JP2009168472A - Calibration device and calibration method of laser scanner - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for calibrating a laser scanner without using any special apparatus. <P>SOLUTION: A target having a vertical plane S<SB>1</SB>whose both right-and-left sides are parallel and vertical is scanned horizontally by the laser scanner 1 from a measuring origin O<SB>m</SB>in front of the vertical plane S<SB>1</SB>, to thereby obtained point group data to measuring points to the number of M. An effective measuring point extraction part 13 specifies end measuring points A, B on the ends of both right-and-left sides of the vertical plane S<SB>1</SB>among the point group data, and extracts a relative coordinate of effective measuring points to the number of m between the end measuring points A, B. A coordinate determination means 14 sets a geodetic coordinate system based on a line (scanning line E<SB>1</SB>E<SB>2</SB>) cutting the vertical plane S<SB>1</SB>by a locus of an optical axis, and determines a relation parameter for relating the relative coordinate to a geodetic coordinate so that the square sum of a distance d<SB>A</SB>from an endpoint E<SB>1</SB>to the end measuring point A, a distance d<SB>B</SB>from an endpoint E<SB>2</SB>to the end measuring point B, and each distance d<SB>i</SB>from each effective measuring point P<SB>i</SB>other than the end measuring points A, B to the scanning line E<SB>1</SB>E<SB>2</SB>becomes minimum. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  TECHNICAL FIELD The present invention relates to a laser scanner calibration technique, and in particular, as a pre-stage for performing a survey by a laser scanner, a rectangular coordinate value of a figure origin serving as a measurement origin and a survey reference is determined, and a standard for subsequent calibration of the survey value It relates to a technique for determining a value.

  In recent years, navigation for pedestrians using mobile phones and mobile terminals has attracted attention. Along with this, indoor map data such as underground malls and station premises is required. Conventionally, a surveying technique using a total station (electronic angle measuring finder) or an electronic flat plate has been widely used as a method for maintaining such indoor map data. However, since the data required for the indoor map is wide and detailed, such a conventional survey would require a great deal of cost and time.

  Therefore, a measurement method using a laser scanner has been researched and developed as a method for preparing map data.

  A laser scanner is a type of optical wave rangefinder that detects the reflected light reflected from the measurement target point by irradiating a near-infrared pulse laser beam while rotating the irradiation direction with a rotating mirror inside the laser header. By measuring the pulse propagation time of the laser beam by the phase comparison method (see Non-Patent Document 1), etc., the distance to the measurement target point is measured, and at the same time, the direction in which the pulse laser beam is irradiated is measured. A measuring device that acquires the coordinates of a target point. The irradiated pulse laser beam is reflected when it hits an object at a measurement target point. This reflected light is detected by the light receiving portion of the laser scanner. The time difference between the light projecting and receiving of the pulse is detected by the phase difference of the modulation frequency. Since this time difference is proportional to the distance between the laser scanner and the object, the distance to the measurement target point can be obtained from this time difference. The direction of the pulse laser beam is sequentially changed by a rotating mirror inside the laser scanner, and the surrounding area is scanned in a fan shape with a predetermined angular resolution. Angular resolution varies from model to model. Currently, models with an angular resolution of 0.01 to 0.5 ° have been developed for high angular resolutions. Point cloud data representing a laser distance image in a two-dimensional plane drawn by the trajectory of the laser beam can be obtained by one sector scan.

  In a map data method using a laser scanner, measurement is performed while moving the laser scanner, and one map data is created by superimposing time-series point cloud data obtained from the laser scanner. At this time, the position and direction of the measuring device are estimated based on time-series point cloud data obtained from the laser scanner. This is called self-position estimation.

  In outdoor measurement, GPS is widely used for self-position estimation (see Non-Patent Documents 2 and 3). However, since GPS cannot be used for indoor measurement, a method using an inertial measurement unit (IMU: Inertial Measurement Unit) (see Non-Patent Document 4) and a method using scan matching (Non-Patent Document 5) -7) is used. The scan matching method is executed by superimposing point cloud data acquired in time series by a laser scanner. In practice, the time series point cloud data is acquired by continuously measuring while moving the place where the carriage equipped with the laser scanner is to be measured. One map point cloud data is created by superimposing these acquired time series point cloud data. Various methods have been devised for this superposition method (see Non-Patent Documents 5-7). For example, in the method of Non-Patent Document 7, execution is performed by superimposing point cloud data obtained at a certain time on point cloud data at the previous time by transformation (Helmart transformation) using translation and rotation. Is done.

  The point cloud data acquired by the laser scanner is coordinates centered on the laser scanner. Therefore, if the measurement start position and the orientation of the laser scanner are different for each measurement, point group data of a different coordinate system is obtained for each measurement. Therefore, when trying to create map data by synthesizing point cloud data obtained by a plurality of measurements, it is not possible to synthesize correctly without information on absolute position coordinates for each measurement. In general, laser scanners generate errors in measured values due to temperature, location, etc., and errors due to atmospheric influences. Therefore, in the measurement by the laser scanner, it is necessary to perform calibration after the laser scanner is horizontally installed.

  In the calibration of the laser scanner, the actual measurement target dimension is compared with the measurement value, and the measurement value is corrected. Usually, in calibration, scanning is performed at the position where measurement is started, and several measurement target points whose coordinate values are known are used to start the laser header (measurement origin) and the front of the laser header. Determine the geodetic coordinates of the measurement target point (gazing point) in the direction. Then, the position and orientation are calibrated based on the relationship between the point cloud data obtained from the laser scanner and the position coordinates of the measurement origin and gazing point.

  As a laser scanner calibration method, those described in Non-Patent Documents 8 and 9 are known.

  The method described in Non-Patent Document 8 describes a method in which a CoBit (Compact Battery-less Information Terminal) is pasted as a target on a vertically standing glass surface, and a laser scanner is calibrated using this CoBit.

  In this method, paying attention to the point that the laser beam is transmitted through the glass surface, a target serving as a reference point is attached to the glass surface. In order to accurately capture the reference point, it is necessary that the center of the spot diameter of the laser beam coincides with the center of the target attached to the glass surface. Since the laser beam is invisible and the spot diameter increases with distance, the spot of the laser beam cannot be confirmed with the naked eye.

Therefore, CoBit is used to capture the spot of the laser beam. CoBit is a device that allows you to listen to sound with earphones directly connected to solar cells. By using CoBit as a target, the laser beam is stronger at the center of the laser beam spot, and the generated sound is larger. This feature is used to capture the center of the spot. Next, a plurality of reference points installed so as to be in the center of the laser spot are measured, and the coordinates (X _l , Y _l , Z _l ) of each reference point in the laser scanner coordinate system are obtained. Here, since the locus of the laser is a straight line, the values of Zl are all 0. Next, in order to obtain the true value (absolute value) of the position coordinates of each reference point, the position coordinates of each reference point are measured by the total station, and the coordinates (X _t , Y _t , Z) of each reference point in the geodetic coordinate system are measured. _t ). From these results, a conversion formula from the laser scanner coordinate system to the geodetic coordinate system is obtained. Specifically, the transformation parameters of the Helmat transform for correcting the three-dimensional rotation, the translation of the origin, and the scale are obtained. The Helmat transform is expressed by the following equation.

ここで、Ｒは三次元回転行列、ｓは縮尺、（Ｔ_ｘ，Ｔ_ｙ，Ｔ_ｚ）^ｔは平行移動量である。

Here, R is a three-dimensional rotation matrix, s is a scale, and (T _x , T _y , T _z ) ^t is a translation amount.

In Non-Patent Document 9, four or more reference points are set on the feature to be measured, reflection sheets are set at the reference points, and the laser scanner is calibrated with these reflection sheets. The method of doing is described. Also in this method, the position coordinates of each reference point are measured and determined by the total station, measured by the laser scanner, and the coordinates of each reference point in the laser scanner coordinate system and each reference in the geodetic coordinate system measured by the total station. Calibration is performed by obtaining affine transformation parameters for transforming from the laser scanner coordinate system to the geodetic coordinate system based on the coordinates of the points.
Oshima Taichi, “Surveying Science <Basics>”, first edition, Kyoritsu Publishing Co., Ltd., April 10, 1997, pp. 152-158. Tsuji, Ryosuke Shibasaki, “Automatic construction of 3D urban space model using vehicle-mounted laser range sensor”, 8th Symposium on Image Sensing, 2002, pp.121-126. H. Zhao and R. Shibasaki, "A Vehicle-borne Urban 3D Acquisition System using Single-row Laser Range Scanners", IEEE Trans.SMC Part B: Cybernetics, 2003, Vol.33, No.4, pp .. Masahiko Nagai et al., “Construction of 3D model by integrating laser scanner and CCD sensor using IMU”, Photographic Society of Japan Annual Meeting 2004, Photogrammetric Society of Japan, 2004, pp.5 -8. PJBesl and NDMckay, "A Method for Registration of 3-D Shapes", IEEE Transaction on Pattern Analysis and Maching Intelligence, 1992, Vol.14, No.2, pp.239-256. Atsumi Nakamoto, Atsushi Yamashita, Toru Kaneko, “Creating a three-dimensional map of a dynamic environment using a mobile robot equipped with a laser range finder”, Technical Report of ITE, 2006, Vol.30, No.36, pp.263-271 . Toru Hiraoka, Koji Wakamatsu, Naoto Ozaki, “Robust Superposition of Time Series Laser Scanner Data for Indoor Mapping”, Photogrammetry and Remote Sensing, 2007, Vol.46, No.2, pp.37-41 . Masahiko Nagai, Mananda Dinesh, Ryosuke Shibasaki, “Research on Laser Scanner Calibration Method Using“ Cobit ””, Proceedings of 2004 Annual Conference of the Japan Society of Photogrammetry, 2004, pp.1-4. Koji Ujiie, “Tracking landslide displacement by object matching using laser scanner”, Kochi University of Technology 2004 (2004) Master thesis, [online], January 2005, Kochi University of Technology Library, [2007 October Search on the 1st of the month], Internet <http://www.kochi-tech.ac.jp/library/ron/2004/g9/M/1075013.pdf>

  However, when the calibration is performed using the CoBit or the reflection plate as described above, it is inconvenient because it is necessary to prepare a special device dedicated for calibration such as the CoBit or the reflection plate or the total station every time measurement is performed.

  Accordingly, an object of the present invention is to provide a technique for calibrating a laser scanner without using a special device dedicated for calibration.

A first configuration of the present invention according to the calibration apparatus of a laser scanner, the target left and right sides sides with parallel and perpendicular vertical plane, and the horizontal scanning by the front of the measuring laser scanner from the origin O _m of the vertical plane Point cloud data storage means for storing point cloud data consisting of relative coordinates based on the measurement origin O _m for the M measurement points (M is an integer) obtained;
The end points including the end points A and B are identified from the point cloud data by specifying end points (hereinafter referred to as “end points”) A and B on the left and right sides of the vertical plane. Effective station extraction means for extracting relative coordinates of m stations between A and B (hereinafter referred to as “effective stations”) P _i (i = n,..., N + m−1);
A geodetic coordinate system is set with reference to a scanning line E ₁ E _{2 in} which the locus of the optical axis of the laser scanner cuts the vertical plane, and the end measurement is performed from _one end point E _{1 of the} scanning line E ₁ E _2. the distance _{d a} to the point a, the distance _{d B} from the other end point _{E 2} of the scanning lines _E 1 _{E 2} to the end stations B, and the end stations a, wherein each active stations _{P i} other than B A relation parameter that relates the relative coordinate to the geodetic coordinate is determined so that the sum or square sum of the distances d _i from (i = n + 1,..., N + m−2) to the scanning line E ₁ E ₂ is minimized. And a coordinate determining means.

  According to this configuration, an object having a vertical plane in which both left and right sides are parallel and vertical may be used as a target used for laser scanner calibration. Therefore, for example, a prism of a prismatic building standing upright can be used as a target. Therefore, the laser scanner can be calibrated without using a special device dedicated to calibration as in the prior art.

Further, when determining the relational parameter relating the relative coordinates to the geodetic coordinates, the distance d _A , the distance d _B , and the distance d _i (i = n + 1,..., N + m−2) are used as the convex evaluation function J to be minimized. Sum of

又は自乗和

Or square sum

を用いることで、走査時における測定方位角の離散化による量子化誤差の影響を最小限に抑えることができる。

By using, it is possible to minimize the influence of quantization error due to discretization of the measurement azimuth angle during scanning.

Here, the “vertical plane” may be a plane in which the left and right sides are parallel and vertical, and the shape of the upper and lower sides is not limited. However, it is assumed that both sides of the vertical plane have convex angles. “Measurement origin” refers to the point of the rotation center of the optical axis of the laser beam of the laser scanner. “Measurement point” refers to a spatial point at which coordinate measurement is performed by a laser scanner. The number M of measurement points is not particularly limited, and is a value determined by the angular resolution and scanning angle of the laser scanner. Here, “angular resolution” refers to a difference in azimuth between adjacent measurement points. “Scanning angle” refers to the difference angle between the azimuth angle of the optical axis at which the laser scanner starts scanning in one scan and the azimuth angle of the optical axis at which scanning ends. “Relative coordinates with reference to the measurement origin O _m ” refers to coordinates fixed to the laser scanner with the measurement origin as the origin. Relative coordinates can be expressed in terms of coordinates such as Cartesian coordinates, polar coordinates, cylindrical coordinates, etc. However, since the values directly measured by the laser scanner are the azimuth angle φ of the measurement point and the distance l from the measurement origin to the measurement point, For the relative coordinates, polar coordinate representation (1, φ) is used. The “geodetic coordinate system” refers to a coordinate system in which the locus of the optical axis of the laser scanner is set based on a line (scanning line E ₁ E ₂ ) that cuts a vertical plane. Since the target is fixed at a predetermined position in the measurement space, the geodetic coordinate system is a coordinate system fixed in the measurement space. The coordinate representation of the geodetic coordinate system is not particularly limited, but orthogonal coordinate representation is usually used in consideration of convenience in creating a map. “End measurement point” refers to a measurement point at the end of a plane to be measured. Here, since the azimuth angle between each measurement point is discrete in the scanning of the laser scanner, in general, the end measurement point does not exactly coincide with the end of the plane to be measured. “Relational parameters that relate relative coordinates to geodetic coordinates” refer to parameters that associate relative coordinates with geodetic coordinates. Although the method of taking the parameters of the correspondence is arbitrary, for example, the position coordinates (x _O , y _O ) of the measurement origin O _m in the geodetic coordinate system and the gazing point G (on the optical axis in the center direction of the scanning angle range) Position coordinates (x _G , y _G ) can be used.

According to a second configuration of the present invention relating to a laser scanner calibration apparatus, in the first configuration, the coordinate determining means includes:
The end points A and B are located on the scanning line E ₁ E ₂ , and the distance from the end point E ₁ to the end point A is equal to the distance from the end point E ₂ to the end point B. First initialization means for calculating initial values of the position coordinates of the end stations A and B in the geodetic coordinate system;
Based on the initial value of the position coordinates of the end stations A and B and the relative coordinates of the effective stations, the effective stations P _i other than the end stations A and B (i = n + 1,..., N + m−2). and a second initialization means for calculating the initial value of the position coordinates in the geodetic coordinate system of the measurement origin O _m,
Using the sum or square sum of the distance d _A , the distance d _B , and the distance d _i (i = n + 1,..., N + m−2) as an evaluation function, the measurement origin O _m and the end measurement points A and B are Starting from the initial value of the position coordinates of each of the effective measurement points P _i (i = n,..., N + m−1) including the measurement origin O _m and each of the evaluation functions so that the value of the evaluation function is minimized. Conversion parameter calculation means for calculating each conversion parameter of Helmert conversion that performs rotation and translation of the effective measuring point P _i by an iterative method;
Origin coordinate determining means for performing a Helmart transform on the initial value of the position coordinate of the measurement origin O _{m using} the calculated conversion parameters, and calculating the position coordinate of the measurement origin O _{m in} the geodetic coordinate system;
The relative coordinates of the gazing point G, which is a measurement point on the optical axis in the bisecting direction of the scanning angle between the two directions of the scanning start direction and the scanning end direction of the laser scanner that rotationally scans a predetermined angle range, are described above. extracted from the point cloud data, on the basis of the position coordinates in the geodetic coordinate system and the relative coordinates of the noted viewpoints G the measurement origin O _m, gazing point for calculating the position coordinates in the geodetic coordinate system of the gaze point G And a coordinate determination means.

