 2 is a functional block diagram showing a schematic function of the vehicle exterior environment recognition device. FIG. 3 is a flowchart showing the flow of the method for recognizing the environment outside the vehicle. FIG. 4 is an explanatory diagram for explaining a luminance image and a distance image. FIG. 5 is an explanatory diagram for explaining the degree of overlap. FIG. 6 is a graph for explaining how to obtain the threshold value referred to by the specific object determination unit. FIG. 7 is an explanatory diagram for explaining the change mode of the threshold value. FIG. 8 is an explanatory diagram for explaining a change mode of the threshold value.

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The dimensions, materials, and other specific numerical values shown in such an embodiment are merely examples for facilitating the understanding of the invention, and do not limit the present invention unless otherwise specified. In the present specification and drawings, elements having substantially the same function and configuration are designated by the same reference numerals to omit duplicate description, and elements not directly related to the present invention are not shown. do.

(External environment recognition system 100)
FIG. 1 is a block diagram showing a connection relationship of the vehicle exterior environment recognition system 100. The vehicle exterior environment recognition system 100 includes an image pickup device 110, a vehicle exterior environment recognition device 120, and a vehicle control device 130.

The image pickup device 110 is configured to include an image pickup element such as a CCD (Charge-Coupled Device) or a CMOS (Complementary Metal-Oxide Semiconductor), images the outside environment in front of the own vehicle 1, and includes at least brightness information. A brightness image (color image or monochrome image) can be generated. Further, the image pickup device 110 is arranged so as to be separated from each other in the substantially horizontal direction so that the optical axes of the two image pickup devices 110 are substantially parallel to each other on the traveling direction side of the own vehicle 1. The image pickup apparatus 110 continuously generates a luminance image of a three-dimensional object existing in the detection region in front of the own vehicle 1 every frame (60 fps) of, for example, 1/60 second.

Further, the vehicle exterior environment recognition device 120 recognizes the boundary outside the vehicle through the luminance image acquired from the image pickup apparatus 110 and the distance image based on the luminance image of 2. Then, the vehicle exterior environment recognition device 120 performs speed control and steering angle control in the traveling of the own vehicle 1 based on the recognized external environment and the traveling condition of the own vehicle 1. The vehicle exterior environment recognition device 120 will be described in detail later.

The vehicle control device 130 is composed of an ECU (Electronic Control Unit), receives a driver's operation input through a steering wheel 132, an accelerator pedal 134, and a brake pedal 136, refers to a speed sensor 138 provided on the axle, and is a steering mechanism. The own vehicle 1 is controlled by transmitting the information to 142, the drive mechanism 144, and the braking mechanism 146. Further, the vehicle control device 130 controls the steering mechanism 142, the drive mechanism 144, and the braking mechanism 146 in accordance with the instructions of the vehicle exterior environment recognition device 120. Further, when traveling in the rain, the vehicle control device 130 operates a wiper 148 that wipes off dirt and impurities on the windshield and the rear glass of the own vehicle 1 in response to the driver's operation.

(External environment recognition device 120)
FIG. 2 is a functional block diagram showing a schematic function of the vehicle exterior environment recognition device 120. As shown in FIG. 2, the vehicle exterior environment recognition device 120 includes an I / F unit 150, a data holding unit 152, and a central control unit 154.

The I / F unit 150 is an interface for bidirectional information exchange with the image pickup device 110 and the vehicle control device 130. The data holding unit 152 is composed of a RAM, a flash memory, an HDD, and the like, and holds various information necessary for processing of each of the following functional units.

The central control unit 154 is composed of a semiconductor integrated circuit including a central processing unit (CPU), a ROM in which programs and the like are stored, a RAM as a work area, and the like, and is an I / F unit 150 and a data holding unit through a system bus 156. It controls 152 and so on. Further, in the present embodiment, the central control unit 154 includes an image acquisition unit 160, a distance image generation unit 162, a three-dimensional object identification unit 163, a first width derivation unit 164, a second width derivation unit 166, and an overlap degree derivation unit 168. It also functions as a specific object determination unit 170. Hereinafter, the method of recognizing the external environment, which is characteristic of the present embodiment, is to extract a three-dimensional object in front of the own vehicle 1 and determine a specific object such as a preceding vehicle, based on the operation of each functional unit of the central control unit 154. It will be described in detail.

