JPS6154577A - Labeling device for contour feature point - Google Patents

Labeling device for contour feature point

Info

Publication number
JPS6154577A
JPS6154577A JP17708584A JP17708584A JPS6154577A JP S6154577 A JPS6154577 A JP S6154577A JP 17708584 A JP17708584 A JP 17708584A JP 17708584 A JP17708584 A JP 17708584A JP S6154577 A JPS6154577 A JP S6154577A
Authority
JP
Japan
Prior art keywords
contour
point
points
feature
base 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.)
Granted
Application number
JP17708584A
Other languages
Japanese (ja)
Other versions
JPH0377549B2 (en
Inventor
Masao Nito
正夫 仁藤
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.)
Fuji Electric Co Ltd
Original Assignee
Fuji Electric 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 Fuji Electric Co Ltd filed Critical Fuji Electric Co Ltd
Priority to JP17708584A priority Critical patent/JPS6154577A/en
Publication of JPS6154577A publication Critical patent/JPS6154577A/en
Publication of JPH0377549B2 publication Critical patent/JPH0377549B2/ja
Granted legal-status Critical Current

Links

Landscapes

  • Image Analysis (AREA)

Abstract

PURPOSE:To decide feature points regardless of the position of a subject form by deciding a base point based on an amount that has no relation with the position of the subject form in a screen and using this base point as a start point to put numbers successively on the feature points in the prescribed direction along a contour. CONSTITUTION:Both corner points P1-P4 and bottom points N1-N8 are detected as the feature points of the contour of a subject form F. These bottom points are connected successively for approximation to a polygon. The maximum side length has no relation with the change of position of the form F due to shift, rotation, etc. Therefore the length of each side is checked and an end of the longest side is defined as a base point in the fixed direction (clockwise), and the numbers are put successively in a prescribed direction A. In case plural longest sides exist (a, c), the adjacent sides (b) and (d) are compared with each other. Then the longer side (d) replaced with the longest side to decide the base point. As a result, the point N1 is defined as a base point and thereafter the numbers are put to N2-N8 successively. For the corner points, the point P1 having the longest distance from the centroid is defined as a base point. Then the numbers are put to points P2-P4 respectively.

Description

【発明の詳細な説明】 〔発明の属する技術分野〕 本発明は例えば組立ロボットや半導体の自動ボンディン
グ装置等、Iこ設けられ2次元の撮像手段を用いて操作
あるいは加工の対象となる部品(対象部品という)等の
対象物を撮像し、その輪郭一形状中の特徴点とし、での
ハンドリング点やボンデング点などを検出する装置tこ
関する。
[Detailed Description of the Invention] [Technical Field to which the Invention Pertains] The present invention relates to, for example, an assembly robot, an automatic semiconductor bonding device, etc., which uses a two-dimensional imaging means to detect parts (objects) to be manipulated or processed. It is related to a device that captures an image of an object such as a component, uses it as feature points in its contour, and detects handling points, bonding points, etc.

〔従来技術とその問題点〕[Prior art and its problems]

以下各図の説明において同一の符号は同一又は相当部分
を示す。
In the following description of each figure, the same reference numerals indicate the same or corresponding parts.

従来、前記対象部品としての被撮像物体を撮像し、該部
品中の操作あるいは加工の対象点となる、その画像の輪
部特徴点を検出する技術さしては、本出願人蚤こよる特
願昭58−76771号「輪郭特徴検出方式」(以下文
献(1)と呼ぶ)がある。
Conventionally, a technique for capturing an image of an object to be imaged as the target part and detecting a limbal feature point of the image, which becomes a target point for manipulation or processing in the part, was disclosed in a patent application filed by the present applicant, Koyo Flea. No. 58-76771 "Contour feature detection method" (hereinafter referred to as document (1)).

第3図はこの方式によって求められた特徴点の例を示す
。すなわち同図tこおいてFは対象部品20。
FIG. 3 shows an example of feature points determined by this method. That is, in the figure t, F is the target part 20.

の2次元形状を示す2値化画像、Pはこの2値化画像F
の輪郭上の特徴点としての突起(以下角点という)、N
は同じく特徴点としてのくぼみ(以下谷点という)であ
る。なおGは画像Fの重心(図心)、X,Yは画像走査
の軸を示す。
A binarized image showing the two-dimensional shape of , P is this binarized image F
Protrusions as feature points on the contour (hereinafter referred to as corner points), N
is also a depression (hereinafter referred to as a valley point) as a feature point. Note that G indicates the center of gravity (centroid) of the image F, and X and Y indicate the axes of image scanning.

