JP6924607B2 - Expression method and evaluation method of three-dimensional solid - Google Patents

Expression method and evaluation method of three-dimensional solid Download PDF

Info

Publication number
JP6924607B2
JP6924607B2 JP2017084753A JP2017084753A JP6924607B2 JP 6924607 B2 JP6924607 B2 JP 6924607B2 JP 2017084753 A JP2017084753 A JP 2017084753A JP 2017084753 A JP2017084753 A JP 2017084753A JP 6924607 B2 JP6924607 B2 JP 6924607B2
Authority
JP
Japan
Prior art keywords
dimensional
solid
group data
point group
dimensional solid
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
JP2017084753A
Other languages
Japanese (ja)
Other versions
JP2018181271A (en
Inventor
新一 前田
新一 前田
保 藤川
保 藤川
隆朗 新川
隆朗 新川
正巳 瀬谷
正巳 瀬谷
継彦 京免
継彦 京免
黒田 千歳
千歳 黒田
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sato Kogyo Co Ltd
Original Assignee
Sato Kogyo Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sato Kogyo Co Ltd filed Critical Sato Kogyo Co Ltd
Priority to JP2017084753A priority Critical patent/JP6924607B2/en
Publication of JP2018181271A publication Critical patent/JP2018181271A/en
Application granted granted Critical
Publication of JP6924607B2 publication Critical patent/JP6924607B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Image Analysis (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Geophysics And Detection Of Objects (AREA)
  • Image Processing (AREA)
  • Image Generation (AREA)

Description

本発明は、トンネル切羽面や法面(切り土)工事などにおける地盤面のほか、各種三次元人工物あるいは自然の立体部分などの三次元立体の表現方法及び評価方法に関する。 The present invention relates to a method for expressing and evaluating a three-dimensional solid such as various three-dimensional artificial objects or natural three-dimensional parts, in addition to the ground surface in tunnel face surface and slope (cutting soil) construction.

トンネル工事や切り土法面工事などの工事現場において、安全で経済的な施工を行うために、トンネルの切羽面や切り土法面の掘削面などの地盤面、特に岩盤が露出している面(岩盤面)については、目視観察によってその特徴を抽出し、その特徴に基づき対象とする岩盤面の安定性に関する評価を行った上で、適切な施工方法を選択している。 In order to carry out safe and economical construction at construction sites such as tunnel construction and cut soil slope construction, the ground surface such as the face surface of the tunnel and the excavation surface of the cut soil slope, especially the surface where the rock is exposed. The characteristics of (rock surface) are extracted by visual observation, and the stability of the target rock surface is evaluated based on the characteristics, and then an appropriate construction method is selected.

一般的には、経験が豊富な技術者による目視観察による岩盤面の特徴の抽出、岩盤面の写真撮影及びその保管に基づき行っている。 In general, it is carried out based on extraction of rock surface features by visual observation by an experienced technician, photography of the rock surface, and storage thereof.

しかし、人が実施する目視観察は比較的長い作業時間を必要とし、必要な作業時間が確保できない場合には適切な目視観察ができないこともある。
また、岩盤面の特徴の抽出については熟練技術者を必要とするとともに、目視観察で得られた情報における個々の担当者の技量・経験に起因するバラツキが、集積されたデータの解析、有効利用を阻害する懸念がある。
さらに、時間が経ってから見直そうとした場合、平面的な写真資料だけからでは岩盤面の特徴を十分に把握できない。
However, visual observation performed by a person requires a relatively long working time, and if the required working time cannot be secured, appropriate visual observation may not be possible.
In addition, a skilled technician is required to extract the features of the bedrock surface, and the variation due to the skill and experience of each person in charge in the information obtained by visual observation is analyzed and effectively used. There is a concern that it will hinder.
Furthermore, if you try to review it after a while, you cannot fully grasp the characteristics of the bedrock surface only from the flat photographic materials.

他方で、可能な限り評価を客観的に行おうとする試みもなされている(特許文献1)。 On the other hand, attempts have been made to make the evaluation as objective as possible (Patent Document 1).

特許第2996835号公報Japanese Patent No. 2996835

しかし、特許文献1のものは、最終的には「線画像」を求めるものであり、特に地盤の割れ目、割れ目の状態(形態)などについて評価材料を与えてくれるものではない。 However, the one of Patent Document 1 finally obtains a "line image", and does not give an evaluation material particularly regarding a crack in the ground, a state (morphology) of the crack, and the like.

そこで、本発明の主たる課題は、地盤面などの3次元立体の形態及びその特徴に関する表現・評価を的確に行い得る3次元立体の表現方法及び評価方法を提供するものである、 Therefore, a main object of the present invention is to provide a three-dimensional solid expression method and an evaluation method capable of accurately expressing and evaluating the morphology of a three-dimensional solid such as a ground surface and its features.