(Method of recognizing the environment outside the vehicle)
FIG. 3 is a flowchart showing the flow of the method for recognizing the environment outside the vehicle. The vehicle exterior environment recognition device 120 executes the vehicle exterior environment recognition method at predetermined interruption times. In the vehicle exterior environment recognition method, first, the image acquisition unit 160 acquires a plurality of luminance images (S200). The distance image generation unit 162 generates a distance image (S202). The three-dimensional object identification unit 163 identifies the three-dimensional object (S203). The first width derivation unit 164 derives the first width of the first three-dimensional object based on a luminance image capable of specifying the luminance of the imaging target (S204). The second width derivation unit 166 derives the second width of the second three-dimensional object based on the distance image in which the distance to be imaged can be specified (S206). The overlap degree derivation unit 168 obtains the overlap width, which is the width of the region where the first width and the second width overlap in the horizontal direction of the image. Then, the overlap degree derivation unit 168 sets the larger of the overlap width / first width and the overlap width / second width as the overlap degree (S208). If the degree of overlap is equal to or greater than a predetermined threshold value, the specific object determination unit 170 determines that the first three-dimensional object and the second three-dimensional object are the same specific object (S210). Hereinafter, each process of the vehicle exterior environment recognition method will be described in detail, and description of the process unrelated to the features of the present embodiment will be omitted. In the present embodiment, "/" means division, and for example, the overlap width / first width indicates that the overlap width is divided by the first width.

(Image acquisition process S200)
FIG. 4 (FIGS. 4A to 4C) is an explanatory diagram for explaining a luminance image and a distance image. The image acquisition unit 160 acquires a plurality of (here, 2) luminance images imaged by the image pickup apparatus 110 with different optical axes. Here, the image acquisition unit 160 uses the first luminance image 180a captured by the imaging device 110 located on the relatively right side of the own vehicle 1 shown in FIG. 4A and the own vehicle shown in FIG. 4B as the luminance image 180. It is assumed that the second luminance image 180b imaged by the image pickup apparatus 110 located on the relatively left side of 1 is acquired.

With reference to FIG. 4, it can be understood that the image positions of the three-dimensional objects included in the images differ in the horizontal direction between the first luminance image 180a and the second luminance image 180b due to the difference in the imaging position of the imaging device 110. Here, horizontal indicates the horizontal direction of the screen of the captured image, and vertical indicates the vertical direction of the screen of the captured image.

(Distance image generation process S202)
The distance image generation unit 162 is based on the first luminance image 180a shown in FIG. 4A and the second luminance image 180b shown in FIG. 4B acquired by the image acquisition unit 160, and the distance to be imaged as shown in FIG. 4C. Generates a identifiable distance image 182.

Specifically, the distance image generation unit 162 derives parallax information including parallax and an image position indicating a position in an image of an arbitrary block by using so-called pattern matching. Specifically, the block corresponding to the block arbitrarily extracted from one luminance image (here, the first luminance image 180a) is searched from the other luminance image (here, the second luminance image 180b). Here, the block is represented by, for example, an array of 4 horizontal pixels × 4 vertical pixels. Further, pattern matching is a method of searching for a block corresponding to a block arbitrarily extracted from one luminance image from the other luminance image.

For example, as a function for evaluating the degree of matching between blocks in pattern matching, SAD (Sum of Absolute Difference) that takes the difference in brightness, SSD (Sum of Squared intensity Difference) that squares the difference, and the brightness of each pixel There is a method such as NCC (Normalized Cross Correlation) that takes the similarity of the variance value obtained by subtracting the average value from.

The distance image generation unit 162 performs such a block-based parallax derivation process for all the blocks projected in the detection area of, for example, 600 pixels × 200 pixels. Here, the block is 4 pixels × 4 pixels, but the number of pixels in the block can be set arbitrarily.