このように前記方式では、特徴点の角点P,谷点Nなど
の種類,X,Y軸上の位置。数は検出しでいるが,対象
部品200従って2値化画像Fの重心Gの移動や゛重心
Gを中心とする回転(以下このような部品200又は画
像Fの変位を単に5配置1又は1移動や回転1という)
によって、特徴点の検出順が変った場合にも、輪郭上の
特定の特徴点を同定し.ていない。従ってこの従来技術
では種々の配置で作業台上音装置かれる対象部品200
上の特定の操作あるいは加工対象点の位置を検出するこ
とができないという問題点がある。
In this way, in the above method, the types of feature points, such as corner points P and valley points N, and their positions on the X and Y axes. Although the number has already been detected, the movement of the center of gravity G of the target part 200 and thus the binarized image F and the rotation around the center of gravity G (hereinafter, the displacement of such a part 200 or image F is simply expressed as 5 arrangement 1 or 1 (referred to as movement or rotation 1)
, even if the detection order of feature points changes, specific feature points on the contour can be identified. Not yet. Therefore, in this prior art, the target parts 200 that are mounted on the workbench sound device in various locations are
There is a problem in that the above specific operation or the position of the point to be processed cannot be detected.

〔発明の目的〕[Purpose of the invention]

本発明の目的は前記の問題点を除去し、対象部品の配置
に拘わらず、その特定の操作あるいは加工対象点、従っ
てその画像の輪郭上の所定の特徴点の位置を特定できる
輪郭特徴点の番号付(ラベリングともいう)装置を提供
することを目的とする。
An object of the present invention is to eliminate the above-mentioned problems and to provide a contour feature point that can specify the specific operation or processing target point, and therefore the position of a predetermined feature point on the contour of the image, regardless of the arrangement of the target part. The purpose is to provide a numbering (also called labeling) device.

〔発明の袂点〕[The point of invention]

この発明の要点は、対象部品の2次元画像中の対象形状
の輪郭を一定方向fζ全周瞭こわたり追跡し、このとき
得られる輪郭情報により輪郭部−こおける形状特徴点を
検出し、前記対象形状を前記特徴点により多角形に近似
し、前記特徴点のうちの1つを基点々し,で、前記多角
形の周上を所定方向曇こたどりつつ、出現する各特徴点
tこ対し順次番号付けを行う点シこある。
The gist of this invention is to clearly track the contour of the target shape in a two-dimensional image of the target part in a fixed direction fζ all around, detect shape feature points in the contour part using the contour information obtained at this time, and Approximate the target shape to a polygon using the feature points, use one of the feature points as a base point, trace the circumference of the polygon in a predetermined direction, and compare each feature point t that appears. There are points for sequential numbering.