本発明は、3次元立体を再構成して表す表現方法のほか、その方法によって表現された情報に基づいて目的の評価を行うものである。
本発明は、3次元立体の3次元立体情報に基づき、繰返し大局的平面近似法(IPGF法、Iterative Global Plane Fitting Method)を用いて、限られた任意の枚数の平面にて3次元立体の表面形状の再構成、例えば岩盤面の表面形状の再構成を行い、その結果から、例えば岩盤面である場合、その亀裂面に関する特徴を抽出し、亀裂面の方位、見掛けの大きさ、卓越する亀裂群の有無、亀裂群における亀裂面同士の間隔等を評価しようとうするものである。
In the present invention, in addition to an expression method for reconstructing and expressing a three-dimensional solid, a target evaluation is performed based on the information expressed by the method.
The present invention uses a repeating global plane approximation method (IPGF method, Iterative Global Plane Fitting Method) based on 3D solid information of a 3D solid, and uses a limited number of planes to cover the surface of the 3D solid. Reconstruction of the shape, for example, the surface shape of the bedrock surface, and from the result, for example, in the case of the bedrock surface, the characteristics related to the crack surface are extracted, and the orientation of the crack surface, the apparent size, and the outstanding crack This is an attempt to evaluate the presence or absence of a group, the distance between crack surfaces in a crack group, and the like.

すなわち、本発明の3次元立体の表現方法は、
3次元立体の3次元の点群データを取得し、
取得した3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求め、
この3次元立体再構成情報を構成する平面の幾何学的性質に基づいて、3次元立体を表現する3次元立体の表現方法であって、
3次元の点群データを、主成分分析及びとグラフ理論を用いて、前記3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求める
ことを特徴とするものである。
That is, the method for expressing a three-dimensional solid of the present invention is
Acquire 3D point group data of 3D solid,
Obtained 3D 3D reconstruction information described by a limited number of planes so that the difference from the acquired 3D point group data is minimized in the whole.
It is an expression method of a three-dimensional solid that expresses a three-dimensional solid based on the geometrical properties of the plane that constitutes this three-dimensional solid reconstruction information.
Three-dimensional point group data is described by a limited number of planes so that the difference from the three-dimensional point group data is minimized by using principal component analysis and graph theory. It is characterized by obtaining three-dimensional reconstruction information.

3次元立体に対向配置し幅方向に離隔して配置した撮像手段により得た撮像情報により、3次元の点群データを得ることができる。 Three-dimensional point group data can be obtained from the imaging information obtained by the imaging means arranged to face the three-dimensional solid and separated in the width direction.

3次元の点群データを、三次元レーザースキャナまたはTOFカメラにより得ることもできる。 Three-dimensional point group data can also be obtained by a three-dimensional laser scanner or a TOF camera.

3次元立体の3次元の点群データの取得に際し、撮像手段を移動物体に搭載して、移動物体の移動過程で、撮像手段による撮像情報による3次元の点群データを得ることができる。 When acquiring the three-dimensional point group data of the three-dimensional solid, the imaging means can be mounted on the moving object, and the three-dimensional point group data based on the imaging information by the imaging means can be obtained in the process of moving the moving object.

再構成法として、より具体的に、3次元の点群データを、主成分分析及びとグラフ理論を用いて、前記3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求めることができる。 As a reconstruction method, more specifically, the three-dimensional point group data is limited so as to minimize the difference from the three-dimensional point group data by using principal component analysis and graph theory. It is possible to obtain the three-dimensional solid reconstruction information described by the number of planes.

他方で、本発明の3次元立体の評価方法は、
3次元立体の3次元の点群データを取得し、
取得した3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求め、
この3次元立体再構成情報を構成する平面の幾何学的性質に基づいて、3次元立体を表現するとともに、
3次元立体の形状、3次元立体の構成要素の形状、3次元立体の構成要素の配置の少なくとも一つによって3次元立体の評価を行う3次元立体の評価方法であって、
前記3次元立体が地盤面であり、
前記地盤面の評価を行う
ことを特徴とするものである。
On the other hand, the method for evaluating a three-dimensional solid of the present invention is:
Acquire 3D point group data of 3D solid,
Obtained 3D 3D reconstruction information described by a limited number of planes so that the difference from the acquired 3D point group data is minimized in the whole.
Based on the geometrical properties of the planes that make up this 3D solid reconstruction information, the 3D solid is expressed and
It is an evaluation method of a three-dimensional solid that evaluates the three-dimensional solid by at least one of the shapes of the three-dimensional solids, the shapes of the components of the three-dimensional solids, and the arrangement of the components of the three-dimensional solids.
The three-dimensional solid is the ground surface,
It is characterized by evaluating the ground surface.

本発明の地盤面の評価方法は、
3次元の点群データを取得し、
取得した3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求め、
この3次元立体再構成情報を構成する平面の幾何学的性質に基づいて、地盤面を表現するとともに、
地盤面に発達する亀裂面の情報を表現し、亀裂面の方向、大きさ、亀裂群の有無及び間隔の少なくとも一つによって地盤面の評価を行う、ことを特徴とするものである。
The method for evaluating the ground surface of the present invention is
Obtain 3D point group data and
Obtained 3D 3D reconstruction information described by a limited number of planes so that the difference from the acquired 3D point group data is minimized in the whole.
Based on the geometrical properties of the planes that make up this three-dimensional three-dimensional reconstruction information, the ground surface is expressed and at the same time.
It expresses information on the crack surface that develops on the ground surface, and evaluates the ground surface based on at least one of the direction, size, presence / absence of crack groups, and spacing of the crack surface.

より具体的に、各平面の法線ベクトルについて、類似した方位の法線ベクトルを選別することにより、地盤面において卓越する亀裂面群を抽出し、
卓越した亀裂面群の方位、見掛けの大きさ及び亀裂面同士の間隔の少なくとも一つによって地盤面の評価を行うことができる。
More specifically, by selecting normal vectors with similar orientations for the normal vectors of each plane, a group of predominant crack planes on the ground surface is extracted.
The ground surface can be evaluated by at least one of the outstanding orientation of the crack surface group, the apparent size, and the distance between the crack surfaces.

