JP4815506B2 - Vehicle periphery monitoring device - Google Patents

Description

  The present invention relates to an apparatus for monitoring the periphery of the vehicle using an image of the periphery of the vehicle obtained by an imaging device mounted on the vehicle.

  A method for determining whether or not an object extracted from an image corresponds to an artificial structure has been proposed (see Patent Document 1).

JP 2008-071154 A

  However, when occlusion occurs, the feature amount of the object in the image is greatly different from the original feature amount, which may make it difficult to determine the type of the object.

  Accordingly, an object of the present invention is to provide a vehicle periphery monitoring device that can reduce the influence of occurrence of occlusion and improve the accuracy of determination of the type of an object.

  The vehicle periphery monitoring device of the present invention for solving the above-mentioned problem is a device that monitors the periphery of the vehicle using an image of the periphery of the vehicle obtained by an imaging device mounted on the vehicle, A first image processing means configured to extract an image area corresponding to an object existing around the vehicle from the image as the object area, and a plurality of the image areas arranged in a row direction constituting the object area For each of the cell rows, a first cell group including a plurality of cells whose luminance is included in the specified luminance range and adjacent in the row direction is defined. The number of overlapping cells in one cell group is calculated as a first index, a pair of first cell groups having the maximum first index is defined as a first pair, and an i-th pair (i = 1, 2,... N-1) one of the first And the first cell group that is not defined as the first to i-th pairs, and the first cell group having the maximum first index with the one first cell group is the (i + 1) th pair. And all of the first indices for the pair of first cell groups constituting each of the first to j1 pairs (where “j1” is a first constant equal to or less than N) is a first reference value. With the above requirements, the image processing apparatus includes a second image processing unit configured to determine that the object corresponds to an artificial structure.

  According to the vehicle periphery monitoring device of the present invention, it is determined whether or not all the first indices are equal to or greater than the first reference value for the pair of first cell groups constituting each of the first to j1 pairs. . “First index” means the number of overlapping cells in the column direction in which a plurality of cell rows are arranged.

  Since the first cell group is composed of a plurality of cells that are included in the designated luminance range and are continuously arranged, the length of the first cell group and the length of the first cell group may include the object in the image. The degree of “brightness uniformity” can be reflected. Since the i-th pair is composed of a pair of first cell groups that are sequentially selected in descending order of the first index, the number of overlapping cells in the column direction of the pair of first cell groups constituting a pair having a somewhat lower rank To some extent, the degree of “regularity of the outer shape” of the object in the image can be reflected.

  Therefore, it is highly determined by the above determination whether the object corresponds to an artificial structure having the properties that the luminance is uniform throughout the image obtained by the imaging apparatus and the outer shape is regular. Can be determined with accuracy. In addition, even if a part of the object is hidden behind other objects, the remaining part exhibits the above-mentioned properties, so the influence of the occurrence of occlusion is reduced, and as a result, the object corresponds to an artificial structure. Whether or not to do so can be determined with high accuracy.

  The second image processing means defines a second cell group composed of a plurality of cells continuous across the boundary between the first cell group and the background cell group in the row direction. Then, the degree of correlation between the two second cell groups that overlap in the column direction is calculated as a second index, and the first to j2 pairs ("j2" is a second constant equal to or less than the first constant "j1"). With respect to a pair of second cell groups corresponding to each of (2), it is determined that all the second indices are equal to or greater than a second reference value, and that the object corresponds to an artificial structure. It may be configured.

  According to the vehicle periphery monitoring device having the configuration, it is determined whether or not all the second indices are equal to or greater than the second reference value for the pair of second cell groups corresponding to the first to j2 pairs. . The “second index” means the degree of correlation between two second cell groups that overlap in the column direction.

  Since the second cell group is composed of a plurality of cells that are continuous across the boundary of the first cell group, the “contour” in the column direction of the object in the image is used for the degree of correlation of the second cell group. The degree of “uniformity of the extending direction” can be reflected.

  Therefore, according to the above determination, whether or not the object corresponds to an artificial structure having a regular outer shape whose contour extends uniformly in the row direction is highly accurate while reducing the influence of occurrence of occlusion. Can be determined.

Explanatory drawing regarding the mounting aspect to the vehicle of an imaging device and a vehicle periphery monitoring apparatus. The structure explanatory drawing of a vehicle periphery monitoring apparatus. The flowchart which shows the whole function of a vehicle periphery monitoring apparatus. The flowchart which shows the procedure of an artificial structure determination process. Explanatory drawing regarding the extraction method of a target object area | region. Explanatory drawing regarding the definition method of a 1st cell group and a 2nd cell group.

