JP2007256101A - Unevenness determination device, and unevenness determination method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily and quickly determine the presence of unevenness in a measuring object, without measuring directly a three-dimensional shape of the measuring object. <P>SOLUTION: A measuring objective face is irradiated with a slit beam (100), a reflected beam reflected on the measuring objective face is imaged to generate a slit beam reflection image (102), and the slit beam reflection image is processed into fine lines to extract a slit constitution point sequence (104). Coordinates of each point in the slit constitution point sequence are Haugh-transformed (106), a representative point of an area in an point crossed with Haugh curves of a preset number or more is extracted as a line segment candidate, on a ρ-θ plane (108), and the presence of the unevenness measuring objective face is determined and a degree of the unevenness thereon is measured (112), by calculating the number of the representative points with a vote number of a grid area exceeding the threshold value, the maximum vote number thereof, the minimum vote number, and an average vote number, as line segment parameters (110). <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an unevenness determination apparatus and an unevenness determination method, and more particularly to an unevenness determination apparatus and an unevenness determination method for determining the presence or absence of an unevenness to be measured.

  Conventionally, when moving a mobile robot, the unevenness of the floor surface has a large effect on the movement stability. Therefore, the shape of the floor surface is measured with a 3D sensor and expressed in 3D coordinates. Movement control has been performed by judging features and unevenness. In this method, since it is necessary to measure a three-dimensional shape whether or not the floor surface is uneven, it takes time to perform the shape measurement process even when the floor surface is not uneven. Further, since it takes time to judge the unevenness from the obtained shape, there is an adverse effect that it is necessary to slow down the moving speed for stable movement.

  Here, in order to increase the measurement processing speed of the three-dimensional shape with the height measuring device by the light cutting method using the slit light source and the TV camera, the image processing area of the TV camera that captured the reflected image of the slit light is previously set. A height measuring device that is set narrowly is known (Patent Document 1).

In this height measuring apparatus, since a narrow range is scanned, the processing time is shortened, but the measurement range is narrow. The target three-dimensional shape is used as coordinate data.
JP 2001-188008 A

  However, in the technique described in Patent Document 1, it is necessary to determine the presence or absence of unevenness from the obtained three-dimensional shape, and there is a problem that wasteful processing time is required when there is no unevenness on the floor surface. is there.

  The present invention has been made to solve the above-described problems, and can provide an unevenness determination device that can easily and quickly determine the presence or absence of unevenness in a measurement object without directly measuring the three-dimensional shape of the measurement object. And it aims at providing the unevenness | corrugation determination method.

  In order to achieve the above object, an unevenness determination apparatus according to the present invention is provided with a slit light irradiating unit that irradiates a measurement target with slit-shaped light, and is irradiated by the slit light irradiating unit, and from the measurement target. An imaging unit that captures reflected reflected light to generate a slit light reflection image, a calculation unit that calculates the number of line segments constituting the slit light reflection image generated by the imaging unit, and the calculation unit And determining means for determining the presence or absence of the unevenness of the measurement object based on the calculated number of the line segments.

  Further, the unevenness determination method according to the present invention generates a slit light reflection image by irradiating a measurement target with slit-shaped light, and imaging the reflected light that is irradiated and reflected from the measurement target. Then, the number of the line segments constituting the generated slit light reflection image is calculated, and the presence / absence of the unevenness of the measurement target is determined based on the calculated number of line segments.

  According to the present invention, a slit-shaped light is emitted to a measurement target, and the reflected light that is irradiated and reflected from the measurement target is imaged to generate a slit light reflection image. And the number of the line segments which comprise the produced | generated slit light reflected image is calculated, and the presence or absence of the unevenness | corrugation of a measuring object is determined based on the calculated number of line segments.

  Therefore, the slit light reflection image is determined by irradiating the measurement target with slit-shaped light and determining the presence / absence of the unevenness of the measurement target based on the number of line segments constituting the generated slit light reflection image. Since it is only necessary to calculate whether or not it is composed of line segments, it is possible to determine the presence or absence of unevenness in the measurement target easily and at high speed without directly measuring the three-dimensional shape of the measurement target.

  In addition, the line segment which comprises a slit light reflected image is a line segment which comprises a slit light reflected image, Comprising: The line segment which connects the edge part and edge part which adjoin each other, and edge part which adjoins It is a line segment connecting the bending points, or a line segment connecting adjacent bending points and bending points.