本発明によれば、3次元立体、例えば地盤面の形態及びその特徴に関する把握・評価を的確に行うことができる。 According to the present invention, it is possible to accurately grasp and evaluate a three-dimensional solid, for example, the morphology of the ground surface and its characteristics.

3次元の点群データの取得状態のトンネル縦断面図である。It is a vertical cross-sectional view of a tunnel in a state where three-dimensional point group data is acquired. トンネル切羽面を睨んだ状態で3次元の点群データの取得状態の説明図である。It is explanatory drawing of the acquisition state of the three-dimensional point group data in the state of staring at the tunnel face surface. トンネル切羽面の元データを示す撮像図である。It is an image drawing which shows the original data of a tunnel face surface. スケッチ図である。It is a sketch diagram. 5枚の平面で構成された単純化3次元立体再構成情報図である。It is a simplified three-dimensional reconstruction information diagram composed of five planes. 20枚の平面で構成された単純化3次元立体再構成情報図である。It is a simplified three-dimensional reconstruction information diagram composed of 20 planes. 本発明に係る再構成法を対比的に示す説明図である。It is explanatory drawing which shows the reconstruction method which concerns on this invention in contrast. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 異常値の発生説明図である。It is explanatory drawing of occurrence of an abnormal value. 異常値による不具合の説明図である。It is explanatory drawing of the trouble by an abnormal value. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method. 再構成法のアルゴリズム説明図である。It is an algorithm explanatory drawing of the reconstruction method.

本発明の実施形態を、図面を参照しながら以下に説明する。 Embodiments of the present invention will be described below with reference to the drawings.

実施の形態例の概要は、次のとおりである。
(1)計測器(たとえばカメラ)により、3次元立体、例えば岩盤面の3次元立体情報を取得する。
(2)取得した3次元立体情報を、後述する「再構成法」(以下、仮に「IGPF法」と称する。)を適用して、単純化した「3次元立体再構成情報」に再構成する。
(3)「3次元立体再構成情報」に基づいて、岩盤に発達する亀裂面の情報(方向、大きさ、亀裂群の有無、間隔)などの岩盤評価を行う。
The outline of the embodiment is as follows.
(1) A three-dimensional solid, for example, a three-dimensional solid information of a rock surface is acquired by a measuring instrument (for example, a camera).
(2) The acquired three-dimensional stereoscopic information is reconstructed into simplified "three-dimensional stereoscopic reconstruction information" by applying the "reconstruction method" (hereinafter, tentatively referred to as "IGPF method") described later. ..
(3) Based on the "three-dimensional three-dimensional reconstruction information", the rock mass is evaluated such as information on the crack surface (direction, size, presence / absence of crack group, interval) that develops in the rock mass.

実施の形態では、従来法では表現がきわめて困難な、岩盤に発達する亀裂について特徴を捉えて表現することが可能となる。 In the embodiment, it is possible to capture and express the characteristics of the cracks that develop in the bedrock, which are extremely difficult to express by the conventional method.

岩盤面の3次元立体情報として、トンネルの切羽面を対象とする例における3次元立体情報を取得する形態例を図1及び図2に示した。 As the three-dimensional three-dimensional information of the bedrock surface, a form example of acquiring the three-dimensional three-dimensional information in the example of targeting the face surface of the tunnel is shown in FIGS. 1 and 2.

すなわち、トンネル切羽10の手前に、架台12を有する三脚11を設置する。架台12上の中央にはカラー画像撮像器、たとえばカラーCCDカメラ13を、両側には白黒画像撮像器、たとえば白黒CCDカメラ14A、14Bを設けられる。また、架台12上にはレーザー距離計15及び水準器16も設けられている。各機器からの情報は、オンラインであるいはUSBメモリなどを介してパソコンなどに伝送される。符号17はカメラの視野を示している。 That is, a tripod 11 having a gantry 12 is installed in front of the tunnel face 10. A color image imager, for example, a color CCD camera 13, is provided in the center of the gantry 12, and a black and white image imager, for example, black and white CCD cameras 14A and 14B are provided on both sides. A laser rangefinder 15 and a spirit level 16 are also provided on the gantry 12. Information from each device is transmitted to a personal computer or the like online or via a USB memory or the like. Reference numeral 17 indicates the field of view of the camera.