また前記基点とし,では、例えば前記多角形の辺のうち
、最大辺長を持つ1辺の一端を用いたり、あるいは前記
対象形状の重心と各特徴点との距離のうち.最大flQ
 /ill tこある点を用いたりする点にある、 換言すれば本発明の要点は、(対象部品等を2次元撮像
装置等で撮像し,この画像信号をA/D変換器等を介し
2値化して得た)対象となる2値化画像の輪郭画素を、
(輪郭追跡回路等を介し)順次該2値化画像を所定方向
(例えば反時計方向、なおこの方向を以下輪郭追跡方向
と呼ぶ)Iこ1周するよう曇ζ追跡し、このとき得られ
る各輪郭画素の座標値と、隣接する輪郭画素への追跡方
向を示す追跡方向コードとから、(特徴検出回路等を介
し)輪郭上の突起又はくはみ等、所定の特徴部/I3を
代表する(角点又は谷点などの)輪郭画素(以下特徴点
と呼ぶ)の位置を、前記輪郭を所定方向(例えば反時計
方向)に1周する順)こ検出し、その111μ序で記憶
する装置tCおいで、前記特徴点をその全Cからなる第
1の群(例えば角点及び谷点の全Cを含む#)又はそれ
が共有する性質で群別された+12数の第2の群(例え
は角点のみ又は谷点のみからなる?)+こ区分するとき
、前記第1もしくは第2の群ごとfこ、前記輪郭を1周
するようfこ前記の同一の群内の各特徴点を順次結んで
得られる多角形の各辺長を算出する辺長算出手段、又は
(及び)(重心算出回路などを介しで得た前記2値化画
像の重心と前記の同一の群内の各特徴点との距tm: 
(以下重心距離と呼ぶ)を算出するMI心距RI[算出
手段と、 算出された前記の各辺長又は(及び)各重心距離−こ基
づいて当該群内の特徴点を代表する点(以下基点と呼ぶ
)を決定する基点決定手段と。
In addition, for the base point, for example, one end of the side with the maximum side length among the sides of the polygon may be used, or the distance between the center of gravity of the target shape and each feature point. Maximum flQ
In other words, the main point of the present invention is to use certain points. The outline pixels of the target binarized image (obtained by digitizing) are
The binarized image is sequentially tracked in a predetermined direction (for example, counterclockwise, hereinafter referred to as the contour tracking direction) (via a contour tracking circuit, etc.), and each image obtained at this time is Based on the coordinate value of the contour pixel and the tracking direction code indicating the tracking direction to the adjacent contour pixel, (via a feature detection circuit, etc.) a predetermined feature such as a protrusion or indentation on the contour is determined to represent a predetermined feature /I3. A device that detects the positions of contour pixels (hereinafter referred to as feature points) (such as corner points or valley points) in the order of going around the contour once in a predetermined direction (for example, counterclockwise) and stores them in the 111μ order. At tC, the feature points are divided into a first group consisting of all Cs (e.g. # including all Cs of corner points and valley points) or a second group of +12 numbers (e.g. consists of only corner points or only valley points?) When dividing, each feature point in the same group is divided so that each of the first or second groups goes around the contour once. A side length calculation means for calculating each side length of a polygon obtained by sequentially connecting the polygons, or (and) a center of gravity of the binarized image obtained through a center of gravity calculation circuit, etc., and each feature in the same group. Distance to point tm:
(hereinafter referred to as the center of gravity distance) [calculating means; base point determining means for determining a base point (referred to as a base point).

前記輪郭を所定方向(例えば反時計方向)に1周するよ
う醗こ、当該群内の各特徴点に順次番号性を行う特徴点
ラベリング手段と、(の各手段)を(特徴点番号付回路
内などシこ)備えた点、もし、くけざらに前記基点決定
手段は前記各辺長の6ち最大値を持つ辺の両側中前記輪
郭追跡方向に対し所定の側(例えは時計方向の側)にあ
る特徴点を前記基点とする手段を含むようをこした点、
もしくはさらtこ前記基点決定手段は前記各重心距離の
うち最大のもの蚤こ対応する特徴点を前記基点とする手
段を含むよう−こした点、もしくはさら蚤こ前記基点決
定手段は、前記の同一の群+C属する基点が複数となっ
たときは、該麹数の各基点を前記の所定側(例えば時計
方向all )とする最大辺長の各辺Eこ対し、前記輪
郭追跡方向;2店準とする所定の側(例えば反時計方向
側)lこ、隣接する前記多角形の辺長を比較する手段を
設けるとともfこ、前記隣接辺長のうち最大値を持つ辺
の両側中、前記輪郭追跡方向に対し所定の側(例えば時
計方向1!+ ) )こある特徴点を新たな基点として
決定する手段を含むようにした点、又はざらtこ前記基
点決定手段は、前記の同一の併重こ属する基点が複数と
なったときは、該複数の各基点を前記輪郭追跡回路に対
し所定の側(例えば時計方向側)とする前記多角形の辺
が、前記重心lこ対し、で張る角度(辺対向角など)を
比較する手段を設けるととも(こ、前記角度のうち最大
のものに対応する基点を新たな基点として決定する手段
を含むようにした窯番こある。
feature point labeling means for sequentially assigning numbering to each feature point in the group so as to go around the contour once in a predetermined direction (for example, counterclockwise); If the base point determining means is set at a predetermined side (for example, a clockwise side) with respect to the contour tracing direction on both sides of the side having the maximum value among the six side lengths, ) including means for making the feature point located in ) the base point,
Alternatively, the base point determining means may include means for setting the feature point corresponding to the maximum of the respective center of gravity distances as the base point. When there are multiple base points belonging to the same group + C, each base point of the koji number is on the predetermined side (for example, clockwise all), and each side E of the maximum side length is set in the contour tracing direction; A predetermined side (for example, the counterclockwise side) to be referred to as a standard, and means for comparing the side lengths of the adjacent polygons are provided, and on both sides of the side having the maximum value among the adjacent side lengths, A point including a means for determining a feature point located on a predetermined side (for example clockwise 1!+) with respect to the contour tracing direction as a new base point, or a feature point located on a predetermined side (for example clockwise 1!+) with respect to the contour tracing direction, When there are multiple base points to which the multiple base points belong, the side of the polygon that makes each of the multiple base points a predetermined side (for example, clockwise side) with respect to the contour tracking circuit is opposite to the center of gravity l, The kiln number includes a means for comparing the angles (side facing angles, etc.) between the two angles, and a means for determining the base point corresponding to the largest of the angles as a new base point.

