JPS5953969A - Linear figure vector-converting system - Google Patents

Linear figure vector-converting system

Info

Publication number
JPS5953969A
JPS5953969A JP57164723A JP16472382A JPS5953969A JP S5953969 A JPS5953969 A JP S5953969A JP 57164723 A JP57164723 A JP 57164723A JP 16472382 A JP16472382 A JP 16472382A JP S5953969 A JPS5953969 A JP S5953969A
Authority
JP
Japan
Prior art keywords
points
frame
tracking
point
vector
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
JP57164723A
Other languages
Japanese (ja)
Other versions
JPH0420221B2 (en
Inventor
Hiroaki Harada
裕明 原田
Masashige Yamamoto
山本 正成
Mitsuo Ishii
石井 光雄
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.)
Fujitsu Ltd
Original Assignee
Fujitsu 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 Fujitsu Ltd filed Critical Fujitsu Ltd
Priority to JP57164723A priority Critical patent/JPS5953969A/en
Publication of JPS5953969A publication Critical patent/JPS5953969A/en
Publication of JPH0420221B2 publication Critical patent/JPH0420221B2/ja
Granted legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/002D [Two Dimensional] image generation
    • G06T11/20Drawing from basic elements, e.g. lines or circles

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)

Abstract

PURPOSE:To make a linear figure approximate easily to a vector, by extracting a tracking start point, setting a prescribed frame as to this point, tracking the contour line of a linear pattern in the frame, connecting the middle point of an intersection point to the frame side and the center of the frame as a result of tracking to set the vector. CONSTITUTION:A picture element data read out of a picture file 11 is stored temporarily in a buffer 12 and a register 13 in a tracking start point detecting part 25 of the upper stage through a bus, and tracking start points P, Q are detected by a deciding device 14. Subsequently, a buffer 15 in the contour line tracking part 26 is turned on, all data in the frame around the center R of the points P, Q are stored directly in a two-dimensional buffer 16 from the file 11, and tracking is executed clockwise or counter-clockwise from the points P, Q. Subsequently, the address of a contour point is sent to a picture element deciding part 19, tracking is executed to the frame side, and the middle point between the intersection points is set and the branch check is executed. Also, a vector data for connecting the points P, Q and the middle point between the intersection points is obtained by a vector-converting part 27.

Description

【発明の詳細な説明】 (す発明の技術分野 本発明は図面上の線図形のみからなる図形を2値画家と
して読取少データ圧縮のためベクトル化する方式に関す
るものである。
DETAILED DESCRIPTION OF THE INVENTION (Technical Field of the Invention) The present invention relates to a method for vectorizing a figure consisting only of line figures on a drawing as a binary painter in order to compress a small amount of data.

(2)従来技術と問題点 従来、CAD(計X機匣用設計法)関連の作業の中で設
計者が描いた手書き図面(回路図1機械設計図、フロー
チャート等)の情報と計算機内部表現に変換する部分が
最も自動化が遅れている。CADK限らず、一般図面の
読取シ、伝達、認識を行−なう場合、その図形の圧縮が
最も大きな問題である。
(2) Prior art and problems Conventionally, information on hand-drawn drawings (circuit diagrams, mechanical designs, flow charts, etc.) drawn by designers during CAD (design method for total x machine box) related work and computer internal representations The part that converts to When reading, transmitting, and recognizing not only CADK but also general drawings, the biggest problem is compression of the figures.

前述の各種図面のうちで、線図形のみから構成された回
路図、プリント板パター7図、機械設計図等を対象とし
た場合、データ圧縮の方法とじてはベクトル化方式が用
いられている。
Among the various drawings mentioned above, vectorization is used as a data compression method for circuit diagrams, printed board pattern diagrams, mechanical design drawings, etc. that are composed only of line figures.

直線2曲線から成る図形を有限数の有向線分(ベクトル
)で近似し、情報を圧縮する方式はこれらの図形の意味
付け、すなわちパターン認識を行なう上で有効なもので
ある。
A method of compressing information by approximating a figure consisting of a straight line and two curves with a finite number of directed line segments (vectors) is effective in giving meaning to these figures, that is, in pattern recognition.

