JP5743635B2 - Foreign object detection device - Google Patents

Description

  The present invention relates to a foreign object detection device that detects a foreign object present in a three-dimensional image.

For example, in the foreign object detection device disclosed in Patent Document 1 below, a sensor that can acquire the distance value of the scanning point is installed by scanning a laser pulse, and the distance acquired by the sensor The altitude value is calculated from the value.
In the foreign object detection device, the altitude value is compared with a preset threshold value, and if the number of scanning points whose altitude value is equal to or greater than the threshold value is equal to or greater than the set number, the foreign object (vehicle) is present. to decide.

JP 2004-272842 A (paragraph number [0016])

  Since the conventional foreign matter detection apparatus is configured as described above, it is possible to accurately detect foreign matter if the distance value of the scanning point can be accurately obtained under a high SNR environment. For example, if the SNR environment is low due to the influence of the above, and the distance value of the scanning point cannot be obtained accurately, the foreign matter detection accuracy deteriorates.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to obtain a foreign matter detection device that can accurately detect foreign matter even in a low SNR environment.

The foreign matter detection device according to the present invention is a means for capturing a three-dimensional image of an observation region by scanning a laser beam, and before starting the foreign matter detection process, the observation region in a situation where no foreign matter is present. The three-dimensional image is captured and the reflection intensity value of each pixel of the three-dimensional image is output, and the three-dimensional image of the observation area is captured during the foreign object detection process, and the reflection intensity value of each pixel of the three-dimensional image Of the reflection intensity value of each pixel output from the three-dimensional image pickup means before starting the foreign substance detection processing in units of pixels in which the vertical component of the three-dimensional image is the same. The threshold setting means for acquiring the maximum reflection intensity value and setting the maximum reflection intensity value as the upper limit threshold, and the three-dimensional image in the same pixel unit as the vertical component of the three-dimensional image during the foreign object detection processing Reflection intensity value and threshold value of each pixel output from the imaging means A pixel detection unit that compares the upper threshold value set by the setting unit and detects a pixel whose reflection intensity value is higher than the upper threshold value, and the foreign object detection unit sets the total number of pixels detected by the pixel detection unit in advance. When the number of pixels is larger than the number of pixels, it is determined that a foreign object exists in the three-dimensional image.

According to the present invention, the laser beam is scanned to capture a three-dimensional image of the observation region, and the three-dimensional image of the observation region can be obtained in the absence of foreign matter before starting the foreign matter detection process. The image is captured and the reflection intensity value of each pixel of the 3D image is output. At the same time, the 3D image of the observation area is captured during the foreign object detection process, and the reflection intensity value of each pixel of the 3D image is calculated. Among the reflection intensity values of each pixel output from the 3D image pickup means before starting the foreign substance detection processing in the same pixel unit as the output 3D image pickup means and the vertical component of the 3D image Threshold value acquisition means for acquiring the maximum reflection intensity value and setting the maximum reflection intensity value as the upper limit threshold, and 3D image imaging during the detection of foreign matter in the same pixel unit of the vertical component of the 3D image Reflection intensity value of each pixel output from the means and threshold setting The upper limit threshold set by the means is compared, and a pixel detection means for detecting a pixel whose reflection intensity value is higher than the upper limit threshold is provided. The foreign object detection means is preset with the total number of pixels detected by the pixel detection means. If the number of pixels is larger than the number of pixels, it is determined that a foreign object is present in the three-dimensional image. Therefore, there is an effect that the foreign object can be accurately detected even in a low SNR environment.

It is a block diagram which shows the foreign material detection apparatus by Embodiment 1 of this invention. It is a block diagram which shows the threshold value setting part 11 of the foreign material detection apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows the installation state of the three-dimensional image imaging device. It is a block diagram which shows the foreign material detection apparatus by Embodiment 2 of this invention. It is a block diagram which shows the threshold value calculation part 31 of the foreign material detection apparatus by Embodiment 2 of this invention. It is a block diagram which shows the threshold value processing part 32 of the foreign material detection apparatus by Embodiment 2 of this invention. It is explanatory drawing which shows the example of a setting of the foreign material detection window at the time of threshold value generation. It is explanatory drawing which shows the example of a setting of the foreign material detection window at the time of a threshold value process. It is a block diagram which shows the foreign material detection apparatus by Embodiment 4 of this invention. It is explanatory drawing which shows the example of a setting of the window for false detection removal set by the false detection recognition part. It is a block diagram which shows the foreign material detection apparatus by Embodiment 5 of this invention.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a foreign object detection apparatus according to Embodiment 1 of the present invention.
The foreign object detection device of FIG. 1 is an apparatus that is applied to, for example, a gate entrance monitoring system or a road traffic monitoring system, and detects a person, a vehicle, or the like as a foreign object.
In the first embodiment, for convenience of explanation, an example in which a vehicle is detected as a foreign object will be described. However, an object other than a vehicle may be detected as a foreign object.

In FIG. 1, for example, a receiving scanless type laser image measuring device corresponds to the three-dimensional image pickup device 1, and before the foreign object detection process is started, the observation region is observed in a situation where no foreign object vehicle exists. A three-dimensional image (background image) is imaged, and the reflection intensity value I _{i, j} of each pixel of the three-dimensional image is output. An image that may be displayed) is captured, and the reflection intensity value I _{i, j} of each pixel of the three-dimensional image is output. Note that the 3D image capturing apparatus 1 constitutes a 3D image capturing unit.
The foreign matter detection processing device 2 detects a vehicle (foreign matter) existing in the three-dimensional image using the reflection intensity values I _{i, j} of each pixel of the three-dimensional image output from the three-dimensional image pickup device 1. It is a device to do.

The threshold value setting unit 11 of the foreign object detection processing device 2 is constituted by, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like, and the data direction (i) that is a vertical component of the three-dimensional image is the same. The maximum reflection intensity value I _imax and the minimum reflection intensity among the reflection intensity values I _{i, j} of each pixel output from the three-dimensional image pickup apparatus 1 before starting the foreign substance detection process for each pixel of A _process of obtaining the value I _imin and setting the maximum reflection intensity value I _imax to the upper limit threshold TH _UP and setting the minimum reflection intensity value I _imin to the lower limit threshold TH _LOW is performed. The threshold setting unit 11 constitutes a threshold setting unit.