計測に際しては、例えば次の手順で行う。
a 切羽10の所定の位置に三脚11にて設置する。
b 水準器16にてカメラ架台12の水平を出す。
c レーザー距離計15を使用してカメラと掘削面の距離を概略の推奨距離に設定する。
d カラーCCDカメラ13により岩質の判定用カラー画像を取得する。
e ステレオ配置の白黒CCDカメラ14A、14Bにより、視野内の人と物をよけて評価する面の白黒画像を2枚取得する。
f 取得した2枚の白黒画像の歪補正を行う。
g 補正後の2枚の白黒画像の視差量の計算を行い、その値から三次元画像(凹凸像)を作製する。
h 三次元画像(凹凸像)から平面構造による近似を行い、パソコン上に凸凹像及び平面近似画像を表示し、記録する。
i 検査員の目視による岩盤の節理の深さ量などをパソコンC上に記録し、ステップhと共に機械学習の教師データとする。
j 測定したステレオ・モノクロ画像、カラー画像、凸凹像、平面近似画像、目視による節理の高さ量をUSBメモリ等にコピーし、送付することによりデジタル・データを回収する。
For measurement, for example, the following procedure is performed.
a A tripod 11 is installed at a predetermined position on the face 10.
b Level the camera mount 12 with the spirit level 16.
c Use the laser rangefinder 15 to set the distance between the camera and the excavated surface to the approximate recommended distance.
d A color image for determining rock quality is acquired by the color CCD camera 13.
e Two black-and-white images of the surface to be evaluated by avoiding people and objects in the field of view are acquired by the black-and-white CCD cameras 14A and 14B arranged in stereo.
f Distortion correction of the two acquired black-and-white images is performed.
g The amount of parallax between the two black-and-white images after correction is calculated, and a three-dimensional image (concavo-convex image) is created from the value.
h Approximate by a planar structure from a three-dimensional image (concavo-convex image), and display and record the uneven image and the planar approximate image on a personal computer.
i Record the amount of joint depth of the bedrock visually by the inspector on the personal computer C, and use it as the teacher data for machine learning together with step h.
j Digital data is collected by copying the measured stereo / monochrome image, color image, uneven image, plane approximation image, and visual amount of knot height to a USB memory or the like and sending it.

前述のとおり、実施の形態においては、取得した3次元立体情報を、以下に例を詳述する「再構成法」(「IGPF法」)を適用して、単純化した「3次元立体再構成情報」に再構成する。 As described above, in the embodiment, the acquired three-dimensional stereoscopic information is simplified by applying the "reconstruction method" ("IGPF method") described in detail below to "three-dimensional stereoscopic reconstruction". Reconstruct into "information".

本発明者が創案した「再構成法」(「IGPF法」:Iterative Global Plane Fitting Method)について説明する。 The "reconstruction method" ("IGPF method": Iterative Global Plane Fitting Method) devised by the present inventor will be described.

かかる「再構成法」による利点の実例を示す。図3は岩盤面の詳細な3次元立体情報の表示例、図4は手書きによる岩盤面のスケッチ、図5及び図6は、図3に示した3次元立体情報について、再構成法によりそれぞれ5枚と20枚の平面で構成された単純化3次元立体再構成情報を表示したものである。
図5及び図6を参照すると、5枚平面の単純化3次元立体再構成情報であっても、岩盤に発達する亀裂を把握できる。しかしながら、20枚平面の単純化3次元立体再構成情報であると、岩盤に発達する亀裂を、より的確に把握できるようになることが分かる。
An example of the advantages of such a "reconstruction method" is shown. FIG. 3 shows an example of displaying detailed three-dimensional three-dimensional information of the bedrock surface, FIG. 4 shows a hand-drawn sketch of the bedrock surface, and FIGS. 5 and 6 show the three-dimensional three-dimensional information shown in FIG. It displays the simplified three-dimensional three-dimensional reconstruction information composed of one sheet and 20 planes.
With reference to FIGS. 5 and 6, even with the simplified three-dimensional three-dimensional reconstruction information of the five-plane plane, the cracks that develop in the bedrock can be grasped. However, it can be seen that the simplified three-dimensional three-dimensional reconstruction information of the 20-plane plane makes it possible to more accurately grasp the cracks that develop in the bedrock.

一方、「再構成法」は、大局的に再構成するものであり、図7に示すように、元データに対する、従来一般のフィッティング方式(局所的最適化)による最適化の場合に対し、本発明の再構成法では大局的な最適化によるフィッティングを行っているので、例えば山と山との間に谷をはっきり認識するのに適していることが分かる。 On the other hand, the "reconstruction method" reconstructs the original data from a broad perspective, and as shown in FIG. 7, this method is different from the case of optimization by the conventional general fitting method (local optimization) for the original data. Since the reconstruction method of the present invention performs fitting by global optimization, it can be seen that it is suitable for clearly recognizing a valley between mountains, for example.

次いで、本発明の「再構成法」アルゴリズムを、図8以降の図面を参照しながら概念的に説明する。
以下に処理の流れを記述する。
Next, the "reconstruction method" algorithm of the present invention will be conceptually described with reference to the drawings after FIG.
The processing flow is described below.

(S1) 平面枚数Nを指定する。以下の説明では簡明を目的として4枚の平面の場合の説明を行う。 (S1) The number of planes N is specified. In the following description, the case of four planes will be described for the purpose of simplicity.

(S2) 前述の手法例によって、3次元立体情報を得て、図8のように、各点の位置情報を取得する。このときの点の総数はMである。 (S2) Three-dimensional stereoscopic information is obtained by the above-mentioned method example, and the position information of each point is acquired as shown in FIG. The total number of points at this time is M.

(S3) 全M個の点の中から、N個の点V1,V2,・・・,VNを選択する。図9に示す例では、4個の点である。
この場合、N個の点の選択法として、次の2つの方法を挙げることができる。
3a. N個の点をランダムに選択する。
3b. x−y平面上で交点がN個となるようなグリッドを設定し、その各交点に最も近いx−y平面への投影点を持つ点を選択する。
(S3) N points V1, V2, ..., VN are selected from all M points. In the example shown in FIG. 9, there are four points.
In this case, the following two methods can be mentioned as a method for selecting N points.
3a. N points are randomly selected.
3b. A grid is set so that there are N intersections on the xy plane, and a point having a projection point on the xy plane closest to each intersection is selected.