According to this configuration, first, the first initialization means sets initial values of the position coordinates of the end measuring points A and B. In this specification, “positional coordinates” refers to positional coordinates in the geodetic coordinate system unless otherwise specified. Since the relative coordinates of the end stations A and B are known from the point cloud data, the distance r _AB between the end stations A and B is also known. Therefore, the end measurement points A and B are arranged on the scanning line E ₁ E ₂ that is separated from the midpoint of the scanning line E ₁ E _{2 by} the distance r _AB / 2. Next, the second initialization means calculates the initial values of the position coordinates of the effective measurement point P _i (i = n + 1,..., N + m−2) and the measurement origin O _m . End measurement point A, the relative coordinates of the effective measurement point P _i and the measurement origin O _m to B is also known by the point cloud data, the end measurement point A, the effective measurement point P if the initial value of the position coordinates Sadamare of B the initial value of the position coordinates of the _i and the measurement origin O _m also uniquely determined. Next, the conversion parameter calculation means moves the measurement origin O _m and each effective measurement point P _i (i = n,..., N + m−1) to an optimal position by Helmert conversion. The Helmat transform parameter at the time of this movement is obtained by an iterative method using the sum of the distances or the sum of squares expressed by the above formula (2) or (3) as the evaluation function J. Since Expression (2) or Expression (3) is a convex function, convergence in the iterative method is guaranteed. Next, the origin coordinate defining means, as one of the relevant parameters, and calculates the position coordinates in geodetic coordinate system of the measuring origin O _m. Finally, the gazing point coordinate determining means calculates the position coordinates of the gazing point G in the geodetic coordinate system as another related parameter. As described above, if the position coordinates of the measurement origin O _m and the gazing point G in the geodetic coordinate system are obtained as the related parameters, the relative coordinates of other measurement points can be converted into the absolute coordinate system. In addition, when point cloud data is acquired by repeatedly scanning while moving the laser scanner from the front of the target to another location, the position coordinates in the geodetic coordinate system are determined while superimposing the point cloud data obtained by the previous scanning. I can go.

  Here, as the algorithm of the “iterative method”, various known methods such as the steepest descent method, the Jacobian method, the Gauss-Seidel method, and the SOR (Successive Over Relaxation) method can be used.

According to a third configuration of the present invention relating to a laser scanner calibration apparatus, in the first configuration, the point cloud data storage means includes a plurality of vertical planes whose left and right sides are parallel and perpendicular to each other. the target are parallel to each other, wherein the front of the measuring laser scanner from the origin O _m of the vertical plane obtained by horizontally scanning, the measurement origin O _m for measurement point M (M is an integer greater than 2) It stores point cloud data consisting of relative coordinates as a reference,
The effective station extraction means specifies end stations A _j and B _j (j is a subscript for distinguishing each vertical plane) for each vertical plane, and each effective station P _j, extract the relative coordinates of _i ,
The coordinate determination means sets a geodetic coordinate system based on scanning lines E _{p, 1} E _{p, 2 in} which the optical axis of the laser scanner cuts the vertical planes, and the scanning in the vertical planes line _{E j, 1 E} j, the distance ₂ of one end point _{E j,} from ₁ to the end stations _{a j d j, a,} the scan lines _{E j, 1 E} j, ₂ of the other end point _{E j,} distance from ₂ to the end stations _{B j d j, B,} and said end stations _a j, wherein each active stations _P j other than _{B j, i (i = n} j + 1, ..., n j + m j −2) to the scanning lines E _{j, 1} E _{j, 2} , the relative coordinates are set so that the sum of the distances d _{j, i} or sums of squares of all the vertical planes is minimized. It is characterized in that a relation parameter related to the geodetic coordinates is determined.

  According to this configuration, even when the target has a plurality of vertical planes on the same plane, the laser scanner can be calibrated in the same manner as in the first configuration.

The fourth configuration of the present invention relating to the laser scanner calibration apparatus is that the left side is a straight line or curved line whose left side is vertical and the right side is inclined, and the left side is a straight line whose right side is vertical and the left side. A right vertical plane that is an inclined straight line or an inclined curve parallel to the right side of the left vertical plane, and the left vertical plane and the right vertical plane are parallel and coplanar, It was obtained by scanning a target whose right side and the left side of the right vertical plane are separated by a certain horizontal width from a measurement origin O _{m in} front of the left vertical plane and the right vertical plane with a laser scanner. Point cloud data storage means for storing point cloud data composed of relative coordinates based on the measurement origin O _m of M measurement points (M is an integer greater than 2);
From the point cloud data, the left end point (hereinafter referred to as “end point”) A and right end point C _{L of} the left vertical plane, and the right end point of the right vertical plane are measured. identify the point B and the left end measuring point C _R, the end stations a, horizontal width between C _L r _L and the end stations B, a width calculation means for calculating the horizontal width r _R between C _R ,
The horizontal width of the left vertical plane at height z is f _L (z), the horizontal width of the right vertical plane is f _R (z), and the inverse functions of f _L (z) and f _R (z) are f _L , respectively. ⁻¹ (x), f _R ⁻¹ (x), the height H of the measurement origin O _m of the laser scanner is set to a height f _L ⁻¹ (r ′ _L ) or a height f _R ⁻¹ ( any one of r ′ _R ) or a height calculating means for calculating an average value of the height f _L ⁻¹ (r _L ) and the height f _R ⁻¹ (r _R );
It is provided with.

According to this configuration, an object having the left vertical plane and the right vertical plane as described above may be used as a target used for calibration of the laser scanner. For example, two congruent right triangle plates (wood board, cardboard, metal plate, etc.) are fixed with a fixed distance apart with their hypotenuses aligned in parallel on the same plane. It is possible to use a target having such a shape that it stands vertically with respect to the floor surface. Such targets, it is possible to make easily without using a special material, calibration of the height H of the measurement origin O _m of the laser scanner without using a conventional special instrument calibration only as Can be performed.

According to a fifth configuration of the present invention relating to a laser scanner calibration apparatus, in the fourth configuration, the horizontal width between the right side of the left vertical plane and the left side of the right vertical plane is r 0 , When the horizontal width between the left side of the left vertical plane and the right side of the right vertical plane is r + r 0 , the horizontal widths r L and r R measured at the measurement origin are expressed by equations (4a) and (4), respectively. 4b) comprising horizontal width correcting means for calculating corrected horizontal widths r ′ L and r ′ R by correcting
The height calculation means, the height H of the measurement origin O m of the laser scanner, one or heights of f L -1 (r 'L) or the height f R -1 (r' R) It is calculated as an average value of f L −1 (r ′ L ) and height f R −1 (r ′ R ).

According to this arrangement, the end measurement point A, the horizontal width between _{C L r L} and Tanhaka point B, wherein a horizontal width _{r R} between _{C R} (4a), by _{(4b), r L + r} R = r by correcting such that, it is possible to reduce the influence of quantization errors due to discretization of the measured azimuth angle during scanning, is possible to perform calibration of the height H of the more accurate the measurement origin O _m It becomes possible.

According to a sixth configuration of the present invention relating to a laser scanner calibration apparatus, in the fourth or fifth configuration, the point cloud data storage means is in front of the left vertical plane and the right vertical plane and has a height. N points of the point cloud data measured from the measurement origins of N points that are the same and have different positions on the horizontal plane are stored.
Said width calculating means, for N sets of the point group data, each of the end stations A, horizontal width between C _L and the end stations B, calculates a horizontal width between C _R, wherein each point group said end stations a for data, C the average value of the horizontal width between _L horizontal width r _L, the said end stations B for each point cloud data, the average value of the horizontal width horizontal between C _R and characterized in that to calculate the width r _R.

With this configuration, by performing the calibration of the height H of the measurement origin O _m based on the point cloud data which is output by the scanning from the laser scanner is moved horizontally a plurality of measurement origin O _m, more accurate measurement it is possible to calibrate the height H of the origin O _m.

The seventh configuration of the present invention relating to the laser scanner calibration apparatus is a left vertical plane whose horizontal width changes according to a monotonic function f _L (z) whose height is z, and a monotone function f _R whose horizontal width is a height z. A right vertical plane that varies according to (z), the left vertical plane and the right vertical plane being parallel and spaced apart from each other by a fixed distance d, and the right side of the left vertical plane and the right vertical plane M targets obtained by scanning a target separated from the left side of the right vertical plane by a certain horizontal width from the measurement origin O _{m in} front of the left vertical plane and the right vertical plane with a laser scanner ( Point cloud data storage means for storing point cloud data composed of relative coordinates with reference to the measurement origin O _m of the measurement point of M (an integer greater than 2);
From the point cloud data, the left end point (hereinafter referred to as “end point”) A and right end point C _{L of} the left vertical plane, and the right end point of the right vertical plane are measured. identify the point B and the left end measuring point C _R, the end stations a, horizontal width between C _L r _L and the end stations B, a width calculation means for calculating the horizontal width r _R between C _R ,
When the inverse functions of the functions f _L (z) and f _R (z) are f _L ⁻¹ (x) and f _R ⁻¹ (x), respectively, the height h _L of the scanning line in the left vertical plane is f _L ⁻¹ (r _L ), a height calculating means for calculating the height h _R of the scanning line in the right vertical plane as f _L ⁻¹ (r _R );
An elevation angle calculating means for calculating an elevation angle θ of the optical axis of the racer scanner based on the height h _L , the height h _R , and the distance d, using Equation (5);
It is provided with.

  According to this configuration, an object having the left vertical plane and the right vertical plane as described above may be used as a target used for calibration of the laser scanner. As such, for example, two congruent right triangle plates (wood board, cardboard, metal plate, etc.) are fixed with a certain distance apart with their hypotenuses parallel and shifted back and forth. However, it is possible to use a target having such a shape that it stands vertically to the floor surface. Since such a target can be easily made without using a special material, the elevation angle of the optical axis of the laser scanner can be calibrated without using a special instrument dedicated for calibration as in the past. It becomes possible.

The eighth configuration of the present invention relating to the laser scanner calibration apparatus comprises a left vertical plane whose left side and right side are vertical and parallel, and a right vertical plane whose left side and right side are vertical and parallel, The left vertical plane and the right vertical plane are parallel to each other and separated from each other by a fixed distance d, and the right side of the left vertical plane and the left side of the right vertical plane are separated by a fixed horizontal width. was to have targeted, the obtained by scanning by the left vertical plane and in front of the measuring laser scanner from the origin O _m of the right vertical plane, M (M is an integer greater than 2) measurement origin O of the measuring points point cloud data storage means for storing point cloud data consisting of relative coordinates based on _m ;
From the point cloud data, the left end point (hereinafter referred to as “end point”) A and right end point C _{L of} the left vertical plane, and the right end point of the right vertical plane are measured. identify the point B and the left end measuring point C _R, the end stations a, the end stations a containing C _L, m _L-number of stations between C _L (hereinafter referred to as "left effective stations". _{) P i (i = n L} , ..., n L + m L -1) of relative coordinates, and said end stations B, said end stations B, _{m R-number} of stations between _{C R} containing _{C R} ( Hereinafter referred to as “right effective station”.) Effective station extracting means for extracting relative coordinates of P _i (i = n _R ,..., N _R + m _R −1);
Based on said relative coordinates of each left effective stations and the respective right enabled stations, two parallel slope and intercept of the scanning line L _L and L _R together from said scanning line L _L each left enabled stations P _{i (i = n L, ...} , n L + m L -1) to the sum of the distance to, from said scanning line _{L R} each right effective measurement point _{P i (i = n R,} ..., n R + m R - Scanning line calculation means for calculating by a least square method so that the sum of the distances up to 1) is minimized;
A scan line distance calculation means for calculating a distance d 'from the scanning line L _L to the scanning line L _R,
An elevation angle calculating means for calculating an elevation angle θ of the optical axis of the racer scanner based on the distance d ′ and the distance d by the equation (6);
It is provided with.

  According to this configuration, an object having the left vertical plane and the right vertical plane as described above may be used as a target used for calibration of the laser scanner. For example, a rectangular plate (wood board, cardboard, metal plate, etc.) is fixed with a certain distance apart with one side aligned in parallel, and this is placed vertically on the floor. Targets with different shapes can be used. Since such a target can be easily made without using a special material, the elevation angle of the optical axis of the laser scanner can be calibrated without using a special instrument dedicated for calibration as in the past. It becomes possible.

According to a ninth configuration of the present invention relating to a laser scanner calibration apparatus, in the eighth configuration, the point cloud data storage means is in front of the left vertical plane and the right vertical plane and has the same height. And N sets of the point cloud data measured from the N measurement origins at different positions on the horizontal plane,
The effective point extraction means extracts the relative coordinates of the left effective point and the right effective point for each point cloud data,
The scanning line calculation means calculates the slope and intercept of each of the scanning line L _L and the scanning line L _R wherein for each point cloud data,
The scanning line distance calculation means calculates a distance d 'of the from each of the scanning line L _L for each point cloud data to the scanning line L _R,
The elevation angle calculating means calculates the elevation angle θ of the optical axis of the racer scanner for each point cloud data,
An averaging means is provided for outputting the average value of the elevation angle θ calculated by the elevation angle calculation means as the elevation angle of the optical axis of the racer scanner.

  With this configuration, the average value of the elevation angle θ measured from a plurality of measurement origins is used as the elevation angle of the optical axis of the racer scanner, so that influences such as quantization errors during scanning of the laser scanner are proposed, and more accurate The elevation angle θ can be calibrated.

In the tenth aspect of the invention according to the calibration apparatus of a laser scanner, a target having a vertical plane ABCD that issued rectangular convex, three-dimensionally scanned by the laser scanner from the front of the measurement origin O _m of the vertical plane Point cloud data storage means for storing point cloud data composed of relative coordinates based on the measurement origin O _m with respect to M × N measurement points (M and N are integers) obtained,
Of the point cloud data, the stations closest to the vertices A, B, C, D at the four corners of the vertical plane ABCD (hereinafter referred to as “corner stations”) P _A , P _B , P _C , P _D And m in a region surrounded by a rectangle including the corner measuring points P _A , P _B , P _C , and P _D including the corner measuring points P _A , P _B , P _C , and P _D. Relative coordinates of n stations (hereinafter referred to as “effective stations”) P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ + n−1) Effective station extraction means for extracting
4 corners of the vertex A of the vertical plane ABCD, B, C, and set standards geodetic coordinate system D, the distance d _A between the vertex A and the corner measurement point P _A, the said vertex B corner measurement point _{P B} and the distance _{d B} between the distance _{d C} between the apex C and the corner measurement point _{P C,} the distance _{d D} between the vertex D and the corner measurement point _{P D,} and the The sum or the square sum of the distances d _{i, j} from each of the effective measurement points P _{i, j} other than the corner measurement points P _A , P _B , P _C , P _D to the vertical plane ABCD is minimized. Coordinate determining means for determining a relationship parameter relating a relative coordinate to the geodetic coordinate;
It is provided with.

  According to this configuration, an object having a rectangular protruding vertical plane ABCD may be used as a target used for calibration of the three-dimensional laser scanner. Therefore, the calibration of the three-dimensional laser scanner can be performed without using a special device dedicated for calibration as in the prior art.

According to an eleventh configuration of the present invention relating to a laser scanner calibration apparatus, in the tenth configuration, the coordinate determining means is the corner measuring points P _A , P _B , P _C , P _D in the geodetic coordinate system. First initialization means for setting an initial value of the position coordinates of
The corner measurement point _{P A, P B,} on the basis of the P C, the initial value and the relative coordinates of each active stations coordinates of _{P D,} the corner measurement point _{P A, P B, P C} , other than _{P D} Second initialization means for calculating initial values of position coordinates in the geodetic coordinate system of each of the effective measurement points P _{i, j} and the measurement origin O _m ;
The distance _{d A, d B, d C} , d D, and the respective distances _{d i,} as the evaluation function the sum or square sum of _j, the measurement origin _{O m} and the corner measurement point _{P A, P B, P C} , each valid stations including _{P D P i, j (i} = m 0, ..., m 0 + m-1, j = n 0, ..., n 0 + n-1) starting from an initial value of the position coordinates of the Thus, transformation parameters for calculating the transformation parameters for geometric transformation for rotating and translating the measurement origin O _m and the effective measurement points P _{i, j} so that the value of the evaluation function is minimized Parameter calculation means;
Origin coordinate determining means for performing a Helmart transform on the initial value of the position coordinate of the measurement origin O _{m using} the calculated conversion parameters, and calculating the position coordinate of the measurement origin O _{m in} the geodetic coordinate system;
Relative coordinates of the gazing point G, which is a measuring point on the optical axis in the central direction of the scanning range of the laser scanner that rotates and scans a predetermined azimuth angle range and elevation angle range, are extracted from the point cloud data, and based on the position coordinates in the relative coordinates and the geodetic coordinate system of the measurement origin O _m, and gazing point coordinate defining means for calculating the position coordinates in the geodetic coordinate system of the gaze point G,
It is characterized by having.