The threshold processing unit 12 is configured by, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like. The reflection intensity value I _{i, j of} each pixel output from the original image pickup apparatus 1 is compared with the upper limit threshold TH _UP and the lower limit threshold TH _LOW set by the threshold setting unit 11, and the reflection intensity value I _{i, j} is upper threshold _{TH UP} higher pixel or reflected intensity values _{I i, j} is carried out a process of detecting a lower pixel than the lower limit threshold _{TH lOW.}
In addition, the threshold processing unit 12 is, for example, a fixed threshold (upper limit threshold TH _UP , lower limit) set in advance at the start-up before the reflection intensity value I _{i, j of} each pixel is output from the 3D image capturing apparatus 1. When a threshold value switching flag of “0” instructing the output of the threshold value TH _LOW is output to the threshold value setting unit 11 and the output of the reflection intensity value I _{i, j} is started from the three-dimensional image pickup device 1, the three-dimensional image is displayed. A threshold value switching flag of “1” for instructing the output of threshold values (upper threshold TH _UP , lower threshold TH _LOW ) set from the reflection intensity values I _{i, j} of each pixel of (background image) is given to the threshold setting unit 11. Perform the output process. The threshold processing unit 12 constitutes a pixel detection unit.

  The foreign object detection unit 13 is composed of, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like, and the total number of pixels detected by the threshold processing unit 12 is larger than a preset effective pixel number Then, a process for determining that a foreign substance is present in the three-dimensional image is performed. In addition, the foreign material detection part 13 comprises the foreign material detection means.

In the example of FIG. 1, it is assumed that each of the threshold setting unit 11, the threshold processing unit 12, and the foreign matter detection unit 13 that are components of the foreign matter detection processing device 2 in the foreign matter detection device is configured by dedicated hardware. However, the foreign object detection processing device 2 may be configured by a computer.
In this case, a program describing the processing contents of the threshold setting unit 11, the threshold processing unit 12, and the foreign matter detection unit 13 is stored in the memory of the computer, and the CPU of the computer executes the program stored in the memory. What should I do?

FIG. 2 is a block diagram showing the threshold setting unit 11 of the foreign object detection device according to Embodiment 1 of the present invention.
In FIG. 2, the threshold switching flag acquisition unit 21 performs a process of acquiring a threshold switching flag output from the threshold processing unit 12.
The threshold generation unit 22 stores preset fixed thresholds (upper limit threshold TH _UP , lower limit threshold TH _LOW ), and if the threshold switching flag acquired by the threshold switching flag acquiring unit 21 is “0”. If the threshold switching flag acquired by the threshold switching flag acquisition unit 21 is “1” while the fixed thresholds (upper threshold TH _UP , lower threshold TH _LOW ) are output to the threshold processing unit 12, the three-dimensional image is displayed. The maximum reflection intensity among the reflection intensity values I _{i, j} of each pixel output from the three-dimensional image pickup apparatus 1 before starting the foreign substance detection process in units of pixels having the same data direction (i). to obtain the value _{I imax} and the minimum of the reflected intensity values _{I imin,} and sets the maximum reflection intensity values _{I imax} the upper threshold _{TH uP,} it sets the minimum of the reflected intensity values _{I imin} the lower threshold _{TH LOW} It carries out a process of outputting the upper threshold _{TH UP} and lower threshold _{TH LOW} to the threshold processing unit 12.

Next, the operation will be described.
FIG. 3 is an explanatory diagram showing an installation state of the three-dimensional image pickup apparatus 1.
In the example of FIG. 3, the direction directly above the ground (upward in the figure) is defined as the altitude direction (Z), and the direction perpendicular to the altitude direction (rightward in the figure) is defined as the depth direction (X). doing.
In the example of FIG. 3, a direction perpendicular to the altitude-depth plane is defined as the traveling direction.

When the beam is scanned from the 3D imaging device 1, the distance value in the line-of-sight direction obtained from the beam irradiation region is D, the installation height of the 3D imaging device 1 is H, the beam scanning angle is φ, and the sensor is off. Assuming that the Nadia angle θ, the altitude value Z and the depth distance value X at the beam scanning point are expressed by the following equation (1).
Z = HD-cos (θ + φ)
X = D · sin (θ + φ) (1)
The three-dimensional image capturing apparatus 1 scans the beam in the one-dimensional direction in the depth direction, and acquires a two-dimensional cross section of the observation region from the multipoint distance value X and the scanning angle φ in the scanning range.
In addition, the three-dimensional image capturing apparatus 1 acquires a three-dimensional image when a vehicle that is a foreign object passes through the observation region.
The image shown in FIG. 3 is an example of an image captured by the three-dimensional image capturing apparatus 1. The horizontal direction of the image is defined as the line direction (j), and the vertical direction is defined as the data direction (i). ing.

The three-dimensional image capturing apparatus 1 captures a three-dimensional image (background image) of an observation region in a situation where no foreign vehicle exists before starting the foreign object detection process, and the three-dimensional image and it outputs the reflection intensity value I _i of each pixel of _{the j} foreign matter detecting device 2.
The three-dimensional image capturing apparatus 1 captures a three-dimensional image of the observation region (an image in which a vehicle may be reflected on the background) during the foreign object detection process, and each pixel of the three-dimensional image is captured. The reflection intensity value I _{i, j} is output to the foreign object detection processing device 2.
The three-dimensional image pickup apparatus 1 can acquire an altitude value and a distance value in addition to the reflection intensity value I _{i, j} of each pixel. Since the altitude value and the distance value are not used _, only the reflection intensity value I _{i, j of} each pixel is output to the foreign object detection processing device 2.

The threshold value processing unit 12 of the foreign object detection processing device 2 is, for example, a fixed threshold value set in advance at the start-up before the reflection intensity value I _{i, j of} each pixel is output from the three-dimensional image capturing device 1. A threshold switching flag of “0” instructing output of (upper limit threshold TH _UP , lower limit threshold TH _LOW ) is output to the threshold setting unit 11.
When the output of the reflection intensity value I _{i, j} is started from the three-dimensional image capturing apparatus 1, a threshold (upper limit) set from the reflection intensity value I _{i, j} of each pixel of the three-dimensional image (background image). Threshold value switching flag “1” instructing output of threshold value TH _UP and lower limit threshold value TH _LOW is output to threshold value setting unit 11.

