[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

JP4595957B2 - 3D position and orientation detection device for circular features - Google Patents

3D position and orientation detection device for circular features Download PDF

Info

Publication number
JP4595957B2
JP4595957B2 JP2007112876A JP2007112876A JP4595957B2 JP 4595957 B2 JP4595957 B2 JP 4595957B2 JP 2007112876 A JP2007112876 A JP 2007112876A JP 2007112876 A JP2007112876 A JP 2007112876A JP 4595957 B2 JP4595957 B2 JP 4595957B2
Authority
JP
Japan
Prior art keywords
arc
circle
feature
point sequence
candidate point
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.)
Expired - Lifetime
Application number
JP2007112876A
Other languages
Japanese (ja)
Other versions
JP2007234042A (en
Inventor
伸行 藤原
清秀 阿部
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.)
Meidensha Corp
Original Assignee
Meidensha Corp
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 Meidensha Corp filed Critical Meidensha Corp
Priority to JP2007112876A priority Critical patent/JP4595957B2/en
Publication of JP2007234042A publication Critical patent/JP2007234042A/en
Application granted granted Critical
Publication of JP4595957B2 publication Critical patent/JP4595957B2/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Landscapes

  • Length Measuring Devices By Optical Means (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Description

本発明は、いわゆる物体認識処理に関するものである。特にモデルベーストマッチング法に用いる特徴を抽出する手段を提供するものである。   The present invention relates to so-called object recognition processing. In particular, the present invention provides a means for extracting features used in the model-based matching method.

閉曲線と抽出する方法として、特願平8−100974号(以降これを特許文献1と呼ぶ。)、円特徴を検出する方法として、特願平9−136452号(以降これを特許文献2と呼ぶ。)、円弧近似方法として、特願平6−113217号(以降これを特許文献3と呼ぶ。)、特願平6−113218号(以降これを特許文献4と呼ぶ。)がある。
特願平8−100974号 特願平9−136452号 特願平6−113217号 特願平6−113218号
As a method for extracting a closed curve, Japanese Patent Application No. 8-100974 (hereinafter referred to as Patent Document 1), and as a method for detecting a circle feature, Japanese Patent Application No. 9-136452 (hereinafter referred to as Patent Document 2). As an arc approximation method, there are Japanese Patent Application No. 6-113217 (hereinafter referred to as Patent Document 3) and Japanese Patent Application No. 6-113218 (hereinafter referred to as Patent Document 4).
Japanese Patent Application No.8-110084 Japanese Patent Application No. 9-136452 Japanese Patent Application No. 6-113217 Japanese Patent Application No. 6-113218

特許文献2では部品が傾いても穴や円筒の端点の円特徴の三次元位置姿勢を検出することができ、部品の様々な設置状態に対応できる。
しかしながら、円特徴検出の際の円特徴の候補として特許文献1の方法による閉曲線を用いているため、部品上の円特徴部分の表面に油や埃が付着していて抽出したエッジ線分が小さく切れてしまい、閉曲線とならない場合は円特徴を検出することができない。
本発明の目的は、工業部品の「穴」や「円筒部分の端点」といった円特徴の三次元位置姿勢を検出することである。
In Patent Document 2, the three-dimensional position and orientation of the circular feature of the end point of a hole or cylinder can be detected even if the component is inclined, and it is possible to cope with various installation states of the component.
However, since the closed curve according to the method of Patent Document 1 is used as a circle feature candidate when detecting a circle feature, oil and dust are attached to the surface of the circle feature portion on the part, and the extracted edge line segment is small. A circle feature cannot be detected if it is cut and does not become a closed curve.
An object of the present invention is to detect a three-dimensional position and orientation of a circular feature such as a “hole” or “end point of a cylindrical portion” of an industrial part.

斯かる目的を達成する本発明の請求項1に係る円特徴の三次元位置姿勢検出装置は、検出すべき対象部品の穴がカメラの光軸に対してほぼ傾かない設置状態に限定して前記カメラで前記対象部品を撮影し、撮影された前記対象部品の画像からエッジ線分を円の一部である円弧特徴として抽出する際、設定した半径範囲に入るもののみを円弧特徴とする円弧抽出部と、前記円弧抽出部により抽出された円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作る円弧グループ設定部と、前記円弧グループ設定部により作られた円弧グループのそれぞれの円弧の円周角の総計が設定した値以上であることを確認することにより積算角度チェックを行う積算角度検査部と、前記積算角度検査部により積算角度チェック後の円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を基に円近似を行い、その近似円の円周上の点の位置を計算し、近似円の円周として求められた点列を円候補点列として設定する円弧グループ近似円円候補点列設定部と、前記円弧グループ近似円円候補点列設定部により設定された円候補点列に基づいて円特徴を計算する円特徴データ計算部とを有することを特徴とする。 The three-dimensional position / orientation detection device for a circular feature according to claim 1 of the present invention that achieves such an object is limited to an installation state in which the hole of the target component to be detected is not substantially inclined with respect to the optical axis of the camera. When the target part is photographed with a camera and an edge line segment is extracted from the photographed image of the target part as an arc feature that is a part of a circle, only those that fall within a set radius range are extracted with an arc feature. And an arc group setting unit for creating an arc group with a center position in the vicinity of the arc extracted by the arc extracting unit and having a short radius, and the arc group setting unit. An integrated angle inspection unit that performs an integrated angle check by confirming that the total circumferential angle of each arc in the arc group is greater than or equal to a set value, and an integrated angle check by the integrated angle inspection unit Approximate a circle based on a set of point sequences that are the origin of each arc in each arc group, calculate the position of the point on the circumference of the approximate circle, and An arc group approximate circle / circle candidate point sequence setting unit that sets the point sequence obtained as a circle candidate point sequence and a circle feature based on the circle candidate point sequence set by the arc group approximate circle / circle candidate point sequence setting unit And a circle feature data calculation unit for calculating .