  In addition, the unevenness determination apparatus according to the present invention includes a slit center position in an orthogonal direction based on pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of the slit light reflection image generated by the imaging unit. Extraction means for extracting a plurality of central pixels in the longitudinal direction, and the calculation means can calculate the number of line segments based on the plurality of central pixels extracted by the extraction means. Thus, the number of line segments can be calculated with high accuracy by calculating the number of line segments constituting the slit light reflection image based on a plurality of central pixels in the longitudinal direction of the slit light reflection image.

  Further, the extraction means according to the present invention uses the pixel of the pixel signal indicating the maximum amount of reflected light as the central pixel based on the pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of the slit light reflected image. A plurality of samples can be extracted in the longitudinal direction.

  Further, the extraction means according to the present invention is based on pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of the slit light reflection image, and the barycentric position of the amount of reflected light indicated by the pixel signal of each pixel. A plurality of pixels can be extracted in the longitudinal direction with the pixel to be the center pixel.

  Further, the calculation means according to the present invention includes a Hough conversion means for converting each of the coordinates indicating the positions of the plurality of central pixels extracted by the extraction means into a Hough curve on the Hough plane, and the conversion result by the Hough conversion means Based on this, the number of line segments can be calculated. Thereby, since the number of line segments constituting the slit light reflection image may be calculated using the Hough transform, it is possible to determine the presence or absence of the unevenness of the measurement target more easily and at high speed.

  Further, the Hough transforming unit according to the present invention converts each of the coordinates indicating the positions of the plurality of central pixels into a Hough curve, and each of the lattice regions when the Hough plane is divided into a plurality of predetermined lattice regions. On the other hand, vote for the grid area at the passing position or the intersection position of the Hough curve to calculate the number of votes of each grid area, and the calculating means is based on the number of votes calculated by the Hough transform means, By detecting a line segment, the number of line segments can be calculated.

  In addition, the unevenness determination apparatus according to the present invention can measure the flatness or unevenness of the measurement target based on the number of line segments calculated by the calculation unit and the number of votes calculated by the Hough conversion unit. Thereby, not only the determination of the presence or absence of unevenness but also the flatness or unevenness of the measurement object can be measured easily and at high speed.

  As described above, according to the unevenness determination apparatus and unevenness determination method of the present invention, by irradiating the measurement target with slit-shaped light, the number of line segments constituting the generated slit light reflection image, Since it is only necessary to calculate how many line segments the slit light reflection image consists of by determining the presence or absence of unevenness on the measurement target, it is easy and fast without directly measuring the three-dimensional shape of the measurement target. In addition, it is possible to determine the presence or absence of unevenness to be measured.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  As shown in FIG. 1, the unevenness determination apparatus 10 according to the present embodiment includes a slit light source 12 that irradiates a measurement target surface with one slit light, and slit light irradiation that controls irradiation of slit light by the slit light source 12. The control unit 14, the imaging unit 16 that captures the reflected light of the slit light from the measurement target surface, generates an image signal of one slit light reflection image indicating the reflected light of the slit light, and the imaging unit 16 generates the image signal. An A / D converter 18 for A / D converting the image signal and an image memory 20 for temporarily storing the A / D converted image signal are provided.

  Further, the unevenness determination apparatus 10 further includes a slit extraction image processing unit 22 that extracts a center pixel that is a slit center position from an image signal stored in the image memory 20, and a slit extracted by the slit extraction image processing unit 22. Based on the line segment calculation unit 24 that calculates the number of line segments that form one slit light reflection image from the central pixel that is the center position, and based on the number of line segments calculated by the line segment calculation unit 24 A measurement unit 26 that determines whether or not the surface is uneven and measures the degree of unevenness, and an output unit 28 that outputs the presence and absence of the unevenness determined by the measurement unit 26 and the measured unevenness degree are provided.

  Here, the line segment constituting one slit light reflection image is a line segment constituting one slit light reflection image, and an adjacent end portion and end portion of one slit light reflection image Is a line segment connecting the adjacent end points and the bending point, or a line segment connecting the adjacent bending point and the bending point.

  Next, the effect | action of the unevenness | corrugation determination apparatus 10 which concerns on this Embodiment is demonstrated.

  When the slit light source 12 of the unevenness determination device 10 is directed to the measurement target surface, processing for measuring the unevenness degree shown in FIG. 2 is performed.

  First, in step 100, the slit light irradiation control unit 14 causes the slit light source 12 to irradiate the measurement target surface with slit light, and in step 102, the imaging unit 16 irradiates the slit light source 12 and from the measurement target surface. The reflected reflected light is imaged, and a one-dimensional slit light reflection image is generated.