第1図(α)の画像データを、同図<b)に示すように
有向線分(ベクトル)によシ近似することによシ、デー
タ量は格段に圧縮される。
By approximating the image data in FIG. 1 (α) by a directed line segment (vector) as shown in FIG. 1 (<b), the amount of data can be significantly reduced.

(3)発明の目的 本発明の目的は図面上の線図形のみから成る図形を2値
画像として読取りデータ圧縮のためベクトル化する場合
、少ないデータ量で処理の容易な線図形ベクトル化方式
を提供することである。
(3) Purpose of the Invention The purpose of the present invention is to provide a line graphic vectorization method that uses a small amount of data and is easy to process when a figure consisting only of line figures on a drawing is read as a binary image and vectorized for data compression. It is to be.

(4)発明の構成 前記目的を達成するため、本発明の線図形ベクトル化方
式は図面上の線図形のみから成る図形を2値画像データ
として読取シデータ圧縮のためベクトル化する方式にお
いて、図面上で走査線によp線の両側に対応する2点を
追跡開始点として検出する手段と、該2点間の第1の中
点を求め、これを中心として所定の枠を設定し、追跡開
始点よシ互に反対回シに輪郭線を追跡し枠辺との交点を
求める輪郭線追跡手段と、該枠辺との2交点間の第2の
中点を求め、前記第1の中点と第2の中点を結ぶベクト
ルを設定するベクトル化手段とを具え、以下用2の中点
を基準にして同様の枠を設定し、輪郭線追跡手段による
追跡とベクトル化手段によるベクトル設定を繰返し行な
い、線図形を逐次ベクトル化するととvil−特徴とす
るものである。
(4) Structure of the Invention In order to achieve the above-mentioned object, the line figure vectorization method of the present invention reads a figure consisting only of line figures on a drawing as binary image data and vectorizes it for data compression. means to detect two points corresponding to both sides of the p-line on the scanning line as the tracking start point, find the first midpoint between the two points, set a predetermined frame around this point, and start tracking. a contour tracing means that traces the contour line in opposite directions from point to point to find the point of intersection with the frame side, and a second midpoint between the two intersections with the frame side, and the first midpoint. and vectorization means for setting a vector connecting the second midpoint, and a similar frame is set based on the second midpoint in the following, and tracing by the contour tracing means and vector setting by the vectorization means are performed. If this is repeated and the line figure is sequentially converted into a vector, it becomes a vil-feature.

(5)発明の実施例 第2図(a)〜(c)は本発明の概略説明図である。(5) Examples of the invention FIGS. 2(a) to 2(c) are schematic illustrations of the present invention.

同図(α)は主要機能の流れ図でろ9.6段階に大別し
て繰返えされ、同図(6)はこれを図形で示したもので
ある。
Figure (α) is a flowchart of the main functions, which are roughly divided into 9.6 stages and repeated, and Figure (6) is a graphical representation of this.

第1段階では、図面上の線図形1に対し走査線によp走
査し追跡開始点を抽出する(■)。次に第2段階では輪
郭追跡に移シ、まず追跡開始点に関連して所定の枠2を
設定し、この枠内の線図形の輪郭線追跡を行なう(@)
。第6段階ではベクトル化であシ、輪郭線追跡の結果枠
2の辺との2交点の中点と枠の中心とを結びベクトルを
設定する(0)。次(枠辺との2交点の中点を中心とし
て枠を設定し、その枠内の線図形の輪郭線追跡を行ない
、同様にしてベクトルを設定する。なお後述する手法に
よ多分岐線図形のベクトルの設定も行ない、全体として
同図(6)の線図形1を同図(0)のベクトル6で近似
させることができる。
In the first step, the line figure 1 on the drawing is scanned by p scanning lines to extract the tracing start point (■). Next, in the second stage, we move on to contour tracing. First, we set a predetermined frame 2 in relation to the tracing start point, and perform contour tracing of the line figure within this frame (@)
. In the sixth step, vectorization is performed, and a vector is set by connecting the center of the frame and the midpoint of the two intersections with the sides of frame 2 as a result of contour tracing (0). Next (set a frame centered on the midpoint of the two intersections with the frame sides, trace the outline of the line figure within that frame, and set the vector in the same way. , and the line figure 1 in (6) of the same figure can be approximated as a whole by the vector 6 of (0) in the same figure.