However, although the distance image generation unit 162 can derive the parallax for each block, which is a detection resolution unit, it cannot recognize what kind of three-dimensional object the block is a part of. Therefore, the parallax information is independently derived not in units of three-dimensional objects but in units of detection resolution in the detection region, for example, in units of blocks. Here, for convenience of explanation, the block from which the parallax is derived is represented by a black dot.

The distance image generation unit 162 converts the parallax information for each block in the distance image 182 into three-dimensional position information using the so-called stereo method, and derives the relative distance. Here, the stereo method is a method of deriving the relative distance of the block to the imaging device 110 from the parallax of the block by using the triangulation method.

The vehicle exterior environment recognition device 120 recognizes the vehicle exterior environment through the luminance image 180 and the distance image 182 derived in this way, and determines, for example, a three-dimensional object in front of the own vehicle 1 as a specific object such as a preceding vehicle. However, when the three-dimensional object 184 in FIG. 4A and the three-dimensional object 186 in FIG. 4C are located far from the own vehicle 1, the area occupied by the three-dimensional object itself in the image is small, and the shape and distance in the luminance image 180. Since the relative distance in the image 182 fluctuates, the determination accuracy of the specific object is lowered. Therefore, the degree of overlap in which the three-dimensional objects specified in the luminance image 180 and the distance image 182 each overlap is derived, and it is determined whether or not the compatible objects are the same specific object according to the degree of overlap.

(Three-dimensional object identification process S203)
The three-dimensional object identification unit 163 first identifies the road surface in front of the own vehicle 1. Then, the three-dimensional object identification unit 163 identifies a three-dimensional object having a height vertically above the specified road surface. Specifically, the three-dimensional object identification unit 163 sets a block whose height from the road surface is at a predetermined distance (for example, 0.3 m) or more as a candidate for a three-dimensional object protruding from the road surface in the height direction. The three-dimensional object identification unit 163 groups blocks having the same relative distance to the own vehicle 1 from among a plurality of blocks that are candidates for a three-dimensional object having a height vertically above the road surface, and identifies them as a three-dimensional object.

(First width derivation process S204)
The first width derivation unit 164 sets the first width, which is the horizontal width of a predetermined three-dimensional object 184 (first three-dimensional object) in the first luminance image 180a, for example, of one of the two luminance images 180. Derived. The first width is represented by, for example, a converted value of the number of pixels or the distance.

A machine learning technique is used when extracting a three-dimensional object from the luminance image 180. For example, the edge pattern and the temporal change of the luminance image 180 are input, and a three-dimensional object that can be regarded as integrally formed is output. Since various existing techniques can be applied to such machine learning techniques, detailed description thereof will be omitted here.

(Second width derivation process S206)
The second width derivation unit 166 derives the second width, which is the horizontal width of the predetermined three-dimensional object 186 (second three-dimensional object) in the distance image 182. The second width is represented by, for example, a converted value of the number of pixels or the distance.

Here, both the first width derivation unit 164 and the second width derivation unit 166 derive the horizontal widths of the three-dimensional objects 184 and 186. This is due to the following reasons. The own vehicle 1 may swing in the pitch direction and the roll direction due to the unevenness of the road. On the other hand, the yaw direction does not swing as much as the pitch direction and the roll direction. Therefore, by targeting the width in the horizontal direction corresponding to the stable yaw direction, the determination accuracy of the specific object can be improved. Needless to say, on a flat road, the width in the vertical direction corresponding to the pitch direction can be targeted instead of or in addition to the width in the horizontal direction corresponding to the yaw direction.

(Multiplicity derivation process S208)
5 (FIGS. 5A to 5F) are explanatory views for explaining the degree of overlap. The multiplicity derivation unit 168 first acquires the first width 190 derived by the first width derivation unit 164 and the second width 192 derived by the second width derivation unit 166, and obtains the first width and the second width. Derivates the overlap width 194, which is the width of the overlapping regions in the horizontal direction of the image.