以上、詳細に説明したように本発明によれば、以下の効果を奏する。
(1)入力画像において、部品表面に油や埃が付着している場合でも、「穴」や「円筒の端点」といった円となる部分を抽出することができる。
(2)入力画像において、部品表面に油や埃が付着している場合でも、部品上の「穴」や「円筒の端点」といった円特徴の位置と姿勢を検出することができる。
(3)入力画像において、部品表面に油や埃が付着している場合でも、部品上の「穴」や「円筒の端点」といった円特徴を用いて、部品の認識を行うことができる。
(4)入力画像において、部品表面に油や埃が付着している場合でも、部品上の「穴」や「円筒の端点」といった円特徴を用いて、認識の間違いを減らすことができる。
(5)円特徴を抽出することで、対象部品が三次元的にどのような位置にあるかを検出することができる。
As described above, according to the present invention, the following effects can be obtained.
(1) In the input image, even when oil or dust adheres to the component surface, it is possible to extract a circular portion such as “hole” or “cylinder end point”.
(2) In the input image, even when oil or dust adheres to the surface of the component, the position and orientation of circular features such as “holes” and “cylindrical end points” on the component can be detected.
(3) In the input image, even when oil or dust adheres to the surface of the part, the part can be recognized using circular features such as “hole” and “end of cylinder” on the part.
(4) In the input image, even when oil or dust adheres to the surface of the component, recognition errors can be reduced by using circular features such as “holes” and “cylindrical end points” on the component.
(5) By extracting the circle feature, it is possible to detect what position the target part is three-dimensionally.

(1)基本的な考え方
本発明では、検出すべき対象部品の穴がカメラの光軸に対してほぼ傾かない設置状態に限定し、部品上に付着した油や埃のために抽出したエッジ線分が小さく切れてしまい閉曲線とならない場合でも、エッジ線分を円弧近似することで画面上で円弧を抽出し、特許文献2の閉曲線の代わりに円弧を基に円候補点列を設定し、最終的に円特徴を検出する。
即ち、特許文献2では、検出すべき対象物品がカメラの光軸に対して傾いた場合も想定していため、対象物品の穴の画像が閉曲線或いは閉曲線と見なしうるものでなければ、円特徴の三次元的な姿勢を検出することができなかった。
これに対し、本発明では、検出すべき対象部品の穴がカメラの光軸に対してほぼ傾かない設置状態にあることを前提とするので、カメラにより撮影された対象部品の穴の画像は、エッジ線分が小さく切れた場合でも、円の一部である円弧であるから、閉曲線か否かに係わらず、円特徴を抽出することができる。
円弧近似としては、特許文献3,4に記載されるように、画像データ中にある点列を円または円弧に近似する際、対象とする点列の2つの端点と、他の1点の合計3点を通る円弧を仮定し、その点列の更に他の1点が仮定した円弧の円周上にあれば、点列の全ての要素の点が仮定した円弧の円周上にあると判定し、対象となる点列を仮定した円弧として認識する。
(1) Basic concept In the present invention, the edge line extracted for oil or dust adhering to the part is limited to an installation state in which the hole of the target part to be detected is not substantially inclined with respect to the optical axis of the camera. Even if the minute cuts out and does not become a closed curve, an arc is extracted on the screen by approximating the edge line with an arc, and a circle candidate point sequence is set based on the arc instead of the closed curve of Patent Document 2, To detect circular features.
That is, in Patent Document 2, it is assumed that the target article to be detected is inclined with respect to the optical axis of the camera. Therefore, if the image of the hole of the target article cannot be regarded as a closed curve or a closed curve, The 3D posture could not be detected.
On the other hand, in the present invention, since it is assumed that the hole of the target part to be detected is in an installation state that is not substantially inclined with respect to the optical axis of the camera, the image of the hole of the target part photographed by the camera is Even when the edge line segment is cut into small pieces, it is an arc that is a part of a circle, so that it is possible to extract a circle feature regardless of whether it is a closed curve or not.
As arc approximation, as described in Patent Documents 3 and 4, when approximating a point sequence in image data to a circle or arc, the sum of the two end points of the target point sequence and the other one point Assuming an arc passing through 3 points, and if another point in the sequence is on the assumed arc circumference, it is determined that the points of all elements in the sequence are on the assumed arc circumference Then, the target point sequence is recognized as an assumed circular arc.

(2)円弧半径範囲の設定
対象部品がカメラの視野内に設置される範囲(奥行き)Lが分かっている場合は、検出する円特徴の実際の半径から、図1のように相似関係を用いて画面上で抽出する円弧半径の範囲、即ち、最大円弧半径DMAX及び最小円弧半径DMINを設定することが出来る。
円弧特徴を抽出する際、このようにして設定した半径範囲(許容誤差)に入るもののみを円弧特徴とする。
(2) Setting of arc radius range When the range (depth) L in which the target part is installed in the field of view of the camera is known, the similarity relationship is used as shown in FIG. 1 from the actual radius of the detected circular feature. The range of the arc radius to be extracted on the screen, that is, the maximum arc radius DMAX and the minimum arc radius DMIN can be set.
When extracting the arc feature, only those within the radius range (allowable error) set in this way are set as the arc feature.