(S4) V1,V2,・・・,VNの各々に対して、近傍のQ個の点を求める。そのQ点の(x,y,z)の3変数に関しての主成分分析(PCA)を行い、近似平面P1,P2,・・・,PNを求める。図11に近似平面P1を図示してある。
図10及び図11に示す例では、Q=8で実施しているが、必要に応じて変更可である。
(S4) For each of V1, V2, ..., VN, find Q points in the vicinity. Principal component analysis (PCA) is performed on the three variables (x, y, z) of the Q point, and the approximate planes P1, P2, ..., PN are obtained. The approximate plane P1 is shown in FIG.
In the examples shown in FIGS. 10 and 11, Q = 8 is used, but it can be changed as needed.

(S5) 図12に示すように、全ての点Uj(1≦j≦M)から平面Pk(1≦k≦N)に下した垂線の長さをD(Uj,Pk)とする。
各Ujに対して、D(Uj,Pk)を最小にするPkを求める。この時、図13に示すように、Ujにkというラベルを付与する。
(S5) As shown in FIG. 12, the length of the perpendicular line drawn from all the points Uj (1 ≦ j ≦ M) to the plane Pk (1 ≦ k ≦ N) is defined as D (Uj, Pk).
For each Uj, find the Pk that minimizes D (Uj, Pk). At this time, as shown in FIG. 13, Uj is labeled with k.

(S6) 各k(1≦k≦N)に対して、ラベルkの全ての点を用いた主成分分析(PCA)を行い、その結果を、図14に示すように、新しい平面Pkの傾きとオフセット位置として修正する。 (S6) For each k (1 ≦ k ≦ N), a principal component analysis (PCA) using all the points of the label k is performed, and the result is the slope of the new plane Pk as shown in FIG. And correct it as an offset position.

(S7) 上記(S5)〜(S6)をW1回繰り返す。図15を参照。
1については、(S5)〜(S6)の結果が収束したことを確認するまで繰り返す。
主成分分析(PCA)による平面近似の結果を図16に示した。
(S7) repeating the above the (S5) ~ (S6) W 1 times. See FIG.
W 1 is repeated until it is confirmed that the results of (S5) to (S6) have converged.
The result of plane approximation by principal component analysis (PCA) is shown in FIG.

(S8) (S7)の結果において、3次元立体情報の中に異常値があった場合(図17参照)、その異常値に起因する極めて狭い平面が形成されたり(図18)、平面同士の境界線の形状が不規則になったりすることがある。このようなことを避けるため(結果の平滑化)、グラフ理論を用いて以下の処理を行う。
まず、3次元立体情報(図19参照)から全M個の点のグラフ構造の情報(どの2点間に辺があるか=どの点と点が隣接しているか)を取得する。
3次元立体情報がSTLデータ(ファイルフォーマット)の場合は、グラフ構造を修正するため8a〜8cの手順を追加する。
8a. 補正係数εの値を指定する(通常、ε=0.02)。
8b. 最も離れた2点間の距離をLとする(立体モデルの大きさ)。
8c. 上記8のグラフで、頂点UiとUj(1≦i≦M, 1≦j≦M)が辺で連結されていない場合、Ui,Uj,からx−y平面に垂線を降ろし(=投影)、垂線とx−y平面との交点をそれぞれHi,Hj,とする。Hi−Hj間の距離がε×L未満ならば、Ui−Uj間に辺を追加し、グラフ構造の情報を修正する。
(S8) In the results of (S7), when there is an abnormal value in the three-dimensional stereoscopic information (see FIG. 17), an extremely narrow plane is formed due to the abnormal value (FIG. 18), or the planes are connected to each other. The shape of the boundary line may be irregular. In order to avoid such a situation (smoothing the result), the following processing is performed using graph theory.
First, information on the graph structure of all M points (which two points have an edge = which point and the point are adjacent to each other) is acquired from the three-dimensional solid information (see FIG. 19).
When the 3D stereoscopic information is STL data (file format), the procedures 8a to 8c are added to correct the graph structure.
8a. Specify the value of the correction coefficient ε (usually ε = 0.02).
8b. Let L be the distance between the two farthest points (the size of the three-dimensional model).
8c. In the graph of 8 above, when the vertices Ui and Uj (1 ≦ i ≦ M, 1 ≦ j ≦ M) are not connected by the sides, a perpendicular line is drawn from Ui, Uj, to the xy plane (= projection). Let the intersections of the perpendicular and the xy plane be Hi and Hj, respectively. If the distance between Hi and Hj is less than ε × L, an edge is added between Ui and Uj to correct the information in the graph structure.

(S9) λの値を指定する(平滑化のための重み)。 (S9) Specify the value of λ (weight for smoothing).