A first configuration of the present invention relating to a laser scanner calibration method is a calibration method for calibrating a laser scanner by a computer,
The target to which the left and right sides sides with parallel and perpendicular vertical plane, the horizontal scanning by the laser scanner from the front of the measurement origin O _m of the vertical plane, the measurement origin O _m for measuring points of M (M is an integer) A scanning step of capturing point cloud data consisting of relative coordinates as a reference into a computer and storing it in a storage device;
From the point cloud data stored in the storage device, the measurement points A and B (hereinafter referred to as “edge measurement points”) A and B on both the left and right sides of the vertical plane are specified. Effective station extraction for extracting the relative coordinates of m stations (hereinafter referred to as “effective stations”) P _i (i = n,..., N + m−1) between the end stations A and B including B Steps,
A geodetic coordinate system is set with reference to a scanning line E ₁ E _{2 in} which the locus of the optical axis of the laser scanner cuts the vertical plane, and the end measurement is performed from _one end point E _{1 of the} scanning line E ₁ E _2. the distance _{d a} to the point a, the distance _{d B} from the other end point _{E 2} of the scanning lines _E 1 _{E 2} to the end stations B, and the end stations a, wherein each active stations _{P i} other than B A relation parameter that relates the relative coordinate to the geodetic coordinate is determined so that the sum or square sum of the distances d _i from (i = n + 1,..., N + m−2) to the scanning line E ₁ E ₂ is minimized. A coordinate determining step is performed.

According to a second configuration of the present invention relating to a laser scanner calibration method, in the first configuration, in the coordinate determination step,
The end measuring points A and B are located on the scanning line E ₁ E ₂ , and the distance from the end point E ₁ to the end measuring point A is equal to the distance from the end point E ₂ to the end measuring point B. A first initialization step of calculating initial values of the position coordinates of the end stations A and B in the geodetic coordinate system,
Based on the initial value of the position coordinates of the end stations A and B and the relative coordinates of the effective stations, the effective stations P _i other than the end stations A and B (i = n + 1,..., N + m−2). and a second initializing step of calculating the initial value of the position coordinates in the geodetic coordinate system of the measurement origin O _m,
Using the sum or square sum of the distance d _A , the distance d _B , and the distance d _i (i = n + 1,..., N + m−2) as an evaluation function, the measurement origin O _m and the end measurement points A and B are Starting from the initial value of the position coordinates of each of the effective measurement points P _i (i = n,..., N + m−1) including the measurement origin O _m and each of the evaluation functions so that the value of the evaluation function is minimized. each transformation parameters Helmert transformation for rotating and translation of the active stations P _i, a conversion parameter calculating step of calculating iteratively,
An origin coordinate determination step of performing Helmart transform on the initial value of the position coordinate of the measurement origin O _m by the calculated conversion parameters and calculating the position coordinate of the measurement origin O _{m in} the geodetic coordinate system;
The relative coordinates of the gazing point G, which is a measurement point on the optical axis in the bisecting direction of the scanning angle between the two directions of the scanning start direction and the scanning end direction of the laser scanner that rotationally scans a predetermined angle range, are extracted from the point cloud data, on the basis of the position coordinates in the geodetic coordinate system and the relative coordinates of the noted viewpoints G the measurement origin O _m, gazing point for calculating the position coordinates in the geodetic coordinate system of the gaze point G And a coordinate determining step.

According to a third configuration of the present invention relating to a laser scanner calibration method, in the first configuration, the scanning step includes a plurality of vertical planes whose left and right sides are parallel and perpendicular to each other. the target is parallel, the obtained by horizontal scanning by the laser scanner from the front of the measurement origin O _m of the vertical plane, M (M is an integer greater than 2) and relative to the measurement origin O _m for measurement points Point cloud data consisting of relative coordinates to be captured in a computer and stored in a storage device,
In the effective station extraction step, end stations A _j and B _j (j is a subscript for distinguishing each vertical plane) are specified for each vertical plane, and each effective station P _j is specified. _{, I} to extract the relative coordinates,
In the coordinate determination step, a geodetic coordinate system is set with reference to scanning lines E _{p, 1} E _{p, 2 in} which the optical axis of the laser scanner cuts the vertical planes. scan line _{E j, 1 E} j, ₂ for one end point _{E j, 1} from the distance _{d j} to the end stations _{a j, a,} the scan lines _{E j, 1 E} j, ₂ of the other end point _{E j ,} the distance from ₂ to the end stations _{B j d j, B,} and said end stations _a j, _{B j} other than the respective effective stations _{P j, i (i = n} j + 1, ..., n j + m _j- 2) to the scanning lines E _{j, 1} E _{j, 2} , the relative coordinates such that the sum of the distances d _{j, i} or sums of squares of all the vertical planes is minimized. Determining a relational parameter relating the to the geodetic coordinates.

A fourth configuration of the present invention relating to a laser scanner calibration method is a calibration method for calibrating a laser scanner by a computer,
A left vertical plane that is a straight line or a curved line whose left side is vertical and whose right side is inclined; and a straight line or a curved line whose right side is vertical and whose left side is parallel to the right side of the left vertical plane. A right vertical plane, and the left vertical plane and the right vertical plane are parallel and coplanar, and the right side of the left vertical plane and the left side of the right vertical plane have a certain horizontal width. the spaced apart from being a target, the scanning by the left vertical plane and in front of the measurement origin O _m from the laser scanner of the right vertical plane, of the measuring points of M (M is an integer greater than 2) the measurement origin O _m A scanning step of capturing point cloud data consisting of relative coordinates as a reference into a computer and storing it in a storage device;
Among the point cloud data stored in the storage device, the left end point (hereinafter referred to as “end point”) A and the right end point C _{L of} the left vertical plane, and the right identify right end measuring point B and the left end measuring point C _R of the vertical plane, the end stations a, C horizontal width between _L r _L and the end stations B, horizontal width between C _R r _R A width calculating step for calculating
The horizontal width of the left vertical plane at height z is f _L (z), the horizontal width of the right vertical plane is f _R (z), and the inverse functions of f _L (z) and f _R (z) are f _L , respectively. ⁻¹ (x), f _R ⁻¹ (x), the height H of the measurement origin O _m of the laser scanner is set to a height f _L ⁻¹ (r ′ _L ) or a height f _R ⁻¹ ( any one of r ′ _R ) or a height calculating step for calculating as an average value of the height f _L ⁻¹ (r _L ) and the height f _R ⁻¹ (r _R );
It is characterized by performing.

According to a fifth configuration of the present invention relating to a laser scanner calibration method, in the fourth configuration, after executing the width calculating step, the right side of the left vertical plane and the left side of the right vertical plane r 0 the horizontal width between said when the horizontal width between the left vertical plane of the left side and the right side of the right vertical plane was r + r 0, wherein measured by the measuring origin horizontal width r L, r R Are corrected by equations (4a) and (4b), respectively, and a horizontal width correction step for calculating corrected horizontal widths r ′ L and r ′ R is executed,
Wherein the height calculation step, the height H of the measurement origin O m of the laser scanner, one or height of the height f L -1 (r 'L) or the height f R -1 (r' R) It is calculated as an average value of the height f L −1 (r ′ L ) and the height f R −1 (r ′ R ).

According to a sixth configuration of the present invention relating to a laser scanner calibration method, in the fourth or fifth configuration, in the scanning step, the height is the same in front of the left vertical plane and the right vertical plane. In addition, horizontal scanning is performed by a laser scanner from N measurement origins at different positions on the horizontal plane, and N sets of the point cloud data are captured in a computer and stored in a storage device.
In the width calculation step, the N sets of the point group data stored in the storage device, each said end stations A, horizontal width between C _L and the end stations B, and horizontal width between C _R to calculate, the said end stations a for each point group data, C _L between the average value of the horizontal width of the horizontal width r _L, the said end stations B for each point cloud data, between C _R and calculates an average value of the horizontal width as the horizontal width r _R.

A seventh configuration of the present invention relating to a laser scanner calibration method is a calibration method for calibrating a laser scanner by a computer,
A left vertical plane whose horizontal width changes according to a monotonic function f _L (z) having a height z, and a right vertical plane whose horizontal width changes according to a monotone function f _R (z) whose height is z; The plane and the right vertical plane are parallel to each other and are separated from each other by a certain distance d, and the right side of the left vertical plane and the left side of the right vertical plane are separated by a certain horizontal width. The target is scanned with a laser scanner from the measurement origin O _{m in} front of the left vertical plane and the right vertical plane, and the target is composed of relative coordinates based on the measurement origins O _m of M (M is an integer) measurement points. A scanning step of capturing group data into a computer and storing it in a storage device;
Among the point cloud data stored in the storage device, the left end point (hereinafter referred to as “end point”) A and the right end point C _{L of} the left vertical plane, and the right identify right end measuring point B and the left end measuring point C _R of the vertical plane, the end stations a, C horizontal width between _L r _L and the end stations B, horizontal width between C _R r _R A width calculating step for calculating
When the inverse functions of the functions f _L (z) and f _R (z) are f _L ⁻¹ (x) and f _R ⁻¹ (x), respectively, the height h _L of the scanning line in the left vertical plane is f _L ⁻¹ (r _L ), a height calculating step for calculating the height h _R of the scanning line in the right vertical plane as f _L ⁻¹ (r _R );
An elevation angle calculating step for calculating an elevation angle θ of the optical axis of the racer scanner based on the height h _L , the height h _R , and the distance d, according to Equation (5);
It is characterized by performing.

An eighth configuration of the present invention relating to a laser scanner calibration method is a calibration method for calibrating a laser scanner by a computer,
A left vertical plane whose left side and right side are vertical and parallel; and a right vertical plane whose left side and right side are vertical and parallel. The left vertical plane and the right vertical plane are parallel and constant with each other. A target that is on a plane separated by a distance d and separated by a certain horizontal width from the right side of the left vertical plane and the left side of the right vertical plane is positioned in front of the left vertical plane and the right vertical plane. of scanning by the laser scanner from the measurement origin O _m, M (M is an integer) as a step of storing in the storage device takes in the point group data composed of relative coordinates relative to the measurement origin O _m of measuring points in the computer ,
Among the point cloud data stored in the storage device, the left end point (hereinafter referred to as “end point”) A and the right end point C _{L of} the left vertical plane, and the right identify right end measuring point B and the left end measuring point C _R of the vertical plane, the end stations a, the end stations a containing C _L, m _L-number of stations between C _L (hereinafter " left enable stations _{"hereinafter.) P i (i = n} L, ..., n L + m relative coordinates L -1), and said end stations B, said end stations B containing _{C R,} between _{C R} an effective point extraction step of extracting relative coordinates of m _R points (hereinafter referred to as “right effective points”) P _i (i = n _R ,..., n _R + m _R −1);
Based on said relative coordinates of each left effective stations and the respective right enabled stations, two parallel slope and intercept of the scanning line L _L and L _R together from said scanning line L _L each left enabled stations P _{i (i = n L, ...} , n L + m L -1) to the sum of the distance to, from said scanning line _{L R} each right effective measurement point _{P i (i = n R,} ..., n R + m R - A scanning line calculation step of calculating by a least square method so that the sum of the distances up to 1) is minimized;
A scanning line distance calculating step of calculating a distance d ′ from the scanning line L _L to the scanning line LR _;
And an elevation angle calculating step of calculating an elevation angle θ of the optical axis of the racer scanner based on the distance d ′ and the distance d by the equation (6).

According to a ninth configuration of the present invention relating to a laser scanner calibration method, in the eighth configuration, in the scanning step, the height is the same in front of the left vertical plane and the right vertical plane and has a horizontal plane. A horizontal scanning is performed by a laser scanner from N measurement origins at different positions on the upper position, and N sets of the point cloud data are captured in a computer and stored in a storage device.
In the effective point extraction step, the relative coordinates of the left effective point and the right effective point are extracted for each point cloud data,
Wherein in the scanning line calculating step, and calculates the slope and intercept of each of the scanning line L _L and the scanning line L _R wherein for each point cloud data,
Wherein the distance calculation step between the scan lines, and calculates the distance d 'of the from respectively each point cloud data of the scanning lines L _L to the scanning line L _R,
In the elevation angle calculating step, the elevation angle θ of the optical axis of the racer scanner is calculated for each point cloud data,
Thereafter, an averaging step is performed in which an average value of the elevation angle θ calculated in the elevation angle calculation step is calculated and output as an elevation angle of the optical axis of the racer scanner.

A tenth configuration of the present invention relating to a laser scanner calibration method is a calibration method for calibrating a laser scanner by a computer,
A target having a rectangular protruding vertical plane ABCD is three-dimensionally scanned from a measurement origin O _{m in} front of the vertical plane by a laser scanner, and the above-mentioned measurement points for M × N (M and N are integers) are measured. a scanning step of storing the point cloud data consisting of relative coordinates as a reference to the storage device takes in the computer measurement origin O _m,
From among the point cloud data stored in the storage device, the four corners of the vertex A of the vertical plane ABCD, B, C, nearest stations (hereinafter referred to as "corner measuring point".) To D P _A, P _B , P _C , P _D are specified, and the corner measuring points P _A , P _B , P _C , P _D including the corner measuring points P _A , P _B , P _C , P _D M × n measuring points (hereinafter referred to as “effective measuring points”) P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ in the enclosed area. An effective station extraction step of extracting the relative coordinates of + n-1);
4 corners of the vertex A of the vertical plane ABCD, B, C, and set standards geodetic coordinate system D, the distance d _A between the vertex A and the corner measurement point P _A, the said vertex B corner measurement point _{P B} and the distance _{d B} between the distance _{d C} between the apex C and the corner measurement point _{P C,} the distance _{d D} between the vertex D and the corner measurement point _{P D,} and the The sum or the square sum of the distances d _{i, j} from each of the effective measurement points P _{i, j} other than the corner measurement points P _A , P _B , P _C , P _D to the vertical plane ABCD is minimized. A coordinate determination step for determining a relationship parameter relating a relative coordinate to the geodetic coordinate;
It is characterized by performing.

According to an eleventh configuration of the present invention relating to a laser scanner calibration method, in the tenth configuration, in the coordinate determining step, the corner measurement points P _A , P _B , P _C , P in the geodetic coordinate system are used. _A first initialization step for setting an initial value of a position coordinate of _D ;
The corner measurement point _{P A, P B,} on the basis of the P C, the initial value and the relative coordinates of each active stations coordinates of _{P D,} the corner measurement point _{P A, P B, P C} , other than _{P D} A second initialization step of calculating initial values of position coordinates in the geodetic coordinate system of each of the effective measurement points P _{i, j} and the measurement origin O _m ;
The distance _{d A, d B, d C} , d D, and the respective distances _{d i,} as the evaluation function the sum or square sum of _j, the measurement origin _{O m} and the corner measurement point _{P A, P B, P C} , each valid stations including _{P D P i, j (i} = m 0, ..., m 0 + m-1, j = n 0, ..., n 0 + n-1) starting from an initial value of the position coordinates of the Thus, transformation parameters for calculating the transformation parameters for geometric transformation for rotating and translating the measurement origin O _m and the effective measurement points P _{i, j} so that the value of the evaluation function is minimized A parameter calculation step;
An origin coordinate determination step of performing Helmart transform on the initial value of the position coordinate of the measurement origin O _m by the calculated conversion parameters and calculating the position coordinate of the measurement origin O _{m in} the geodetic coordinate system;
Relative coordinates of the gazing point G, which is a measuring point on the optical axis in the central direction of the scanning range of the laser scanner that rotates and scans a predetermined azimuth angle range and elevation angle range, are extracted from the point cloud data, and based on the position coordinates in the relative coordinates and the geodetic coordinate system of the measurement origin O _m, and the gazing point coordinate determination step of calculating the position coordinates of the geodetic coordinate system of the gaze point G,
It is characterized by performing.

  As described above, according to the present invention, as a target used for calibration of a laser scanner, a pillar of a building, a vertical plate with a specific shape, or the like is used, and this target only needs to be scanned once or several times. Therefore, it becomes possible to calibrate the laser scanner without using a special device dedicated for calibration.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  FIG. 1 is an external view of a laser scanner 1 according to Embodiment 1 of the present invention that is grounded to a carriage. The laser scanner 1 is mounted horizontally in front of the carriage 2. The cart 2 is a hand-held four-wheel cart, and it is possible to scan surrounding features while moving the laser scanner 1 by pushing the cart manually. A computer 3 is installed on the upper surface of the carriage 2. The point cloud data obtained by scanning with the laser scanner 1 is transferred to the computer 3 via the connection cable and stored in a storage device inside the computer 3. The computer 3 is installed with a calibration program for the laser scanner 1. By executing this program, the computer 3 functions as a laser scanner calibration apparatus according to the first embodiment of the present invention.