The threshold switching flag acquisition unit 21 of the threshold setting unit 11 acquires the threshold switching flag output from the threshold processing unit 12 and outputs the threshold switching flag to the threshold generation unit 22.
The threshold generation unit 22 of the threshold setting unit 11 stores, for example, fixed thresholds (upper limit threshold TH _UP , lower limit threshold TH _LOW ) set in advance by the user, and is output from the threshold switching flag acquisition unit 21. If the threshold switching flag is “0”, the fixed thresholds (upper threshold TH _UP , lower threshold TH _LOW ) are output to the threshold processing unit 12.

If the threshold value switching flag output from the threshold value switching flag acquisition unit 21 is “1”, the threshold value generation unit 22 starts the foreign object detection process for each pixel unit in which the data direction (i) of the three-dimensional image is the same. Among the reflection intensity values I _{i, j} of each pixel previously output from the three-dimensional image capturing apparatus 1, the maximum reflection intensity value I _imax and the minimum reflection intensity value I _imin are acquired.
For example, when the number of pixels in the line direction (j) of the three-dimensional image is M and the number of pixels in the data direction (i) is N (M × N image), the vertical component is M in the same pixel unit. Among the reflection intensity values I _{i, j} of the pixels, the maximum reflection intensity value I _imax and the minimum reflection intensity value I _imin are obtained.
The threshold generation unit 22 sets the maximum reflection intensity value I _imax to the upper limit threshold TH _UP and sets the minimum reflection intensity value I _imin to the lower limit threshold TH _LOW in units of pixels having the same vertical component. The upper limit threshold TH _UP and the lower limit threshold TH _LOW are output to the threshold processing unit 12.
In this example, since the number of pixels in the data direction (i), which is the vertical component, is N, N upper limit threshold values TH _UP and N lower limit threshold values TH _LOW are output to the threshold processing unit 12.

When the threshold processing unit 12 receives the upper limit threshold TH _UP and the lower limit threshold TH _LOW in the same pixel unit from the threshold setting unit 11 in the data direction (i), the threshold processing unit 12 is output from the 3D image capturing apparatus 1 during the foreign object detection processing. The reflection intensity value I _{i, j of} each pixel is compared with the upper limit threshold TH _UP and the lower limit threshold TH _LOW set by the threshold setting unit 11. However, the threshold value for comparing the reflection intensity values _{I i,} and _j, the reflection intensity value _{I i, j} and the data direction (i) is the same upper threshold _{TH UP} and a lower threshold _{TH LOW.}
Inequality if the threshold processing unit 12, when the reflection intensity value _{I i} of the _{pixel, j} is higher than the upper threshold _{TH UP,} or reflection intensity value _{I i} of the _{pixel, j} is less than the lower threshold _{TH LOW} (formula (2) Is established), the pixel is determined to be a pixel constituting a part of the vehicle (foreign matter), and the pixel count value C indicating the number of detected foreign matter is incremented. However, the initial value of the pixel count value C is zero.
I _{i, j} <TH _LOW or I _{i, j} > TH _UP (2)

When the pixel count value C output from the threshold processing unit 12 is greater than the preset effective pixel number, the foreign object detection unit 13 determines that there is a foreign object in the three-dimensional image.
For example, if a foreign object exists in the 3D image, a foreign object detection flag of “1” is output, and if no foreign object exists in the 3D image, a foreign object detection flag of “0” is output. To do.
Note that the number of effective pixels does not have to be a fixed threshold value, and erroneous detection can be reduced by setting an upper limit value and a lower limit value to be a band pass.

As is apparent from the above, according to the first embodiment, before starting the foreign object detection process, a three-dimensional image of the observation region is captured in a situation where no foreign object is present, and the three-dimensional image is obtained. outputs the reflection intensity value I _{i, j} of each pixel, when the foreign matter detection processing by imaging the three-dimensional image of the observation area, and outputs the reflection intensity value I _{i, j} of each pixel of the three-dimensional image The reflection intensity value I _{i of} each pixel output from the 3D image pickup device 1 before starting the foreign substance detection processing in the same pixel unit as that of the 3D image pickup device 1 and the vertical component of the 3D image _{. In j} , the maximum reflection intensity value I _imax and the minimum reflection intensity value I _imin are obtained, the maximum reflection intensity value I _imax is set to the upper limit threshold TH _UP , and the minimum reflection intensity value I _imin is _set to threshold setting unit that sets a lower threshold _{TH LOW} 1, the vertical direction component the same pixel of the three-dimensional image, is set by the reflection intensity value I _{i, j} and the threshold setting unit 11 of each pixel output from the three-dimensional imaging apparatus 1 when the foreign matter detection processing comparing the upper threshold _{TH UP} and lower threshold _{TH lOW,} the threshold and the reflected intensity values _{I i, j} is the upper threshold _{TH UP} higher pixel or the reflection intensity value _{I i, j} detects the lower pixels than the lower limit threshold _{TH lOW} When the total number of pixels detected by the threshold processing unit 12 is greater than the preset effective pixel number, a vehicle that is a foreign object exists in the three-dimensional image. Therefore, the upper limit threshold TH _UP and the lower limit threshold TH _LOW are dynamically set according to the imaging environment of the three-dimensional image capturing apparatus 1, and as a result, even in a low SNR environment , Different An effect of vehicle can be detected accurately it is.

In the first embodiment, the threshold processing unit 12 includes the reflection intensity value I _{i, j of} each pixel output from the 3D image capturing apparatus 1 and the upper limit threshold TH _UP and the lower limit set by the threshold setting unit 11. The threshold TH _LOW is compared, and a pixel whose reflection intensity value I _{i, j} is higher than the upper limit threshold TH _UP or a pixel whose reflection intensity value I _{i, j} is lower than the lower limit threshold TH _LOW is shown. When the vehicle that is a foreign object is not detected by the unit 13, the threshold processing unit 12 uses, for example, a combination of a LOG (Laplacian Of Gaussian Filter) filter and a zero-cross filter, a Canny filter, or the like by the 3D image capturing apparatus 1. When edge detection processing is performed on the captured 3D image and an edge is detected from the 3D image, The detection unit 13 may determine that there is a foreign object in the three-dimensional image.

In the first embodiment, the threshold processing unit 12 is a fixed value set in advance at the start-up before the reflection intensity value I _{i, j of} each pixel is output from the 3D image capturing apparatus 1. When a threshold value switching flag of “0” instructing the output of the threshold value is output and the output of the reflection intensity value I _{i, j} is started from the 3D image pickup apparatus 1, each pixel of the 3D image (background image) is started. Although the output of the threshold switching flag of “1” instructing the output of the threshold set from the reflection intensity value I _{i, j} is shown as an example, for example, in an environment where the background does not change A “0” threshold switching flag is output to instruct the output of fixed threshold values (upper threshold TH _UP and lower threshold TH _LOW ) set in advance, and a three-dimensional image (background image) in an environment where the background changes. reflection intensity value I _i of each pixel of _the threshold that is set from _j (Upper threshold _{TH UP,} lower threshold _{TH LOW)} may output a threshold switching flags indicating the output of "1".