In FIGS. 5A to 5C, the first width 190 of the three-dimensional object 184 is longer than the second width 192 of the three-dimensional object 186. In FIG. 5A, the three-dimensional object 186 is included in the three-dimensional object 184 in the horizontal direction. Therefore, the overlap width 194 is equal to the second width 192. In FIG. 5B, a part of the three-dimensional object 184 and the three-dimensional object 186 overlap each other in the horizontal direction. Therefore, the overlapping width 194 is shorter than the first width 190 and the second width 192. In FIG. 5C, the three-dimensional object 184 and the three-dimensional object 186 do not overlap in the horizontal direction. Therefore, the overlap width 194 becomes 0.

In FIGS. 5D to 5F, the first width 190 of the three-dimensional object 184 is shorter than the second width 192 of the three-dimensional object 186. In FIG. 5D, the three-dimensional object 184 is included in the three-dimensional object 186 in the horizontal direction. Therefore, the overlap width 194 is equal to the first width 190. In FIG. 5E, a part of the three-dimensional object 184 and the three-dimensional object 186 overlap each other in the horizontal direction. Therefore, the overlapping width 194 is shorter than the first width 190 and the second width 192. In FIG. 5F, the three-dimensional object 184 and the three-dimensional object 186 do not overlap in the horizontal direction. Therefore, the overlap width 194 becomes 0.

Subsequently, the multiplicity derivation unit 168 determines that the overlap width shows a value other than 0, that is, at least overlaps, the overlap width 194 / first width 190, and the overlap width 194 / second. The width 192 is derived, and the larger of the overlap width 194 / first width 190 and the overlap width 194 / second width 192 is set as the degree of overlap.

In the example of FIG. 5A, since the overlap width 194 / first width 190 <overlap width 194 / second width 192, the degree of overlap is 194 overlap width / 192 second width (= 1). In the example of FIG. 5B, since the overlap width 194 / first width 190 <overlap width 194 / second width 192, the degree of overlap is 194 overlap width / 192 second width. In the example of FIG. 5C, the overlap width 194 is 0, that is, since there is no overlap, the degree of overlap is also 0.

In the example of FIG. 5D, since the overlap width 194 / first width 190> overlap width 194 / second width 192, the degree of overlap is 194 overlap width / 190 first width (= 1). In the example of FIG. 5E, since the overlap width 194 / first width 190> overlap width 194 / second width 192, the degree of overlap is 194 overlap width / 190 first width. In the example of FIG. 5F, since the overlap width 194 is 0, that is, there is no overlap, the degree of overlap is also 0.

(Specific object determination process S210)
The specific object determination unit 170 determines whether or not the three-dimensional object 184 and the three-dimensional object 186 are the same specific object with reference to the threshold value.

FIG. 6 is a graph for explaining how to obtain the threshold value referred to by the specific object determination unit 170. In the graph of FIG. 6, the horizontal axis indicates the relative distance, and the vertical axis indicates the degree of overlap. Here, the degree of overlap is expressed in the range of 0 to 1. Such a multiplicity graph is prepared in advance.

For example, the own vehicle 1 travels in a predetermined area in advance, identifies only a specific object corresponding to the vehicle among a plurality of three-dimensional objects extracted during that period, and plots the measured relative distance and the measured value of the degree of overlap. do. Then, the point cloud of FIG. 6 can be obtained. Here, it can be understood that the longer the relative distance, the smaller the value of the multiplicity. In FIG. 6, a straight line that is equal to or less than the lower limit of the point cloud at all relative distances is generated. Such a straight line becomes the threshold value 196. Therefore, the threshold value 196 can be expressed as a linear function of the relative distance. The threshold value 196 is not limited to a straight line but may be represented by a curve, that is, a function of a plurality of orders as long as it is equal to or less than the lower limit value of the point cloud at all relative distances. Here, the example in which the own vehicle 1 itself plots the measured values of the relative distance and the multiplicity has been described, but the actual measurement values of the relative distance and the multiplicity plotted by another vehicle are not limited to this case. May be reflected in the own vehicle 1.