  An embodiment of a vehicle periphery monitoring device of the present invention will be described.

  First, the structure of the vehicle periphery monitoring apparatus of this embodiment is demonstrated.

  The vehicle periphery monitoring device includes an image processing unit 1 shown in FIGS. 1 and 2. The image processing unit 1 is connected to a pair of left and right infrared cameras (corresponding to the “imaging device” of the present invention) 2R and 2L that image the front of the vehicle 10 and sensors that detect the traveling state of the vehicle 10. The yaw rate sensor 3 for detecting the yaw rate of the vehicle 10, the vehicle speed sensor 4 for detecting the traveling speed (vehicle speed) of the vehicle 10, and the brake sensor 5 for detecting the presence or absence of a brake operation of the vehicle 10 are connected. Further, the image processing unit 1 has a speaker 6 for outputting auditory report information such as voice and the vicinity of the front window for displaying captured images taken by the infrared cameras 2R and 2L and visual report information. HUD7 arrange | positioned is connected.

  Instead of or in addition to the HUD, a display panel that displays a traveling state such as the vehicle speed of the vehicle 10 or a display that is a component of a navigation device mounted on the vehicle 10 may be employed as the image display device.

  The image processing unit 1 includes an ECU (electronic control unit) that includes an A / D conversion circuit, a memory such as a CPU, a RAM, and a ROM, an I / O circuit, and a bus that connects these.

  The image processing unit 1 includes first image processing means 11 and second image processing means 12. Each means includes a memory and a CPU that reads one or both of a program and data stored in the memory and executes arithmetic processing according to the read program or the like.

  Analog signals output from the infrared cameras 2R, 2L, the yaw rate sensor 3, the vehicle speed sensor 4, and the brake sensor 5 are converted into digital data by the A / D conversion circuit, and based on the data digitized by the CPU, humans (pedestrians) Or an object such as an occupant of a bicycle or a two-wheeled vehicle) or an artificial structure (such as a wall or a building), and when the detected object satisfies a predetermined notification requirement, the presence of the object through the speaker 6 or the HUD 7 Alternatively, the driver is notified of the possibility of contact between the vehicle 10 and the object.

  As shown in FIG. 2, the infrared cameras 2 </ b> R and 2 </ b> L are attached to the front portion of the vehicle 10 (the front grill portion in the figure) in order to image the front of the vehicle 10. The right infrared camera 2R is disposed at a position to the right of the center of the vehicle 10 in the vehicle width direction, and the left infrared camera 2L is disposed at a position to the left of the center of the vehicle 10 in the vehicle width direction. Their positions are symmetrical with respect to the center of the vehicle 10 in the vehicle width direction. The infrared cameras 2R and 2L are fixed so that their optical axes extend in the front-rear direction of the vehicle 10 in parallel with each other, and the heights of the respective optical axes from the road surface are equal to each other. The infrared cameras 2R and 2L have sensitivity in the far-infrared region, and have a characteristic that the higher the temperature of an object to be imaged, the higher the level of the output video signal (the higher the luminance of the video signal). is doing.

  When the distance to the object is measured by a radar, only one infrared camera may be mounted on the vehicle 10. Further, instead of or in addition to the infrared camera, a camera having sensitivity to light in various frequency bands, such as a CCD camera or a CMOS image sensor, may be mounted on the vehicle 10 as an imaging device.

  Next, functions of the vehicle periphery monitoring device having the above-described configuration will be described. Details of the image processing are disclosed in Japanese Patent Application Laid-Open Nos. 2001-006096 and 2007-310705, and will be described briefly.

  First, the “object extraction process” is executed by the first image processing means 11 (FIG. 3 / STEP 100). Specifically, the infrared image signal input to the image processing unit 1 from the infrared cameras 2R and 2L is A / D converted, and a grayscale image is generated based on the A / D converted infrared image signal. A reference grayscale image (right image) is binarized. In addition, an area where the object exists in the binarized image is extracted as the object area.

  Specifically, the pixel group constituting the high-luminance region of the binarized image is converted into run-length data, and a label (identifier) is attached to each of the line groups that overlap in the vertical direction of the reference image. Each is extracted as an object area. As a result, a rectangular region surrounding the high luminance region constituted by a group of high luminance pixels (pixel value = 1 pixel) as shown in FIG. 5 is extracted as the object region.