Embodiment 2. FIG.
4 is a block diagram showing a foreign object detection device according to Embodiment 2 of the present invention. In the figure, the same reference numerals as those in FIG.
The three-dimensional image pickup device 3 corresponds to, for example, a reception scanless type laser image measurement device, and before starting the foreign object detection process, the three-dimensional image ( A background image is captured and the altitude value Z _{i, j} of each pixel of the three-dimensional image is output, and a three-dimensional image of the observation area (a vehicle can be shown on the background) during foreign object detection processing The image is captured and the altitude value Z _{i, j} of each pixel of the three-dimensional image is output. Note that the three-dimensional image capturing apparatus 3 constitutes a three-dimensional image capturing unit.
The foreign matter detection processing device 4 detects a vehicle (foreign matter) present in the three-dimensional image using the altitude value Z _{i, j} of each pixel of the three-dimensional image output from the three-dimensional image pickup device 3. Device.

The threshold value calculation unit 31 of the foreign object detection processing device 4 is composed of, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like. For each foreign object detection window (predetermined region) in a three-dimensional image, a foreign object is detected. Before starting the detection process _, the average value Zm and standard deviation value Zstd of the altitude values Z _{i, j} output from the three-dimensional image pickup device 3 are obtained, and the foreign matter is obtained from the average value Zm and the standard deviation value Zstd. Processing for calculating the threshold value THb _{(i, j)} is performed. The threshold value calculation unit 31 constitutes a threshold value calculation unit.

The threshold processing unit 32 is configured by, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like. For each foreign object detection window in the three-dimensional image, from the 3D image pickup device 3 during the foreign object detection processing. An average value Zm _win and standard deviation value Zstd _win of the output altitude values Z _{i, j} of each pixel are obtained, and an index representing a vehicle (foreign matter) other than the background from the average value Zm _win and standard deviation value Zstd _win. A process for calculating the foreign matter index S _{(i, j)} is performed.
Further, the threshold processing unit 32 compares the foreign matter index S _{(i, j)} with the foreign matter threshold THb _{(i, j)} calculated by the threshold calculating unit 31, and the foreign matter index S _{(i, j)} is the foreign matter threshold THb _{( i, j)} A process of detecting a region (foreign object detection window) higher than that is performed.
In addition, the threshold value processing unit 32 has a fixed foreign substance threshold value THb _{(i, j)} set in advance at the start-up time before the altitude value Z _{i, j of} each pixel is output from the 3D image pickup device 3, for example. When a threshold value switching flag of “0” instructing output is output to the threshold value calculation unit 31 and output of the altitude value Z _{i, j} is started from the three-dimensional image pickup device 3, a three-dimensional image (background image) is displayed. Processing for outputting a threshold value switching flag of “1” to the threshold value calculation unit 31 for instructing output of the foreign substance threshold value THb _{(i, j)} set from the altitude value Z _{i, j of} each pixel is performed. Note that the threshold processing unit 32 constitutes an area detection unit.

  The foreign object detection unit 33 is composed of, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like, and the total number of areas detected by the threshold processing unit 32 is greater than a preset number of effective areas Then, a process for determining that a foreign substance is present in the three-dimensional image is performed. The foreign matter detection unit 33 constitutes foreign matter detection means.

In the example of FIG. 4, it is assumed that each of the threshold value calculation unit 31, the threshold value processing unit 32, and the foreign material detection unit 33 that are components of the foreign material detection processing device 4 in the foreign material detection device is configured by dedicated hardware. However, the foreign object detection processing device 4 may be configured by a computer.
In this case, a program describing the processing contents of the threshold calculation unit 31, the threshold processing unit 32, and the foreign matter detection unit 33 is stored in the memory of the computer, and the CPU of the computer executes the program stored in the memory. What should I do?

FIG. 5 is a block diagram showing a threshold value calculation unit 31 of the foreign object detection device according to Embodiment 2 of the present invention.
In FIG. 5, the threshold value switching flag acquisition unit 41 performs processing for acquiring the threshold value switching flag output from the threshold value processing unit 32.
The threshold generation unit 42 stores a fixed foreign substance threshold THb _{(i, j)} set in advance. If the threshold switching flag acquired by the threshold switching flag acquisition unit 41 is “0”, the fixed threshold value THb _{(i, j)} is stored. If the threshold value switching flag acquired by the threshold value switching flag acquisition unit 41 is “1” while the foreign object threshold value THb _{(i, j)} is output to the threshold value processing unit 32 _{, the} foreign object threshold value THb _{(i, j)} Before starting the detection process _, the average value Zm and standard deviation value Zstd of the altitude values Z _{i, j} output from the three-dimensional image pickup device 3 are obtained, and the foreign matter is obtained from the average value Zm and the standard deviation value Zstd. The threshold THb _{(i, j)} is calculated, and the foreign substance threshold THb _{(i, j)} is output to the threshold processing unit 32.

6 is a block diagram showing a threshold processing unit 32 of a foreign object detection device according to Embodiment 2 of the present invention.
In FIG. 6, the foreign matter index calculation unit 51 performs, for each foreign matter detection window in the three-dimensional image _, an average value Zm _win of the altitude values Z _{i, j} of each pixel output from the three-dimensional image pickup device 3 during the foreign matter detection process. A standard deviation value Zstd _win is obtained, and a foreign matter index S _{(i, j),} which is an index representing a vehicle (foreign matter) other than the background, is calculated from the average value Zm _win and standard deviation value Zstd _{win of} the altitude value Z _{i, j.} Perform the process. The foreign matter index calculating unit 51 constitutes a foreign matter index calculating means.
The region detection unit 52 compares the foreign matter index S _{(i, j)} calculated by the foreign matter index calculation unit 51 with the foreign matter threshold THb _{(i, j)} calculated by the threshold calculation unit 31, and the foreign matter index S _{(i, j, j)} is carried out a process of detecting the higher regions (foreign matter detection window) foreign object threshold _{THb (i, j).} The region detection unit 52 constitutes a region detection unit.