The specific object determination unit 170 refers to the threshold value 196 prepared in this way, and refers to the three-dimensional object 184 and the three-dimensional object 186 at the relative distance where the three-dimensional object 184 and the three-dimensional object 186 derived by the multiplicity derivation unit 168 are located. If the degree of overlap is 196 or more, it is determined that the three-dimensional object 184 and the three-dimensional object 186 are the same specific object.

Here, in order to strictly determine that the three-dimensional object 184 and the three-dimensional object 186 are the same specific object, a high value is adopted as the threshold value 196. However, it is not necessary to increase the threshold value 196 until after it is determined that the three-dimensional object 184 and the three-dimensional object 186 are the same specific object. Therefore, a hysteresis function is provided so that once it is determined that the specific object is the same, it can be continuously determined as the same specific object even if the degree of overlap is slightly reduced.

7 and 8 are explanatory views for explaining the change mode of the threshold value 196. Specifically, when the specific object determination unit 170 determines that the three-dimensional object 184 and the three-dimensional object 186 are the same specific object, the threshold value 196 for comparing the degree of overlap between the three-dimensional object 184 and the three-dimensional object 186 is set, for example, in FIG. The threshold value 196a indicated by the broken line in No. 7 is lowered to the threshold value 196b indicated by the alternate long and short dash line.

In this way, even if the degree of overlap between the three-dimensional object 184 and the three-dimensional object 186 determined to be the same specific object fluctuates in the vicinity of the threshold value 196a, the threshold value is at least 196b or more. It becomes easier to judge. Therefore, it is possible to stably determine a specific object.

The specific object determination unit 170 continuously counts the number of times that the same specific object is determined. Then, the specific object determination unit 170 sets the threshold value 196 for the degree of overlap according to the number of times that the same specific object is determined as the threshold value 196a shown by the broken line in FIG. 7 → the threshold value 196b shown by the alternate long and short dash line → the threshold value shown by the solid line. The threshold value of a plurality of steps prepared in advance such as 196c may be lowered stepwise.

Specifically, when the same specific object is determined a predetermined number of times (for example, 10 times) by the threshold value 196a, the specific object determination unit 170 determines that the three-dimensional object 184 and the three-dimensional object 186 are determined to be the same specific object. The threshold value 196 for the degree of overlap is lowered from, for example, the threshold value 196a in FIG. 7 to the threshold value 196b. Further, when the same specific object is determined by the threshold value 196b a predetermined number of times (for example, 10 times), the specific object determination unit 170 overlaps the three-dimensional object 184 determined to be the same specific object and the three-dimensional object 186. The threshold 196 for degrees is lowered, for example, from the threshold 196b in FIG. 7 to the threshold 196c.

In this way, when the three-dimensional object 184 and the three-dimensional object 186 are determined to be the same specific object, the threshold value 196 is gradually lowered, and thereafter, it is easy to be continuously determined to be the same specific object. Therefore, it becomes possible to determine a specific object more stably.

Further, if the degree of overlap is less than the threshold value 196 even though the threshold value 196 is lowered and it is determined that the three-dimensional object 184 and the three-dimensional object 186 are not the same specific object, the threshold value 196 is maintained anymore. should not do. Therefore, even if it is no longer determined to be the same specific object, a hysteresis function is provided, and once it is determined that the specific object is not the same, and if the degree of duplication does not increase, the determination is not made as the same specific object.

Specifically, when the threshold value 196 is the threshold value 196c shown by the solid line in FIG. 8, the specific object determination unit 170 determines that the three-dimensional object 184 and the three-dimensional object 186 are not the same specific object. The threshold value 196 for comparing the degree of overlap between the object 184 and the three-dimensional object 186 is increased from, for example, the threshold value 196c to the threshold value 196b shown by the alternate long and short dash line in FIG.

In this way, even if the degree of overlap between the three-dimensional object 184 determined not to be the same specific object and the three-dimensional object 186 becomes the threshold value 196c or more, it becomes difficult to determine that they are the same specific object until at least the threshold value 196b or more. .. Therefore, a three-dimensional object that is not a specific object can be stably excluded.