  Subsequently, the position of the center of gravity of each object (position on the reference image), the area, and the aspect ratio of the circumscribed rectangle are calculated. Further, tracking of the target object is performed, and the identity of the target object is determined for each calculation processing cycle of the image processing unit 1. Further, the outputs (vehicle speed detection value and yaw rate detection value) of the vehicle speed sensor 4 and the yaw rate sensor 5 are read. Further, in parallel with the calculation of the aspect ratio of the circumscribed rectangle and the tracking of the object during the time, the area corresponding to each object (for example, the area of the circumscribed rectangle of the object) is extracted as a search image from the reference image.

  Furthermore, by executing the correlation calculation, an image (corresponding image) corresponding to the search image is searched and extracted from the left image. Further, a real space distance (distance in the front-rear direction of the vehicle 10) z from the vehicle 10 to the object is calculated. Moreover, the real space position (relative position with respect to the vehicle 10) of each object is calculated. Further, the X component of the real space position (X, Y, Z) (see FIG. 2) of the object is corrected according to the time-series data of the turning angle. Further, a relative movement vector of the object with respect to the vehicle 10 is calculated.

  Subsequently, the “artificial structure determination process” is executed by the second image processing means 12 (FIG. 3 / STEP 200). As described later, when it is determined that the object corresponds to an artificial structure, the flag k is set to “1”, whereas when it is determined that the object does not correspond to an artificial structure, the flag k is set to “0”. "Is set.

  Thereafter, the image processing unit 1 executes the following series of processes. It is determined whether or not the flag k is 1 (FIG. 3 / STEP 300). If the determination result is affirmative (FIG. 3 / STEP 200... YES), that is, it is determined that the object corresponds to an artificial structure. In such a case, the alerting process described later is not executed, and the process after the object extraction process is repeated (see FIG. 3 / STEP 100, etc.).

  When it is determined that the flag k is not 1 (0) (FIG. 3 / STEP200... NO), the object is a human such as a pedestrian or a bicycle occupant, or a quadruped such as a deer, horse or dog. It is determined whether it falls under the attention object (FIG. 3 / STEP 400). Specifically, whether the object corresponds to a human or a quadruped animal based on the shape or size of the high-luminance area (see FIG. 5) in the object area on the gray scale image or the feature quantity such as the luminance distribution. It is determined whether or not.

  It is determined whether or not the real space distance z of the object is equal to or less than the threshold value zth, and the type of the object is determined as a requirement that the real space distance z of the object is determined to be equal to or less than the threshold value zth. Also good. The threshold value zth is set from the viewpoint of improving the object type determination accuracy.

  Further, when it is determined that the object corresponds to the attention object (FIG. 3 / STEP 400... NO), the level of possibility of contact between the vehicle 10 and the object is determined (FIG. 3 / STEP 500).

  Since this determination method is also disclosed in Japanese Patent Laid-Open Nos. 2001-006096 and 2007-310705, it will be briefly described. First, it is determined whether or not the real space distance z of the object is equal to or less than the product of the relative speed Vs and the margin time T. When it is determined that the real space distance z is equal to or less than the value, it is determined whether or not the object exists in the approach determination region (first contact determination process). The approach determination area extends symmetrically in the left-right direction of the vehicle 10 in front of the vehicle 10 and has a width obtained by adding an offset value to the width of the vehicle 10. When the determination result in the first contact determination process is affirmative, it is determined that there is a high possibility that the object is in contact with the vehicle 10. On the other hand, when the determination result in the first contact determination process is negative, the object exists in the entry determination area outside the approach determination area, and the relative movement vector of the object moves toward the approach determination area. It is further determined whether or not there is (second contact determination process). When the determination result in the second contact determination process is affirmative, it is determined that there is a high possibility that the target object is in contact with the vehicle 10.

  When it is determined that the possibility of contact between the vehicle 10 and the object is high (when the determination result in the first or second contact determination process is affirmative) (FIG. 3 / STEP 500... YES), “attention processing Is executed (FIG. 3 / STEP 600). As a result, sound is output from the speaker 6 as “beep”. In addition, an object and a rectangular frame surrounding the object are displayed on the HUD 7.