Next, the operation will be described.
The three-dimensional image pickup device 3 picks up a three-dimensional image (background image) of the observation region in a situation where no foreign vehicle exists before starting the foreign object detection process, and the three-dimensional image outputting altitude values Z _i of each pixel of _{the j} foreign matter detecting device 4.
The three-dimensional image capturing apparatus 3 captures a three-dimensional image of the observation area (an image in which a vehicle may be reflected on the background) during the foreign object detection process, The altitude value Z _{i, j} is output to the foreign object detection processing device 4.
The three-dimensional image pickup device 3 can acquire the reflection intensity value and the distance value in addition to the altitude value Z _{i, j} of each pixel. In the second embodiment, the foreign object detection processing device 4 Since no reflection intensity value or distance value is used _, only the altitude value Z _{i, j of} each pixel is output to the foreign object detection processing device 4.

The threshold value processing unit 32 of the foreign object detection processing device 4 is, for example, a fixed foreign object threshold value THb set in advance at the start-up time before the altitude value Z _{i, j of} each pixel is output from the 3D image pickup device 3. A threshold value switching flag of “0” instructing the output of _{(i, j)} is output to the threshold value calculation unit 31.
When the output of the altitude value Z _{i, j} is started from the 3D image pickup device 3 _{, the} foreign substance threshold THb _(i) set from the altitude value Z _{i, j} of each pixel of the 3D image (background image). _{, J)} is output to the threshold value calculation unit 31 as a threshold value switching flag of “1”.

The threshold value switching flag acquisition unit 41 of the threshold value calculation unit 31 acquires the threshold value switching flag output from the threshold value processing unit 32 and outputs the threshold value switching flag to the threshold value generation unit 42.
The threshold value generation unit 42 of the threshold value calculation unit 31 stores, for example, a fixed foreign object threshold value THb _{(i, j)} set in advance by the user, and the threshold value switching flag output from the threshold value switching flag acquisition unit 41 is stored. If “0”, the fixed foreign substance threshold THb _{(i, j)} is output to the threshold processing unit 32.

If the threshold value switching flag output from the threshold value switching flag acquisition unit 41 is “1”, the threshold value generation unit 42 sets a foreign object detection window in the three-dimensional image.
Here, FIG. 7 is an explanatory diagram illustrating an example of setting of a foreign object detection window at the time of threshold generation.
In the example of FIG. 7, an area of 3 × 3 pixel size is set as the foreign object detection window, and the foreign object detection window is shifted by one pixel at a time in the data direction (i) or the line direction (j). Set to cover the whole.
In the example of FIG. 7, the pixel located at the center of the foreign object detection window is set as the pixel of interest.

When the foreign matter detection window is set in the three-dimensional image, the threshold generation unit 42 sets the altitude value Z _{i, j} of each pixel output from the three-dimensional image pickup device 3 before starting the foreign matter detection process. The altitude value Z _{i, j} of the pixel in the foreign object detection window is acquired. In the example of FIG. 7, the altitude values Z _{i, j} of nine pixels are acquired.
When the threshold value generation unit 42 obtains the altitude value Z _{i, j} of the pixels in the foreign object detection window, the average value Zm and the standard value of the altitude values Z _{i, j} of each pixel in units of the same data direction (i). A deviation value Zstd is calculated.
In the example of FIG. 7, assuming that the coordinate of the target pixel is (i, j), the average value Zm and the standard deviation value Zstd of the altitude values of three pixels whose data direction coordinate is “i−1” The average value Zm and standard deviation value Zstd of the altitude values of the three pixels whose data direction coordinate is “i”, and the average value Zm of the altitude values of the three pixels whose data direction coordinate is “i + 1” And a standard deviation value Zstd.

When the threshold value generation unit 42 calculates the average value Zm and the standard deviation value Zstd of the altitude values Z _{i, j} of each pixel in a pixel unit having the same data direction (i), the threshold value generating unit 42 determines the maximum value among the average values Zm. The average value Zm _max is specified, and the maximum standard deviation value Zstd _max among the standard deviation values Zstd is specified.
The altitude value Z _{i, j} of each pixel used by the threshold generation unit 42 is related to a three-dimensional image captured in a situation where no vehicle (foreign object) exists. Since there is always a difference in the altitude true value of the background, the standard deviation value Zstd becomes high in a situation where the background and foreign objects are mixed.

The threshold value generator 42 uses the maximum average value Zm _max and the maximum standard deviation value Zstd _max to calculate the foreign object threshold value THb _{(i, j)} related to the foreign object detection window as shown in the following equation (3). calculate.
THb _{(i, j)} = K × (Zm _max + Zstd _max ) (3)
In Expression (3), in order to specify the position of the foreign object detection window, the position of the target pixel in the foreign object detection window is represented by (i, j). K is a predetermined coefficient.
After calculating the foreign object threshold THb _{(i, j)} , the threshold generation unit 42 outputs the foreign object threshold THb _{(i, j)} to the threshold processing unit 32.

The foreign substance index calculation unit 51 of the threshold processing unit 32 sets a foreign object detection window in the three-dimensional image, similarly to the threshold generation unit 42.
FIG. 8 is an explanatory diagram showing a setting example of the foreign object detection window at the time of threshold processing.
In the example of FIG. 8, an area of 3 × 3 pixel size is set as the foreign object detection window, and the foreign object detection window is shifted one pixel at a time in the data direction (i) or the line direction (j) and the 3D image is displayed. Set to cover the whole.
In the example of FIG. 8, the pixel located at the center of the foreign object detection window is set as the target pixel.

When the foreign object index calculation unit 51 sets a foreign object detection window in the 3D image, the foreign object detection is performed among the altitude values Z _{i, j} output from the 3D image pickup device 3 during the foreign object detection process. Obtain the altitude value Z _{i, j} of the pixels in the window. In the example of FIG. 8, the altitude values Z _{i, j} of nine pixels are acquired.
Foreign object index calculating unit 51, the altitude value _{Z i} of the pixel in the object detection _window and obtain a _j, altitude value _{Z i} of each pixel in the foreign matter detection _window, the average value of _{j Zm win} and standard deviation _{Zstd win} Is calculated. In the example of FIG. 8, the average value Zm _win and the standard deviation value Zstd _win of the altitude values Z _{i, j} of nine pixels are calculated.