  FIG. 2 is a block diagram illustrating a functional configuration of the laser scanner calibration apparatus 10 according to the first embodiment of the present invention. In FIG. 2, the laser scanner 1 and the computer 3 are the same as those in FIG. The computer 3 includes a communication interface 9, a calibration device 10, and a calibration parameter storage unit 11. Actually, the calibration apparatus 10 is provided as a calibration program, and is functionally realized by causing the computer 3 to execute the calibration program. The calibration parameter storage unit 11 is configured by a storage device (such as a magnetic disk device) inside the computer 3.

The calibration apparatus 10 includes a point cloud data storage unit 12, an effective measurement point extraction unit 13, and a coordinate determination unit 14. The point cloud data storage unit 12 stores the point cloud data taken into the computer 3 from the laser scanner 1. The point cloud data storage unit 12 is configured by a storage device such as a magnetic disk or a RAM (Random Access Memory). The effective measurement point extraction unit 13 extracts effective measurement points necessary for calibration calculation from the point cloud data stored in the point cloud data storage unit 12. The coordinate determination unit 14 sets a geodetic coordinate system based on the scanning line E ₁ E ₂ , which is a line in which the locus of the optical axis of the laser scanner cuts the vertical plane, and is relative to the measurement origin O _m of the laser scanner. A relational parameter relating the coordinates to the geodetic coordinates is determined.

  The coordinate determination unit 14 includes a first initialization unit 16, a second initialization unit 17, a conversion parameter calculation unit 18, an origin coordinate determination unit 19, and a gazing point coordinate determination unit 20. The functions of these units will be described together with the following operation description.

  Next, a laser scanner calibration method using the calibration apparatus 10 of FIG. 2 will be described.

FIG. 3 is an external perspective view of a target used in the laser scanner calibration method according to the first embodiment of the present invention. The target T ₁ has a vertical plane S ₁ whose left and right sides are parallel and vertical. Horizontal width of the vertical plane _{S 1} is _{r 1.} Left and right side edges of the vertical plane S ₁ has a right angle. It should be noted that the bend angle on both sides of the usable target is not necessarily a right angle, but may be a convex angle.

Such targets T _1, for example, can be used as pillars in a building. Of course, a box having a shape as shown in FIG. 3 may be prepared and used for calibration. Since the target T ₁ is a simple shape, for example, it can be readily prepared using cardboard and plate.

The target T ₁ is scanned horizontally by the laser scanner 1. At this time, the line segment where the locus plane (horizontal plane spreading in a fan shape) S _{locus of} the laser beam irradiated by the laser scanner 1 cuts the vertical plane S ₁ is referred to as “scanning line” and is denoted as E ₁ E _2. . The rotation center point of the optical axis of the laser scanner is called “measurement origin” and is denoted as O _m . The difference angle between the azimuth angle of the optical axis at which the laser scanner 1 starts scanning in one scan and the azimuth angle of the optical axis at which scanning ends is referred to as “scan angle” and is denoted as θ. A measuring point on the optical axis in the center direction of the scanning angle range of the laser scanner 1 is referred to as a “gaze point” and is denoted as G. The relationship among the target T ₁ , the vertical plane S ₁ , the scanning line E ₁ E ₂ , the measurement origin O _m , the optical axis locus plane S _locus , the scanning angle θ, and the gaze point G is as shown in FIG.

In the calibration, the position coordinates of the measurement origin O _m and the gazing point G in the geodetic coordinate system fixed to the target T ₁ are determined. Here, in the geodetic coordinate system, the origin O is the end point E _{1 on} the left side of the scanning line E ₁ E ₂ , the scanning line direction is the x axis, the direction perpendicular to the scanning line direction and within the _locus plane S _locus of the optical axis is the y axis. , The vertical upper direction is set as the z axis.

  FIG. 4 is a flowchart of the laser scanner calibration method according to the first embodiment of the present invention.

First, in step S1, the carriage 3 the laser scanner 1 is mounted, is placed on level surface of the vertical plane S ₁ in front of the target T _1, the trajectory plane of the optical axis during the scanning with a level or the like Adjust so that the S _locus is horizontal. The height of the measurement origin of the laser scanner 1 is also measured in advance with a rectangular scale or the like. Since the internal structure of the laser scanner 1 is known from the design drawing, this height can be measured accurately. Further, the horizontal width r of the vertical plane S ₁ of the target T ₁ is also measured in advance with a rectangular scale or the like and input to the computer 3.

Then, by the laser scanner 1 to scan a range including a vertical plane S _1. The laser scanner 1 performs scanning while rotating the scan head within a certain azimuth range. Therefore, the optical axis of the laser beam rotates in a horizontal plane, and distance information to the measurement point is sampled at every fixed rotation angle. The point cloud data obtained by sampling is transferred to the computer 3 and stored in the point cloud data storage unit 12.

FIG. 5 shows the positional relationship between each measurement point P _i (i = 1, 2,..., M) obtained by scanning the target T ₁ and the measurement origin O _m . Let M be the number of samples in one scan. The sampling interval (azimuth angle interval) of the laser scanner 1 is referred to as “angular resolution” and is denoted as θ _s . Laser scanner 1 starts scanning from the azimuth angle theta _1. The azimuth angle θ _i in the i-th (i = 1, 2,..., M) sampling direction is

である。また、各サンプリング方向の光軸が計測対象物に当たりレーザービームが反射される点を「測点」といい、Ｐ_ｉと記す。また、計測原点Ｏ_ｍから測点Ｐ_ｉまでの距離（光軸の長さ）をｌ_ｉと記す。

It is. Also, the point at which the optical axis of each sampling direction the laser beam is reflected Upon measurement object called "survey point", referred to as P _i. The distance from the measurement origin _{O m} to stations _{P i} (the length of the optical axis) denoted as _{l i.}

A measuring point in the sampling direction in which the optical axis intersects with the scanning lines E ₁ E ₂ is referred to as an “effective measuring point”. In FIG. 5, the n-th to (n + m−1) -th station P _j (j = n, n + 1,..., N + m−1) is an effective station. Of all effective stations, those at the right end and the left end are referred to as “end stations” and denoted as A and B, respectively. A = P _n and B = P _{n + m−1} . In addition, the distance from the measurement origin Qm to each of the A and B ends is _denoted as l _A = ln, l _B = ln _{+ m−1} .

The laser scanner 1 acquires point cloud data {(l _i , θ _i ) | i = 1, 2,..., M} by one scan. The point cloud data {(l _i , θ _i ) | i = 1, 2,..., M} is stored in the point cloud data storage unit 12.

Next, in step S2, the effective measurement point extraction unit 13, the point cloud data _{{(l i, θ i)} | i = 1,2, ..., M} from the, vertical plane _{S 1} left and right sides of the End stations A and B are identified, and relative coordinates {(l _j , θ _j ) of m effective stations P _j (j = n,..., N + m−1) between the end stations A and B are specified. ) | J = n, n + 1,..., N + m−1} is extracted.

As it can be seen from FIG. 5, the distance l _i from the measurement origin O _m to each measuring point P _i is the difference since discontinuous in the lateral sides of the projecting angle of the vertical plane S _1, the distance data {l _i} By extracting this discontinuous point, for example, by determining a threshold value, the end measurement points A and B can be easily extracted.

  The extraction of the end points A and B is configured such that the point cloud data is displayed on the display of the computer 3 and the operator directly designates the end points A and B by an input device such as a pointing device. Also good.

  Next, in step S3, the first initialization unit 16 determines initial values of the position coordinates of the end stations A and B in the geodetic coordinate system.

Here, the position coordinates of the end points E ₁ and E ₂ of the scanning line E ₁ E ₂ in the geodetic coordinate system are (0, 0) and (r, 0) (note that the z coordinate is the locus plane S of the optical axis. _Since they are all equal on the _locus , they are omitted below). On the other hand, the point cloud data actually output from the laser scanner 1 slightly deviated from the scanning lines E ₁ E ₂ as shown in FIG. 6 due to quantization errors caused by discretizing the measurement distance and the measurement angle. It will be a thing. Further, the end measurement points A and B are not necessarily exactly coincident with the end points E ₁ and E _{2 of the} scanning line. Therefore, as shown in FIG. 7, the initial values of the position coordinates of the end stations A and B are as follows. The end stations A and B are located on the scanning line E ₁ E ₂ and the end stations E ₁ to the end stations. is the distance from the distance and the end point E ₂ from a to the end measuring point B is set to be equal.

  Now, the coordinate values of the edge measuring points A and B in the laser scanner coordinate system

とする。光軸Ｏ_ｍＡと光軸Ｏ_ｍＢとがなす角θ’はθ_ＡＢ＝｜θ_Ａ−θ_Ｂ｜である。また、測地座標系における端測点Ａ，Ｂの位置座標の初期値をそれぞれ、

And An angle θ ′ formed by the optical axis O _m A and the optical axis O _m B is θ _AB = | θ _A −θ _B |. In addition, the initial values of the position coordinates of the end stations A and B in the geodetic coordinate system are respectively

とする。ここで、Ａ’，Ｂ’は測地座標系の初期位置に配置された端測点Ａ，Ｂを表している。端点Ｅ_１から端測点Ａ’までの距離をｄ_Ａ、端点Ｅ_２から端測点Ｂ’までの距離ｄ_Ｂと記す。いま、ｄ_Ａ＝ｄ_Ｂなので、ｘ’_Ａ＝ｒ−ｘ’_Ｂの関係が成り立つ。また、端測点Ａ’，Ｂ’は走査線Ｅ_１Ｅ_２上なので、ｙ_Ａ＝ｙ_Ｂ＝０である。端測点Ａと端測点Ｂとの間の距離をｒ_ＡＢとする。

And Here, A ′ and B ′ represent end stations A and B arranged at the initial position of the geodetic coordinate system. 'The distance to the _{d A,} from the end point _{E 2} end measuring point B' from the end point _{E 1} end measuring point A referred to the distance _{d B} to. Now, since d _A = d _B , the relationship x ′ _A = r−x ′ _B holds. Further, since the end measurement points A ′ and B ′ are on the scanning line E ₁ E ₂ , y _A = y _B = 0. The distance between the end measurement point A and the Tanhaka point B and r _AB.

According to the cosine theorem, (x ′ _A , y ′ _A ) and (x ′ _B , y ′ _B ) can be calculated by the following equations.

Next, in step S4, the second initialization unit 17 performs measurement based on the initial values (x ′ _A , y ′ _A ) and (x ′ _B , y ′ _B ) of the position coordinates of the end measurement points A and B. An initial value (x ′ _O , y ′ _O ) of position coordinates in the geodetic coordinate system of the origin O _m is calculated. Hereinafter, the point of the position specified by the initial value (x ′ _O , y ′ _O ) of the position coordinates of the measurement origin O _m is denoted as O _m ′. The measurement origin O _m ′ is one of two intersections of a circle with a radius l _A centered on the end measurement point A ′ and a circle with a radius l _B centered on the end measurement point B ′. Therefore, the following simultaneous equations hold.

Now, since the geodetic coordinate system is set as shown in FIG. 3, y _O ′ <0. Accordingly, the initial values (x ′ _O , y ′ _O ) of the position coordinates of the measurement origin O _m are obtained as follows using the equations (12a) and (12b).

Further, the second initialization unit 17 initializes the position coordinates in the geodetic coordinate system (x ′ _j , y) of the effective stations P _j (j = n + 1,..., N + m−2) other than the end stations A and B. ' _j ) is calculated. The azimuth angle of the end measuring point B = _Pn is expressed by the following equation.

計測原点Ｏ_ｍから有効測点Ｐ_ｊに向かう光軸Ｏ_ｍＰ_ｊの方位角θ_ｊは式（７）と同様に表される。従って、次式が成り立つ（図８参照）。

The azimuth angle θ _j of the optical axis O _m P _j from the measurement origin O _m toward the effective measurement point P _j is expressed in the same manner as in the equation (7). Therefore, the following equation holds (see FIG. 8).

Therefore, the initial value (x ′ _j , y ′ _j ) of the position coordinates in the geodetic coordinate system of the effective measurement point P _j (j = n + 1,..., N + m−2) is calculated by the following equation.

Next, in step S5, the conversion parameter calculation unit 18 uses the sum of squares of the distance d _A , the distance d _B , and the distance d _j (j = n + 1,..., N + m−2) as the evaluation function f, and then the measurement origin O _m Starting from the initial value of the position coordinates of each effective measurement point P _j (j = n,..., N + m−1), the measurement origin O _m and each effective measurement point are set so that the value of the evaluation function f is minimized. Each transformation parameter of Helmert transformation that performs rotation and translation of P _j is calculated by an iterative method. Here, the distance d _A is the distance from the end point E ₁ of the scanning line E ₁ E ₂ to the end measuring point A, and the distance d _B is the distance from the end point E ₂ to the end measuring point B of the scanning line E ₁ E _2. d _j is a distance from the scanning line E ₁ E ₂ to the effective measuring point P _j (see FIG. 9).

This calculation can be performed as follows. Now, the rotational points A ′, B ′, P _j ′ (j = n + 1, n + 2,..., N + m−2) having geodetic coordinates set as initial values are rotated around the origin O = (0, 0). After that, it is assumed that a parallel translation (Helmart transform) is performed. The rotation amount of the rotational movement is Δθ, and the movement amount of the parallel movement is (Δx, Δy). The points where the stations A ′, B ′, P _j ′ are moved by the Helmert transformation are denoted as A ″, B ″, P _j ″ respectively. The coordinates of the stations A ″, B ″, P _j ″ after the movement are Each is expressed as follows.

Therefore, the distance d _A , the distance d _B , and the distance d _j (j = n + 1,..., N + m−2) are expressed by the following equations.

  Therefore, the evaluation function f (Δx, Δy, Δθ) to be minimized is expressed by the following equation.

  here,

とおくと、上記評価関数ｆは次式のように書き直すことができる。

Then, the evaluation function f can be rewritten as the following equation.

  The evaluation function f is minimized by an iterative solution. Here, the steepest descent method is used. That is, by making the result of partial differentiation of the evaluation function f by Δx, Δy, a, and b to be 0, a recurrence formula for each parameter is obtained as follows. Here, η is the number of iterations.

In this embodiment, in order to simplify the calculation, the evaluation function f is the sum of squares of the distance d _A , the distance d _B , and the distance d _j (j = n + 1,..., N + m−2). The sum of these distances may be used as the function f.

Next, in step S6, the origin coordinate determination unit 19 performs Helmart transformation on the initial value (x ′ _O , y ′ _O ) of the position coordinate of the measurement origin O _m by the calculated conversion parameters Δx, Δy, a, b. Then, the final position coordinates (x _O , y _O ) in the geodetic coordinate system of the measurement origin O _m are calculated. The calculated position coordinates (x _O , y _O ) are stored in the calibration parameter storage unit 11. Here, the position coordinates (x _O , y _O ) can be calculated as follows.

Finally, in step S7, the gazing point coordinate determination unit 20 uses the relative coordinates (l _G , θ _G ) of the gazing point G, which is a measurement point on the optical axis in the center direction of the scanning angle range of the laser scanner, as point cloud data. Based on the relative coordinates (l _G , θ _G ) of the gazing point G and the final position coordinates (x _O , y _O ) in the geodetic coordinate system of the measurement origin O _m. The position coordinates (x _G , y _G ) in the coordinate system are calculated. The calculated position coordinates (x _G , y _G ) are stored in the calibration parameter storage unit 11. Here, the position coordinates (x _G , y _G ) can be calculated as follows.

Through the above processing, the position coordinates (x _O , y _O ) of the measurement origin O _m and the position coordinates (x _G , y _G ) of the gazing point G in the geodetic coordinate system are determined as calibration data. Based on the determined position coordinates, the position coordinates of other measurement data in the geodetic coordinate system can also be sequentially determined.

  FIG. 10 is an external perspective view of a target used in the laser scanner calibration method according to the second embodiment of the present invention. The configurations of the laser scanner and the calibration device used in this embodiment are the same as those shown in FIGS.

The target T _{2 in} FIG. 10 has vertical planes S ₁ and S ₂ whose left and right sides are parallel and vertical. The vertical planes S ₁ and S ₂ are parallel to each other. In this embodiment, the number of vertical planes is two in order to simplify the description, but the number of vertical planes may be three or more.

The target T ₂ is scanned horizontally by the laser scanner 1. The scanning line on the vertical plane _S 1, _{S 2,} respectively E _1,1 E _1,2, referred to as E _2,1 E _2,2. The origin O of the geodetic coordinate system and the left end point E _{1, 1} scan line E _{1, 1} E _{1, 2,} the scanning lines E _{1, 1} E _{1, 2} direction x-axis, scanning lines E _{1, 1} E _{1 , The} direction in the locus plane S _locus of the optical axis perpendicular to _{the two} directions is the y-axis, and the upper vertical direction is the z-axis. The widths of the vertical planes S ₁ and S ₂ are r ₁ and r ₂ . Further, the distance from the y axis to the left end of the vertical plane S _k (k = 1, 2) is x _{k, 0} , and the distance from the x axis to the vertical plane S _k is D _k .