〔発明の実施例〕[Embodiments of the invention]

以下第1図、第2図を用いて本発明の詳細な説明する。 The present invention will be described in detail below with reference to FIGS. 1 and 2.

第1図は本発明の実施例の構成を示すブロック、図、第
2図は第1図に基づき、検出、番号付さnた対象部品の
2値化画像の輪郭上の特徴点の例を示す図である。
Fig. 1 is a block diagram showing the configuration of an embodiment of the present invention, and Fig. 2 is an example of feature points on the contour of a binary image of a detected and numbered target part based on Fig. 1. FIG.

第1図において、200は対象部品であり、201はこ
れを2次元的に撮像し、その画像を前記X、Y軸方向暑
こ走査して画像18号に変換する2次元撮像装置である
。202はA/D変換器で、対象部品200の表面IC
お1ブる明るさや、色相)こよって変化する、2次元撮
像装ff1201から出力されるアナログの映像信号を
適当なS/N比を得るレベル、または対象部品200I
こ応じたレベルでデジタル化(z値化)し、背景と対象
部品200の領域とを区分し、対象部品の2値化画像F
を得る回路である。203は輪郭追跡回路で、A/D変
換器202から得た2値化画像Fのデータ中の対象形状
のデータをもとに、対象形状の輪郭画素(輪郭点ともい
う)を一定方向(例えば反時計方向)+C追跡し1、各
輪郭点の座標値と各輪郭点)こ隣接する次輪郭点への方
向コードをもとめる回路である。この回路の詳細は本出
願人訃こよる特願昭58−9108号「パターンの輪郭
追跡方法」にその例が説明されている。204は特徴検
出回路で、輪郭追跡回路203から得られる情報をもと
蚤こ前記対象形状1cおける輪郭部の特徴点(角点P、
谷点N等)の位置を検出し、その検出順に記憶する回路
である。この回路の詳細は前記文献(1)にその例が説
明されている。205は特徴点番号付回路で、特徴検出
回路204で検出された各特徴点に対し、画面中の画像
Fの移動や回転に無関係となるような、一定の番号性を
行う回路であり、本発明の核心となるものである。20
6は重心算出回路で、特徴点番号付回路205の機能を
補強するために対象形状千重心を算出する。
In FIG. 1, 200 is a target part, and 201 is a two-dimensional imaging device that images this two-dimensionally, scans the image in the X and Y axis directions, and converts it into image No. 18. 202 is an A/D converter, and the surface IC of the target part 200
The level at which an appropriate S/N ratio can be obtained for the analog video signal output from the two-dimensional imaging device ff1201, which changes depending on the brightness and hue of the target component 200I
The area of the target part 200 is digitized (z-valued) at the corresponding level, the background and the area of the target part 200 are separated, and a binarized image F of the target part is created.
This is a circuit that obtains the following. 203 is a contour tracking circuit that moves contour pixels (also referred to as contour points) of the target shape in a certain direction (for example This is a circuit that performs counterclockwise tracking (1) and obtains the coordinate value of each contour point and the direction code to the next adjacent contour point. The details of this circuit are described in Japanese Patent Application No. 58-9108 entitled "Pattern Contour Tracing Method" by the present applicant, Koyoru Mori. 204 is a feature detection circuit which detects feature points (corner points P,
This is a circuit that detects the positions of valley points (N, etc.) and stores them in the order of detection. An example of this circuit is explained in detail in the above-mentioned document (1). Reference numeral 205 denotes a feature point numbering circuit, which assigns a fixed numbering to each feature point detected by the feature detection circuit 204 so as to be unrelated to the movement or rotation of the image F on the screen. This is the core of the invention. 20
Reference numeral 6 denotes a centroid calculation circuit which calculates the centroid of the object shape in order to reinforce the function of the feature point numbering circuit 205.