The foreign object index calculating unit 51 uses the average value Zm _win and the standard deviation value Zstd _win of the altitude values Z _{i, j} of each pixel in the foreign object detection window, as shown in the following formula (4), A foreign material index S _{(i, j),} which is an index representing the vehicle (foreign material), is calculated.
S _{(i, j)} = Zm _win + Zstd _win (4)
In Expression (4), in order to specify the position of the foreign object detection window, the position of the target pixel in the foreign object detection window is represented by (i, j).
In the situation where the background and foreign matter are mixed, as described above, the standard deviation value Zstd becomes high, and the foreign matter index S _{(i, j)} becomes large.

Region detecting unit 52 of the threshold processing unit 32, a foreign object index calculating unit 51 calculates the foreign object index _{S (i, j),} the foreign object index _{S (i, j)} as foreign threshold THb calculated by the threshold calculation unit 31 Compare _{(i, j)} .
When the foreign matter index S _{(i, j)} is higher than the foreign matter threshold THb _{(i, j)} (S _{(i, j)} > THb _{(i, j)} ), the area detection unit 52 is in the foreign matter detection window. It is determined that a pixel constituting a part of (foreign matter) is included, and an area count value E indicating the number of foreign object detection areas is incremented. However, the initial value of the area count value E is 0.

The foreign object detection unit 33 determines that there is a foreign object in the three-dimensional image when the region count value E output from the threshold processing unit 32 is greater than the preset number of effective regions.
For example, if a foreign object exists in the 3D image, a foreign object detection flag of “1” is output, and if no foreign object exists in the 3D image, a foreign object detection flag of “0” is output. To do.
Note that the number of effective areas may not be a constant threshold value, and erroneous detection can be reduced by setting an upper limit value and a lower limit value to be a band pass.

As apparent from the above, according to the second embodiment, before starting the foreign object detection process, a three-dimensional image of the observation region is captured in a situation where no foreign object is present, altitude value Z _i of each _pixel, and outputs the _j, when the foreign matter detection processing by imaging the three-dimensional image of the observation area, three-dimensional imaging for outputting a high value Z _{i, j} of each pixel of the three-dimensional image The average value Zm and standard deviation value of the altitude values Z _{i, j} of each pixel output from the 3D image pickup device 3 before starting the foreign object detection process in a predetermined region unit in the 3D image. Zstd is obtained, a threshold value calculation unit 31 for calculating the foreign substance threshold value THb _{(i, j)} from the average value Zm and standard deviation value Zstd of the altitude values Z _{i, j} , and a predetermined area unit in the three-dimensional image, Output from 3D imaging device 3 during detection processing Altitude value _{Z i} of each _pixel, the average value _{Zm win} and standard deviation values _{Zstd win} of _j, the average value _{Zm win} and debris index _S is an index from the standard deviation _{Zstd win} representative of the foreign matter other than the background _{( The} foreign matter index calculation unit 51 for calculating _{i, j)} , the foreign matter index S _{(i, j)} calculated by the foreign matter index calculation unit 51 and the foreign matter threshold THb _{(i, j)} calculated by the threshold calculation unit 31 are compared. And an area detection unit 52 that detects an area where the foreign object index S _{(i, j)} is higher than the foreign object threshold THb _{(i, j)} , and the foreign object detection unit 33 detects the area detected by the area detection unit 52. When the total number is larger than the preset number of effective areas, it is determined that a foreign substance is present in the three-dimensional image. Therefore, the foreign substance threshold THb _{( i, j)} Dynamically come to be set, as a result, even under a low SNR environment, an effect which can accurately detect the vehicle as a foreign body.
In the second embodiment, since the presence / absence of foreign matter is determined for each foreign matter detection window using the statistic of the altitude value Z _{i, j} of each pixel, the reflection intensity value I _i, Compared with the first embodiment in which _j and the threshold value are compared pixel by pixel, the foreign object detection accuracy is improved. Moreover, the required distance accuracy with respect to the three-dimensional image imaging device 3 can be eased.

Embodiment 3 FIG.
In the second embodiment, the foreign object index calculation unit 51 calculates the average value Zm _win and the standard deviation value Zstd _win of the altitude values Z _{i, j} of each pixel in the foreign object detection window, and the average value Zm _win and the standard The deviation value Zstd _win is used to calculate a foreign matter index S _{(i, j),} which is an index representing a vehicle (foreign matter) other than the background. The altitude value Z _i, of _j, reflection intensity value I _{i, j} is set to only the pixel is above a predetermined threshold value calculation target of an average value and standard deviation value, the reflection intensity value I _{i, j} is the pixel above a predetermined threshold value The average value Zm _win and standard deviation value Zstd _win of the altitude value Z _{i, j} are calculated, and the foreign material index, which is an index representing a vehicle (foreign material) other than the background, using the average value Zm _win and standard deviation value Zstd _win. S _{(i, )} May be calculated.

In the third embodiment, it is assumed that the three-dimensional image capturing apparatus 3 outputs not only the altitude value Z _{i, j} of each pixel of the three-dimensional image but also the reflection intensity value I _{i, j} of each pixel.
Since only the processing content of the foreign material index calculation unit 51 is different from that of the second embodiment, the processing content of the foreign material index calculation unit 51 will be described below.

The foreign object index calculation unit 51 sets a foreign object detection window in the three-dimensional image as in the second embodiment.
When the foreign object index calculation unit 51 sets a foreign object detection window in the 3D image, the foreign object detection is performed among the altitude values Z _{i, j} output from the 3D image pickup device 3 during the foreign object detection process. Obtain the altitude value Z _{i, j} of the pixels in the window.
Further, foreign matter index calculating unit 51, the reflection intensity value I _i of each pixel output from the three-dimensional imaging apparatus 3 when the foreign matter detection _processing, among the _j, the reflection intensity value I _i of the pixels within the object detection window _{, J.}
In the example of FIG. 8, the altitude value Z _{i, j} and the reflection intensity value I _{i, j} of nine pixels are acquired.

Foreign object index calculating unit 51, the altitude value Z _i of the pixel in the object detection _{window, j} and the reflection intensity value I _i, obtains the _j, reflection intensity value I _i of each _{pixel, j} and a predetermined threshold (e.g., FIG. The threshold value TH _UP set by the second threshold value generator 22 can be used, but other threshold values may be used) to identify pixels whose reflection intensity values I _{i, j} are higher than a predetermined threshold value.
When the foreign matter index calculation unit 51 identifies pixels in the foreign matter detection window whose reflection intensity value I _{i, j} is higher than a predetermined threshold _, the average value Zm _win and standard deviation value Zstd of the altitude values Z _{i, j} of those pixels _Win is calculated. For example, if the number of pixels in the foreign object detection window in which the reflection intensity value I _{i, j} is higher than a predetermined threshold value is 5 _, the average value Zm _win and the standard deviation value Zstd _win of the altitude values Z _{i, j} of the 5 pixels. Is calculated.