  The processing flow of the laser scanner calibration method of the present embodiment is basically the same as that of the first embodiment, and will be described with reference to the flowchart of FIG.

Installation of the laser scanner 1 in step S1, scanning by the laser scanner 1, and taking in point cloud data into the computer 3 are the same as in the first embodiment. In this case, the positional relationship between each measurement point P _i (i = 1, 2,..., M) obtained by scanning the target T ₂ and the measurement origin O _m is as shown in FIG.

Next, in step S2, the effective measurement point extraction unit 13 specifies the end measurement points (A ₁ , B ₁ ) and (A ₂ , B ₂ ) at the left and right ends of the vertical planes S ₁ and S _2. , Relative coordinates {(l _j , θ _j ) | j of m ₁ effective measurement points P _j (j = n ₁ ,..., N ₁ + m ₁ −1) between end stations (A ₁ , B ₁ ) = N ₁ , n ₁ +1,..., N ₁ + m ₁ -1} and m ₂ effective measuring points P _j (j = n ₂ ,..., N ₂ ) between the end measuring points (A ₂ , B ₂ ) + M ₂ −1) relative coordinates {(l _j , θ _j ) | j = n ₂ , n ₂ +1,..., N ₂ + m ₂ −1} are extracted. The extraction method is the same as in the first embodiment.

Next, in step S3, the first initialization unit 16 determines initial values of the position coordinates of the end stations (A ₁ , B ₁ ), (A ₂ , B ₂ ) in the geodetic coordinate system. The coordinate values in the laser scanner coordinate system, the position coordinates in the geodetic coordinate system, and the length of the scanning line at each end point on the vertical planes S ₁ and S ₂ are set as in the following equation, as in the first embodiment.

_{Incidentally, [r A1B2] x} is the distance on the x-axis from the leftmost end measuring point _{A 1} to the most right end measuring point _{B 2.}

If r = x _2,0 + r ₂ , the initial position coordinates (x ′ _{k, A} , y ′ _{k, A} ), (x ′ _{k, B} , y ′ _{k, B} ) (k = 1, 2) can be calculated by the same calculation as in the first embodiment as shown in the following equation.

Here, if the distance from the end point E _1,1 to the end station A ′ ₁ is d _{1, A} and the distance d _{2, B} from the end point E _2,2 to the end station B ′ ₂ is d _{1, A.} = Position coordinates of each end point were determined so as to be d2 _{, B.}

Next, in step S4, the second initialization unit 17 calculates initial values (x ′ _O , y ′ _O ) of position coordinates in the geodetic coordinate system of the measurement origin O _m . Similar to the equations (13a) and (13b) of the first embodiment, this calculation can be performed by the following equation.

In addition, the second initialization unit 17 uses the effective measurement points P _j (j = n ₂ +1,..., N ₂ + m ₂ −2) other than the end measurement points (A ₁ , B ₁ ) and (A ₂ , B ₂ ). , N ₁ +1,..., N ₁ + m ₁ -2), the initial position coordinates (x ′ _j , y ′ _j ) in the geodetic coordinate system are calculated. Similar to the equations (16a) and (16b) of the first embodiment, this calculation can be performed by the following equation.

Next, in step S5, the conversion parameter calculation unit 18, a distance _{d 1, A, d 1,} B, d 2, A, d 2, B and the distance _{d j (j = n 2 +} 1, ..., n 2 + m _{2 -2, n 1 + 1,} ..., n 1 + m 1 the sum of squares of -2) as the evaluation function f, the measurement origin _{O m} and each active stations _{P j (j = n 2,} ..., n 2 + m 2 - 1, n ₁ ,..., N ₁ + m ₁ −1), starting from the initial position coordinates, and rotating the measurement origin O _m and each effective measurement point P _j so that the value of the evaluation function f is minimized. And each transformation parameter of the Helmat transform that performs translation is calculated by an iterative method. Here, the distance _{d k, A (k = 1,2} ) is the distance from the scanning line E _{k, 1} E _{k, 2} of the end point E _{k, 1} to the end measurement point A _k, the distance d _{k, B} scan line A distance d _{k, j} (j = n _k ,..., N _k + m _k −1) from the end point E _{k, 2 of} E _{k, 1} E _{k, 2} to the end measuring point B _k is the scanning line E _k, The distance from ₁ E _{k, 2} to the effective measuring point P _j (see FIG. 11). The distances d _{k, A} , d _{k, B} , d _{k, j} and the evaluation function f in this case are as follows.

反復法の計算も、評価関数の形が若干異なるが、実施例１と同様にして行うことができる。

Calculation of the iterative method can be performed in the same manner as in the first embodiment, although the shape of the evaluation function is slightly different.

Next, in step S6, the origin coordinate determination unit 19 performs Helmart transformation on the initial value (x ′ _O , y ′ _O ) of the position coordinate of the measurement origin O _m by the calculated conversion parameters Δx, Δy, a, b. Then, the final position coordinates (x _O , y _O ) in the geodetic coordinate system of the measurement origin O _m are calculated. The calculated position coordinates (x _O , y _O ) are stored in the calibration parameter storage unit 11.

Finally, in step S7, the gazing point coordinate determination unit 20 uses the relative coordinates (l _G , θ _G ) of the gazing point G, which is a measurement point on the optical axis in the center direction of the scanning angle range of the laser scanner, as point cloud data. Based on the relative coordinates (l _G , θ _G ) of the gazing point G and the final position coordinates (x _O , y _O ) in the geodetic coordinate system of the measurement origin O _m. The position coordinates (x _G , y _G ) in the coordinate system are calculated. The calculated position coordinates (x _G , y _G ) are stored in the calibration parameter storage unit 11.

  FIG. 12 is a block diagram illustrating a functional configuration of a laser scanner calibration apparatus according to the third embodiment of the present invention. The laser scanner used in this embodiment is the same as that shown in FIG. In FIG. 12, the same components as those in FIG. 2 are denoted by the same reference numerals and description thereof is omitted.

The calibration apparatus 10 of this embodiment includes a point cloud data storage unit 12, a width calculation unit 25, a horizontal width correction unit 26, and a height calculation unit 27. The calibration apparatus 10, when exact height H of the measurement origin O _m of the laser scanner 1 is unknown, calibration of the height H.

  Next, a laser scanner calibration method using the calibration apparatus 10 of FIG. 12 will be described.

FIG. 13 is an external perspective view of a target used in the laser scanner calibration method according to the third embodiment of the present invention. Target T ₃ is a left vertical plane S _L left side is a straight line and the right side is inclined in a vertical straight line and the left side the right side is a vertical straight line parallel to the right side of the left vertical plane S _L and a right vertical plane S _R is an inclined straight line. The left vertical plane S _L and the right vertical plane S _R are parallel and on the same plane. The right side of the left vertical plane S _{L and} the left side of the right vertical plane S _R are separated by a certain horizontal width r ₀ .

The right vertical plane S _R is a right triangle with a base r and a height h, and the left vertical plane S _L is a right triangle that is rotated by 180 ° jointly with the right vertical plane S _R. Therefore, the horizontal width _{f R,} the horizontal width _{f L} of the left vertical plane _{S L} of the right vertical plane _{S R} at a height z is expressed by the following equation.

  FIG. 14 is a flowchart of the laser scanner calibration method according to the third embodiment of the present invention.

First, in step S11, the carriage 3 the laser scanner 1 is mounted, is placed on level surface of the vertical plane S ₁ in front of the target T _1, the trajectory plane of the optical axis during the scanning with a level or the like Adjust so that the S _locus is horizontal. Then, the target T ₃ is scanned by the laser scanner 1. The laser scanner 1 performs scanning while rotating the scan head within a certain azimuth range. Therefore, the optical axis of the laser beam rotates in a horizontal plane, and distance information to the measurement point is sampled at every fixed rotation angle. The point cloud data obtained by sampling is transferred to the computer 3 and stored in the point cloud data storage unit 12.

At this time, S _locus scan line segment is cut left vertical plane S _{L E L1} E L2 (horizontal plane extends in a fan shape) the locus plane of the optical axis of the laser beam laser scanner 1 is _irradiated, is the locus surface S _locus A line segment that cuts the right vertical plane S _R is called a scanning line E _R1 E _R2 (see FIG. 13).

Next, in step S12, the width calculation unit 25 _selects the left end point A and the right end point C _{L of the} left vertical plane SL from the point cloud data stored in the point cloud data storage unit 12. extract the end measurement point a, and calculates the distance _{r L} between _{C L.} The calculation of the distance r _L can be performed in the same manner as the equation (11e).

Next, in step S13, the width calculating unit 25, from the point group data stored in the point cloud data storage unit 12, point measuring left end of the right vertical plane S _R C _R and right end measuring point B And the distance r _R between the end points C _R and B is calculated. The distance _{r R} calculations may be performed in the same manner as in the formula (11e).

Next, in step S14, the horizontal width correction unit 26 calculates the corrected horizontal widths r ′ _L and r ′ _R by correcting the measured horizontal widths r _L and r _{R according} to the following equations, respectively.

Finally, in step S15, the height calculation means 27, the height H of the measurement origin O _m of the laser scanner, was calculated by the following equation, is stored in the calibration parameter storage unit 11. Here, f _L ⁻¹ (x) and f _R ⁻¹ (x) are inverse functions of the functions f _L (z) and f _R (z), respectively.

By the above processing, it is possible to calculate the calibration parameters in the height of the measurement origin O _m of the laser scanner.

In the present embodiment, as is the gap between the left edge of the right side and a right vertical plane S _R of the left vertical plane S _L of the target T ₃ and a linear gap, shown in FIG. 15, It may be curved. In this case, when the horizontal widths of the left vertical plane S _L and the right vertical plane S _{R at} the height z are expressed by functions f _L (z) and f _R (z), f _L (z) = r−f _R (z ) And needs to be a monotonically decreasing function or a monotonically increasing function with f _R (z). When such a target T ₃ is used, the shapes of f _R (z) and f _L (z) are different from those in the case of FIG. 13, but are otherwise the same.

FIG. 16 is a flowchart of a laser scanner calibration method according to the fourth embodiment. The laser scanner 1 used in this embodiment is the same as that shown in FIG. 1, and the block diagram of the laser scanner calibration device 10 is the same as that shown in FIG. The calibration apparatus 10, similarly to Example 3, when exact height H of the measurement origin O _m of the laser scanner 1 is unknown, calibration of the height H.

First, in step S21, the carriage 3 the laser scanner 1 is mounted, is placed on level surface of the vertical plane S ₁ in front of the target T _1, the trajectory plane of the optical axis during the scanning with a level or the like Adjust so that the S _locus is horizontal. Then, by the laser scanner 1, to scan the target T ₃ as shown in FIGS. 13 and 15. Point cloud data obtained by sampling as a result of scanning is transferred to the computer 3 and stored in the point cloud data storage unit 12.

Next, if the Nth measurement is not completed in step S22, the installation position of the laser scanner 1 is moved in step S24, and the process returns to step S21 again. Here, the installation position of the laser scanner 1 is good anywhere in position and width of the target T ₃ in front of the target T ₃ is included in the track surface S _locus of the optical axis during the scanning, from the target T ₃ It is preferable to change the distance to the laser scanner 1. If the Nth measurement is completed in step S22, the process proceeds to the next step S24.

In step S <b> 24, the width calculation unit 25 sequentially reads from the first point group data from among the N sets of point group data stored in the point group data storage unit 12. Here, the read point cloud data is set as the k-th (kε {1,..., N}) point cloud data {(l _{k, i} , θ _{k, i} ) | i = 1, 2,. M}. Here, (l _{k, i} , θ _{k, i} ) is polar coordinates in the laser scanner coordinate system of the i-th measurement point P _{k, i in} the _k- th point group data.

Next, in step S25, the width calculating unit 25, the point group data _{{(l k, i, θ} k, i) | i = 1,2, ..., M} from the left of the left vertical plane _{S L} The end point A and the right end point C _L are extracted, and the distance r _{L, k} between the end points A and C _L is calculated. The calculation of the distance r _{L, k} can be performed in the same manner as in the equation (11e).

Next, in step S26, the width calculating unit 25, the point group data _{{(l k, i, θ} k, i) | i = 1,2, ..., M} from the left of the right vertical plane _{S R} end extracts measurement points C _R and right end measurement point B, and calculates the edge measuring point C _R, the distance between B r _R, a _k. The distance r _{R, k} can be calculated in the same manner as in the equation (11e).

Next, in step S27, the horizontal width correction unit 26 corrects the measured horizontal widths r _{L, k} , r _{R, k} by calculation similar to the equations (31a) and (31b), respectively, and corrects the horizontal width. r ′ _{L, k} and r ′ _{R, k} are calculated.

Finally, in step S28, the height calculation means 27, the height _{H k} of the measurement origin _{O m} of the laser scanner, is calculated by a calculation similar to that of equation (32). The calculated height H _k is stored in the calibration parameter storage unit 11.

  In step S29, if the processes of steps S24 to S28 are not completed for all the N sets of point cloud data stored in the point cloud data storage unit 12, the process returns to step S24 again. If completed, the process proceeds to step S30.

In step S30, the height calculating unit 27 reads N heights H _k (k = 1,..., N) stored in the calibration parameter storage unit 11, calculates an average value H thereof, and calculates this. to save the calibration parameter storage unit 11 as the height of the final measurement origin O _m.

By the above processing, it is possible to calculate the calibration parameters in the height of the measurement origin O _m of the laser scanner. Further, by performing the average of the N measurements can be performed more calibration of the height of the small measurement origin O _m of error.

  FIG. 17 is a block diagram illustrating a functional configuration of a laser scanner calibration apparatus according to the fifth embodiment of the present invention. The laser scanner used in this embodiment is the same as that shown in FIG. In FIG. 17, the same components as those in FIG.

  The calibration apparatus 10 of this embodiment includes a point cloud data storage unit 12, a width calculation unit 30, a height calculation unit 31, and an elevation angle calculation unit 32. The calibration device 10 calibrates the elevation angle φ when the optical axis direction of the laser beam of the laser scanner 1 is deviated from the horizontal.

  Next, a laser scanner calibration method using the calibration apparatus 10 of FIG. 12 will be described.

Figure 18 is an external perspective view of a target T ₄ used in the calibration method of the laser scanner according to Example 5 of the present invention. Target T ₄ includes a left vertical plane S _L that varies in accordance with monotonic function f _L of the horizontal width height z _(z), the right vertical plane changes the horizontal width according monotonic function f _{R (z)} of the height z S _R. In this embodiment, as an example, the left vertical plane S _L and the right vertical plane S _R and congruent right-angled triangles, right triangle right vertical plane S _R, the length of the hook is r, the length of the crotch is in h , The heel is placed along a horizontal floor, and the crotch is arranged perpendicular to the floor. The left vertical plane S _L is a right triangle obtained by rotating the right vertical plane S _R 180 °, touches the floor at the apex between the crotch and the string, the crotch is perpendicular to the floor, and the heel is horizontal to the floor. It is arranged to become. The left vertical plane S _L and the right vertical plane S _R are parallel to each other and are separated from each other by a certain distance d. Moreover, are arranged spaced apart by a predetermined horizontal width r ₀ is the right side of the left vertical plane S _K (chord) and the right vertical plane S _R of the left side (chord). Accordingly, the width f _L (z) at the height z of the left vertical plane S _L and the width f _R (z) at the height z of the left vertical plane S _R are expressed by the following equations.

  FIG. 19 is a flowchart of a laser scanner calibration method according to the fifth embodiment.

First, in step S31, the carriage 3 the laser scanner 1 is mounted, is placed on level surface of the left vertical plane S _L and the right vertical plane S _R in front of the target T _4, the scan using a spirit level or the like The locus surface S _locus of the optical axis is adjusted so as to be horizontal. Then, the target T ₄ is scanned by the laser scanner 1. The point cloud data obtained by scanning is transferred to the computer 3 and stored in the point cloud data storage unit 12.

At this time, the line locus surface S _locus of the optical axis of the laser beam laser scanner 1 is irradiated left vertical plane S scan line segment to cut _L E _{L1 E L2,} the trajectory plane S _locus off right vertical plane S _R The minute is called a scanning line E _R1 E _R2, and the elevation angle of the _locus plane S _{locus from} the horizontal plane is denoted by φ (see FIG. 18).

Next, in step S32, the width calculation unit 30 _selects the left end point A and the right end point C _{L of the} left vertical plane SL from the point cloud data stored in the point cloud data storage unit 12. extract the end measurement point a, to calculate the horizontal width _{r L} between _{C L.} The calculation of the horizontal width r _L can be performed in the same manner as in the equation (11e).