第3図、第4図、第5図(a) + <b)は上述の追
跡開始点抽出と輪郭線追跡の段階の詳細説明図である。
FIG. 3, FIG. 4, and FIG. 5 (a) + <b) are detailed explanatory diagrams of the above-mentioned steps of extracting the tracing start point and tracing the contour line.

第6図tよ追跡開始点抽出の手順を示す。FIG. 6t shows the procedure for extracting the tracking start point.

いま、図面上の画素を白を60”、黒を11”を示すも
のとし、線図形1は黒”1”の連なシとして表現されて
いる。
Now, regarding the pixels on the drawing, white indicates 60" and black indicates 11", and line figure 1 is expressed as a series of black "1"s.

まず、走査線によって、白から黒へ変わる点Pと、黒か
ら白に変わる点Qとを検出する。このよりなP、 Qの
組はいくつか見つけられるが、このうちPQの距離があ
る一定値以下のものを採用する。
First, a point P where the color changes from white to black and a point Q where the color changes from black to white are detected using a scanning line. Several such pairs of P and Q can be found, but among them, the one whose distance of PQ is less than a certain value is adopted.

第4図は輪郭線追跡の手順を示す。FIG. 4 shows the procedure for contour tracing.

追跡開始点P、 Qが見つかった後、この2点から輪郭
線追跡を行なう。P、 Q点の中心Rを求めRを中心と
する所定の大きさの枠2を設定する。P点から反時計回
シに、Q点から時計回シに輪郭線を追跡し、枠2の枠辺
との交点Le、’Reに至った場合、そこで追跡を中止
する。交点Le、Reの中点をSとする。またPから時
計回シに、Q点から反時計回シに追跡し枠辺との交点L
′e、 Rhおよびその中点S′も記録する。そしてL
e、Reの中点SとRとを結び、またL’e 、 R’
eの中点S′と、Rとを結びそれぞれベクトルを作る。
After the tracing start points P and Q are found, contour tracing is performed from these two points. The center R of points P and Q is determined, and a frame 2 of a predetermined size with R as the center is set. The contour is traced counterclockwise from point P and clockwise from point Q, and when it reaches the intersections Le and 'Re with the frame sides of frame 2, the tracing is stopped there. Let S be the midpoint between the intersections Le and Re. Also, track from point P clockwise and counterclockwise from point Q to the intersection point L with the frame side.
'e, Rh and their midpoint S' are also recorded. And L
Connect midpoints S and R of e, Re, and also L'e, R'
Connect the midpoint S' of e and R to create vectors.

次に枠2に隣接し、Sを基準にした枠2′を設定し、L
e、Reから輪郭線追跡を行ない、次のベクトルを作る
Next, set a frame 2' adjacent to frame 2 and based on S, and
Contour tracing is performed from e and Re to create the next vector.

このようにしてできたベクトルの集合で線図形がベクト
ル図形として近似される。
A line figure is approximated as a vector figure by a set of vectors created in this way.

第5図(α) 、 Cb)は輪郭線追跡の手順で分岐、
屈曲がある場合の処理手順を示す。
Figure 5 (α), Cb) branches in the contour tracing procedure,
The processing procedure when there is a bend is shown.

同図(a)は分岐線図形の場合である。すなわち、追跡
開始点Ls 、 Rsから枠2内の線図形1に対し輪郭
線追跡を行なった結果、枠辺との交点Le■、 fle
■が線幅に比し大きく広がった場合には、枠2上を走査
して黒の端点■、■を検出し、端点■から時=を回シに
輪郭線追跡を行ない、端点■に一致すれば図のように分
岐していることが分る。
Figure (a) shows the case of a branch line figure. That is, as a result of contour tracing from the tracing start points Ls and Rs to the line figure 1 in the frame 2, the intersections with the frame sides Le■, fl
If ■ has expanded greatly compared to the line width, scan the frame 2 to detect the black end points ■ and ■, trace the outline from the end point ■ by turning the time =, and match the end point ■. You can see that it is branched as shown in the figure.