The foreign matter index calculation unit 51 uses the average value Zm _win and the standard deviation value Zstd _win of the altitude values Z _{i, j} of the pixels in the foreign matter detection window in which the reflection intensity value I _{i, j} is higher than a predetermined threshold. As shown in Expression (5), a foreign matter index S _{(i, j)} that is an index representing a vehicle (foreign matter) other than the background is calculated.
S _{(i, j)} = Zm _win + Zstd _win (5)
In Expression (5), in order to specify the position of the foreign object detection window, the position of the target pixel in the foreign object detection window is represented by (i, j).

In the third embodiment, when calculating the foreign matter index S _{(i, j) ,} only the altitude value Z _{i, j} of the pixel in the foreign matter detection window in which the reflection intensity value I _{i, j} is higher than a predetermined threshold is used. Therefore, foreign matter detection is performed in consideration of the reflection intensity values I _{i, j of} the pixels, and the foreign matter detection accuracy is further improved.

Embodiment 4 FIG.
FIG. 9 is a block diagram showing a foreign object detection apparatus according to Embodiment 4 of the present invention. In the figure, the same reference numerals as those in FIG.
The false detection recognition unit 60 is composed of, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like, and is three-dimensionally formed by the foreign matter detection unit 13 in FIG. 1 (or the foreign matter detection unit 33 in FIG. 4). When it is determined that a vehicle that is a foreign object exists in the image (when a foreign object detection flag of “1” is output), the continuity of the foreign object in the three-dimensional image is confirmed, If the size is not recognized to be greater than or equal to the predetermined value, the foreign matter determined to be present by the foreign matter detector 13 in FIG. 1 (or the foreign matter detector 33 in FIG. 4) is recognized as a false detection. Perform the process. In addition, the false detection authorization part 60 comprises the false detection authorization means.

Next, the operation will be described.
Except for the fact that the false detection authorization unit 60 is mounted on the foreign matter detection processing devices 2 and 4, it is the same as in the first to third embodiments.
In this Embodiment 4, the processing content of the false detection authorization part 60 is demonstrated.

When the foreign object detection unit 13 in FIG. 1 or the foreign object detection unit 33 in FIG. 4 determines that a vehicle that is a foreign object exists in the three-dimensional image, as in the first to third embodiments, 1 "foreign object detection flag is output.
However, it is assumed that the foreign object detection flag of “1” includes coordinates (i, j) indicating the position of a pixel constituting a part of the foreign object.

First, the false detection recognition unit 60 has a predetermined width in the data direction (i) in the three-dimensional image (for example, a width corresponding to n pixels: the number of pixels in the vertical axis direction of the three-dimensional image N> n). Set the removal window.
FIG. 10 is an explanatory view showing a setting example of a false detection removal window set by the false detection authorization unit 60.

When receiving the foreign object detection flag “1” from the foreign object detection unit 13 in FIG. 1 or the foreign object detection unit 33 in FIG. 4, the false detection recognition unit 60 determines the coordinates (i, j) included in the foreign object detection flag. Referring to FIG. 10, a pixel located at the coordinates (i, j) is defined as a “detection pixel”.
Since a vehicle that is a foreign object has a certain size, one isolated detection pixel (a pixel whose foreign object detection flag is “1”) is a false detection and constitutes a part of the foreign object. The possibility of a pixel that is doing is very low.

Therefore, the false detection recognition unit 60 has a plurality of false detection pixels having continuity in the false detection removal window having a predetermined width in the data direction (i) and continuous in the line direction (j). If detection pixel continuity is recognized in all of the detection removal windows (for example, m false detection removal windows: the number of pixels M> m in the horizontal axis direction of the three-dimensional image), Finally, it is determined that there is a foreign object vehicle, and a foreign object detection flag of “1” is output.
On the other hand, if the detection pixels in the false detection removal window having a predetermined width in the data direction (i) are not continuous, or one of a plurality of false detection removal windows continuous in the line direction (j) On the other hand, when the continuity of the detection pixels is not recognized, it is determined that there is no foreign object in the three-dimensional image, and a foreign object detection flag of “0” is output.
However, for example, in the case where the continuity of the detection pixel is recognized in the right region of the three-dimensional image even if the continuity of the detection pixel is not recognized in the left region of the three-dimensional image, the three-dimensional It may be determined that a vehicle that is a foreign object exists in the image, and a foreign object detection flag of “1” may be output.

  As apparent from the above, according to the fourth embodiment, when the foreign object detection unit 13 in FIG. 1 determines that a vehicle that is a foreign object exists in the three-dimensional image, If it is not confirmed that the size of the foreign material is not less than a predetermined value by checking the continuity of the foreign material, the foreign material determined to be present by the foreign material detection unit 13 in FIG. Since it has been configured so as to be recognized, there is an effect that it is possible to eliminate erroneous detection of foreign matter.

Embodiment 5 FIG.
FIG. 11 is a block diagram showing a foreign object detection apparatus according to Embodiment 5 of the present invention.
In FIG. 11, the integrated processing unit 70 is composed of, for example, a semiconductor integrated circuit on which a CPU is mounted, a one-chip microcomputer, or the like, and a foreign object detection flag output from the foreign object detection processing device 2 of the first embodiment. From the foreign object detection flag output from the foreign object detection processing device 4 according to the second embodiment, the foreign object detection flag output from the foreign object detection processing device 4 according to the third embodiment, and the foreign object detection processing device 2 according to the fourth embodiment. A logical sum with the output foreign object detection flag is calculated, and if any foreign object detection flag is “1”, a process of outputting a foreign object detection flag of “1” is performed. The integration processing unit 70 constitutes a logical sum calculation means.

Next operation will be described.
In the first embodiment, a foreign object is detected when there is a difference between the reflection intensity value of the background and the reflection intensity value of the foreign object. In the second embodiment, the background altitude value and the foreign object altitude value are detected. A foreign object is detected when there is a difference between
In the third embodiment, the foreign matter is detected when the reflection intensity value is high and there is a difference between the altitude value of the background and the altitude value of the foreign matter.
In the fourth embodiment, a foreign object is detected when the size is a predetermined value or more.