  When it is determined that the object does not correspond to the attention object (FIG. 3 / STEP 400... NO), or when it is determined that the possibility of contact between the vehicle 10 and the object is low (FIG. 3 / STEP 500... NO). The processing after the object extraction processing is repeated without executing the alert processing (see FIG. 3 / STEP 100, etc.).

  Here, the artificial structure determination process (FIG. 3 / STEP 200) will be described in detail.

  First, for each of a plurality of cell rows constituting the object region, a first cell group including a plurality of adjacent cells whose luminance is included in the designated luminance range is defined (FIG. 4 / STEP 202).

  For example, the average luminance of the object area (see FIG. 5) or a range equal to or higher than a predetermined luminance can be adopted as the designated luminance range. Thereby, for example, in the image coordinate system (x, y) defined as shown in FIG. 6A, a plurality of cells (or pixels) arranged in succession in the x direction (row direction). And a plurality of first cell groups y1 to y19 arranged in the y direction (column direction) are defined.

  Each of the row direction and the column direction may match each of the x direction and the y direction, or may match each of different arbitrary directions. For example, the row direction may coincide with the y direction while the column direction may coincide with the x direction. The row direction may coincide with a direction inclined 30 ° counterclockwise with respect to the x-axis, while the column direction may coincide with a direction inclined 30 ° counterclockwise with respect to the y-axis. Further, the row direction and the column direction may not be orthogonal to each other. For example, the row direction may coincide with a direction inclined 30 ° counterclockwise with respect to the x axis, while the column direction may coincide with a direction inclined 45 ° counterclockwise with respect to the y axis.

  Further, the number of overlapping cells in the pair of first cell groups in the column direction in which a plurality of cell rows are arranged is calculated as the first index C1 (FIG. 4 / STEP 204). Note that it is not necessary to calculate the first index C1 for every combination of two different first cell groups, and the two first items that are likely to form a pair in view of the coordinate values at both ends in the image coordinate system. The first index C1 may be calculated for the cell group.

  Furthermore, the i-th pair (i = 1 to N) is defined based on the first index C1 (FIG. 4 / STEP 206).

  Specifically, a pair of first cell groups with the maximum first index C1 is defined as “first pair”. Then, among the first cell group of one of the i-th pair (i = 1, 2,..., N−1) and the first cell group not defined as the first to i-th pairs, One first cell group having the maximum first index with the cell group is defined as the (i + 1) th pair.

  Thereby, for example, among the first cell groups y1 to y19 shown in FIG. 6A, the first cell groups y2 and y3 are defined as the first pair. Further, one first cell group y2 of the first pair and the first cell group y4 having the maximum first index are defined as the second pair. Hereinafter, the 18th pair is sequentially defined in the same manner. The first cell group y8 to y12 is shorter than the other first cell groups, but this suggests that part of the object may be hidden behind other objects due to the occurrence of occlusion. is doing.

  After some or all of the first cell groups y1 to y19 are displaced in the x direction (row direction) so that the left ends or right ends (ends in the row direction) of the first cell groups y1 to y19 are aligned, As described above, the i-th pair may be defined.

  With the requirement that the dispersion of the end positions is not more than an allowable value for at least one of the first cell groups y1 to y19 in the x direction, the deviation of the length in the x direction, that is, the number of cells constituting the first cell group The i-th pair may be defined after the first index C1 is evaluated so that the first index C1 becomes higher as the deviation is smaller.

  Note that, as illustrated in FIG. 6B, the first cell groups y1 to y19 may be rearranged in the descending order of the first index C1 of the affiliation pair (may be sorted).

  Furthermore, a second cell group composed of a plurality of cells that are continuous across the boundary of the first cell group is defined (FIG. 4 / STEP 208). Thereby, as shown in FIG. 6C, the second cell groups Δy1 to Δy19 configured by a plurality of cells that are continuous across the right boundary of each of the first cell groups y1 to y19 are defined. The Note that a second cell group including a plurality of cells that are continuous across the left boundary of each of the first cell groups y1 to y19 may be defined alternatively or additionally.

  Further, the degree of correlation between the pair of second cell groups is calculated as the second index (FIG. 4 / STEP 210). As a method for calculating the correlation degree, a known method such as normalized correlation is employed. Note that the second index C2 need not be calculated for every combination of two different second cell groups, but a pair of second cell groups corresponding to each of the pair of first cell groups constituting the pair is targeted. The second index C2 may be calculated as follows. As a result, the degree of correlation between the pair of second cell groups Δy3 and Δy2 corresponding to the first pair is calculated as the second index C2. In addition, the degree of correlation between the pair of second cell groups Δy2 and Δy4 corresponding to the second pair is calculated as the second index C2.