(3)円候補点列の設定
本発明では円候補点列について特許文献2における閉曲線の代わりに、円弧特徴を用いて円候補点列を設定する。
具体的には、(3.1)〜(3.6)に示す方法により行う。
(3.1)抽出円弧を円候補点列とする方法
抽出した円弧の元となったエッジ線分の点列を円候補点列として設定する。
(3.2)抽出円弧の仮想円上の点列を円候補点列とする方法
抽出した円弧の中心と半径と同じデータをもつ仮想的な円を考え、その仮想円の円周上の点の位置を計算し、仮想円の円周として求められた点列を円候補点列として設定する。
(3.3)円弧グループを円候補点列とする方法
抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作り、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を円候補点列とする。
(3.4)円弧グループの近似円上の点列を円候補点列とする方法
抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作り、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を基に円近似を行い、その近似円の円周上の点の位置を計算し、近似円の円周として求められた点列を円候補点列として設定する。
(3.5)積算角度チェックを行い円弧グループを円候補点列とする方法
抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作り、それぞれの円弧の円周角の総計が設定した値以上である場合、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を円候補点列とする。
(3.6)積算角度チェックを行い円弧グループの近似円上の点列を円候補点列とする方法
抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作り、それぞれの円弧の円周角の総計が設定した値以上である場合、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を基に円近似を行い、その近似円の円周上の点の位置を計算し、近似円の円周として求められた点列を円候補点列として設定する。
(3) Setting of Circle Candidate Point Sequence In the present invention, a circle candidate point sequence is set using a circular arc feature instead of the closed curve in Patent Document 2 for the circle candidate point sequence.
Specifically, it is carried out by the methods shown in (3.1) to (3.6).
(3.1) Method of Using Extracted Arc as Circle Candidate Point Sequence The point sequence of the edge line segment that is the basis of the extracted arc is set as the circle candidate point sequence.
(3.2) Method of using a point sequence on the virtual circle of the extracted arc as a circle candidate point sequence Considering a virtual circle having the same data as the center and radius of the extracted arc, the points on the circumference of the virtual circle And the point sequence obtained as the circumference of the virtual circle is set as a circle candidate point sequence.
(3.3) Method of using arc group as circle candidate point sequence An arc group is created from the extracted arcs whose center position is near and whose radius is close, and the origin of each arc of the arc group A set of point sequences of the plurality of edge line segments that become the circle candidate point sequence is used.
(3.4) Method of using point sequence on approximate circle of arc group as circle candidate point sequence An arc group is created from the extracted arcs whose center position is near and whose radius is close, and arc Approximate a circle based on the set of point sequences that are the origin of each arc of the group, calculate the position of the point on the circumference of the approximate circle, and find the circumference of the approximate circle The obtained point sequence is set as a circle candidate point sequence.
(3.5) Method of checking accumulated angle and making arc group as circle candidate point sequence An arc group is created from the extracted arcs whose center position is near and whose radius is close. When the sum of the circumferential angles of the circles is equal to or greater than the set value, a set of point sequences of a plurality of edge line segments that are the origins of the respective arcs of the arc group is set as a circle candidate point sequence.
(3.6) Checking the accumulated angle and using the point sequence on the approximate circle of the arc group as the circle candidate point sequence The arc is determined based on the extracted arc whose center position is near and whose radius is close. When a group is created and the total circumferential angle of each arc is greater than or equal to the set value, a circle approximation is performed based on the set of point sequences of the edge line segments that are the origin of each arc in the arc group. Then, the position of the point on the circumference of the approximate circle is calculated, and the point sequence obtained as the circumference of the approximate circle is set as the circle candidate point sequence.

(4)円特徴データの計算
点列の三次元位置計測、三次元円計算、円特徴確認等の円特徴データ計算処理は、特許文献2の図17、図24及び図25に示す装置の「点列データ抽出部」以降の処理を用いることとする。
即ち、図14に示すように、入力画像から得たエッジデータのうち円になりそうな点列部分を抽出し、その点列の三次元位置データを基に計算される平面上で点列データに関して円近傍を行い、円特徴の三次元位置姿勢を検出するものである。
また、図15に示すように、入力画像から得たエッジデータのうち円になりそうな点列部分を抽出し、その点列の三次元位置計測において、計測しようとする点の前後数点から作られる直線とエピポーララインのなす角度を求め、その角度と設定した値とを比較することで、その点を三次元位置計測するか否かを判断し、その判断を経て計測した点列の三次元位置データを基に計算される平面上で点列データに関して円近似を行い、円特徴の三次元位置姿勢を検出するものである。
更に、図16に示すように、入力画像から得たエッジデータのうち円になりそうな点列部分を抽出し、その点列の三次元位置データを基に計算される平面上で点列データに関して円近傍を行い、抽出した円と同心円上の近傍の円弧を抽出し、円と近傍の円弧の点列の三次元位置データから再度円平面を計算し、その円平面上で円を構成していた点列データに関して再度円近似を行い、円特徴の三次元位置姿勢を検出するものである。
(4) Calculation of circle feature data Circle feature data calculation processing such as three-dimensional position measurement of point sequences, three-dimensional circle calculation, circle feature confirmation, etc. is performed by “ The processing after “point sequence data extraction unit” will be used.
That is, as shown in FIG. 14, a point sequence portion that is likely to be a circle is extracted from the edge data obtained from the input image, and the point sequence data is calculated on the plane calculated based on the three-dimensional position data of the point sequence. 3 is performed to detect the three-dimensional position and orientation of the circle feature.
In addition, as shown in FIG. 15, a point sequence portion that is likely to be a circle is extracted from the edge data obtained from the input image, and in the three-dimensional position measurement of the point sequence, from several points before and after the point to be measured. The angle between the straight line to be created and the epipolar line is obtained, and the angle is compared with the set value to determine whether or not to measure the point in 3D. A circle approximation is performed on the point sequence data on a plane calculated based on the original position data, and the three-dimensional position and orientation of the circle feature is detected.
Further, as shown in FIG. 16, a point sequence portion that is likely to be a circle is extracted from the edge data obtained from the input image, and the point sequence data is calculated on the plane calculated based on the three-dimensional position data of the point sequence. The circle near the circle is extracted, arcs near the extracted circle and concentric circles are extracted, the circle plane is calculated again from the three-dimensional position data of the point sequence of the circle and the nearby arc, and the circle is constructed on the circle plane. The circle approximation is performed again with respect to the point sequence data, and the three-dimensional position and orientation of the circle feature is detected.

(5)円特徴の傾き検査及び補正
円候補点列を方法(3.1),(3.2)を用いて設定した場合、円特徴データ計算のデータとなる元のデータ量が少ないため、検出した円特徴の傾きが実際とは大きく異なる場合がある。
そこで、これらに関しては傾き角度の検査及び補正を行う。
(5.1)円特徴の傾き角度検査
検出した円特徴の傾き角度と予め設定して置いた傾き角度の許容値を比較し、円特徴の傾き角度が許容値以内である場合は「正しい」と、それ以外の場合は「正しくない」と判断する。
(5.2)円特徴の傾き角度補正方法
検出した円特徴の放線ベクトルを鉛直上向きに設定し、円特徴の姿勢データを再計算する。
[参考例1
(5) Inclination inspection and correction of circle feature When a circle candidate point sequence is set using the methods (3.1) and (3.2), the amount of original data that becomes the data of circle feature data calculation is small. In some cases, the inclination of the detected circle feature is significantly different from the actual one.
In view of this, the inclination angle is inspected and corrected.
(5.1) Inclination angle inspection of circle feature The inclination angle of the detected circle feature is compared with an allowable value of a preset inclination angle. If the inclination angle of the circle feature is within the allowable value, “correct” Otherwise, it is determined as “incorrect”.
(5.2) Method of correcting inclination angle of circle feature The ray vector of the detected circle feature is set vertically upward, and the posture data of the circle feature is recalculated.
[Reference Example 1 ]