次に第2図を用いて1図の構成装置による各特徴点呼こ
対する番号性の動作を説明する。第2図では検出された
特徴点としての角点PをPi−P4で、同じく谷点Nを
N1−N3で表わし、である。
Next, using FIG. 2, the operation of numbering for each feature point call by the component device shown in FIG. 1 will be explained. In FIG. 2, the corner point P as a detected feature point is represented by Pi-P4, and the valley point N is similarly represented by N1-N3.

なおこの例では角点P、(P 1〜P4)と谷点N(N
1〜N8)とは、特徴点番号付回路205内の別々の手
段憂こより番号性を行っている。
Note that in this example, corner points P, (P 1 to P4) and valley points N (N
1 to N8), numbering is performed using separate means within the feature point numbering circuit 205.

まず谷点Nの番号性はシこついて説明する。対象形状(
2値化画像Fの外形の輪郭)から検出された谷点N1〜
N8を輪郭点の追跡方向(反時計方向)Aの順に結び多
角形近似を行い、同時―ここの多角形の各辺長を調・べ
、最大の辺長をもとめ、この最大辺長となる辺の所定方
向(この方向は前記追跡方向Aを基準としで定められる
。ここでは例えば時計方向)の側の一端を基点として、
所定方向(例えば反時計方向A)+こ順欠番号性を行う
First, the numerical nature of the valley point N will be explained in detail. Target shape (
Valley points N1~ detected from the outer contour of the binarized image F)
Connect N8 in the tracing direction of contour points (counterclockwise) in order of A to perform polygonal approximation, and at the same time check the length of each side of this polygon, find the maximum side length, and use this maximum side length. With one end of the side in a predetermined direction (this direction is determined based on the tracking direction A, here, for example, clockwise) as a base point,
A predetermined direction (for example, counterclockwise direction A) + missing numbering is performed in this order.

ただし第2図の例では最大辺長の辺aと、2番目の長さ
を持つ辺Cとの長さがほぼ等しく有意差をもって両者を
区別できないので、このような場合には例えば次の規則
を適用する。すなわち辺a。
However, in the example shown in Figure 2, the longest side a and the second longest side C are almost equal in length and cannot be distinguished from each other with a significant difference, so in such a case, for example, the following rule apply. That is, side a.

Clこ対しそれぞれ所定方向(例えば反時計方向A)の
(ti tこ隣接する辺す、dの辺長を比較し長い方の
辺(この場合d)を前記最大辺長tC置換えて、前記の
基点N1の決定及びこれfこ次ぐ各点N2〜N8の前記
順次番号性を行うものである。このよう)こ各特徴点に
番号性を行うこと−こより、画面中の画像十の前記移動
や回転に関係なく、一義的tこ番号性を行うことができ
る。
Compare the side lengths of adjacent sides d and Cl in a predetermined direction (for example, counterclockwise A), replace the longer side (d in this case) with the maximum side length tC, and The purpose is to determine the base point N1 and perform the sequential numbering of each of the points N2 to N8 following this point.Thus, by performing the numbering of each feature point, the movement of the images in the screen and Unambiguous t-numberality can be performed regardless of rotation.

次)こ角点P+ごついては、谷点Nと同様の方法で番号
性を行ってもよいし、また谷点N及び角点Pのすべての
特徴点を前記追跡方向Aにおける出現の順に結んで、谷
、角点を共曇こ含む多角形近似を行い、この新たな多角
形について前記と同様な方法により基点を求め、この基
点が谷点てあればNl。
Next) For this corner point P + bump, numbering may be done in the same way as for the valley point N, or by connecting all the feature points of the valley point N and corner point P in the order of appearance in the tracking direction A. , troughs, and corner points, and calculate the base point of this new polygon by the same method as described above. If this base point is a trough point, then Nl.

角点てあれはPIと定め、以後新たな多角形の各辺を反
時計方向(追跡方向A))こたどりつつ、谷点、角点の
種類別tこそれぞれ順番の番号性を行うこととしてもよ
い。
The corner point is defined as PI, and from now on, while tracing each side of the new polygon in a counterclockwise direction (tracing direction A), numbering is performed in order for each type of valley point and corner point. Good too.