In step 104, the slit extraction image processing unit 22 performs thinning processing on the slit light reflected image, and a slit configuration point sequence (X ₁ , Y) indicating the center pixel serving as the slit center position of the slit light reflected image. ₁ ), (X ₂ , Y ₂ ). . . Extract (X _n , Y _n ). For example, based on the pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of one slit light reflection image, a plurality of pixels in the longitudinal direction are extracted with the pixel of the pixel signal indicating the maximum amount of reflected light as the central pixel. A plurality of center pixels arranged in the longitudinal direction are defined as slit constituent point sequences.

In the next step 106, the line segment calculation unit 24 performs Hough transform on the coordinates (coordinates on the XY plane) of each point of the extracted slit composing point sequence to the ρ-θ plane as the Hough plane. Project. Here, the Hough transform of the slit composing point sequence is performed by the following equation.
ρ ₁ = X ₁ · cos θ ₁ + Y ₁ · sin θ ₁
ρ ₂ = X ₂ · cos θ ₂ + Y ₂ · sin θ ₂
...
ρ _n = X _n · cos θ _n + Y _n · sin θ _n
By this Hough transform, the coordinates of the N points in the slit composing point sequence are expressed as N Hough curves after the transformation on the ρ-θ plane. When M of these N Hough curves intersect at one point, it is determined that the points before conversion of the M Hough curves are on the same straight line, as shown in FIG. In addition, when all the Hough curves intersect at one point, it is determined that each point of the slit composing point sequence is a point on one line segment. In addition, when each point of the slit composing point sequence is a point on two line segments, as shown in FIG. 4, the converted N Hough curves mainly intersect at two points. It has become.

  Then, in step 108, representative points in the area where the Hough curves of a predetermined number or more intersect on the projected ρ-θ plane are extracted as line segment candidates. For example, in the voting plane obtained by previously quantizing the ρ-θ plane (p × q) and dividing it into lattice areas, all the intersection points (ρ, θ) of the Hough curves are voted to the lattice area corresponding to the position of the intersection point. Then, the number of votes for each lattice region is calculated, and based on the calculated number of votes for each lattice region, representative points predetermined for the lattice region are extracted as line segment candidates. At this time, by extracting the representative points of the lattice area where the intersections of the Hough curves are concentrated as line segment candidates, the points on the same straight line that should intersect at one point are within a certain range due to the influence of the quantization error during imaging. You can consider the exchanges inside.

  Here, p and q are predetermined parameters for quantizing the ρ-θ plane composed of values having different dimensions such as ρ and θ, and indicate the size of the lattice region. For example, p = q q = 1. In addition, if there are subtle irregularities on the surface to be measured, the range of intersection becomes wider depending on the size of the irregularities, so the threshold of this intersection range is set by p and q. The uneven range is determined as the same straight line.

  In the next step 110, among the extracted representative points, the number Num of representative points where the number of votes R in the lattice area of the representative points exceeds a threshold Min (for example, Min = 100) is calculated as a line segment parameter. Further, the maximum vote number Rmax, the minimum vote number Rmin, and the average vote number Rmean of the representative points exceeding the threshold value Min are calculated as line segment parameters. Here, Num is a value corresponding to the number of line segments constituting one slit light reflection image, and Rmax is the longest line segment among line segments constituting one slit light reflection image. Rmin is a value corresponding to the length, and Rmin is a value corresponding to the length of the shortest line segment among the line segments constituting one slit light reflection image. Originally, the representative points exceeding the threshold value are returned to the straight lines before the Hough transform, and these straight lines are reconstructed to examine the unevenness of the measurement target surface. However, from the simplified Nun, Rmax, Rmin, Rmean, The unevenness of the measurement target surface can be roughly estimated.

  In step 112, the measurement unit 26 determines the presence or absence of unevenness on the measurement target surface and measures the unevenness level. For example, when the measurement target surface is a plane, one slit light irradiated from the slit light source 12 is reflected from the measurement target surface, and as shown in FIG. Since a light reflection image is generated, if Num is 1, it is determined that there is no unevenness. In addition, when the measurement target surface is a surface having a step, the irradiated slit light is reflected from the measurement target surface as intermittent slit reflected light at the step portion, and as shown in FIG. Is composed of two or more line segments, so that Num is 2 or more and the distance of the coordinates (ρ, θ) of the representative point exceeding the threshold Min is greater than or equal to a preset value Z If it is, it will determine with an unevenness | corrugation, and it will determine with the measurement object surface being comprised by the several surface.