The specific object determination unit 170 indicates the threshold value 196 for the degree of overlap according to the number of times it is determined that they are not the same specific object: the threshold value 196c shown by the solid line in FIG. 8 → the threshold value 196b shown by the alternate long and short dash line → the broken line. The threshold value may be increased stepwise, such as 196a.

When the three-dimensional object 184 and the three-dimensional object 186 determined to be the same specific object are determined not to be the same specific object a predetermined number of times (for example, 10 times) by the threshold value 196c, the specific object determination unit 170 is the same. The threshold value 196 for the degree of overlap between the three-dimensional object 184 determined to be the specific object and the three-dimensional object 186 is increased from the threshold value 196c in FIG. 8 to the threshold value 196b, for example. Further, when it is determined by the threshold value 196b that the specific objects are not the same a predetermined number of times (for example, 10 times), the specific object determination unit 170 determines that the three-dimensional object 184 and the three-dimensional object 186 are determined to be the same specific object. The threshold value 196 for the degree of overlap of the above is raised from the threshold value 196b in FIG. 8 to the threshold value 196a, for example.

In this way, when it is determined that the three-dimensional object 184 and the three-dimensional object 186 determined to be the same specific object are not the same specific object, the threshold value 196 is gradually increased, and thereafter it is determined to be the same specific object. It becomes difficult. Therefore, a three-dimensional object that is not a specific object can be eliminated more stably.

When traveling in the rain, raindrops may adhere to, for example, the windshield or the lens in front of the image pickup device 110, and the luminance image 180 or the distance image 182 may be blurred. In that case, the degree of overlap between the three-dimensional object 184 and the three-dimensional object 186 may be smaller than the original value. Then, the specific object determination unit 170 may determine that the three-dimensional object 184 and the three-dimensional object 186, which should be originally determined to be the same specific object, are not the same specific object. Therefore, the specific object determination unit 170 may lower the threshold value 196 when raindrops are expected to adhere, for example, when the wiper 148 is operating and when the wiper 148 is not operating. With such a configuration, it is possible to appropriately determine a specific object even when traveling in the rain.

Further, when the speed of the own vehicle 1 is high, it is necessary to more appropriately determine the preceding vehicle as a specific object located far away, but when the speed of the own vehicle 1 is low, the higher the speed of the own vehicle 1, the higher the speed. It is not necessary to make a reliable judgment. Therefore, the specific object determination unit 170 changes the threshold value 196 according to the speed of the own vehicle 1. For example, when the speed of the own vehicle 1 detected by the speed sensor 138 is high, the threshold value 196 is raised as compared with when the speed is low, and when the speed of the own vehicle 1 is low, the threshold value 196 is lowered as compared with when the speed is high. Here, the threshold value 196 may be changed linearly or stepwise in proportion to the speed of the own vehicle 1. In this way, it is possible to appropriately determine a specific object regardless of the speed of the own vehicle 1.

Then, the specific object determination unit 170 identifies which specific object is the same specific object whose degree of overlap is equal to or greater than a predetermined threshold value. For example, in the distance image 182, the specific object determination unit 170 first has the same relative distance to the own vehicle 1 among the plurality of blocks located at a height from the road surface equal to or higher than a predetermined distance, and is perpendicular and horizontal to each other. Blocks that are close in direction are grouped and used as candidates for the preceding vehicle, for example. Next, the specific object determination unit 170 specifies a rectangular region in the first luminance image 180a including all the three-dimensional objects specified in this way as a three-dimensional object region. Here, the rectangle is composed of two straight lines extending in the vertical direction and contacting the left and right edges of the three-dimensional object, and two straight lines extending in the horizontal direction and contacting the upper and lower edges of the three-dimensional object, respectively.

The specific object determination unit 170 identifies the specific object included in the three-dimensional object region thus formed as the preceding vehicle based on various conditions indicating the vehicle-likeness. In this way, when the three-dimensional object region is identified as the preceding vehicle, the vehicle exterior environment recognition device 120 controls to reduce the damage caused by the collision with the preceding vehicle and keep the distance between the preceding vehicle and the vehicle at a safe distance. do.