(6.1)本発明の参考例1に係る円特徴の三次元位置姿勢を検出する装置1の例を図2に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、抽出された円弧特徴を基に(3.1)による「抽出円弧を円候補点列とする方法」を用いて、円になりそうな点列部分を設定する。
即ち、本装置は、図2に示すように、円弧抽出部110及び抽出円弧円候補点列設定部120よりなる点列データ抽出部100を備える。
円弧抽出部110は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧円候補点列設定部120は、抽出した円弧の元となったエッジ線分の点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
[参考例2
(6.1) FIG. 2 shows an example of a device 1 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 1 of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extracting arc features that fall within the set radius range, and using the extracted arc features as the method of making the extracted arc as a circle candidate point sequence according to (3.1), a point sequence that is likely to be a circle Set the part.
That is, the apparatus includes a point sequence data extraction unit 100 including an arc extraction unit 110 and an extracted arc circle candidate point sequence setting unit 120, as shown in FIG.
When extracting the arc feature, the arc extracting unit 110 sets only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc circle candidate point sequence setting unit 120 sets the point sequence of the edge line that is the origin of the extracted arc as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
[Reference Example 2 ]

(6.2)本発明の参考例2に係る円特徴の三次元位置姿勢を検出する装置2の例を図3に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円瓜特徴を抽出し、その円弧特徴を基に(3.2)による「抽出円弧の仮想円上の点列を円候補点列とする方法」を用いて、円になりそうな点列部分を設定する。
即ち、本装置は、図3に示すように、円弧抽出部210及び抽出円弧仮想円円候補点列設定部220よりなる点列データ抽出部200を備える。
円弧抽出部210は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧仮想円円候補点列設定部220は、抽出した円弧の中心と半径と同じデータをもつ仮想的な円を考え、その仮想円の円周上の点の位置を計算し、仮想円の円周として求められた点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
[実施例1]
(6.2) FIG. 3 shows an example of the device 2 for detecting the three-dimensional position and orientation of the circular feature according to Reference Example 2 of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Using the “Method of making the point sequence on the virtual circle of the extracted arc a circle candidate point sequence” according to (3.2) based on the arc feature extracted from the circle radius feature that falls within the set radius range, Set the point sequence that is likely to become.
That is, the apparatus includes a point sequence data extraction unit 200 including an arc extraction unit 210 and an extracted arc virtual circle circle candidate point sequence setting unit 220 as shown in FIG.
When extracting the arc feature, the arc extracting unit 210 sets only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc virtual circle circle candidate point sequence setting unit 220 considers a virtual circle having the same data as the center and radius of the extracted arc, calculates the position of the point on the circumference of the virtual circle, A point sequence obtained as a circumference is set as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
[ Example 1]

(6.3)本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置3の例を図4に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.3)による「円弧グループを円候補点列とする方式」を用いて、円になりそうな点列部分を設定する。
即ち、本装置は、図4に示すように、円弧抽出部310、円弧グループ設定部320及び円弧グループ円候補点列設定部330よりなる点列データ抽出部300を備える。
円弧抽出部310は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
円弧グループ設定部320は、抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作る。
円弧グループ円候補点列設定部330は、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を円候補点列とする。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
[実施例2]
(6.3) FIG. 4 shows an example of the device 3 for detecting the three-dimensional position and orientation of the circular feature according to the first embodiment of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extract arc features that fall within the set radius range, and use the arc feature to select a point sequence that is likely to be a circle, using the method described in (3.3). Set.
That is, the apparatus includes a point sequence data extraction unit 300 including an arc extraction unit 310, an arc group setting unit 320, and an arc group circle candidate point sequence setting unit 330, as shown in FIG.
When extracting the arc feature, the arc extracting unit 310 sets only an arc feature that falls within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The arc group setting unit 320 creates an arc group based on the extracted arcs whose center positions are in the vicinity and whose radius is close.
The arc group circle candidate point sequence setting unit 330 sets a set of point sequences of a plurality of edge line segments that are the origin of each arc of the arc group as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
[ Example 2]

(6.4)本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置4の例を図5に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.4)による「円弧グループの近似円上の点列を円候補点列とする方式」を用いて、円になりそうな点列部分を設定する。
即ち、本装置は、図5に示すように、円弧抽出部410、円弧グループ設定部420及び円弧グループ近似円円候補点列設定部430よりなる点列データ抽出部400を備える。
円弧抽出部410は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
円弧グループ設定部420は、抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作る。
円弧グループ近似円円候補点列設定部430は、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を基に円近似を行い、その近似円の円周上の点の位置を計算し、近似円の円周として求められた点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
[実施例3]
(6.4) An example of the device 4 for detecting the three-dimensional position and orientation of the circle feature according to the second embodiment of the present invention is shown in FIG.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extracting arc features that fall within the set radius range, and using the arc feature as a circle candidate point sequence based on the arc feature (3.4) Set the sequence of possible dot sequences.
That is, as shown in FIG. 5, the apparatus includes a point sequence data extraction unit 400 including an arc extraction unit 410, an arc group setting unit 420, and an arc group approximate circle candidate point sequence setting unit 430.
When the arc extraction unit 410 extracts an arc feature, only an arc feature that falls within the set radius range is used as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The arc group setting unit 420 creates an arc group based on the extracted arcs whose center position is in the vicinity and whose radius is close.
The arc group approximate circle / circle candidate point sequence setting unit 430 performs circle approximation based on a set of point sequences of a plurality of edge line segments that are the origin of each arc of the arc group, and on the circumference of the approximate circle The point positions are calculated, and the point sequence obtained as the circumference of the approximate circle is set as the circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
[ Example 3]

(6.5)本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置5の例を図6に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.5)による「積算角度チェックを行い円弧グループを円候補点列とする方式」を用いて、円になりそうな点列部分を設定する。
即ち、本装置は、図6に示すように、円弧抽出部510、円弧グループ設定部520、積算角度検査部530及び円弧グループ円候補点列設定部540よりなる点列データ抽出部500を備える。
円弧抽出部510は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
円弧グループ設定部520は、抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作る。
積算角度検査部530は、それぞれの円弧の円周角の総計が設定した値以上であることを確認することにより、積算角度チェックを行う。
円弧グループ円候補点列設定部540は、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を円候補点列とする。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
[実施例4]
(6.5) an example of a device 5 for detecting the three-dimensional position and orientation of the circle feature according to a third embodiment of the present invention shown in FIG.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). An arc feature that falls within the set radius range is extracted, and based on the arc feature, it is likely to be a circle by using the method of (3.5) Checking the accumulated angle and making the arc group a circle candidate point sequence Set the point sequence part.
That is, this apparatus includes a point sequence data extraction unit 500 including an arc extraction unit 510, an arc group setting unit 520, an integrated angle inspection unit 530, and an arc group circle candidate point sequence setting unit 540, as shown in FIG.
When extracting the arc feature, the arc extracting unit 510 sets only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The arc group setting unit 520 creates an arc group based on the extracted arcs whose center position is in the vicinity and whose radius is close.
The accumulated angle inspection unit 530 performs an accumulated angle check by confirming that the total circumferential angle of each arc is equal to or greater than a set value.
The arc group circle candidate point sequence setting unit 540 sets a set of point sequences of a plurality of edge line segments that are the origin of each arc of the arc group as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
[ Example 4]