  Z represents a distance threshold value, for example, Z = 2, and can be fixed to Z = 1 by adjusting the above p and q.

  The degree of unevenness represents the size of the unevenness. For example, when the measurement target surface is a surface having many small irregularities, the irradiated slit light is reflected from the measurement target surface as intermittent slit reflected light, and as shown in FIG. The image is composed of many line segments, and even in the Hough plane as a result of the Hough transform, as shown in FIG. 8, the regions where the passage positions of the Hough curve are concentrated are dispersed in a plurality, and many representative points are line segments. Since they are extracted as candidates, the difference between Rmax and Rmin is small, and if Num is large, it is estimated that the surface has many small unevenness, and a low unevenness is measured. On the other hand, when the Num is small, it is estimated that the Num is gentle and has large unevenness, so that a high unevenness degree is measured.

  Then, the output unit 28 outputs the presence / absence of unevenness and the measured unevenness degree.

  As described above, according to the unevenness determination apparatus according to the embodiment of the present invention, a single slit light reflection image is formed by irradiating a measurement target surface with one slit light. Since it is only necessary to calculate the number of line segments in the slit light reflection image by determining the presence or absence of irregularities on the measurement target surface based on the number of line segments, the three-dimensional shape of the measurement target surface can be directly calculated. The processing can be simplified without measuring, and the presence or absence of unevenness on the measurement target surface can be determined easily and at high speed.

  In addition, by thinning one slit light reflection image, a plurality of central pixels are extracted from the slit light reflection image in the longitudinal direction to form a slit constituent point sequence, and at the position of each point of the extracted slit constituent point sequence. Based on this, by calculating the number of line segments, the number of line segments constituting the slit light reflection image can be calculated with high accuracy.

  Further, since the number of line segments constituting one slit light reflection image may be calculated using the Hough transform, it is possible to determine the presence or absence of unevenness on the measurement target surface more easily and at high speed.

  Moreover, since the degree of unevenness of the measurement target surface is measured based on Num corresponding to the number of line segments and Rmax and Rmin related to the number of votes, not only the presence / absence of unevenness but also the unevenness degree of the measurement target surface can be easily And it can measure at high speed.

  Further, normally, the three-dimensional position of the irradiated slit light is calculated based on the principle of triangulation, and the shape of the measurement target surface is measured. In the unevenness determination apparatus according to the present embodiment, one irradiated single light is measured. The reflected light of the slit light is imaged, the number of line segments constituting the generated slit light reflected image and the state of bending are measured, and the presence or absence of irregularities on the measurement target surface can be easily and quickly determined. The size of the unevenness can be measured.

  In the above-described embodiment, the case where the unevenness degree is measured has been described as an example. However, the flatness indicating the degree of flatness may be measured.

  In addition, the case has been described as an example in which representative points of a lattice area where many intersections are located are extracted as line segment candidates by Hough transform. However, the present invention is not limited to this. You may make it extract the representative point of the lattice area | region on the Hough plane where the passage position of the Hough curve obtained by conversion concentrates as a line segment candidate. In this case, it is only necessary to vote for the lattice area that is the passing position of each Hough curve and calculate the number of votes for each lattice area.

  Moreover, although the case where the unevenness degree is measured based on the line segment parameters such as Rmax and Rmin regarding the number of votes has been described as an example, the number of extracted representative points and the extracted representative points on the ρ-θ plane The unevenness degree may be measured from the variation of the projection points by extracting the variation of the projection points in the ρ and θ directions in the peripheral region as a line segment parameter.

  Moreover, although the case where p and q which show the threshold value of the intersection range which is the uneven | corrugated range judged to be the same straight line segment becomes the same value was demonstrated to the example, as a feature of the ρ-θ plane, ρ P and q may be set so that the range of the direction (corresponding to the difference in Y intercept on the XY plane) is wide, and the range of the θ direction (corresponding to the difference in inclination on the XY plane) is narrow.

  In addition, the line segment of the representative point extracted on the Hough plane is reconstructed into a straight line on the XY plane, and the extracted line segment is reconfirmed by comparing it with the captured slit light reflection image. It may be.

  Moreover, although the case where the number of slit lights to irradiate is one was demonstrated to the example, you may make it irradiate two or more slit lights. In this case, a slit constituent point sequence may be extracted for each of the plurality of slit lights from the slit light reflection image generated by imaging the reflected light, and the number of line segments may be calculated.