(S10) 頂点Uj(1≦j≦M)から平面Pk(1≦k≦N)に下した垂線の長さD(Uj,Pk)を求める。
(S11) 図21に示すように、頂点Uj(1≦j≦M)に付与したラベルがkの場合、D(Uj,Pk)をコストX(適合コスト)とする。図22に示すように、辺で連結されている(=隣接する)2点Ui,Ujのラベルが異なる場合のコストY(切断コスト)をλとする(「辺」が平面と平面の境界線で切断されたとみなす)。
これらのコストX、Yの和(適合コスト+切断コスト)を最小化するようなラベルを全ての点について求める(各点のラベルを変化させて、下記の式でコストの和を計算する)。
(S10) The length D (Uj, Pk) of the perpendicular line drawn from the apex Uj (1 ≦ j ≦ M) to the plane Pk (1 ≦ k ≦ N) is obtained.
(S11) As shown in FIG. 21, when the label given to the apex Uj (1 ≦ j ≦ M) is k, D (Uj, Pk) is set as the cost X (adaptation cost). As shown in FIG. 22, let λ be the cost Y (cutting cost) when the labels of the two points Ui and Uj connected by the sides (= adjacent) are different (“side” is the boundary line between planes). It is considered to have been disconnected by).
A label that minimizes the sum of these costs X and Y (adaptation cost + cutting cost) is obtained for all points (the label of each point is changed, and the sum of costs is calculated by the following formula).

Figure 0006924607
式1において、第1項はX、第2項はYである。Xは式1のデータ項、Yは式1の平滑化項に相当する。
ただし関数Cは、辺で連結されている2点Ui,Ujのラベルが異なる場合は1、2点Ui,Ujが辺で連結されていないかラベルが同じ場合は0を返す関数。
Figure 0006924607
In Equation 1, the first term is X and the second term is Y. X corresponds to the data term of Equation 1 and Y corresponds to the smoothing term of Equation 1.
However, the function C is a function that returns 1 when the labels of the two points Ui and Uj connected by the sides are different, or 0 when the two points Ui and Uj are not connected by the sides or the labels are the same.

(S12) 全ての点について、(S9)で指定したλを用いて(S11)の最小化計算により求めたラベルと、λ=ゼロとして(S11)の最小化計算により求めたラベルとを比較し、同じラベルを持つ点を選別する。その選別された点の中からラベルがk(1≦k≦N)の点についてそれぞれPCAを行い、平面Pkの傾きとオフセットを修正する。 (S12) For all points, compare the label obtained by the minimization calculation of (S11) using λ specified in (S9) with the label obtained by the minimization calculation of (S11) with λ = zero. , Select points with the same label. PCA is performed on each of the selected points whose label is k (1 ≦ k ≦ N), and the slope and offset of the plane Pk are corrected.

(S13) 上記(S10)〜(S12)をW2回繰り返す。
2については、(S10)〜(S12)の結果が収束したことを確認するまで繰り返す。
(S13) and repeats the above the (S10) ~ (S12) W 2 times.
W 2 is repeated until it is confirmed that the results of (S10) to (S12) have converged.

本発明は、その性質上、用途は広いものである。例えば、トンネルの切羽面のほか、トンネルの掘削面、切り土法面などの地盤面の表現及び評価に使用できる。地盤面には、モルタルやコンクリートなどの被覆材料が被覆された面も含む。
他方で、本発明の3次元立体としては、人間が加工を加えた人工的な三次元立体のほか、手を加えていない自然な3次元立体をも含む。
さらに、人工の各種構造物、建築物を対象としてもよい。
3次元立体の点群データの取得に際して、既述の実施の形態では、撮像手段(装置)を固定してあるが、必要ならばドローン、ヘリコプターなどの飛行物体、又はロボットや自動車などの人工移動物体に撮像手段を搭載し、地表の起伏の3次元形状の把握、構造物内の物体の形状又は配置、通路の認識などに利用できる。
The present invention has a wide range of uses due to its nature. For example, it can be used to express and evaluate the ground surface such as the excavated surface of the tunnel and the cut soil slope in addition to the face surface of the tunnel. The ground surface also includes a surface covered with a covering material such as mortar or concrete.
On the other hand, the three-dimensional solid of the present invention includes not only an artificial three-dimensional solid modified by humans but also a natural three-dimensional solid that has not been modified.
Further, various artificial structures and buildings may be targeted.
In the above-described embodiment, the imaging means (device) is fixed when acquiring the point cloud data of the three-dimensional solid, but if necessary, a flying object such as a drone or a helicopter, or an artificial movement such as a robot or an automobile An imaging means is mounted on an object, and it can be used for grasping the three-dimensional shape of undulations on the ground surface, recognizing the shape or arrangement of an object in a structure, recognizing a passage, and the like.

また、本発明者が創案した「再構成法」(IGPF法)の実現手段としては、コンピュータによる計算のほか、FPGA(プログラム可能で高性能な集積回路)による、リアルタイム性を保持した計算も可能である。その場合には、上記の例で、撮像手段をドローンやロボットに搭載し、リアルタイムで歩行制御や飛行制御を行うことが可能となる。 撮像手段による情報は、静止画情報のほか動画情報であってもよい。 In addition, as a means for realizing the "reconstruction method" (IGPF method) devised by the present inventor, in addition to computer-based calculations, FPGA (programmable and high-performance integrated circuits) can also be used for real-time calculations. Is. In that case, in the above example, it is possible to mount the imaging means on the drone or robot and perform walking control and flight control in real time. The information obtained by the imaging means may be moving image information as well as still image information.

10…切羽面、12…架台、13…カラーCCDカメラ、14A、14B…白黒CCDカメラ。 10 ... face surface, 12 ... mount, 13 ... color CCD camera, 14A, 14B ... black and white CCD camera.