(6.6)本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置6の例を図7に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.6)による「積算角度チェックを行い円弧グループの近似円上の点列を円候補点列とする方式」を用いて、円になりそうな点列部分を設定する。
即ち、本装置は、図7に示すように、円弧抽出部610、円弧グループ設定部620、積算角度設定部630及び円弧グループ近似円円候補点列設定部640よりなる点列データ抽出部600を備える。
円弧抽出部610は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
円弧グループ設定部620は、抽出した円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作る。
積算角度設定部630は、それぞれの円弧の円周角の総計が設定した値以上であることを確認することにより、積算角度チェックを行う。
円弧グループ円候補点列設定部640は、円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を基に円近似を行い、その近似円の円周上の点の位置を計算し、近似円の円周として求められた点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
[参考3]
(6.6) An example of the device 6 for detecting the three-dimensional position and orientation of the circle feature according to the fourth embodiment of the present invention is shown in FIG.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extract the arc features that fall within the set radius range, and based on the arc features, use “3.6 to check the accumulated angle and make the point sequence on the approximate circle of the arc group a circle candidate point sequence” Use to set the point sequence that is likely to be a circle.
That is, as shown in FIG. 7, this apparatus includes a point sequence data extraction unit 600 including an arc extraction unit 610, an arc group setting unit 620, an integrated angle setting unit 630, and an arc group approximate circle / circle candidate point sequence setting unit 640. Prepare.
When extracting the arc feature, the arc extracting unit 610 determines only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The arc group setting unit 620 creates an arc group based on the extracted arcs whose center position is in the vicinity and whose radius is close.
The integrated angle setting unit 630 performs an integrated angle check by confirming that the total of the circumferential angles of the respective arcs is equal to or greater than a set value.
The arc group circle candidate point sequence setting unit 640 performs a circle approximation based on a set of point sequences of a plurality of edge line segments that are the origins of the respective arcs of the arc group, and calculates points on the circumference of the approximate circle. The position is calculated, and the point sequence obtained as the circumference of the approximate circle is set as the circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
[Reference Example 3]

(6.7)本発明の参考に係る円特徴の三次元位置姿勢を検出する装置7の例を図8に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.1)による「抽出円弧を円候補点列とする方法」を用いて、円になりそうな点列部分を設定し、円特徴データを計算した後、(5.1)による「円特徴の傾き角度検査」を行う。
即ち、本装置は、図8に示すように、円弧抽出部710及び抽出円弧円候補点列設定部720よりなる点列データ抽出部700を備える。
円弧抽出部710は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧円候補点列設定部720は、抽出した円弧の元となったエッジ線分の点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
更に、円特徴データを計算した後、検出した円特徴の傾き角度と予め設定して置いた傾き角度の許容値を比較し、円特徴の傾き角度が許容値以内である場合は「正しい」と、それ以外の場合は「正しくない」と判断する円特徴傾き検査部730を備える。
[参考4]
(6.7) An example of a device 7 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 3 of the present invention is shown in FIG.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extract the arc features that fall within the set radius range, and use the “method to make the extracted arc as a circle candidate point sequence” according to (3.1) based on the arc features, After setting and calculating the circle feature data, the “circle feature tilt angle inspection” according to (5.1) is performed.
That is, this apparatus includes a point sequence data extraction unit 700 including an arc extraction unit 710 and an extracted arc circle candidate point sequence setting unit 720, as shown in FIG.
When extracting the arc feature, the arc extracting unit 710 sets only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc circle candidate point sequence setting unit 720 sets the point sequence of the edge line segment that is the origin of the extracted arc as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
Furthermore, after calculating the circle feature data, compare the detected tilt angle of the circle feature with the preset tilt angle tolerance value, and if the circle feature tilt angle is within the tolerance value, it is “correct”. In other cases, a circle feature inclination inspection unit 730 that determines “incorrect” is provided.
[Reference Example 4]

(6.8)本発明の参考に係る円特徴の三次元位置姿勢を検出する装置8の例を図9に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.2)による「抽出円弧の仮想円上の点列を円候補点列とする方法」を用いて、円になりそうな点列部分を設定し、円特徴データを計算した後、(5.1)による「円特徴の傾き角度検査」を行う。
即ち、本装置は、図9に示すように、円弧抽出部810及び抽出円弧仮想円円候補点列設定部820よりなる点列データ抽出部800を備える。
円弧抽出部810は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧仮想円円候補点列設定部820は、抽出した円弧の中心と半径と同じデータをもつ仮想的な円を考え、その仮想円の円周上の点の位置を計算し、仮想円の円周として求められた点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
更に、円特徴データを計算した後、検出した円特徴の傾き角度と予め設定して置いた傾き角度の許容値を比較し、円特徴の傾き角度が許容値以内である場合は「正しい」と、それ以外の場合は「正しくない」と判断する円特徴傾き検査部830を備える。
[参考5]
(6.8) An example of an apparatus 8 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 4 of the present invention is shown in FIG.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extracting arc features that fall within the set radius range, and using the “circular point sequence on the virtual circle of the extracted arc as a circle candidate point sequence” according to (3.2) based on the arc features, A point sequence portion that is likely to be set is set, and circle feature data is calculated. Then, “inclination angle inspection of circle feature” according to (5.1) is performed.
That is, this apparatus includes a point sequence data extraction unit 800 including an arc extraction unit 810 and an extracted arc virtual circle / circle candidate point sequence setting unit 820, as shown in FIG.
When the arc extraction unit 810 extracts an arc feature, only an arc feature that falls within the set radius range is used as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc virtual circle circle candidate point sequence setting unit 820 considers a virtual circle having the same data as the center and radius of the extracted arc, calculates the position of the point on the circumference of the virtual circle, A point sequence obtained as a circumference is set as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
Furthermore, after calculating the circle feature data, compare the detected tilt angle of the circle feature with the preset tilt angle tolerance value, and if the circle feature tilt angle is within the tolerance value, it is “correct”. In other cases, a circle feature inclination inspection unit 830 that determines “incorrect” is provided.
[Reference Example 5]