ただし、第2図の角点P+こついては以下をζ述べるよ
うIC1さらtこ異なった方法で番号性を行う例を示す
。つまり両像Fの重心Gと各角点P(P1〜F4)との
距離(重心用^]Lともいう)のうち、最大距離にある
ものを求めて基点とし、以下前記と同様(反時計方向A
)に順次番号性を行う。ただし第2図の場合重心距離e
(角点P41c対応)。
However, regarding the corner point P+ in FIG. 2, an example of performing numbering in a different manner is shown below. In other words, among the distances (also called center of gravity ^] L) between the center of gravity G of both images F and each corner point P (P1 to F4), the maximum distance is found and used as the base point, and the following is the same as above (counterclockwise) Direction A
) to perform sequential numbering. However, in the case of Fig. 2, the center of gravity distance e
(Corresponding to corner point P41c).

とf(角点Firこ対応)がほぼ等し、く最大距離でた ある+め、さらに次のような規則を適用して基点を決定
する。
Since and f (corresponding to the corner point Fir) are approximately equal and the maximum distance is +, the base point is determined by applying the following rules.

すなわち前記最大圧1ζ(にある各角点P4.PIと各
角点P4.PLのそれぞれ−こ追跡方向A+こ隣接する
角点Pi、P2とを結ぶ辺P4.PI、Pi、P2が重
心G+こ対し、で作る角(辺対向角ともいう)α。
That is, the sides P4.PI, Pi, P2 connecting the corner points P4.PI and P4.PL located at the maximum pressure 1ζ (in the tracking direction A+) and the adjacent corner points Pi, P2 are the center of gravity G+ The angle (also called side-opposite angle) made by this and.

βを比較し、大きい方の角βに対応する角点をPlとし
て、以下前記と同様fこ角点P2〜P4の番号性を行う
ものである。
The angles β are compared, and the angle point corresponding to the larger angle β is set as Pl, and the numbering of the angle points P2 to P4 is performed in the same manner as described above.

〔発明の効果〕〔Effect of the invention〕

以上の説明から明らかなよう多こ、この発明によれば、
対象形状の輪郭上−こ検出された各特徴点のうち基点と
なるものを、各Qi徴点を前記輪郭に沿って結んだ多角
形の辺長、あるいは各特徴点と対象形状の重心との距離
など、画面中の対象形状の配置と無関係となる量に基づ
いて決定し、たつえ、他の特徴点を前記基点を出発点と
し、て前記輪郭に沿って所定方向tこ順次番号性を行う
こととし、だので、全ての特徴点の番号が対象形状の移
動や回転に拘わりなく決定でき、従って前記特徴点を操
作・もしくは加工の対象点としたときも対象部品の配置
と無関係■こ対象部品−こ必要な操作、加工を施すこと
が可能となる。
As is clear from the above description, according to this invention,
On the contour of the target shape - The base point of each detected feature point is defined as the side length of a polygon connecting each Qi feature point along the contour, or the distance between each feature point and the center of gravity of the target shape. It is determined based on a quantity, such as a distance, that is unrelated to the arrangement of the target shape in the screen, and other feature points are sequentially numbered in a predetermined direction t along the contour using the base point as a starting point. Therefore, the numbers of all feature points can be determined regardless of the movement or rotation of the target shape, and therefore, even when the feature points are used as target points for manipulation or processing, they are independent of the arrangement of the target parts. It becomes possible to perform necessary operations and processing on parts.

【図面の簡単な説明】[Brief explanation of the drawing]