As described above, in the first to fourth embodiments, since the foreign matter detection viewpoint is different, for example, even if the foreign matter is not detected in the detection method in the first embodiment, the detection method in the second embodiment is used. A foreign object may be detected.
Therefore, in this fifth embodiment, in order to prevent foreign object detection from being missed, the integrated processing unit 70 and the foreign object detection flag output from the foreign object detection processing device 2 of the first embodiment and the foreign object detection of the second embodiment. The foreign matter detection flag output from the processing device 4, the foreign matter detection flag output from the foreign matter detection processing device 4 of the third embodiment, and the foreign matter detection flag output from the foreign matter detection processing device 2 of the fourth embodiment. A logical sum is calculated, and if any foreign object detection flag is “1”, a foreign object detection flag of “1” is output.
Therefore, if all the foreign matter detection flags are not “0”, a foreign matter detection flag of “1” is output.
Thereby, there is an effect that it is possible to greatly reduce the leakage of foreign object detection.

  In the present invention, within the scope of the invention, any combination of the embodiments, or any modification of any component in each embodiment, or omission of any component in each embodiment is possible. .

  DESCRIPTION OF SYMBOLS 1,3 Three-dimensional image imaging device (three-dimensional image imaging means), 2,4 Foreign object detection processing apparatus, 11 Threshold setting part (threshold setting means), 12 Threshold processing part (pixel detection means), 13 Foreign object detection part (foreign matter) Detection means), 21 threshold switching flag acquisition unit, 22 threshold generation unit, 31 threshold calculation unit (threshold calculation unit), 32 threshold processing unit (region detection unit), 33 foreign matter detection unit (foreign matter detection unit), 41 threshold switching flag Acquisition unit, 42 threshold generation unit, 51 foreign matter index calculation unit (foreign matter index calculation unit), 52 region detection unit (region detection unit), 60 false detection certification unit (false detection certification unit), 70 integrated processing unit (logical sum calculation) means).

Claims (6)

And means for imaging a three-dimensional image of the observation area by scanning the laser beam, before starting the detection process of the foreign matter, in a situation where the foreign matter is not present by imaging a three-dimensional image of the observation area A three-dimensional image that outputs a reflection intensity value of each pixel of the three-dimensional image, captures a three-dimensional image of the observation region during the foreign object detection process, and outputs a reflection intensity value of each pixel of the three-dimensional image Among the reflection intensity values of each pixel output from the three-dimensional image pickup means before starting the foreign substance detection processing in the same pixel unit as the vertical direction component of the three-dimensional image and the image pickup means, A threshold setting means for acquiring the reflection intensity value of the image and setting the maximum reflection intensity value as an upper limit threshold value, and capturing the 3D image at the time of foreign object detection processing in the same pixel unit as the vertical component of the 3D image. Reflection intensity value of each pixel output from the means The pixel detection means for comparing the upper limit threshold set by the threshold setting means and detecting the pixel whose reflection intensity value is higher than the upper limit threshold, and the pixel in which the total number of pixels detected by the pixel detection means is preset A foreign matter detection device comprising foreign matter detection means for determining that foreign matter is present in the three-dimensional image when the number is greater than the number. It said threshold setting means, the same pixel is the vertical direction component of the three-dimensional image, in the reflection intensity value of each pixel output from the three-dimensional image pickup means before starting the detection process of the foreign matter, Obtain the maximum reflection intensity value and the minimum reflection intensity value, set the maximum reflection intensity value to the upper threshold, set the minimum reflection intensity value to the lower threshold,
The pixel detection means is set by the threshold value setting means and the reflection intensity value of each pixel output from the three-dimensional image imaging means during the foreign object detection process in units of pixels having the same vertical component of the three-dimensional image. The foreign substance detection device according to claim 1, wherein the upper limit threshold value and the lower limit threshold value are compared, and a pixel having the reflection intensity value higher than the upper limit threshold value or a pixel having the reflection intensity value lower than the lower limit threshold value is detected. And means for imaging a three-dimensional image of the observation area by scanning the laser beam, before starting the detection process of the foreign matter, in a situation where the foreign matter is not present by imaging a three-dimensional image of the observation area 3D image imaging that outputs the altitude value of each pixel of the 3D image and outputs the altitude value of each pixel of the 3D image by capturing the 3D image of the observation region during the foreign object detection processing A mean value and a standard deviation value of altitude values of each pixel output from the 3D image imaging means before starting the foreign substance detection processing in a predetermined region unit in the 3D image, and the altitude A threshold value calculating means for calculating a threshold value from an average value and a standard deviation value, and an altitude value of each pixel output from the 3D image imaging means during a foreign object detection process in a predetermined region unit in the 3D image; Mean value and standard deviation A foreign substance index calculating means for calculating a foreign substance index, which is an index representing foreign substances other than the background, from the average value and standard deviation value of the altitude value, the foreign substance index calculated by the foreign substance index calculating means, and the threshold value calculating means When the calculated threshold value is compared and the total number of the areas detected by the area detecting means and the area detecting means for detecting the area where the foreign substance index is higher than the threshold value is greater than a preset number of areas, the three-dimensional A foreign matter detection apparatus comprising foreign matter detection means for determining that foreign matter is present in an image. The three-dimensional imaging means, when the foreign matter detection processing by imaging a three-dimensional image of the observation area, and outputs a high value and the reflection intensity value of each pixel of the three-dimensional image,
The foreign matter index calculation means is an average value of altitude values of pixels whose reflection intensity values output from the 3D image pickup means during a foreign matter detection process are greater than or equal to a predetermined threshold for a predetermined region unit in the 3D image. The foreign matter detection device according to claim 3, wherein a foreign matter index that is an index representing foreign matter other than the background is calculated from the average value and standard deviation value of the altitude values. When the foreign matter detection means determines that a foreign matter is present in the three-dimensional image, the size of the foreign matter is equal to or greater than a predetermined value by checking the continuity of the foreign matter in the three-dimensional image. If it is not recognized, there is provided a false detection recognition means for recognizing that the foreign matter determined to be present by the foreign matter detection means is a false detection. The foreign matter detection device according to any one of the above.   A foreign matter detection apparatus comprising: a logical sum calculation means for calculating a logical sum of a foreign matter judgment result by the foreign matter detection means according to claim 1 and a foreign matter judgment result by the foreign matter detection means according to claim 3 .

Images