同図(b)の場合は、枠2上の走査の結果、同図(α)
と同様に黒の端点■、■、■が検出されても、R8から
、または端点■からの輪郭線追跡の結果Rs、■が結ば
れていれば黒の端点■、■の部分はノイズであることが
分る。
In the case of (b) in the same figure, the result of scanning on frame 2 is (α) in the same figure.
Similarly, even if black endpoints ■, ■, and ■ are detected, if Rs and ■ are connected as a result of contour tracing from R8 or endpoint ■, then the black endpoints ■, ■ are noise. I understand something.

第6図は本発明の実施例の構成説明図である。FIG. 6 is an explanatory diagram of the configuration of an embodiment of the present invention.

同図において、画像ファイル11にはベクトル化すべき
線図形のみから成る図形の2値画像データが格納されて
いる。
In the figure, an image file 11 stores binary image data of a figure consisting only of line figures to be vectorized.

上段には追跡開始点検出部25が、中段には輪郭線追跡
部26が、下段にベクトル化部27が設けられ、輪郭線
追跡部26の結果データがCPUに出力され、CPUか
らベクトル化部27に入力される。なお、画像ファイル
11に対する画像データの授受はCPUで制御されるこ
とは言うまでもな10 画像ファイル11から読出された画素データはバスを通
して上段の追跡開始点検出部25内のバッファ12とレ
ジスタ13に一旦格納され、判定器14により第3図に
示したように追跡開始点P、 Qが検出され、これによ
り輪郭線追跡部26内のバッファ15をオンとし、第4
図で説明したよりなRを中心とする枠2内の全データが
画像ファイル11から直接2次元バッファ16に格納さ
れ、追跡開始点P、 Qよシ時計方向または反時計方向
に追跡が行なわ、れる。すなわち、6×6画素の窓ケ2
次元バッファ16の追跡開始点で始まる輪郭線上に設定
し輪郭点を中心とし8近傍カウンタ20によシ調べ、次
の輪郭点のアドレスをレジスタ(X)17. (’i’
)18 に一旦格納して画素列足部19に送シ、輪郭点
追跡を枠辺まで行なう。この枠辺との交点データをCP
Uに送シ、交点間の中心設定と分岐チェックが行なわれ
る。
A tracing start point detection unit 25 is provided in the upper stage, a contour tracing unit 26 is provided in the middle stage, and a vectorization unit 27 is provided in the lower stage.The result data of the contour tracing unit 26 is output to the CPU, and the data is outputted from the CPU to the vectorization unit. 27. It goes without saying that the transfer of image data to and from the image file 11 is controlled by the CPU 10. The pixel data read from the image file 11 is temporarily stored in the buffer 12 and register 13 in the upper tracking start point detection unit 25 through the bus. The determination unit 14 detects the tracking start points P and Q as shown in FIG.
All the data within the frame 2 centered on the straight line R explained in the figure is stored directly from the image file 11 into the two-dimensional buffer 16, and tracking is performed clockwise or counterclockwise from the tracking starting points P and Q. It will be done. In other words, 6x6 pixel window 2
It is set on the contour line starting from the tracing start point of the dimensional buffer 16, and the 8-neighborhood counter 20 is checked with the contour point as the center, and the address of the next contour point is stored in the register (X) 17. ('i'
) 18 and then sent to the pixel row foot section 19, and contour point tracking is performed up to the frame edge. CP the intersection data with this frame side
Then, the center setting between the intersection points and the branch check are performed.