(6.9)本発明の参考に係る円特徴の三次元位置姿勢を検出する装置9の例を図10に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によ「て設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.1)による「抽出円弧を円候補点列とする方法」を用いて、円になりそうな点列部分を設定し、円特徴データを計算した後、(5.2)による「円特徴の傾き角度補正」を行う。
即ち、本装置は、図10に示すように、円弧抽出部910及び抽出円弧円候補点列設定部920よりなる点列データ抽出部900を備える。
円弧抽出部910は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧円候補点列設定部920は、抽出した円弧の元となったエッジ線分の点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
更に、検出した円特徴の放線ベクトルを鉛直上向きに設定し、円特徴の姿勢データを再計算する傾き角度補正部930を備える。
[参考6]
(6.9) FIG. 10 shows an example of a device 9 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 5 of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the 3D position / orientation detection device for a circular feature of Patent Document 2, the present apparatus sets “arc radius range setting” according to (2). The point that is likely to be a circle by extracting the arc feature that falls within the set radius range and using the “method of making the extracted arc as a circle candidate point sequence” according to (3.1) based on the arc feature. After setting the column part and calculating the circle feature data, the “circle angle correction of circle feature” according to (5.2) is performed.
That is, this apparatus includes a point sequence data extraction unit 900 including an arc extraction unit 910 and an extracted arc circle candidate point sequence setting unit 920 as shown in FIG.
When extracting the arc feature, the arc extracting unit 910 sets only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc circle candidate point sequence setting unit 920 sets the point sequence of the edge line segment that is the origin of the extracted arc as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
In addition, an inclination angle correction unit 930 is provided that sets the detected ray feature ray vector vertically upward and recalculates the circle feature posture data.
[Reference Example 6]

(6.10)本発明の参考に係る円特徴の三次元位置姿勢を検出する装置10の例を図11に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.2)による「抽出円弧の仮想円上の点列を円候補点列とする方法」を用いて、円になりそうな点列部分を設定し、円特徴データを計算した後、(5.2)による「円特徴の傾き角度補正」を行う。
即ち、本装置は、図11に示すように、円弧抽出部1010及び抽出円弧仮想円円候補点列設定部1020よりなる点列データ抽出部1000を備える。
円弧抽出部1010は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧仮想円円候補点列設定部1020は、抽出した円弧の中心と半径と同じデータをもつ仮想的な円を考え、その仮想円の円周上の点の位置を計算し、仮想円の円周として求められた点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
更に、検出した円特徴の放線ベクトルを鉛直上向きに設定し、円特徴の姿勢データを再計算する傾き角度補正部1030を備える。
[参考7]
(6.10) FIG. 11 shows an example of an apparatus 10 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 6 of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extracting arc features that fall within the set radius range, and using the “circular point sequence on the virtual circle of the extracted arc as a circle candidate point sequence” according to (3.2) based on the arc features, The point sequence portion that is likely to be set is set, and the circle feature data is calculated. Then, the “circular feature tilt angle correction” according to (5.2) is performed.
That is, this apparatus includes a point sequence data extraction unit 1000 including an arc extraction unit 1010 and an extracted arc virtual circle / circle candidate point sequence setting unit 1020, as shown in FIG.
When extracting the arc feature, the arc extracting unit 1010 sets only those that fall within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc virtual circle circle candidate point sequence setting unit 1020 considers a virtual circle having the same data as the center and radius of the extracted arc, calculates the position of the point on the circumference of the virtual circle, A point sequence obtained as a circumference is set as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
In addition, an inclination angle correction unit 1030 is provided for setting the detected circle feature ray vector vertically upward and recalculating the circle feature posture data.
[Reference Example 7]

(6.11)本発明の参考に係る円特徴の三次元位置姿勢を検出する装置11の例を図12に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.1)による「抽出円弧を円候補点列とする方法」を用いて、円になりそうな点列部分を設定し、円特徴データを計算した後、(5.1)による「円特徴の傾き角度検査」を行い、円特徴の傾き角度が正しくないと判断された場合について(5.2)による「円特徴の傾き角度補正」を行う。
即ち、本装置は、図12に示すように、円弧抽出部1110及び抽出円弧円候補点列設定部1120よりなる点列データ抽出部1100を備える。
円弧抽出部1110は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧円候補点列設定部1120は、抽出した円弧の元となったエッジ線分の点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
また、円特徴データを計算した後、検出した円特徴の傾き角度と予め設定して置いた傾き角度の許容値を比較し、円特徴の傾き角度が許容値以内である場合は「正しい」と、それ以外の場合は「正しくない」と判断する円特徴傾き検査部1130を備える。
更に、円特徴の傾き角度が正しくないと判断された場合について、検出した円特徴の放線ベクトルを鉛直上向きに設定し、円特徴の姿勢データを再計算する傾き角度補正部1140を備える。
[参考8]
(6.11) FIG. 12 shows an example of a device 11 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 7 of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extract the arc features that fall within the set radius range, and use the “method to make the extracted arc as a circle candidate point sequence” according to (3.1) based on the arc features, After setting and calculating the circle feature data, “circle feature tilt angle check” according to (5.1) is performed, and when the circle feature tilt angle is determined to be incorrect “circle” according to (5.2) “Characteristic tilt angle correction”.
That is, this apparatus includes a point sequence data extraction unit 1100 including an arc extraction unit 1110 and an extracted arc circle candidate point sequence setting unit 1120 as shown in FIG.
When extracting the arc feature, the arc extracting unit 1110 sets only an arc feature that falls within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc circle candidate point sequence setting unit 1120 sets, as a circle candidate point sequence, the point sequence of the edge line segment that is the origin of the extracted arc.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
Also, after calculating the circle feature data, compare the detected tilt angle of the circle feature with the preset tilt angle tolerance value, and if the circle feature tilt angle is within the tolerance value, it will be “correct” In other cases, a circle feature inclination inspection unit 1130 that determines “incorrect” is provided.
Further, when it is determined that the inclination angle of the circle feature is not correct, the inclination angle correction unit 1140 is provided that sets the detected circle feature's ray vector vertically upward and recalculates the posture data of the circle feature.
[Reference Example 8]