  Moreover, although the case where a plurality of pixels of the pixel signal indicating the reflected light of the maximum amount of light are extracted in the longitudinal direction as a central pixel to form a slit constituent point sequence has been described as an example, each array arranged in a direction orthogonal to the longitudinal direction A plurality of pixels in the longitudinal direction may be extracted in the longitudinal direction with the pixel at the center of gravity of the amount of reflected light indicated by the pixel signal of the pixel as the center pixel, and may be used as a slit configuration point sequence.

It is a block diagram which shows the unevenness | corrugation determination apparatus which concerns on embodiment of this invention. It is a flowchart which shows the content of the process for performing the determination of the presence or absence of an unevenness | corrugation, and the measurement of an unevenness | corrugation degree of the unevenness | corrugation determination apparatus which concerns on embodiment of this invention. It is a figure for demonstrating the mode of the Hough curve in the case of 1 line segment. It is a figure for demonstrating the mode of the Hough curve in the case of two line segments. It is a figure for demonstrating a slit light reflective image in case a measurement object surface is a plane. It is a figure for demonstrating a slit light reflected image in case a measurement object surface is a level | step difference surface. It is a figure for demonstrating a slit light reflected image in case a measurement object surface is an uneven surface. It is a figure for demonstrating the mode of a Hough curve in case a measurement object surface is an uneven surface and there are many line segments.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Concavity and convexity determination apparatus 12 Slit light source 14 Slit light irradiation control part 16 Imaging part 18 A / D conversion part 20 Image memory 22 Slit extraction image processing part 24 Line segment calculation part 26 Measurement part 28 Output part R Number of votes

Claims (8)

Slit light irradiation means for irradiating the measurement object with slit-shaped light;
An imaging unit that images the reflected light that is irradiated by the slit light irradiation unit and reflected from the measurement target, and generates a slit light reflection image;
Calculation means for calculating the number of line segments constituting the slit light reflection image generated by the imaging means;
A determination unit that determines the presence or absence of unevenness of the measurement target based on the number of the line segments calculated by the calculation unit;
Concavity and convexity determination device including Based on pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of the slit light reflection image generated by the imaging unit, a central pixel serving as a slit center position in the orthogonal direction is defined in the longitudinal direction. A plurality of extraction means for extracting,
The unevenness determination apparatus according to claim 1, wherein the calculation unit calculates the number of the line segments based on a plurality of center pixels extracted by the extraction unit.   The extraction means uses the pixel signal of the pixel signal indicating the maximum amount of reflected light as the central pixel based on pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of the slit light reflected image as the central pixel in the longitudinal direction. The uneven | corrugated determination apparatus of Claim 2 which extracts two or more to.   The extraction unit is configured to determine, based on pixel signals of a plurality of pixels arranged in a direction orthogonal to the longitudinal direction of the slit light reflected image, a pixel that is a barycentric position of the amount of reflected light indicated by the pixel signal of each pixel. The unevenness determination device according to claim 2, wherein a plurality of center pixels are extracted in the longitudinal direction.   The calculation means includes a Hough conversion means for converting each of the coordinates indicating the positions of a plurality of central pixels extracted by the extraction means into a Hough curve on a Hough plane, and based on a conversion result by the Hough conversion means, The uneven | corrugated determination apparatus of any one of Claims 2-4 which calculates the number of the said line segments. The Hough transforming unit transforms each of the coordinates indicating the positions of the plurality of central pixels into the Hough curve, and converts the Hough plane into each of the lattice regions when the Hough plane is divided into a plurality of predetermined lattice regions. On the other hand, vote for the grid area at the passing position or the intersection position of the Hough curve, and calculate the number of votes of each grid area,
The unevenness determination apparatus according to claim 5, wherein the calculation unit calculates the number of the line segments by detecting the line segment based on the number of votes calculated by the Hough conversion unit.   A concavo-convex determination apparatus further comprising a measurement unit that measures the flatness or the concavo-convex degree of the measurement target based on the number of line segments calculated by the calculation unit and the number of votes calculated by the Hough conversion unit. Irradiate the measurement target with slit-shaped light,
Imaging the reflected light that has been irradiated and reflected from the measurement target to generate a slit light reflection image,
Calculating the number of the line segments constituting the generated slit light reflection image;
An unevenness determination method, wherein the presence or absence of the unevenness of the measurement target is determined based on the calculated number of line segments.

Images