第1図は本発明の実施例の構成を示すブロック図、第2
図は第1図の動作を説明するための対象部品の2値化画
像の例を示す図、第3図は従来技術に基づき検出された
2値化画像の輪郭上の特徴点の例を示す図である。 200・・・・・・対象部品、201・・・・・・2次
元撮像装置、202・・・・・・A/D変換器、203
・・・・・・輪郭追跡回路、204・・・・・・特徴検
出回路、205・・・・・・特徴点番号付回路、206
・・・・・・重心算出回路、P(P1〜P4)・・・・
・・角点、N(N1〜N!3)・・・・・・谷点、G・
・・・・・重心、a、a・・・・・・辺、e−g・・・
・・・重心距離、α、β・・・・・・辺対向角。 ↓ 第2図 ψ オ 3図
FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, and FIG.
The figure shows an example of a binarized image of the target part to explain the operation of Fig. 1, and Fig. 3 shows an example of feature points on the contour of the binarized image detected based on the conventional technique. It is a diagram. 200... Target part, 201... Two-dimensional imaging device, 202... A/D converter, 203
... Contour tracking circuit, 204 ... Feature detection circuit, 205 ... Feature point numbering circuit, 206
... Center of gravity calculation circuit, P (P1 to P4) ...
・Corner point, N (N1~N!3)・・・Trough point, G・
...center of gravity, a, a...side, e-g...
... Center of gravity distance, α, β ... Side opposite angle. ↓ Fig. 2 ψ O Fig. 3

Claims (1)

【特許請求の範囲】 1)対象となる2値化画像の輪郭画素を順次該2値化画
像を所定方向(以下輪郭追跡方向と呼ぶ)に1周するよ
うに追跡し、このとき得られる各輪郭画素の座標値と、
隣接する輪郭画素への追跡方向を示す追跡方向コードと
から、輪郭上の突起又はくぼみ等、所定の特徴部分を代
表する輪郭画素(以下特徴点と呼ぶ)の位置を、前記輪
郭を所定方向に1周する順に検出し、その順序で記憶す
る装置において、 前記特徴点をその全てからなる第1の群又はそれが共有
する性質で群別された複数の第2の群に区分するとき、 前記第1もしくは第2の群ごとに、前記輪郭を1周する
ように前記の同一の群内の各特徴点を順次結んで得られ
る多角形の各辺長を算出する辺長算出手段、又は(及び
)前記2値化画像の重心と前記の同一の群内の各特徴点
との距離(以下重心距離と呼ぶ)を算出する重心距離算
出手段と、算出された前記の各辺長又は(及び)各重心
距離に基づいて当該群内の特徴点を代表する点(以下基
点と呼ぶ)を決定する基点決定手段と、前記基点を出発
点とし、前記の順序に基づいて、前記輪郭を所定方向に
1周するように、当該群内の各特徴点に順次番号付を行
う特徴点ラベリング手段と、を備えたことを特徴とする
輪郭特徴点のラベリング装置。 2)特許請求範囲第1項に記載の装置において、前記基
点決定手段は前記各辺長のうち最大値を持つ辺の両側中
前記輪郭追跡方向に対し所定の側にある特徴点を前記基
点とする手段を含むことを特徴とする輪郭特徴点のラベ
リング装置。 3)特許請求範囲第1項又は第2項に記載の装置におい
て、前記基点決定手段は前記各重心距離のうち最大のも
のに対応する特徴点を前記基点とする手段を含むことを
特徴とする輪郭特徴点のラベリング装置。 4)特許請求範囲第2項又は第3項に記載の装置におい
て、前記基点決定手段は、前記の同一の群に属する基点
が複数となつたときは、該複数の各基点を前記の所定側
とする最大辺長の各辺に対し、前記輪郭追跡方向を基準
とする所定の側に、隣接する前記多角形の辺長を比較す
る手段を設けるとともに、前記隣接辺長のうち最大値を
持つ辺の両側中、前記輪郭追跡方向に対し所定の側にあ
る特徴点を新たな基点として決定する手段を含むことを
特徴とする輪郭特徴点のラベリング装置。 5)特許請求範囲第3項又は第4項に記載の装置におい
て、前記基点決定手段は、前記の同一の群に属する基点
が複数となつたときは、該複数の各基点を前記輪郭追跡
方向に対し所定の側とする前記多角形の辺が、前記重心
に対して張る角度を比較する手段を設けるとともに、前
記角度のうち最大のものに対応する基点を新たな基点と
して決定する手段を含むことを特徴とする輪郭特徴点の
ラベリング装置。
[Claims] 1) The contour pixels of the target binarized image are sequentially traced around the binarized image in a predetermined direction (hereinafter referred to as the contour tracking direction), and each of the contour pixels obtained at this time is The coordinate value of the contour pixel,
The position of a contour pixel (hereinafter referred to as a feature point) representing a predetermined characteristic portion such as a protrusion or depression on the contour is determined from the tracking direction code indicating the tracking direction to the adjacent contour pixel, and the contour is moved in a predetermined direction. In a device that detects feature points in the order of one round and stores them in that order, when classifying the feature points into a first group consisting of all of the feature points or a plurality of second groups grouped by characteristics that they share, Side length calculation means for calculating each side length of a polygon obtained by sequentially connecting each feature point in the same group so as to go around the contour once for each first or second group, or ( and) a centroid distance calculation means for calculating the distance between the centroid of the binarized image and each feature point in the same group (hereinafter referred to as centroid distance); ) base point determining means for determining a point (hereinafter referred to as a base point) representative of the feature points in the group based on each center-of-gravity distance; 1. A contour feature point labeling device comprising: feature point labeling means for sequentially numbering each feature point in the group so as to make one round. 2) In the apparatus according to claim 1, the base point determining means determines, as the base point, a feature point located on a predetermined side with respect to the contour tracing direction on both sides of the side having the maximum value among the respective side lengths. 1. A contour feature point labeling device comprising means for labeling contour feature points. 3) In the apparatus according to claim 1 or 2, the base point determining means includes means for setting a feature point corresponding to the largest one among the respective center-of-gravity distances as the base point. Labeling device for contour feature points. 4) In the apparatus according to claim 2 or 3, when a plurality of base points belong to the same group, the base point determining means selects each of the plurality of base points to the predetermined side. For each side of the maximum side length, a means for comparing the side lengths of the adjacent polygons is provided on a predetermined side with respect to the contour tracing direction, and the polygon has the maximum value among the adjacent side lengths. A labeling device for contour feature points, comprising means for determining a feature point located on a predetermined side with respect to the contour tracing direction as a new base point on both sides of the edge. 5) In the apparatus according to claim 3 or 4, when a plurality of base points belong to the same group, the base point determining means selects each of the plurality of base points in the contour tracking direction. means for comparing angles that sides of the polygon make with respect to the center of gravity, and means for determining a base point corresponding to the largest of the angles as a new base point. A contour feature point labeling device characterized by the following.
JP17708584A 1984-08-25 1984-08-25 Labeling device for contour feature point Granted JPS6154577A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP17708584A JPS6154577A (en) 1984-08-25 1984-08-25 Labeling device for contour feature point