  It is determined whether or not the first index C1 (j1) of the pair of first cell rows constituting the j1 pair (“j1” is a first constant less than N) is greater than or equal to the first reference value TH1. (FIG. 4 / STEP 212).

  As described above, the first index (the number of overlapping cells in the column direction) C1 is defined in order from the first to the Nth pairs in descending order, and therefore, the determination is made for all of the first to j1th pairs. This is equivalent to determining whether C1 is equal to or greater than the first reference value TH1. That is, in the determination, whether the object has one or a plurality of portions having a height corresponding to the first index j1 and a width corresponding to the first reference value TH1 in the real space. This corresponds to determining whether or not.

  The first constant j1 is set on the basis of the standard height in the real space of an artificial structure such as a wall or a column in consideration of the degree of hiding behind other objects when occlusion occurs. The first reference value TH1 is set based on the standard width in the real space of the artificial structure in consideration of the degree of occlusion that is hidden behind other objects. Since the size of the object in the image changes according to the length of the real space distance, each of the first constant j1 and the first reference value TH1 is the real space position calculated in the object extraction process (see FIG. 3 / STEP 100). It can be variably set according to.

  For example, as shown in FIG. 6B, the first to eleventh pairs corresponding to the standard height (a value slightly subtracted in consideration of the occurrence of occlusion) Yth of the artificial structure in the image. When the number of overlapping cells (first index C1) is equal to or greater than one threshold TH1 corresponding to the standard width Xth of the artificial structure in the image, the determination result is positive.

  When it is determined that the first index C1 (j1) of the pair of first cell rows constituting the j1 pair is equal to or greater than the first reference value TH1 (FIG. 4 / STEP212... YES), the first to j2th pairs ( “J2” is a second constant less than or equal to the first constant j1.) For each pair of second cell groups, it is further determined whether or not all the second indices C2 are greater than or equal to the second reference value TH2. It is determined (FIG. 4 / STEP 214).

  As described above, since the plurality of cells sandwiching the boundary of the first cell group are defined as the second cell group, the determination is made according to the second constant j2 with respect to the column direction (y direction). This corresponds to determining whether or not one or a plurality of portions that are stably extended by a distance are included.

  For example, as shown in FIG. 6 (c), all the second indices C2 of the number pairs over the range Yth ′ below the standard height Yth of the artificial structure in the image are greater than or equal to the second reference value. In some cases, the determination result is positive.

  When it is determined that all the second indices C2 are equal to or greater than the second reference value TH2 for the pair of second cell groups corresponding to the first to j2 pairs (FIG. 4 / STEP214... YES), the flag k Is set to “1” indicating that the object corresponds to an artificial object (FIG. 4 / STEP 216).

  On the other hand, when one of the determination results is negative (FIG. 4 / STEP 212... NO, STEP 214... NO), the flag k is set to “0” indicating that the object does not correspond to the artificial object. (FIG. 4 / STEP 218).

  According to the vehicle periphery monitoring device that exhibits the above function, whether or not all the first indices C1 are equal to or greater than the first reference value TH1 for the pair of first cell groups constituting each of the first to j1 pairs. Is determined (see FIG. 4 / STEP 212). The first index C1 means the number of overlapping cells in the column direction in which a plurality of cell rows are arranged (see the y direction in FIG. 6).

  Since the first cell group is composed of a plurality of cells that are included in the designated luminance range and are continuously arranged, the length of the first cell group and the length of the first cell group may include the object in the image. The degree of “luminance uniformity” can be reflected (see FIG. 6A). Since the i-th pair (i = 1, 2,... N) is composed of a pair of first cell groups that are sequentially selected in descending order of the first index C1, a pair of first cells constituting a pair having a somewhat lower order. The degree of “regularity of the outer shape” of the object in the image can be reflected on the number of overlapping cells in the column direction of the cell group (see FIG. 6B).

  Therefore, by the determination (see FIG. 4 / STEP 212), the artificial structure having the properties that the luminance is uniform overall and the outer shape is regular in the images obtained by the infrared cameras 2R and 2L. Whether or not the object is applicable can be determined with high accuracy. In addition, even if a part of the object is hidden behind other objects, the remaining part exhibits the above-mentioned properties, so the influence of the occurrence of occlusion is reduced, and as a result, the object corresponds to an artificial structure. Whether or not to do so can be determined with high accuracy.