また、1枠内の処理終了によシ2次元バッファ16から
終了信号が出されベクトル化部27のバッファ21ヲオ
ンとし、CPUからのベクトル関連データをバッファ2
1とレジスタ22に一旦格納してから、   ・追跡開
始点と枠辺の交点間の中点を結ぶ新しいベクトルデータ
が4”4られる。そして角度算出器23で新しいベクト
ルの角度を計算し、分岐処理器24をそのまま通過させ
ベクトルを出力する。これに対し、第5図(α)に示す
ような分岐線図形の場合は、分岐処理器24に用意され
た2次元バッファにより前述の処理が行なわれ、分岐線
図形に対応したベクトルが別に作成され出力される。以
上の手順が枠毎に)繰返えされる。
Also, when the processing within one frame is completed, a termination signal is output from the two-dimensional buffer 16, the buffer 21 of the vectorization unit 27 is turned on, and vector-related data from the CPU is transferred to the buffer 2.
1 and stored in the register 22, new vector data connecting the midpoint between the tracing start point and the intersection of the frame sides is created. Then, the angle calculator 23 calculates the angle of the new vector and branches. It passes through the processor 24 as it is and outputs a vector.On the other hand, in the case of a branch line figure as shown in FIG. Then, a vector corresponding to the branch line figure is separately created and output.The above procedure is repeated for each frame.

(6)発明の詳細 な説明したように、本発明によれば、追跡開始点を抽出
し、この点に関連して所定の枠を設足し、この枠内の線
図形の輪郭線追跡を行ない、追跡の結果枠辺との交点間
の中点と枠の中心とを結びベクトルを設定する。
(6) As described in detail, according to the present invention, a tracing start point is extracted, a predetermined frame is established in relation to this point, and the outline of a line figure within this frame is traced. As a result of tracking, a vector is set by connecting the midpoint between the intersections with the frame sides and the center of the frame.

このようにベクトルの設定に必要なデータは最小限に減
少するから、簡単迅速に線図形をベクトルに近似させる
ことが可能となる。
In this way, the data necessary for setting a vector is reduced to a minimum, so it is possible to easily and quickly approximate a line figure to a vector.

第2図(a、)〜(C)は本発明の概略説明図、第3図
〜第5図(α) 、 (b)は第2図の要部の詳細説明
図、第6図は本発明の実施例の構成説明図であり、図中
、1は線図形、2は枠、3はベクトル、4はノイズ、1
1は画像ファイル、12.15.21はバッファ、16
゜17、18.22はレジスタ、14は判定器、19は
画素判定器、20は8近傍カウンタ、26は角度算出器
、24は分岐処理器、25は追跡開始点演出部、26は
輪郭線追跡部、27はベクトル化部を示す。
Figures 2 (a,) to (C) are schematic explanatory diagrams of the present invention, Figures 3 to 5 (α), and (b) are detailed explanatory diagrams of the main parts of Figure 2, and Figure 6 is the main part of the present invention. It is a configuration explanatory diagram of an embodiment of the invention, in which 1 is a line figure, 2 is a frame, 3 is a vector, 4 is noise, 1
1 is the image file, 12.15.21 is the buffer, 16
17, 18, 22 are registers, 14 is a judge, 19 is a pixel judge, 20 is an 8-neighborhood counter, 26 is an angle calculator, 24 is a branch processor, 25 is a tracking start point production unit, 26 is a contour line A tracking section 27 indicates a vectorization section.

特許出願人 富士通株式会社 復代理人 弁理士  1)坂 善 重 第1図 (a)                (b)(8)
   第2図 ■ 83図 第4図 L−J b
Patent applicant Fujitsu Ltd. Sub-agent Patent attorney 1) Yoshishige Saka Figure 1 (a) (b) (8)
Figure 2 ■ Figure 83 Figure 4 L-J b

Claims (1)