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP17708584A JPS6154577A (en) 1984-08-25 1984-08-25 Labeling device for contour feature point

Publications (2)

Publication Number Publication Date
JPS6154577A true JPS6154577A (en) 1986-03-18
JPH0377549B2 JPH0377549B2 (en) 1991-12-10

Family

ID=16024861

Family Applications (1)

Application Number Title Priority Date Filing Date
JP17708584A Granted JPS6154577A (en) 1984-08-25 1984-08-25 Labeling device for contour feature point

Country Status (1)

Country Link
JP (1) JPS6154577A (en)

Also Published As

Publication number Publication date
JPH0377549B2 (en) 1991-12-10

Similar Documents

Publication Publication Date Title
US6047893A (en) Method of locating an object-applied optical code
JPH03294976A (en) Reference mark pattern detecting device
KR900002509B1 (en) Apparatus for recognizing three demensional object
JPS6154577A (en) Labeling device for contour feature point
JPS6073408A (en) Pattern recognizing device
JP3627249B2 (en) Image processing device
JPH02242382A (en) Defect checking method
JP2562047B2 (en) Position and orientation recognition method of target object
JPH065545B2 (en) Figure recognition device
JPH11132737A (en) Image processing method
JPH05334432A (en) Method for shaping thin line image
JP2897612B2 (en) Work detection method
JPH0624028B2 (en) Position measurement method of feature point corners for pattern recognition
JPS61286704A (en) Method for detecting boundary line of image
JPS6059736B2 (en) Item position detection method
JPS63131272A (en) Decision method for binarization threshold
JPH10275238A (en) Recognition method for fringe part of object
CN115937273A (en) Image alignment method, system, recording medium, and computer program product
CN117808751A (en) Defect detection method, device, electronic equipment and storage medium
CN117808752A (en) Defect detection method, device, electronic equipment and storage medium
JPH05266200A (en) Shape deciding method using chain code
JPH02179401A (en) Measuring method of object
JPH09311014A (en) Inspection apparatus for protruding part of semiconductor integrated circuit device
JPS63101972A (en) Detecting method for pattern
JPH0421193A (en) Device for recognizing object