(6.12)本発明の参考に係る円特徴の三次元位置姿勢を検出する装置12の例を図13に示す。
本装置は、特許文献2の円特徴の三次元位置姿勢検出装置の「円になりそうな点列部分を抽出する」点列データ抽出部において、(2)による「円弧半径範囲の設定」によって設定された半径範囲に入る円弧特徴を抽出し、その円弧特徴を基に(3.2)による「抽出円弧の仮想円上の点列を円候補点列とする方法」を用いて、円になりそうな点列部分を設定し、円特徴データを計算した後、(5.1)による「円特徴の傾き角度検査」を行い、円特徴の傾き角度が正しくないと判断された場合について(5.2)による「円特徴の傾き角度補正」を行う。
即ち、本装置は、図13に示すように、点列データ抽出部1200は、円弧抽出部1210及び抽出円弧仮想円円候補点列設定部1220を備える。
円弧抽出部1210は、円弧特徴を抽出する際、設定した半径範囲に入るもののみを円弧特徴とする。
円弧半径の範囲は、図1に示す相似関係を用いて円弧半径範囲計算部20により予め設定し、その値をメモリ30に記憶しておく。
抽出円弧仮想円円候補点列設定部1220は、抽出した円弧の中心と半径と同じデータをもつ仮想的な円を考え、その仮想円の円周上の点の位置を計算し、仮想円の円周として求められた点列を円候補点列として設定する。
設定された円候補点列に基づいて、円特徴データ計算部40は、円特徴を計算する。
また、円特徴データを計算した後、検出した円特徴の傾き角度と予め設定して置いた傾き角度の許容値を比較し、円特徴の傾き角度が許容値以内である場合は「正しい」と、それ以外の場合は「正しくない」と判断する円特徴傾き検査部1230を備える。
更に、円特徴の傾き角度が正しくないと判断された場合について、検出した円特徴の放線ベクトルを鉛直上向きに設定し、円特徴の姿勢データを再計算する傾き角度補正部1240を備える。
(6.12) FIG. 13 shows an example of a device 12 for detecting a three-dimensional position and orientation of a circle feature according to Reference Example 8 of the present invention.
In the point sequence data extraction unit “extracts a point sequence portion that is likely to be a circle” of the circle feature three-dimensional position and orientation detection device of Patent Document 2, this device performs “arc radius range setting” in (2). Extracting arc features that fall within the set radius range, and using the “circular point sequence on the virtual circle of the extracted arc as a circle candidate point sequence” according to (3.2) based on the arc features, After setting the likely sequence of points and calculating the circle feature data, perform “circle feature tilt angle inspection” according to (5.1) and determine that the circle feature tilt angle is incorrect ( 5. Perform “Circular feature tilt angle correction” according to 5.2).
That is, as shown in FIG. 13, the apparatus includes a point sequence data extraction unit 1200 that includes an arc extraction unit 1210 and an extracted arc virtual circle / circle candidate point sequence setting unit 1220.
When extracting the arc feature, the arc extracting unit 1210 sets only the feature within the set radius range as the arc feature.
The arc radius range is set in advance by the arc radius range calculation unit 20 using the similarity shown in FIG. 1, and the value is stored in the memory 30.
The extracted arc virtual circle circle candidate point sequence setting unit 1220 considers a virtual circle having the same data as the center and radius of the extracted arc, calculates the position of the point on the circumference of the virtual circle, A point sequence obtained as a circumference is set as a circle candidate point sequence.
Based on the set circle candidate point sequence, the circle feature data calculation unit 40 calculates a circle feature.
Also, after calculating the circle feature data, compare the detected tilt angle of the circle feature with the preset tilt angle tolerance value, and if the circle feature tilt angle is within the tolerance value, it will be “correct” In other cases, a circle feature inclination inspection unit 1230 that determines “incorrect” is provided.
Furthermore, when it is determined that the inclination angle of the circle feature is not correct, the inclination angle correction unit 1240 is provided that sets the detected circle feature's ray vector vertically upward and recalculates the posture data of the circle feature.

本発明は、いわゆる物体認識処理に関し、特にモデルベーストマッチング法に用いる特徴を抽出する手段として産業上広く利用可能なものである。   The present invention relates to so-called object recognition processing, and in particular, can be widely used industrially as means for extracting features used in a model-based matching method.

円弧半径範囲の設定を示す説明図である。It is explanatory drawing which shows the setting of an arc radius range. 本発明の参考例1に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 1 of this invention. 本発明の参考例2に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 2 of this invention. 本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on Example 1 of this invention. 本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature based on Example 2 of this invention. 本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature based on Example 3 of this invention. 本発明の実施例に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature based on Example 4 of this invention. 本発明の参考に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 3 of this invention. 本発明の参考に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 4 of this invention. 本発明の参考に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 5 of this invention. 本発明の参考に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature based on the reference example 6 of this invention. 本発明の参考に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 7 of this invention. 本発明の参考に係る円特徴の三次元位置姿勢を検出する装置を示す構成図である。It is a block diagram which shows the apparatus which detects the three-dimensional position and orientation of the circular feature which concerns on the reference example 8 of this invention. 円特徴データの計算の例を示す説明図である。It is explanatory drawing which shows the example of calculation of circle feature data. 円特徴データの計算の例を示す説明図である。It is explanatory drawing which shows the example of calculation of circle feature data. 円特徴データの計算の例を示す説明図である。It is explanatory drawing which shows the example of calculation of circle feature data.

1 三次元位置姿勢を検出する装置
20 円弧半径範囲計算部
30 メモリ
40 円特徴データ計算部
100 点列データ抽出部
110 円弧抽出部
120 抽出円弧円候補点設定部
DESCRIPTION OF SYMBOLS 1 Apparatus which detects 3D position and orientation 20 Arc radius range calculation part 30 Memory 40 Circle feature data calculation part 100 Point sequence data extraction part 110 Arc extraction part 120 Extraction arc circle candidate point setting part

Claims (1)