【特許請求の範囲】[Claims] (1)図面上の線図形のみからなる図形を2値画像デー
タとして読取少データ圧縮のためベクトル化する方式に
おいて、図面上で走査腺によシ線の両側に対応する2点
を追跡開始点として検出する手段と、該2点間の第1の
中点を求め、これを中心として所定の枠を設定し、追跡
開始点よp互に反対回シに輪郭線を追跡し枠辺との交点
を求める輪郭線追跡手段と、該枠辺との29.点間の第
2の中点を求め、前記第1の中点と第2の中点を結ぶベ
クトルを設定するベクトル化手段とを具え、以下第2の
中点を基準にして同様の枠を設定し、輪郭線追跡手段に
よる追跡とベクトル化手段によるベクトル設定を繰返し
行ない、?IM図形を逐次ベクトル化することを%敵と
する線図形ベクトル化方式。 (2ン  前記輪郭線追跡手段により追跡の結果、枠辺
との2父点が線幅に対し大きく広がった場合は枠辺上を
走査してa幅に対応する未追跡点を検出し、該未追跡点
よシ追跡を行なうことにょシ分岐線図形金ベクトル化す
ることを特徴とする特許請求の範囲第1項記載の線図形
ベクトル化方式。
(1) In the method of reading figures consisting only of line figures on a drawing as binary image data and converting them into vectors for small data compression, the two points corresponding to both sides of the scanning line on the drawing are the starting points for tracing. A first midpoint between the two points is found, a predetermined frame is set around this point, and the contour is traced in opposite directions from the tracing start point to the edges of the frame. 29. Contour tracing means for finding intersection points with the frame sides. vectorizing means for determining a second midpoint between the points and setting a vector connecting the first midpoint and the second midpoint; , and then repeatedly perform tracking using the contour tracing means and vector setting using the vectorization means. A line figure vectorization method that targets successive vectorization of IM figures. (2) As a result of tracing by the contour tracing means, if the two points with the frame side have expanded greatly relative to the line width, scan the frame side to detect untracked points corresponding to the width a, and 2. A method for converting line figures into vectors according to claim 1, characterized in that branch line figures are converted into vectors by performing tracking on untracked points.
JP57164723A 1982-09-21 1982-09-21 Linear figure vector-converting system Granted JPS5953969A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP57164723A JPS5953969A (en) 1982-09-21 1982-09-21 Linear figure vector-converting system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP57164723A JPS5953969A (en) 1982-09-21 1982-09-21 Linear figure vector-converting system

Publications (2)

Publication Number Publication Date
JPS5953969A true JPS5953969A (en) 1984-03-28
JPH0420221B2 JPH0420221B2 (en) 1992-04-02

Family

ID=15798666

Family Applications (1)

Application Number Title Priority Date Filing Date
JP57164723A Granted JPS5953969A (en) 1982-09-21 1982-09-21 Linear figure vector-converting system

Country Status (1)

Country Link
JP (1) JPS5953969A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6190273A (en) * 1984-10-01 1986-05-08 テクトロニックス・インコーポレイテッド Data structuring method
JPS61208168A (en) * 1985-03-12 1986-09-16 Toshiba Corp Drawing reader

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6190273A (en) * 1984-10-01 1986-05-08 テクトロニックス・インコーポレイテッド Data structuring method
JPH0338620B2 (en) * 1984-10-01 1991-06-11 Tektronix Inc
JPS61208168A (en) * 1985-03-12 1986-09-16 Toshiba Corp Drawing reader

Also Published As

Publication number Publication date
JPH0420221B2 (en) 1992-04-02

Similar Documents

Publication Publication Date Title
TWI726422B (en) Two-dimensional code recognition method, device and equipment
JP4738469B2 (en) Image processing apparatus, image processing program, and image processing method
JP2542947B2 (en) Image processing device
CN114626118A (en) Building indoor model generation method and device
JPS5953969A (en) Linear figure vector-converting system
CN114495132A (en) Character recognition method, device, equipment and storage medium
CN112464753B (en) Method and device for detecting key points in image and terminal equipment
Guo et al. A fast page outline detection and dewarping method based on iterative cut and adaptive coordinate transform
JP2885476B2 (en) Image processing method and apparatus
JP2001243479A (en) Method and device for image processing and recording medium having image processing program recorded thereon
JPH0420228B2 (en)
KR20210080790A (en) Method of Brush Drawing Effect Data Generation related to Robot Art
JPS622382A (en) Feature extracting devie for pattern
JPH0656625B2 (en) Feature extraction method
JPH0462107B2 (en)
JPS6038755B2 (en) Feature extraction method
JP2803224B2 (en) Pattern matching method
JP2868533B2 (en) Drawing reading device
CN112464753A (en) Method and device for detecting key points in image and terminal equipment
JPH0545991B2 (en)
JPH11120366A (en) Segment contiguity relation determining method
JPH02264374A (en) Broken line extracting method
JPH02266478A (en) Method for recognizing drawing
JPS5837780A (en) Character recognizing method
JPS6262392B2 (en)