Claims (7)

3次元立体の3次元の点群データを取得し、
取得した3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求め、
この3次元立体再構成情報を構成する平面の幾何学的性質に基づいて、3次元立体を表現する3次元立体の表現方法であって、
3次元の点群データを、主成分分析及びとグラフ理論を用いて、前記3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求める
ことを特徴とする3次元立体の表現方法。
Acquire 3D point group data of 3D solid,
Obtained 3D 3D reconstruction information described by a limited number of planes so that the difference from the acquired 3D point group data is minimized in the whole.
It is an expression method of a three-dimensional solid that expresses a three-dimensional solid based on the geometrical properties of the plane that constitutes this three-dimensional solid reconstruction information.
Three-dimensional point group data is described by a limited number of planes so that the difference from the three-dimensional point group data is minimized by using principal component analysis and graph theory. A method for expressing a three-dimensional solid, which comprises obtaining three-dimensional reconstruction information.
3次元立体に対向配置し幅方向に離隔して配置した撮像手段により得た撮像情報により、3次元の点群データを得る請求項1記載の3次元立体の表現方法。 The method for expressing a three-dimensional solid according to claim 1, wherein the three-dimensional point group data is obtained from the imaging information obtained by the imaging means arranged to face the three-dimensional solid and separated from each other in the width direction. 3次元の点群データを、三次元レーザースキャナまたはTOFカメラにより得る請求項1記載の3次元立体の表現方法。 The method for expressing a three-dimensional solid according to claim 1, wherein the three-dimensional point group data is obtained by a three-dimensional laser scanner or a TOF camera. 3次元立体の3次元の点群データの取得に際し、撮像手段を移動物体に搭載して、移動物体の移動過程で、撮像手段による撮像情報による3次元の点群データを得る請求項1記載の3次元立体の表現方法。 The first aspect of claim 1, wherein when acquiring the three-dimensional point group data of a three-dimensional solid, an imaging means is mounted on a moving object, and the three-dimensional point group data based on the imaged information obtained by the imaging means is obtained in the process of moving the moving object. How to express a three-dimensional solid. 3次元立体の3次元の点群データを取得し、
取得した3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求め、
この3次元立体再構成情報を構成する平面の幾何学的性質に基づいて、3次元立体を表現するとともに、
3次元立体の形状、3次元立体の構成要素の形状、3次元立体の構成要素の配置の少なくとも一つによって3次元立体の評価を行う3次元立体の評価方法であって、
前記3次元立体が地盤面であり、
前記地盤面の評価を行う
ことを特徴とする3次元立体の評価方法。
Acquire 3D point group data of 3D solid,
Obtained 3D 3D reconstruction information described by a limited number of planes so that the difference from the acquired 3D point group data is minimized in the whole.
Based on the geometrical properties of the planes that make up this 3D solid reconstruction information, the 3D solid is expressed and
It is an evaluation method of a three-dimensional solid that evaluates the three-dimensional solid by at least one of the shapes of the three-dimensional solids, the shapes of the components of the three-dimensional solids, and the arrangement of the components of the three-dimensional solids.
The three-dimensional solid is the ground surface,
A method for evaluating a three-dimensional solid, which comprises evaluating the ground surface.
地盤面の3次元の点群データを取得し、
取得した3次元の点群データとの差異が大局的に最少となるような、限定された枚数の平面で記述する3次元立体再構成情報を求め、
この3次元立体再構成情報を構成する平面の幾何学的性質に基づいて、地盤面を表現するとともに、
地盤面に発達する亀裂面の情報を表現し、亀裂面の方向、大きさ、亀裂群の有無及び間隔の少なくとも一つによって地盤面の評価を行う、
ことを特徴とする3次元立体の評価方法。
Acquire 3D point group data of the ground surface,
Obtained 3D 3D reconstruction information described by a limited number of planes so that the difference from the acquired 3D point group data is minimized in the whole.
Based on the geometrical properties of the planes that make up this three-dimensional three-dimensional reconstruction information, the ground surface is expressed and at the same time.
It expresses information on the crack surface that develops on the ground surface, and evaluates the ground surface based on at least one of the direction, size, presence or absence of crack groups, and spacing of the crack surface.
A method for evaluating a three-dimensional solid, which is characterized in that.
平面と点との法線ベクトルについて、類似した方位の法線ベクトルを選別することにより、地盤面において卓越する亀裂面群を抽出し、
卓越した亀裂面群の方位、見掛けの大きさ及び亀裂面同士の間隔の少なくとも一つによって地盤面の評価を行う、請求項に記載の3次元立体の評価方法。
By selecting the normal vectors with similar directions for the normal vectors of the plane and the point, the predominant crack planes on the ground surface are extracted.
The three-dimensional solid evaluation method according to claim 6 , wherein the ground surface is evaluated based on at least one of the outstanding orientation of the crack surface group, the apparent size, and the distance between the crack surfaces.
JP2017084753A 2017-04-21 2017-04-21 Expression method and evaluation method of three-dimensional solid Active JP6924607B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2017084753A JP6924607B2 (en) 2017-04-21 2017-04-21 Expression method and evaluation method of three-dimensional solid

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2017084753A JP6924607B2 (en) 2017-04-21 2017-04-21 Expression method and evaluation method of three-dimensional solid

Publications (2)

Publication Number Publication Date
JP2018181271A JP2018181271A (en) 2018-11-15
JP6924607B2 true JP6924607B2 (en) 2021-08-25

Family

ID=64276720

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2017084753A Active JP6924607B2 (en) 2017-04-21 2017-04-21 Expression method and evaluation method of three-dimensional solid