検出すべき対象部品の穴がカメラの光軸に対してほぼ傾かない設置状態に限定して前記カメラで前記対象部品を撮影し、撮影された前記対象部品の画像からエッジ線分を円の一部である円弧特徴として抽出する際、設定した半径範囲に入るもののみを円弧特徴とする円弧抽出部と、前記円弧抽出部により抽出された円弧の中から中心位置が近傍にあり、かつ半径の長さが近いものにより円弧グループを作る円弧グループ設定部と、前記円弧グループ設定部により作られた円弧グループのそれぞれの円弧の円周角の総計が設定した値以上であることを確認することにより積算角度チェックを行う積算角度検査部と、前記積算角度検査部により積算角度チェック後の円弧グループのそれぞれの円弧の元となった複数のエッジ線分の点列の集合を基に円近似を行い、その近似円の円周上の点の位置を計算し、近似円の円周として求められた点列を円候補点列として設定する円弧グループ近似円円候補点列設定部と、前記円弧グループ近似円円候補点列設定部により設定された円候補点列に基づいて円特徴を計算する円特徴データ計算部とを有することを特徴とする円特徴の三次元位置姿勢検出装置。 The target part is photographed with the camera only in an installation state in which the hole of the target part to be detected is not substantially inclined with respect to the optical axis of the camera, and an edge line segment is extracted from the photographed image of the target part. When extracting as an arc feature that is a part, an arc extracting unit having only an arc feature that falls within the set radius range, and the center position is in the vicinity from the arc extracted by the arc extracting unit, and the radius By confirming that the total of the circumference angle of each arc of the arc group setting unit that creates an arc group by the one having a short length and the arc group created by the arc group setting unit is greater than or equal to the set value and integrated angle inspection unit that performs integrated angle check, based on a plurality of sets of dot sequence of edge line segments became each arc of the original arc group after integrated angle checked by the integrated angle inspection unit Perform a circular approximation, and its position of a point on the circumference of the approximate circle was calculated arc Group approximate circle circle candidate point sequence setting section for setting a sequence of points obtained as the circumference of an approximate circle as a circle candidate point sequence A circle feature data calculation unit for calculating a circle feature based on a circle candidate point sequence set by the arc group approximate circle circle candidate point sequence setting unit, and a three-dimensional position and orientation detection device for a circle feature .
JP2007112876A 2007-04-23 2007-04-23 3D position and orientation detection device for circular features Expired - Lifetime JP4595957B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2007112876A JP4595957B2 (en) 2007-04-23 2007-04-23 3D position and orientation detection device for circular features

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2007112876A JP4595957B2 (en) 2007-04-23 2007-04-23 3D position and orientation detection device for circular features

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
JP23245399A Division JP4312891B2 (en) 1999-08-19 1999-08-19 3D position and orientation detection device for circular features

Publications (2)

Publication Number Publication Date
JP2007234042A JP2007234042A (en) 2007-09-13
JP4595957B2 true JP4595957B2 (en) 2010-12-08

Family

ID=38554506

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2007112876A Expired - Lifetime JP4595957B2 (en) 2007-04-23 2007-04-23 3D position and orientation detection device for circular features

Country Status (1)

Country Link
JP (1) JP4595957B2 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS60179881A (en) * 1984-02-15 1985-09-13 Komatsu Ltd Recognizing method of approximately circular outline
JPH04323503A (en) * 1991-04-23 1992-11-12 Toyoda Mach Works Ltd Image processor
JPH0822540A (en) * 1994-07-06 1996-01-23 Meidensha Corp Recognition method for circle and arc
JPH09178427A (en) * 1995-12-27 1997-07-11 Ntn Corp Image position measuring method
JPH10326347A (en) * 1997-05-27 1998-12-08 Meidensha Corp Detection device and detection method for three-dimensional position attitude of circlar feature of parts, and recording medium for the detection device and detection method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS60179881A (en) * 1984-02-15 1985-09-13 Komatsu Ltd Recognizing method of approximately circular outline
JPH04323503A (en) * 1991-04-23 1992-11-12 Toyoda Mach Works Ltd Image processor
JPH0822540A (en) * 1994-07-06 1996-01-23 Meidensha Corp Recognition method for circle and arc
JPH09178427A (en) * 1995-12-27 1997-07-11 Ntn Corp Image position measuring method
JPH10326347A (en) * 1997-05-27 1998-12-08 Meidensha Corp Detection device and detection method for three-dimensional position attitude of circlar feature of parts, and recording medium for the detection device and detection method

Also Published As

Publication number Publication date
JP2007234042A (en) 2007-09-13

Similar Documents

Publication Publication Date Title
US20160239949A1 (en) Workpiece positioning apparatus using imaging unit
CN111156923A (en) Workpiece detection method, workpiece detection device, computer equipment and storage medium
JP4836086B2 (en) 3D shape detector
JP2017126187A (en) Meter reading device
EP2756482A1 (en) Resolving homography decomposition ambiguity based on orientation sensors
CN110569845A (en) test paper image correction method and related device
US10074551B2 (en) Position detection apparatus, position detection method, information processing program, and storage medium
JP3849228B2 (en) 3D position / orientation detection device for circular feature of parts, detection method thereof, and recording medium thereof
US7355725B2 (en) Measuring system
US20060273268A1 (en) Method for detecting 3D measurement data using allowable error zone
JP7226494B2 (en) Contact wire wear detection method
JP6877233B2 (en) Bolt position detection method and system
JP2016136111A5 (en) Shape error determination device, determination result image generation device, and shape error determination method
JP4595957B2 (en) 3D position and orientation detection device for circular features
JP4569595B2 (en) 3D position and orientation detection device for circular features
KR20170142379A (en) Apparatus for detect dimensional welding line of welding robot using image processing
JP2005283267A (en) Through hole measuring device, method, and program for through hole measurement
JP6863946B2 (en) Image processing device
US20080281543A1 (en) Inspection system and methods with autocompensation for edge break gauging orientation
JP4312891B2 (en) 3D position and orientation detection device for circular features
JP2017173992A (en) Inspection device for bundled steel pipes and inspection method for bundled steel pipes
JP6681759B2 (en) Weld bead inspection system
JP2011100223A (en) Image processing apparatus and image processing method
JP4348859B2 (en) Precise 3D position and orientation detection device for circular features of parts
US20220358635A1 (en) Inspection apparatus, measuring method, and computer readable medium

Legal Events

Date Code Title Description
A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20100204

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20100209

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20100312

A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20100413

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20100707

A911 Transfer to examiner for re-examination before appeal (zenchi)

Free format text: JAPANESE INTERMEDIATE CODE: A911

Effective date: 20100722

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: 20100824

A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20100906

R150 Certificate of patent or registration of utility model

Ref document number: 4595957

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150

Free format text: JAPANESE INTERMEDIATE CODE: R150

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20131001

Year of fee payment: 3

EXPY Cancellation because of completion of term