Next, in step S33, the width calculating unit 30, from the point group data stored in the point cloud data storage unit 12, point measuring left end of the right vertical plane S _R C _R and right end measuring point B And a horizontal width r _R between the end points C _R and B is calculated. The horizontal width r _R calculations may be performed in the same manner as in the formula (11e).

Next, in step S34, the height calculation unit 31 determines the height h _L and the right vertical plane of the scanning line E _L1 E _L2 that cuts the left vertical plane S _L according to the following equation based on the horizontal widths r _L and r _R. The height h _R of the scanning line E _R1 E _R2 that cuts S _R is calculated.

Finally, in step S35, the elevation angle calculating unit 32, the height _h L, _{h R,} and based on the distance d of the left vertical plane _{S L} and the right vertical plane _{S R,} the elevation angle φ of the optical axis of the racer scanner 1 It is calculated by equation (35) and stored in the calibration parameter storage unit 11.

  With the above processing, the calibration parameter of the elevation angle φ of the optical axis of the laser scanner can be calculated.

  FIG. 20 is a block diagram illustrating a functional configuration of a laser scanner calibration apparatus according to the sixth embodiment of the present invention. The laser scanner used in this embodiment is the same as that shown in FIG. Further, in FIG. 20, the same components as those in FIG.

  The calibration apparatus 10 of this embodiment includes a point cloud data storage unit 12, an effective measurement point extraction unit 41, a scanning line calculation unit 42, a scanning line distance calculation unit 43, an elevation angle calculation unit 44, and an averaging unit 45. ing. The calibration device 10 calibrates the elevation angle φ when the optical axis direction of the laser beam of the laser scanner 1 is deviated from the horizontal.

  Next, a laser scanner calibration method using the calibration apparatus 10 of FIG. 20 will be described.

Figure 21 is an external perspective view of a target T ₅ for use in the calibration method of the laser scanner according to Example 6 of the present invention. Target T ₅ includes a left side and a right side vertical and parallel left vertical plane S _L, left side and right side is a perpendicular and parallel right vertical plane S _R. The plane including the left vertical plane S _L and the plane including the right vertical plane S _R are parallel and separated from each other by a certain distance d. Target T ₅ in FIG. 21, the left vertical plane S _L and the right vertical plane S _R are the same width horizontal width, it referred to the horizontal width of both the r. Further, the left vertical plane S _L and the right vertical plane S _R are separated by r _{0 in the} horizontal distance.

  FIG. 22 is a flowchart of a laser scanner calibration method according to the sixth embodiment.

First, in step S41, the carriage 3 the laser scanner 1 is mounted, it is placed to the left vertical plane S _L and level surface of the right vertical plane S _R in front of the target T _5, the scanning using the spirit level, etc. The locus surface S _locus of the optical axis is adjusted so as to be horizontal. Then, by the laser scanner 1, to scan the target _{T 5.} The point cloud data {(l _{k, i} , θ _{k, i} ) | i = 1, 2,..., M} (k∈ {1,..., N}) obtained by scanning is transferred to the computer 3 and the points It is stored in the group data storage unit 12.

At this time, a line segment laser beam scan line a line segment locus surface S _locus of the optical axis cut left vertical plane S _L L _L the laser scanner 1 is _irradiated, trajectory plane S _locus off right vertical plane S _R referred to as a scanning line L _R, referred to as the elevation angle from the horizontal plane of the trajectory plane S _locus phi (see FIG. 21).

If the Nth measurement is not completed in step S42, the installation position of the laser scanner 1 is moved in step S43, and the process returns to step S41 again. Here, the installation position of the laser scanner 1 is good anywhere in position the lateral width of and the target T ₅ in front of the target T ₅ is included in the track surface S _locus of the optical axis during the scanning, from the target T ₅ It is preferable to change the distance to the laser scanner 1. If the Nth measurement is completed in step S42, the process proceeds to the next step S44.

In step S <b> 44, the effective point extraction unit 41 sequentially reads out the first point cloud data from the N sets of point cloud data stored in the point cloud data storage unit 12. Here, the read point cloud data is set as the k-th (kε {1,..., N}) point cloud data {(l _{k, i} , θ _{k, i} ) | i = 1, 2,. M}. Here, (l _{k, i} , θ _{k, i} ) is polar coordinates in the laser scanner coordinate system of the i-th measurement point P _{k, i in} the _k- th point group data.

Next, in step S45, the effective point extraction unit 41 selects the vertical plane S _L from the point cloud data {(l _{k, i} , θ _{k, i} ) | i = 1, 2,..., M}. end measurement point _a k of the left and right opposite ends _{of, C k,} identifies _L, and the end measurement point _{a k, C k, m k} , L -number of effective measurement point _{P k} between _L, i (i = n _{k, L, ..., n k} , L + m k, L relative polar coordinates _{{(l k} of _{-1), i, θ k,} i) | i = n k, L, n k, L + 1, ..., n k _{, L} + m _{k, L} −1}. Here, “relative polar coordinates” refers to polar coordinates in the laser scanner coordinate system. In addition, the effective stations on the left vertical plane S _L referred to as "left Enable stations".

Next, in step S46, the effective point extraction unit 41 selects the vertical plane S _R from the point cloud data {(l _{k, i} , θ _{k, i} ) | i = 1, 2,..., M}. End points C _{k, R} , B _k at the left and right ends of the image are identified, and m _{k, R} effective points P _{k, i} (i = n) between the end points C _{k, R} , B _{k k, R 1} ,..., n _{k, R} + m _{k, R} −1) relative polar coordinates {(l _{k, i} , θ _{k, i} ) | i = n _{k, R} , n _{k, R} +1 _{,. , R} + m _{k, R} −1}. Below, the effective stations on the right vertical plane S _R that "right Enable stations".

Next, in step S47, the scanning line calculation unit 42 determines the slopes of the two parallel scanning lines L _{k, L} , L _{k, R and} the parallel coordinates based on the relative coordinates of each left effective point and each right effective point. Calculate the intercept. Here, an orthogonal coordinate system is assumed in which the measurement origin O _{k, m} is the origin, the two orthogonal directions in the _locus plane S _locus are the x axis, the y axis, the x axis, and the z axis the direction perpendicular to the z axis. This is called a “relative orthogonal coordinate system”. In the relative orthogonal coordinate system _, the inclinations of the scanning lines L _{k, L} , L _{k, R} are p _k , the intercepts of the scanning lines L _{k, L} are q _{k, L} , and the intercepts of the scanning lines L _{k, R} are q _{k, R} deep. The scanning line calculation unit 42 sets the slope p _k and the intercepts q _{k, L} , q _{k, R} from the scanning lines L _{k, L} to the left effective measurement points P _{k, i} (i = n _{k, L} ,..., N). _{k, L} + m _{k, L} −1) is added to the right effective measuring points P _i (i = n _{k, R 1} ,..., n _{k, R} + m _{k, R} −) from the scanning lines L _{k, R.} It is calculated by the method of least squares so that a value obtained by adding the sum g _k (p _k , q _{k, L} , q _{k, R} ) of the distances to 1) is minimized.

If the position coordinates (hereinafter simply referred to as “position coordinates”) of the measuring points P _{k, i} in the relative orthogonal coordinate system are (x _{k, i} , y _{k, i} ),

である。また、走査線Ｌ_ｋ，Ｌはｙ＝ｐ_ｋｘ＋ｑ_ｋ，Ｌ、走査線Ｌ_ｋ，Ｒはｙ＝ｐ_ｋｘ＋ｑ_ｋ，Ｒであるので、上記距離和ｇ_ｋ（ｐ_ｋ，ｑ_ｋ，Ｌ，ｑ_ｋ，Ｒ）は次式により表される。

It is. Further, since the scanning lines L _{k and L} are y = p _k x + q _{k, L} and the scanning lines L _{k, R} are y = p _k x + q _{k, R} , the distance sum g _k (p _k , q _{k, L} , Q _{k, R} ) is represented by the following equation.

したがって、この距離和ｇ_ｋ（ｐ_ｋ，ｑ_ｋ，Ｌ，ｑ_ｋ，Ｒ）を評価関数として、最小自乗法により傾きｐ_ｋ、切片ｑ_ｋ，Ｌ，ｑ_ｋ，Ｒを計算することができる。

Accordingly, the slope p _k and the intercepts q _{k, L} , q _{k, R} can be calculated by the least square method using the distance sum g _k (p _k , q _{k, L} , q _{k, R} ) as an evaluation function. .

Next, in step S48, the inter-scan line distance calculation unit 43 calculates the distance d ′ _k from the scan lines L _{k, L} to the scan lines L _{k, R} by the following equation.

Finally, in step S49, the elevation angle calculation unit 44 calculates the elevation angle φ _k of the optical axis of the racer scanner 1 based on the distance d ′ _k and the distance d _k , and stores it in the calibration parameter storage unit 11. .

  If it is determined in step S50 that the processes in steps S44 to S49 have not been completed for all the N sets of point cloud data stored in the point cloud data storage unit 12, the process returns to step S44 again. If completed, the process proceeds to step S51.

In step S51, the averaging unit 45 reads N elevation angles φ _k (k = 1,..., N) stored in the calibration parameter storage unit 11, calculates the average value φ, and finally calculates this value φ. Is stored in the calibration parameter storage unit 11 as the elevation angle φ of the optical axis of the laser scanner 1.

  With the above processing, the calibration parameter of the elevation angle φ of the optical axis of the laser scanner 1 can be calculated. Further, by performing an average of N measurements, calibration of the elevation angle φ of the optical axis of the laser scanner 1 with fewer errors can be performed.

  In this embodiment, a calibration method when the laser scanner 1 is a three-dimensional laser scanner will be described. The functional configuration of the calibration apparatus used in the calibration method of the present embodiment is the same as that shown in FIG. FIG. 23 is an external perspective view of a target used in the laser scanner calibration method according to the seventh embodiment. The calibration target used in this embodiment has a rectangular protruding vertical plane ABCD. The vertical plane ABCD has a width r and a height s.

In this embodiment, the laser scanner 1 is a three-dimensional laser scanner. A three-dimensional laser scanner is a laser scanner that performs three-dimensional scanning by performing horizontal scanning while gradually changing the elevation angle of the optical axis. Therefore, the point cloud data obtained by the laser scanner 1 is a two-dimensionally arranged point cloud, and each measurement point is represented as P _{i, j} . Here, i is an index indicating the order of scanning in the horizontal direction (azimuth direction), and j is an index indicating the order of scanning in the vertical direction (elevation direction). Indeed data outputted from the laser scanner 1, stations P _i, the azimuth angle phi _{i, j} representing the horizontal _j, stations P _i, elevation represent the vertical _j theta _{i, j,} and the measurement origin A distance (length of the optical axis) from O _m to the _measuring point P _{i is} a set of l _{i, j} (φ _{i, j} , θ _{i, j} , l _{i, j} ). The measurement origin _Om is the origin of the optical axis output from the laser scanner.

The point cloud data storage unit 12 stores point cloud data {P _{i, j} = (φ _{i, j} , θ _{i, j} , l) of M × N measurement points measured by the three-dimensional laser scanner as described above. _{i, j} ) | i = 1,..., M, j = 1,.

First, the effective point extraction unit 13 selects the points P _A , P _B , closest to the vertices A, B, C, D at the four corners of the vertical plane ABCD from the point cloud data {P _{i, j} }. P _C, to identify the _{P D.} These measurement points P _A , P _B , P _C , and P _D are called “corner measurement points”. Further, the effective point extraction unit 13 includes m × n number of points P _{i, j} (i = m) in a region surrounded by a quadrangle having corner points P _A , P _B , P _C , and P _D as vertices. ₀ ,..., _{M 0} + m−1, j = n ₀ ,..., N ₀ + n−1) (including corner measurement points P _A , P _B , P _C , P _D ) relative coordinates {P _{i, j} = (Φ _{i, j} , θ _{i, j} , l _{i, j} ) | i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ + n−1} is extracted. These extracted stations P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ + n−1) are referred to as “effective stations”.

In FIG. 24, the corner measurement points P _A , P _B , P _C , P _D and all the effective measurement points P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,. , N ₀ + n−1).

In the calibration apparatus of the present embodiment, the geodetic coordinate system is set based on the vertices A, B, C, and D at the four corners of the vertical plane ABCD. The geodetic coordinates of the vertices A, B, C, and D are (0, 0, 0), (0, 0, s), (r, 0, s), and (r, 0, 0), respectively (FIG. 24). reference). Distance d _A between the vertex A and the Sumihaka point P _A, distance d _B between the vertex B and Sumihaka point P _B, the distance d _C between vertices C and Sumihaka point P _C , the distance between the vertex D and Sumihaka point _{P D} and _{d D.} Further, the distances from the effective measurement points P _{i, j} other than the corner measurement points P _A , P _B , P _C , P _D to the vertical plane ABCD are defined as d _{i, j} .

Next, the coordinate determination unit 14 relates the relative coordinates to the geodetic coordinates so that the sum or the square sum of the distances d _A , d _B , d _C , d _D and the distances d _{i, j} is minimized. To decide. The coordinate determination unit 14 includes a first initialization unit 16, a second initialization unit 17, a conversion parameter calculation unit 18, an origin coordinate determination unit 19, and a gazing point coordinate determination unit 20.

Next, the first initialization unit 16 sets initial values of the position coordinates of the corner measurement points P _A , P _B , P _C , and P _{D in} the geodetic coordinate system. The second initialization unit 17 is based on the initial values of the position coordinates of the corner measurement points P _A , P _B , P _C , P _D and the relative coordinates of each effective measurement point, and the corner measurement points P _A , P _B , P The initial value of the position coordinate in the geodetic coordinate system of each effective measurement point P _{i, j} other than _{C 1} , P _D and measurement origin O _m is calculated.

Next, the conversion parameter calculation unit 18 includes the effective measurement points P _{i, j} (i = m ₀ ,..., M ₀ + m) including the measurement origin O _m and the corner measurement points P _A , P _B , P _C , and P _D. −1, j = n ₀ ,..., N ₀ + n−1), starting from the initial value of the position coordinates, the measurement origin O _m and each effective measurement point P _i, each transformation parameters geometric transformation for rotating and translation of _j, is calculated by the iterative method. At this time, distances d _A , d _B , d _C , d _D , and the sum or square sum of the distances d _{i, j} are used as the evaluation function. For example, when the sum of squares is used for the evaluation function, the amount of rotation of the geometric transformation (expressed in Euler angles) is (ω, δ, κ) and the amount of translation is (Δx, Δy, Δz). The function f (ω, δ, κ, Δx, Δy, Δz) is expressed as the following equation (40).

ここで、隅測点Ｐ_Ａ，Ｐ_Ｂ，Ｐ_Ｃ，Ｐ_Ｄを幾何変換して得られる隅測点Ｐ’_Ａ，Ｐ’_Ｂ，Ｐ’_Ｃ，Ｐ’_Ｄの測地座標を、それぞれ（ｘ’_Ａ，ｙ’_Ａ，ｚ’_Ａ），（ｘ’_Ｂ，ｙ’_Ｂ，ｚ’_Ｂ），（ｘ’_Ｃ，ｙ’_Ｃ，ｚ’_Ｃ），（ｘ’_Ｄ，ｙ’_Ｄ，ｚ’_Ｄ）とし、各有効測点Ｐ_ｉ，ｊを幾何変換して得られる隅測点Ｐ’ _ｉ，ｊの測地座標を（ｘ’ _ｉ，ｊ，ｙ’ _ｉ，ｊ，ｚ’ _ｉ，ｊ）とした。幾何変換後の各有効測点Ｐ_ｉ，ｊ（ｉ＝ｍ_０，…，ｍ_０＋ｍ−１，ｊ＝ｎ_０，…，ｎ_０＋ｎ−１）の測地座標（ｘ’ _ｉ，ｊ，ｙ’ _ｉ，ｊ，ｚ’ _ｉ，ｊ）は次式（４１）で表される。

Here, the geodetic coordinates of the corner measuring points P ′ _A , P ′ _B , P ′ _C , and P ′ _D obtained by geometric transformation of the corner measuring points P _A , P _B , P _C , and P _D are respectively (x _{'A, y' A, z} 'A), (x' B, y 'B, z' B), (x 'C, y' C, z 'C), (x' D, y 'D, z ' _D ), and the geodetic coordinates of the corner stations P' _{i, j} obtained by geometric transformation of each effective station P _{i, j} are (x ' _{i, j} , y' _{i, j} , z ' _{i, j} ). Geodetic coordinates (x ′ _{i, j} , y) of each effective station P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ + n−1) after geometric transformation. ' _{i, j} , z' _{i, j} ) is expressed by the following equation (41).