Country Status (1)

Country Link
JP (1) JP6924607B2 (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109669219A (en) * 2019-01-16 2019-04-23 武汉市工程科学技术研究院 Tunnel prediction method based on three-dimensional reconstruction
CN109816778B (en) * 2019-01-25 2023-04-07 北京百度网讯科技有限公司 Three-dimensional reconstruction method and device for material pile, electronic equipment and computer readable medium
CN110726726A (en) * 2019-10-30 2020-01-24 中南大学 Quantitative detection method and system for tunnel forming quality and defects thereof
JP7328144B2 (en) * 2019-12-27 2023-08-16 五洋建設株式会社 Face property estimation device and estimation method
CN111553008B (en) * 2020-04-23 2023-06-06 深圳市秉睦科技有限公司 Method for converting three-dimensional geological model of side slope into two-dimensional calculation section
JP2021177218A (en) 2020-05-08 2021-11-11 セイコーエプソン株式会社 Control method of image projection system, and image projection system
CN111710027B (en) * 2020-05-25 2021-05-04 南京林业大学 Tunnel three-dimensional geometric reconstruction method
JP7265269B2 (en) * 2020-07-21 2023-04-26 株式会社Zmp Map data correction method and correction program
CN112070816B (en) * 2020-07-29 2024-04-05 深圳瀚维智能医疗科技有限公司 Human back three-dimensional point cloud reconstruction method and device for massage equipment and computer readable storage medium
KR102357109B1 (en) * 2021-02-17 2022-02-08 주식회사 비엘 Tunnel surface mapping system under construction
CN113295095B (en) * 2021-07-27 2021-10-15 成都理工大学 High fill side slope geotechnical centrifugal model measurement control system
JP7550733B2 (en) 2021-08-02 2024-09-13 鹿島建設株式会社 Crack evaluation method

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4479372B2 (en) * 2004-06-25 2010-06-09 ソニー株式会社 Environmental map creation method, environmental map creation device, and mobile robot device
JP5986362B2 (en) * 2011-09-27 2016-09-06 株式会社熊谷組 Method and apparatus for extracting discontinuity of tunnel face
JP2013205130A (en) * 2012-03-28 2013-10-07 Zenrin Co Ltd Method for generating road surface shape data, device for generating road surface shape data and computer program
JP6584236B2 (en) * 2015-09-01 2019-10-02 日立造船株式会社 Method for determining edge of three-dimensional structure and method for determining outer surface of three-dimensional structure

Also Published As

Publication number Publication date
JP2018181271A (en) 2018-11-15

Similar Documents

Publication Publication Date Title
JP6924607B2 (en) Expression method and evaluation method of three-dimensional solid
Hassani Documentation of cultural heritage; techniques, potentials, and constraints
Arias et al. Digital photogrammetry, GPR and computational analysis of structural damages in a mediaeval bridge
Braun et al. Automated progress monitoring based on photogrammetric point clouds and precedence relationship graphs
KR100912715B1 (en) Method and apparatus of digital photogrammetry by integrated modeling for different types of sensors
CN109978791A (en) A kind of bridge monitoring methods merged based on oblique photograph and 3 D laser scanning
Themistocleous et al. 3D documentation and BIM modeling of cultural heritage structures using UAVS: the case of the Foinikaria church
Roncella et al. Photogrammetry for geological applications: automatic retrieval of discontinuity orientation in rock slopes
Kong et al. Preserving our heritage: A photogrammetry-based digital twin framework for monitoring deteriorations of historic structures
JP6817721B2 (en) Topographic change analysis method
CN112288848A (en) Method for calculating engineering quantity through three-dimensional modeling of unmanned aerial vehicle aerial photography
Arza-García et al. A hybrid photogrammetry approach for archaeological sites: Block alignment issues in a case study (the Roman camp of A Cidadela)
Zhang et al. Generation and comparison of TLS and SFM based 3d models of solid shapes in hydromechanic research
Vrublová et al. Documentation of landslides and inaccessible parts of a mine using an unmanned UAV system and methods of digital terrestrial photogrammetry
Torok et al. Post-disaster robotic building assessment: Automated 3D crack detection from image-based reconstructions
JP6296444B2 (en) Plotting method by point cloud image and plotting apparatus by point cloud image
Sørensen et al. Point clouds from oblique stereo-imagery: Two outcrop case studies across scales and accessibility
Sirkiä Requirements for initial data in photogrammetric recording of rock joint surfaces
Mccollum Refining the Data Collection Methodology of Outdoor Forensic Scenes Involving Scattered Human Remains Using Close-Range Photogrammetry
Gonçalves et al. 3D cliff reconstruction by drone: An in-depth analysis of the image network
Ludvigsen et al. Photogrammetric models for marine archaeology
Xu et al. Extraction of Damaged Building's Geometric Features from Multi-source point clouds
Gonçalves et al. Combining Unmanned Aerial Systems and Structure from Motion Photogrammetry To Reconstruct the Geometry of Groins
Teamaa et al. Evaluating The Approaches of the 3D Models Generation of a Soil Surface: A Review Study
Marques et al. Skewness-adjusted robust statistical assessment on googles earth 3D models: Rapplee ridge

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20200415

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20210428

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20210514

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20210708

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20210721

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20210802

R150 Certificate of patent or registration of utility model

Ref document number: 6924607

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150