  Further, it is determined whether or not all second indices C2 are equal to or greater than the second reference value TH2 for the pair of second cell groups corresponding to the first to j2 pairs (see FIG. 4 / STEP 214). . The second index C2 means the degree of correlation between two second cell groups that overlap in the column direction.

  Since the second cell group is composed of a plurality of cells that are continuous across the boundary of the first cell group, the “contour” in the column direction of the object in the image is used for the degree of correlation of the second cell group. The degree of “uniformity of the extending direction” can be reflected (see FIG. 6C).

  Accordingly, whether or not the object corresponds to an artificial structure having a regular outer shape whose contour extends uniformly in the column direction is determined by the above determination (see FIG. 4 / STEP 214). It can be determined with high accuracy while reducing the influence.

  Further, the determination result that the flag k is 0, that is, the object does not correspond to the artificial structure (see FIG. 3 / STEP 300... NO), the brightness of the object is not uniform, and the outer shape is complicated. Therefore, the determination result may be supplementarily used when determining that the object corresponds to an attention object such as a human being.

  As another embodiment of the present invention, the definition of the second cell group, the calculation of the second index C2, and the comparison between the second index C2 and the second reference value TH2 may be omitted (FIG. 4 / STEP 208, 210). And 214). According to this embodiment, when it is determined that the first constant C1 (j1) of the j1st pair is equal to or greater than the first reference value TH1 (FIG. 4 / STEP212... YES), the flag k is set to “1”. (FIG. 4 / STEP 216). When it is determined that the first constant C1 (j1) of the j1st pair is equal to or greater than the first reference value TH1 (FIG. 4 / STEP 212... YES), the flag k is set to “1” (FIG. 4 / STEP 216). .

  Even in the other embodiment, even if a part of the object is hidden behind another object due to the above-described reason, the remaining part exhibits the above-described property, so that the influence of the occurrence of occlusion is reduced. As a result, it can be determined with high accuracy whether or not the object corresponds to an artificial structure.

  In addition, each of the plurality of first cell groups includes a portion constituted by one or a plurality of characteristic cells such as significantly high or low luminance, and each of the portions is balanced with respect to the column direction. When regularly arranged such as forming a certain angle, it may be determined that the plurality of first cell groups correspond to the artificial structure. According to the embodiment, it can be determined whether or not the object corresponds to an artificial structure such as a wall in which a hole is formed or a characteristic shape appears on the surface.

DESCRIPTION OF SYMBOLS 1 ... Image processing unit, 2R, 2L ... Infrared camera (imaging device), 10 ... Vehicle, 11 ... 1st image processing means, 12 ... 2nd image processing means

Claims (2)

An apparatus for monitoring the periphery of the vehicle using an image of the periphery of the vehicle obtained by an imaging device mounted on the vehicle,
A first image processing means configured to extract an image area corresponding to an object existing around the vehicle from the image as an object area;
For each of a plurality of cell rows arranged in the row direction constituting the object region, a first cell group including a plurality of cells whose luminance is included in a specified luminance range and adjacent in the row direction is defined. ,
Calculating the number of overlapping cells of a pair of first cell groups as a first index in the column direction in which the plurality of cell rows are arranged;
A pair of first cell groups having the maximum first index is defined as a first pair, and one first cell group of the i-th pair (i = 1, 2,..., N−1), Of the first cell groups not defined as the i-th pair, one first cell group having the maximum first index with the one first cell group is defined as the i + 1 pair, and
Requirement that all the first indices be greater than or equal to the first reference value for a pair of first cell groups constituting each of the first to j1 pairs ("j1" is a first constant equal to or less than N). And a second image processing means configured to determine that the object corresponds to an artificial structure. The vehicle periphery monitoring device according to claim 1,
The second image processing means defines a second cell group composed of a plurality of cells that are continuous across the boundary between the first cell group and the background cell group in the row direction,
Calculating the degree of correlation between two second cell groups overlapping in the column direction as a second index; and
For the pair of second cells corresponding to each of the first to j2 pairs (“j2” is a second constant equal to or less than the first constant “j1”), all the second indices are the second reference. The vehicle periphery monitoring device is configured to determine that the target object corresponds to an artificial structure, as a further requirement that the value is equal to or greater than a value.