Next, the origin coordinate determination unit 19 geometrically transforms the initial value of the position coordinate of the measurement origin O _m by each conversion parameter (ω, δ, κ, Δx, Δy, Δz) calculated by the conversion parameter calculation unit 18. and calculates the position coordinates in geodetic coordinate system of the measuring origin O _m.

Finally, the gazing point coordinate determination unit 20 extracts the relative coordinates of the gazing point G from the point cloud data {P _{i, j} }, and the relative coordinates of the gazing point G and the position coordinates in the geodetic coordinate system of the measurement origin O _m. Based on the above, the position coordinates of the gazing point G in the geodetic coordinate system are calculated. Here, in the case of a three-dimensional laser scanner, the gazing point G is a measuring point on the optical axis in the center direction of the scanning range (scanning range in the azimuth angle direction and scanning range in the elevation angle direction).

  Through the above calculation process, the position and direction of the three-dimensional laser scanner in the geodetic coordinate system are determined.

It is an external view of the laser scanner which concerns on Example 1 of this invention grounded to the trolley | bogie. It is a block diagram which shows the functional structure of the calibration apparatus of the laser scanner which concerns on Example 1 of this invention. It is an external appearance perspective view of the target used with the calibration method of the laser scanner which concerns on Example 1 of this invention. It is a flowchart of the calibration method of the laser scanner which concerns on Example 1 of this invention. It is a diagram showing the positional relationship between each measuring point P _i obtained by the scanning of the target T ₁ and the measurement origin O _m. It is a figure which shows the point cloud data actually output from a laser scanner. It is a figure showing the setting of the initial value of the edge measuring points A and B. Measurement origin O _m ', an end measuring point A', which is a diagram showing the positional relationship between B 'and effective measurement point P _j'. Scan lines _E 1 _{E 2} and the position and the distance _{d A} of each measuring point, the distance _{d B,} and the distance _{d j (j = n + 1} , ..., n + m-2) is a diagram illustrating a relationship. It is an external perspective view of a target T ₂ to be used in the calibration method of the laser scanner according to the second embodiment of the present invention. It is a diagram showing the positional relationship of the measurement points obtained by scanning the target T ₂ and the measurement origin O _m. It is a block diagram which shows the functional structure of the calibration apparatus of the laser scanner which concerns on Example 3 of this invention. It is an external perspective view of a target T ₃ to be used in the calibration method of the laser scanner according to a third embodiment of the present invention. It is a flowchart of the calibration method of the laser scanner which concerns on Example 3 of this invention. It is another example of the target T ₃ to be used in the calibration method of the laser scanner according to a third embodiment of the present invention. 9 is a flowchart of a laser scanner calibration method according to a fourth embodiment. It is a block diagram which shows the functional structure of the calibration apparatus of the laser scanner which concerns on Example 5 of this invention. It is an external perspective view of a target T ₄ used in the calibration method of the laser scanner according to Example 5 of the present invention. 10 is a flowchart of a laser scanner calibration method according to a fifth embodiment. It is a block diagram which shows the functional structure of the calibration apparatus of the laser scanner which concerns on Example 6 of this invention. It is an external perspective view of a target T ₅ for use in the calibration method of the laser scanner according to Example 6 of the present invention. 12 is a flowchart of a laser scanner calibration method according to a sixth embodiment. FIG. 10 is an external perspective view of a target used in a laser scanner calibration method according to a seventh embodiment. Corner measuring points P _A , P _B , P _C , P _D and all the extracted effective measuring points P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ + N-1).

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Laser scanner 2 Car 3 Computer 9 Communication interface 10 Calibration apparatus 11 Calibration parameter memory | storage part 12 Point cloud data memory | storage part 13 Effective point extraction part 14 Coordinate determination part 16 1st initialization part 17 2nd initialization part 18 Conversion Parameter calculation unit 19 Origin coordinate determination unit 20 Gaze point coordinate determination unit T ₁ Target S ₁ Vertical plane E ₁ E ₂ Scan line O _m Measurement origin S _locus Optical axis locus plane θ Scan angle G Gaze point 25 Width calculation unit 26 Horizontal Width correction unit 27 Height calculation unit 30 Width calculation unit 31 Height calculation unit 32 Elevation angle calculation unit 41 Effective point extraction unit 42 Scan line calculation unit 43 Inter-scan line distance calculation unit 44 Elevation angle calculation unit 45 Averaging unit

Claims (22)

The measurement origins for M (M is an integer) measurement points obtained by horizontally scanning a target having a vertical plane whose left and right sides are parallel and vertical with a laser scanner from the measurement origin O m in front of the vertical plane. a point group data memory means for storing point group data composed of relative coordinates relative to the O m,
The end points including the end points A and B are identified from the point cloud data by specifying end points (hereinafter referred to as “end points”) A and B on the left and right sides of the vertical plane. Effective station extraction means for extracting relative coordinates of m stations between A and B (hereinafter referred to as “effective stations”) P i (i = n,..., N + m−1);
A geodetic coordinate system is set with reference to a scanning line E 1 E 2 in which the locus of the optical axis of the laser scanner cuts the vertical plane, and the end measurement is performed from one end point E 1 of the scanning line E 1 E 2. the distance d a to the point a, the distance d B from the other end point E 2 of the scanning lines E 1 E 2 to the end stations B, and the end stations a, wherein each active stations P i other than B A relation parameter that relates the relative coordinate to the geodetic coordinate is determined so that the sum or square sum of the distances d i from (i = n + 1,..., N + m−2) to the scanning line E 1 E 2 is minimized. Coordinate determining means for
A laser scanner calibration apparatus comprising: The coordinate determining means includes
The end points A and B are located on the scanning line E 1 E 2 , and the distance from the end point E 1 to the end point A is equal to the distance from the end point E 2 to the end point B. First initialization means for calculating initial values of the position coordinates of the end stations A and B in the geodetic coordinate system;
Based on the initial values of the position coordinates of the end stations A and B and the relative coordinates of the effective stations, the effective stations P i (i = n + 1,..., N + m−) other than the end stations A and B are used. a second initialization means for calculating the initial value of the position coordinates in 2) and the geodetic coordinate system of the measurement origin O m,
Using the sum or square sum of the distance d A , the distance d B , and the distance d i (i = n + 1,..., N + m−2) as an evaluation function, the measurement origin O m and the end measurement points A and B are Starting from the initial value of the position coordinates of each of the effective measurement points P i (i = n,..., N + m−1) including the measurement origin O m and each of the evaluation functions so that the value of the evaluation function is minimized. Conversion parameter calculation means for calculating each conversion parameter of Helmert conversion that performs rotation and translation of the effective measuring point P i by an iterative method;
Origin coordinate determining means for performing a Helmart transform on the initial value of the position coordinate of the measurement origin O m using the calculated conversion parameters, and calculating the position coordinate of the measurement origin O m in the geodetic coordinate system;
The relative coordinates of the gazing point G, which is a measurement point on the optical axis in the bisecting direction of the scanning angle between the two directions of the scanning start direction and the scanning end direction of the laser scanner that rotationally scans a predetermined angle range, are described above. extracted from the point cloud data, on the basis of the position coordinates in the geodetic coordinate system and the relative coordinates of the noted viewpoints G the measurement origin O m, gazing point for calculating the position coordinates in the geodetic coordinate system of the gaze point G Coordinate determination means;
The laser scanner calibration apparatus according to claim 1, further comprising: It said point group data memory means, right and left side edges has a plurality of parallel and perpendicular vertical plane, each vertical plane are parallel to each other target, horizontally by the laser scanner from the front of the measurement origin O m of the respective vertical plane Storing point cloud data composed of relative coordinates based on the measurement origin O m with respect to M (M is an integer greater than 2) measurement points obtained by scanning;
The effective station extraction means specifies end stations A j and B j (j is a subscript for distinguishing each vertical plane) for each vertical plane, and each effective station P j, extract the relative coordinates of i ,
The coordinate determination means sets a geodetic coordinate system based on scanning lines E p, 1 E p, 2 in which the optical axis of the laser scanner cuts the vertical planes, and the scanning in the vertical planes line E j, 1 E j, the distance 2 of one end point E j, from 1 to the end stations a j d j, a, the scan lines E j, 1 E j, 2 of the other end point E j, distance from 2 to the end stations B j d j, B, and said end stations a j, wherein each active stations P j other than B j, i (i = n j + 1, ..., n j + m j −2) to the scanning lines E j, 1 E j, 2 , the relative coordinates are set so that the sum of the distances d j, i or sums of squares of all the vertical planes is minimized. The laser parameter according to claim 1, wherein a relation parameter related to the geodetic coordinates is determined. Canner calibration device. A left vertical plane that is a straight line or a curved line whose left side is vertical and whose right side is inclined; and a straight line or a curved line whose right side is vertical and whose left side is parallel to the right side of the left vertical plane. A right vertical plane, and the left vertical plane and the right vertical plane are parallel and coplanar, and the right side of the left vertical plane and the left side of the right vertical plane have a certain horizontal width. Measurement origins of M (M is an integer greater than 2) measurement points obtained by scanning a separated target with a laser scanner from measurement origins O m in front of the left vertical plane and the right vertical plane Point cloud data storage means for storing point cloud data composed of relative coordinates based on O m ;
From the point cloud data, the left end point (hereinafter referred to as “end point”) A and right end point C L of the left vertical plane, and the right end point of the right vertical plane are measured. identify the point B and the left end measuring point C R, the end stations a, horizontal width between C L r L and the end stations B, a width calculation means for calculating the horizontal width r R between C R ,
The horizontal width of the left vertical plane at height z is f L (z), the horizontal width of the right vertical plane is f R (z), and the inverse functions of f L (z) and f R (z) are f L , respectively. −1 (x), f R −1 (x), the height H of the measurement origin O m of the laser scanner is set to a height f L −1 (r ′ L ) or a height f R −1 ( any one of r ′ R ) or a height calculating means for calculating an average value of the height f L −1 (r L ) and the height f R −1 (r R );
A laser scanner calibration apparatus comprising: The horizontal width between the right side of the left vertical plane and the left side of the right vertical plane is r 0 , and the horizontal width between the left side of the left vertical plane and the right side of the right vertical plane is r + r 0 Horizontal width correction means for correcting the horizontal widths r L and r R measured at the measurement origin and calculating the corrected horizontal widths r ′ L and r ′ R by correcting the horizontal widths r L and r R according to equations (1a) and (1b), respectively. Prepared,
The height calculation means, the height H of the measurement origin O m of the laser scanner, one or heights of f L -1 (r 'L) or the height f R -1 (r' R) The laser scanner calibration apparatus according to claim 4, wherein the calibration value is calculated as an average value of f L −1 (r ′ L ) and height f R −1 (r ′ R ).

The point group data storage means includes N sets of point group data measured from N measurement origins at N positions in front of the left vertical plane and the right vertical plane and having the same height and different positions on the horizontal plane. Is memorized,
Said width calculating means, for N sets of the point group data, each of the end stations A, horizontal width between C _L and the end stations B, calculates a horizontal width between C _R, wherein each point group said end stations a for data, C the average value of the horizontal width between _L horizontal width r _L, the said end stations B for each point cloud data, the average value of the horizontal width horizontal between C _R calibration apparatus of a laser scanner according to claim 4 or 5, wherein the and calculates the width r _R. A left vertical plane whose horizontal width changes according to a monotonic function f _L (z) having a height z, and a right vertical plane whose horizontal width changes according to a monotone function f _R (z) whose height is z; The plane and the right vertical plane are parallel to each other and are separated from each other by a certain distance d, and the right side of the left vertical plane and the left side of the right vertical plane are separated by a certain horizontal width. the target, the obtained by scanning by the left vertical plane and in front of the measuring laser scanner from the origin O _m of the right vertical plane, M (M is an integer greater than 2) relative to the measurement origin O _m of measuring points Point cloud data storage means for storing point cloud data consisting of relative coordinates,
From the point cloud data, the left end point (hereinafter referred to as “end point”) A and right end point C _{L of} the left vertical plane, and the right end point of the right vertical plane are measured. identify the point B and the left end measuring point C _R, the end stations a, horizontal width between C _L r _L and the end stations B, a width calculation means for calculating the horizontal width r _R between C _R ,
When the inverse functions of the functions f _L (z) and f _R (z) are f _L ⁻¹ (x) and f _R ⁻¹ (x), respectively, the height h _L of the scanning line in the left vertical plane is f _L ⁻¹ (r _L ), a height calculating means for calculating the height h _R of the scanning line in the right vertical plane as f _L ⁻¹ (r _R );
An elevation angle calculating means for calculating an elevation angle θ of the optical axis of the racer scanner based on the height h _L , the height h _R , and the distance d, using Equation (2);
A laser scanner calibration apparatus comprising:

A left vertical plane whose left side and right side are vertical and parallel; and a right vertical plane whose left side and right side are vertical and parallel. The left vertical plane and the right vertical plane are parallel and constant with each other. A target that is on a plane separated by a distance d and separated by a certain horizontal width from the right side of the left vertical plane and the left side of the right vertical plane is positioned in front of the left vertical plane and the right vertical plane. measurement origin O _m obtained by scanning by the laser scanner from, M (M is an integer greater than 2) point group for storing point group data composed of relative coordinates relative to the measurement origin O _m of measuring points Data storage means;
From the point cloud data, the left end point (hereinafter referred to as “end point”) A and right end point C _{L of} the left vertical plane, and the right end point of the right vertical plane are measured. identify the point B and the left end measuring point C _R, the end stations a, the end stations a containing C _L, m _L-number of stations between C _L (hereinafter referred to as "left effective stations". _{) P i (i = n L} , ..., n L + m L -1) of relative coordinates, and said end stations B, said end stations B, _{m R-number} of stations between _{C R} containing _{C R} ( Hereinafter referred to as “right effective station”.) Effective station extracting means for extracting relative coordinates of P _i (i = n _R ,..., N _R + m _R −1);
Based on said relative coordinates of each left effective stations and the respective right enabled stations, two parallel slope and intercept of the scanning line L _L and L _R together from said scanning line L _L each left enabled stations P _{i (i = n L, ...} , n L + m L -1) to the sum of the distance to, from said scanning line _{L R} each right effective measurement point _{P i (i = n R,} ..., n R + m R - Scanning line calculation means for calculating by a least square method so that the sum of the distances up to 1) is minimized;
A scan line distance calculation means for calculating a distance d 'from the scanning line L _L to the scanning line L _R,
Elevation angle calculating means for calculating the elevation angle θ of the optical axis of the racer scanner based on the distance d ′ and the distance d by the equation (3);
A laser scanner calibration apparatus comprising:

The point group data storage means includes N sets of point group data measured from N measurement origins at N positions in front of the left vertical plane and the right vertical plane and having the same height and different positions on the horizontal plane. Is memorized,
The effective point extraction means extracts the relative coordinates of the left effective point and the right effective point for each point cloud data,
The scanning line calculation means calculates the slope and intercept of each of the scanning line L _L and the scanning line L _R wherein for each point cloud data,
The scanning line distance calculation means calculates a distance d 'of the from each of the scanning line L _L for each point cloud data to the scanning line L _R,
The elevation angle calculating means calculates the elevation angle θ of the optical axis of the racer scanner for each point cloud data,
9. The laser scanner calibration apparatus according to claim 8, further comprising: averaging means for outputting an average value of the elevation angle θ calculated by the elevation angle calculation means as an elevation angle of the optical axis of the racer scanner. M × N (M and N are integers) measurements obtained by three-dimensionally scanning a target having a rectangular protruding vertical plane ABCD with a laser scanner from the measurement origin O m in front of the vertical plane. Point cloud data storage means for storing point cloud data composed of relative coordinates with respect to the measurement origin O m as a reference;
Of the point cloud data, the stations closest to the vertices A, B, C, D at the four corners of the vertical plane ABCD (hereinafter referred to as “corner stations”) P A , P B , P C , P D And m in a region surrounded by a rectangle including the corner measuring points P A , P B , P C , and P D including the corner measuring points P A , P B , P C , and P D. Relative coordinates of n stations (hereinafter referred to as “effective stations”) P i, j (i = m 0 ,..., M 0 + m−1, j = n 0 ,..., N 0 + n−1) Effective station extraction means for extracting
4 corners of the vertex A of the vertical plane ABCD, B, C, and set standards geodetic coordinate system D, the distance d A between the vertex A and the corner measurement point P A, the said vertex B corner measurement point P B and the distance d B between the distance d C between the apex C and the corner measurement point P C, the distance d D between the vertex D and the corner measurement point P D, and the The sum or the square sum of the distances d i, j from each of the effective measurement points P i, j other than the corner measurement points P A , P B , P C , P D to the vertical plane ABCD is minimized. Coordinate determining means for determining a relationship parameter relating a relative coordinate to the geodetic coordinate;
A laser scanner calibration apparatus comprising: It said coordinate determining means, wherein in the geodetic coordinate system corner measurement point P A, P B, P C , a first initialization means for setting an initial value of the position coordinates of P D,
The corner measurement point P A, P B, on the basis of the P C, the initial value and the relative coordinates of each active stations coordinates of P D, the corner measurement point P A, P B, P C , other than P D A second initializing means for calculating an initial value of a position coordinate in the geodetic coordinate system of each effective measurement point P i, j (and the measurement origin O m ;
The distance d A, d B, d C , d D, and the respective distances d i, as the evaluation function the sum or square sum of j, the measurement origin O m and the corner measurement point P A, P B, P C , each valid stations including P D P i, j (i = m 0, ..., m 0 + m-1, j = n 0, ..., n 0 + n-1) starting from an initial value of the position coordinates of the Thus, transformation parameters for calculating the transformation parameters for geometric transformation for rotating and translating the measurement origin O m and the effective measurement points P i, j so that the value of the evaluation function is minimized Parameter calculation means;
Origin coordinate determining means for performing a Helmart transform on the initial value of the position coordinate of the measurement origin O m using the calculated conversion parameters, and calculating the position coordinate of the measurement origin O m in the geodetic coordinate system;
Relative coordinates of the gazing point G, which is a measuring point on the optical axis in the central direction of the scanning range of the laser scanner that rotates and scans a predetermined azimuth angle range and elevation angle range, are extracted from the point cloud data, and based on the position coordinates in the relative coordinates and the geodetic coordinate system of the measurement origin O m, and gazing point coordinate defining means for calculating the position coordinates in the geodetic coordinate system of the gaze point G,
The laser scanner calibration apparatus according to claim 10, further comprising: A calibration method for calibrating a laser scanner by a computer,
The target to which the left and right sides sides with parallel and perpendicular vertical plane, the horizontal scanning by the laser scanner from the front of the measurement origin O m of the vertical plane, the measurement origin O m for measuring points of M (M is an integer) A scanning step of capturing point cloud data consisting of relative coordinates as a reference into a computer and storing it in a storage device;
From the point cloud data stored in the storage device, the measurement points A and B (hereinafter referred to as “edge measurement points”) A and B on both the left and right sides of the vertical plane are specified. Effective station extraction for extracting the relative coordinates of m stations (hereinafter referred to as “effective stations”) P i (i = n,..., N + m−1) between the end stations A and B including B Steps,
A geodetic coordinate system is set with reference to a scanning line E 1 E 2 in which the locus of the optical axis of the laser scanner cuts the vertical plane, and the end measurement is performed from one end point E 1 of the scanning line E 1 E 2. the distance d a to the point a, the distance d B from the other end point E 2 of the scanning lines E 1 E 2 to the end stations B, and the end stations a, wherein each active stations P i other than B A relation parameter that relates the relative coordinate to the geodetic coordinate is determined so that the sum or square sum of the distances d i from (i = n + 1,..., N + m−2) to the scanning line E 1 E 2 is minimized. A coordinate determination step to perform,
A method for calibrating a laser scanner, comprising: In the coordinate determination step,
The end measuring points A and B are located on the scanning line E 1 E 2 , and the distance from the end point E 1 to the end measuring point A is equal to the distance from the end point E 2 to the end measuring point B. A first initialization step of calculating initial values of the position coordinates of the end stations A and B in the geodetic coordinate system,
Said end stations A, based on the initial value and the respective effective stations of relative coordinates of the position coordinates of B, said end stations A, the non-B each active stations P i (i = n + 1 , ..., n + m- a second initialization step of calculating the initial value of the position coordinates in 2) and the geodetic coordinate system of the measurement origin O m,
Using the sum or square sum of the distance d A , the distance d B , and the distance d i (i = n + 1,..., N + m−2) as an evaluation function, the measurement origin O m and the end measurement points A and B are Starting from the initial value of the position coordinates of each of the effective measurement points P i (i = n,..., N + m−1) including the measurement origin O m and each of the evaluation functions so that the value of the evaluation function is minimized. each transformation parameters Helmert transformation for rotating and translation of the active stations P i, a conversion parameter calculating step of calculating iteratively,
An origin coordinate determination step of performing Helmart transform on the initial value of the position coordinate of the measurement origin O m by the calculated conversion parameters and calculating the position coordinate of the measurement origin O m in the geodetic coordinate system;
The relative coordinates of the gazing point G, which is a measurement point on the optical axis in the bisecting direction of the scanning angle between the two directions of the scanning start direction and the scanning end direction of the laser scanner that rotationally scans a predetermined angle range, are described above. extracted from the point cloud data, on the basis of the position coordinates in the geodetic coordinate system and the relative coordinates of the noted viewpoints G the measurement origin O m, gazing point for calculating the position coordinates in the geodetic coordinate system of the gaze point G Coordinate confirmation step;
The laser scanner calibration method according to claim 12, wherein: In the scanning step, a plurality of the left and right both sides parallel and perpendicular vertical plane, the target each vertical plane are parallel to each other, and the horizontal scanning by the laser scanner from the front of the measurement origin O m of the respective vertical plane Point cloud data consisting of relative coordinates based on the measurement origin O m for M (M is an integer greater than 2) measurement points obtained in a computer and stored in a storage device;
In the effective station extraction step, end stations A j and B j (j is a subscript for distinguishing each vertical plane) are specified for each vertical plane, and each effective station P j is specified. , I to extract the relative coordinates,
In the coordinate determination step, a geodetic coordinate system is set with reference to scanning lines E p, 1 E p, 2 in which the optical axis of the laser scanner cuts the vertical planes. scan line E j, 1 E j, 2 for one end point E j, 1 from the distance d j to the end stations a j, a, the scan lines E j, 1 E j, 2 of the other end point E j , the distance from 2 to the end stations B j d j, B, and said end stations a j, B j other than the respective effective stations P j, i (i = n j + 1, ..., n j + m j- 2) to the scanning lines E j, 1 E j, 2 , the relative coordinates such that the sum of the distances d j, i or sums of squares of all the vertical planes is minimized. 13. A laser as claimed in claim 12, characterized in that a relational parameter relating the position to the geodetic coordinates is determined. -How to calibrate the scanner. A calibration method for calibrating a laser scanner by a computer,
A left vertical plane that is a straight line or a curved line whose left side is vertical and whose right side is inclined; and a straight line or a curved line whose right side is vertical and whose left side is parallel to the right side of the left vertical plane. A right vertical plane, and the left vertical plane and the right vertical plane are parallel and coplanar, and the right side of the left vertical plane and the left side of the right vertical plane have a certain horizontal width. the spaced apart from being a target, the scanning by the left vertical plane and in front of the measurement origin O m from the laser scanner of the right vertical plane, of the measuring points of M (M is an integer greater than 2) the measurement origin O m A scanning step of capturing point cloud data consisting of relative coordinates as a reference into a computer and storing it in a storage device;
Among the point cloud data stored in the storage device, the left end point (hereinafter referred to as “end point”) A and the right end point C L of the left vertical plane, and the right identify right end measuring point B and the left end measuring point C R of the vertical plane, the end stations a, C horizontal width between L r L and the end stations B, horizontal width between C R r R A width calculating step for calculating
The horizontal width of the left vertical plane at height z is f L (z), the horizontal width of the right vertical plane is f R (z), and the inverse functions of f L (z) and f R (z) are f L , respectively. −1 (x), f R −1 (x), the height H of the measurement origin O m of the laser scanner is set to a height f L −1 (r ′ L ) or a height f R −1 ( any one of r ′ R ) or a height calculating step for calculating as an average value of the height f L −1 (r L ) and the height f R −1 (r R );
A method for calibrating a laser scanner, comprising: After executing the width calculating step, the horizontal width between the right side of the left vertical plane and the left side of the right vertical plane is r 0 , the left side of the left vertical plane and the right side of the right vertical plane when the horizontal width between was r + r 0, wherein the measurement origin measured horizontal width r L, the r R, respectively formula (4a), the correction horizontal width r is corrected by (4b) 'L, r' Execute a horizontal width correction step to calculate R ;
Wherein the height calculation step, the height H of the measurement origin O m of the laser scanner, one or height of the height f L -1 (r 'L) or the height f R -1 (r' R) The laser scanner calibration method according to claim 15, wherein the laser scanner is calculated as an average value of the height f L −1 (r ′ L ) and the height f R −1 (r ′ R ).

In the scanning step, horizontal scanning is performed by a laser scanner from N measurement origins at N positions which are in front of the left vertical plane and the right vertical plane and have the same height and different positions on the horizontal plane, The point cloud data is captured in a computer and stored in a storage device,
In the width calculation step, the N sets of the point group data stored in the storage device, each said end stations A, horizontal width between C _L and the end stations B, and horizontal width between C _R to calculate, the said end stations a for each point group data, C _L between the average value of the horizontal width of the horizontal width r _L, the said end stations B for each point cloud data, between C _R calibration method for laser scanner according to claim 15 or 16, wherein calculating an average value of the horizontal width as the horizontal width r _R. A calibration method for calibrating a laser scanner by a computer,
A left vertical plane whose horizontal width changes according to a monotonic function f _L (z) having a height z, and a right vertical plane whose horizontal width changes according to a monotone function f _R (z) whose height is z; The plane and the right vertical plane are parallel to each other and are separated from each other by a certain distance d, and the right side of the left vertical plane and the left side of the right vertical plane are separated by a certain horizontal width. The target is scanned with a laser scanner from the measurement origin O _{m in} front of the left vertical plane and the right vertical plane, and the target is composed of relative coordinates based on the measurement origins O _m of M (M is an integer) measurement points. A scanning step of capturing group data into a computer and storing it in a storage device;
Among the point cloud data stored in the storage device, the left end point (hereinafter referred to as “end point”) A and the right end point C _{L of} the left vertical plane, and the right identify right end measuring point B and the left end measuring point C _R of the vertical plane, the end stations a, C horizontal width between _L r _L and the end stations B, horizontal width between C _R r _R A width calculating step for calculating
When the inverse functions of the functions f _L (z) and f _R (z) are f _L ⁻¹ (x) and f _R ⁻¹ (x), respectively, the height h _L of the scanning line in the left vertical plane is f _L ⁻¹ (r _L ), a height calculating step for calculating the height h _R of the scanning line in the right vertical plane as f _L ⁻¹ (r _R );
An elevation angle calculating step for calculating an elevation angle θ of the optical axis of the racer scanner based on the height h _L , the height h _R , and the distance d, according to Equation (5);
A method for calibrating a laser scanner, comprising:

A calibration method for calibrating a laser scanner by a computer,
A left vertical plane whose left side and right side are vertical and parallel; and a right vertical plane whose left side and right side are vertical and parallel. The left vertical plane and the right vertical plane are parallel and constant with each other. A target that is on a plane separated by a distance d and separated by a certain horizontal width from the right side of the left vertical plane and the left side of the right vertical plane is positioned in front of the left vertical plane and the right vertical plane. of scanning by the laser scanner from the measurement origin O _m, M (M is an integer) as a step of storing in the storage device takes in the point group data composed of relative coordinates relative to the measurement origin O _m of measuring points in the computer ,
Among the point cloud data stored in the storage device, the left end point (hereinafter referred to as “end point”) A and the right end point C _{L of} the left vertical plane, and the right identify right end measuring point B and the left end measuring point C _R of the vertical plane, the end stations a, the end stations a containing C _L, m _L-number of stations between C _L (hereinafter " left enable stations _{"hereinafter.) P i (i = n} L, ..., n L + m relative coordinates L -1), and said end stations B, said end stations B containing _{C R,} between _{C R} an effective point extraction step of extracting relative coordinates of m _R points (hereinafter referred to as “right effective points”) P _i (i = n _R ,..., n _R + m _R −1);
Based on said relative coordinates of each left effective stations and the respective right enabled stations, two parallel slope and intercept of the scanning line L _L and L _R together from said scanning line L _L each left enabled stations P _{i (i = n L, ...} , n L + m L -1) to the sum of the distance to, from said scanning line _{L R} each right effective measurement point _{P i (i = n R,} ..., n R + m R - A scanning line calculation step of calculating by a least square method so that the sum of the distances up to 1) is minimized;
A scanning line distance calculating step of calculating a distance d ′ from the scanning line L _L to the scanning line LR _;
An elevation angle calculating step of calculating an elevation angle θ of the optical axis of the racer scanner based on the distance d ′ and the distance d by Equation (6);
A calibration method for a laser scanner, comprising:

In the scanning step, a horizontal scanning is performed by a laser scanner from N measurement origins at N positions which are in front of the left vertical plane and the right vertical plane and have the same height and different positions on the horizontal plane, and N sets of the points The group data is imported into a computer and stored in a storage device.
In the effective point extraction step, the relative coordinates of the left effective point and the right effective point are extracted for each point cloud data,
Wherein in the scanning line calculating step, and calculates the slope and intercept of each of the scanning line L _L and the scanning line L _R wherein for each point cloud data,
Wherein the distance calculation step between the scan lines, and calculates the distance d 'of the from respectively each point cloud data of the scanning lines L _L to the scanning line L _R,
In the elevation angle calculating step, the elevation angle θ of the optical axis of the racer scanner is calculated for each point cloud data,
The laser scanner calibration according to claim 19, further comprising: an averaging step of calculating an average value of the elevation angle θ calculated in the elevation angle calculating step and outputting the average value as an elevation angle of the optical axis of the racer scanner. Method. A calibration method for calibrating a laser scanner by a computer,
A target having a rectangular protruding vertical plane ABCD is three-dimensionally scanned from a measurement origin O _{m in} front of the vertical plane by a laser scanner, and the above-mentioned measurement points for M × N (M and N are integers) are measured. a scanning step of storing the point cloud data consisting of relative coordinates as a reference to the storage device takes in the computer measurement origin O _m,
From among the point cloud data stored in the storage device, the four corners of the vertex A of the vertical plane ABCD, B, C, nearest stations (hereinafter referred to as "corner measuring point".) To D P _A, P _B , P _C , P _D are specified, and the corner measuring points P _A , P _B , P _C , P _D including the corner measuring points P _A , P _B , P _C , P _D M × n measuring points (hereinafter referred to as “effective measuring points”) P _{i, j} (i = m ₀ ,..., M ₀ + m−1, j = n ₀ ,..., N ₀ in the enclosed area. An effective station extraction step of extracting the relative coordinates of + n-1);
4 corners of the vertex A of the vertical plane ABCD, B, C, and set standards geodetic coordinate system D, the distance d _A between the vertex A and the corner measurement point P _A, the said vertex B corner measurement point _{P B} and the distance _{d B} between the distance _{d C} between the apex C and the corner measurement point _{P C,} the distance _{d D} between the vertex D and the corner measurement point _{P D,} and the The sum or the square sum of the distances d _{i, j} from each of the effective measurement points P _{i, j} other than the corner measurement points P _A , P _B , P _C , P _D to the vertical plane ABCD is minimized. A coordinate determination step for determining a relationship parameter relating a relative coordinate to the geodetic coordinate;
A method for calibrating a laser scanner, comprising: In the coordinate determination step, a first initialization step of setting initial values of position coordinates of the corner measurement points P _A , P _B , P _C , and P _{D in} the geodetic coordinate system;
The corner measurement point _{P A, P B,} on the basis of the P C, the initial value and the relative coordinates of each active stations coordinates of _{P D,} the corner measurement point _{P A, P B, P C} , other than _{P D} A second initialization step of calculating initial values of position coordinates in the geodetic coordinate system of each of the effective measurement points P _{i, j} and the measurement origin O _m ;
The distance _{d A, d B, d C} , d D, and the respective distances _{d i,} as the evaluation function the sum or square sum of _j, the measurement origin _{O m} and the corner measurement point _{P A, P B, P C} , each valid stations including _{P D P i, j (i} = m 0, ..., m 0 + m-1, j = n 0, ..., n 0 + n-1) starting from an initial value of the position coordinates of the Thus, transformation parameters for calculating the transformation parameters for geometric transformation for rotating and translating the measurement origin O _m and the effective measurement points P _{i, j} so that the value of the evaluation function is minimized A parameter calculation step;
An origin coordinate determination step of performing Helmart transform on the initial value of the position coordinate of the measurement origin O _m by the calculated conversion parameters and calculating the position coordinate of the measurement origin O _{m in} the geodetic coordinate system;
Relative coordinates of the gazing point G, which is a measuring point on the optical axis in the central direction of the scanning range of the laser scanner that rotates and scans a predetermined azimuth angle range and elevation angle range, are extracted from the point cloud data, and based on the position coordinates in the relative coordinates and the geodetic coordinate system of the measurement origin O _m, and the gazing point coordinate determination step of calculating the position coordinates of the geodetic coordinate system of the gaze point G,
The laser scanner calibration method according to claim 21, wherein:

Images