Further, the specific object determination unit 170 stores the positions and velocities of the three-dimensional object 184 and the three-dimensional object 186 determined to be the same specific object in the data holding unit 152, and stores the three-dimensional object 184 and the three-dimensional object 186 based on the position and speed. Chase. In this way, the specific object determination unit 170 can identify the three-dimensional object 184 and the three-dimensional object 186 that are determined to be the same specific object in the previous frame.

In the present embodiment, it is possible to improve the determination accuracy of a specific object by using a luminance image and a distance image and determining whether or not they are the same specific object according to the degree of overlap of the three-dimensional objects in each. Become.

In addition, a program that causes the computer to function as the vehicle exterior environment recognition device 120, and a storage medium such as a computer-readable flexible disk, magneto-optical disk, ROM, CD, DVD, or BD that records the program are also provided. Here, the program refers to a data processing means described in an arbitrary language or description method.

Although the preferred embodiment of the present invention has been described above with reference to the accompanying drawings, it goes without saying that the present invention is not limited to such an embodiment. It is clear that a person skilled in the art can come up with various modifications or modifications within the scope of the claims, and it is understood that these also naturally belong to the technical scope of the present invention. Will be done.

For example, in the above-described embodiment, an example of determining a preceding vehicle as a specific object located particularly far in front of the own vehicle 1 has been described, but not only in such a case, but also to a specific object located in the vicinity. Needless to say, it can be applied.

Further, in the above-described embodiment, an example is given in which two luminance images 180 captured by the two imaging devices 110 are used, and a first luminance image 180a and a distance image 182 based on both of them are used. explained. However, the luminance image 180 and the distance image 182 may be acquired independently of each other. For example, the luminance image 180 is obtained from one imaging device 110 of the three imaging devices 110, and the distance image 182 is obtained from the two imaging devices 110. May be derived.

It should be noted that each step of the vehicle exterior environment recognition method of the present specification does not necessarily have to be processed in chronological order in the order described as a flowchart, and may include processing in parallel or by a subroutine.

The present invention can be used as an external environment recognition device for identifying a specific object existing in the traveling direction of the own vehicle.

110 Imaging device 120 External environment recognition device 130 Vehicle control device 164 1st width out-licensing unit 166 2nd width out-licensing unit 168 Multiplicity out-licensing unit 170 Specific object judgment unit 180 Brightness image 182 Distance image 190 1st width 192 2nd width 194 Overlap Width 196 threshold

Claims (6)

A first width derivation unit that derives the first width of the first three-dimensional object based on a brightness image that can specify the brightness of the imaged object, and
A second width derivation unit that derives the second width of the second three-dimensional object based on a distance image that can specify the distance to be imaged, and a second width derivation unit.
The overlap width in which the first width and the second width overlap in the horizontal direction is obtained, and the larger of the overlap width / the first width and the overlap width / the second width is defined as the degree of overlap. Derivation part and
If the degree of overlap is equal to or greater than a predetermined threshold value, a specific object determination unit that determines that the first three-dimensional object and the second three-dimensional object are the same specific object, and
An external environment recognition device equipped with. The vehicle exterior environment recognition device according to claim 1, wherein the specific object determination unit lowers the threshold value for the degree of overlap determined to be the same specific object. The vehicle exterior environment recognition device according to claim 2, wherein the specific object determination unit lowers the threshold value for the overlap degree and then raises the threshold value for the overlap degree determined that the specific object is not the same. The vehicle exterior environment recognition device according to claim 2 or 3, wherein the specific object determination unit changes the threshold value for the degree of overlap stepwise according to the number of times that the same specific object is determined. The vehicle exterior environment recognition device according to any one of claims 1 to 4, wherein the specific object determination unit changes the threshold value according to whether or not the wiper is operated. The vehicle exterior environment recognition device according to any one of claims 1 to 5, wherein the specific object determination unit changes the threshold value according to the speed of the own vehicle.

Images