JPH0623751B2 - Base organization identification method - Google Patents
Base organization identification methodInfo
- Publication number
- JPH0623751B2 JPH0623751B2 JP63024252A JP2425288A JPH0623751B2 JP H0623751 B2 JPH0623751 B2 JP H0623751B2 JP 63024252 A JP63024252 A JP 63024252A JP 2425288 A JP2425288 A JP 2425288A JP H0623751 B2 JPH0623751 B2 JP H0623751B2
- Authority
- JP
- Japan
- Prior art keywords
- value
- density
- pixels
- base
- tissue
- 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 - Fee Related
Links
Landscapes
- Investigating And Analyzing Materials By Characteristic Methods (AREA)
Description
【発明の詳細な説明】 産業上の利用分野 本発明は鋳鉄管を形成するダクタイル鋳鉄などの基地組
織判別方法に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for determining a base structure of ductile cast iron or the like for forming a cast iron pipe.
従来の技術 従来、鋳鉄管などの製造工程においては、鋳造された製
品の基地組織の検査を行っており、この検査は、試供さ
れる基地組織中の黒鉛とパーライトを判別し、黒鉛球状
化率、パーライト量、セメンタイト量を計測し、計測さ
れた各値に基づいて基地組織の良否を判定していた。Conventional technology Conventionally, in the manufacturing process of cast iron pipes, etc., the matrix structure of the cast product is inspected, and this inspection discriminates between graphite and pearlite in the matrix structure to be tested, and the graphite spheroidization rate The amount of pearlite and the amount of cementite were measured, and the quality of the base structure was judged based on the measured values.
発明が解決しようとする課題 しかし、試供される基地に対して行うエッチングの度合
や、基地に対して照射する光量の多少により、基地組織
中の黒鉛とパーライトとのコントラストおよび明暗が影
響される。このために、コントラストと明暗に依処して
行う黒鉛とパーライトの判別が困難となり、このこと
が、基地組織の検査の判断に誤りを生む原因となり、問
題点とされていた。However, the contrast and brightness of graphite and pearlite in the matrix are affected by the degree of etching performed on the sample to be tested and the amount of light applied to the matrix. For this reason, it is difficult to distinguish between graphite and pearlite depending on contrast and brightness, which causes an error in the determination of the inspection of the matrix structure, which has been a problem.
本発明は上記問題点を解決するもので、エッチングや光
量の度合に影響されずに、すなわちコントラストと明暗
に依らずに、基地を構成する組織の判別を行える基地組
織判別方法を提供することを目的とする。The present invention solves the above problems and provides a base tissue discrimination method capable of discriminating a tissue constituting a base without being affected by the degree of etching or the amount of light, that is, without depending on contrast and brightness. To aim.
課題を解決するための手段 上記問題点を解決するために、本発明は、試供される基
地の表面を区画して判別対象の画面と成し、この判別対
象の両面を複数に分割して画素と成し、隣り合う画素の
濃淡の比較により得られる濃度の差値を全画面にわたっ
て測定し、測定される差値を、比較した画素のうち濃度
の濃い側の濃度値ごとに和算し、この和算した値を各濃
度値ごとの画素数で割った値を各濃度値の頻度と成す微
分ヒストグラムを形成し、頻度の大きい濃度値を閾値と
して各画素の組織を判別する構成としたものである。Means for Solving the Problems In order to solve the above-mentioned problems, the present invention divides the surface of a base to be tested into a screen to be discriminated, and divides both surfaces of the discriminant to be divided into a plurality of pixels. The difference value of the density obtained by comparing the shades of adjacent pixels is measured over the entire screen, and the measured difference value is summed for each density value on the darker side of the compared pixels, A structure in which a differential histogram in which the value obtained by dividing the summed value by the number of pixels for each density value is the frequency of each density value is formed, and the tissue of each pixel is discriminated using the density value with a high frequency as a threshold value Is.
作用 上記構成において、各画素を構成する組織は、同組織の
間において濃度の差値が小さく、異組織の間において濃
度の差値が大きくなる。このために、各画素の比較によ
って得られる差値を、比較した画素のうち濃度の濃い側
の濃度値ごとに和算し、その値を各濃度値ごとの画素数
で割ると異組織の境界での濃度値の頻度が大きくなる。
したがって、この濃度値が、濃い濃度を有する組織を判
別する織値となる。そして、頻度のピークは、各組織ご
との境界にあらわれるので、各ピークの頻度を与える濃
度値が、各組織を判別する閾値となる。Action In the above-mentioned configuration, the tissue constituting each pixel has a small difference in concentration between the same tissues and a large difference in concentration between different tissues. For this reason, the difference value obtained by comparing each pixel is summed for each density value on the darker density side of the compared pixels, and the value is divided by the number of pixels for each density value to determine the boundary of different tissues. The frequency of the density value at becomes large.
Therefore, this density value becomes a weave value for discriminating a tissue having a high density. Since the frequency peak appears at the boundary of each tissue, the concentration value giving the frequency of each peak becomes the threshold value for discriminating each tissue.
実施例 以下、本発明の一実施例を図面に基づいて説明する。第
１図において、管１はダクタイル鋳鉄で形成されてお
り、この管１に当接して自動研磨装置２が配置されてい
る。また自動研磨装置２の近傍にはエッチング装置３が
配置されている。そして、管１の基地を画像に撮り込む
テレビカメラ４が、画像メモリー５を介してＣＰＵ６に
電気的に接続されている。また、画像メモリー５にはモ
ニターＴＶ７が電気的に接続されている。Embodiment An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, a pipe 1 is made of ductile cast iron, and an automatic polishing device 2 is arranged in contact with the pipe 1. An etching device 3 is arranged near the automatic polishing device 2. A television camera 4 that captures an image of the base of the tube 1 is electrically connected to the CPU 6 via the image memory 5. A monitor TV 7 is electrically connected to the image memory 5.
以下、上記構成における作用について説明する。まず、
管１の基地の表面を自動研磨装置２により研磨し、その
後エッチング装置３により腐食させる。このとき、第２
図に示すように、基地を構成する組織には黒鉛Ｘとパー
ライトＹとフェライトＺがあり、これらの組織の濃淡の
境界は、黒鉛ＸとパーライトＹの間、およびパーライト
ＹとフェライトＺとの間において存在する。そして、エ
ッチング最適の状態においては、各組織の間における濃
淡差が同程度となる。The operation of the above configuration will be described below. First,
The surface of the base of the tube 1 is polished by the automatic polishing device 2 and then corroded by the etching device 3. At this time, the second
As shown in the figure, there are graphite X, pearlite Y, and ferrite Z in the structure that constitutes the matrix, and the shade boundaries of these structures are between graphite X and pearlite Y, and between pearlite Y and ferrite Z. Exists in. Then, in the optimum state of etching, the difference in density between the tissues is about the same.
次に、研磨し、エッチングした基地の表面をテレビカメ
ラ４で撮像し、撮像された画像を判別対象画面として画
像メモリー５に記録する。このとき画像メモリー５にお
いては、判別対象画面を複数に分割して画素となし、各
画素の濃淡階調を、複数段階に設定された濃度値として
記録する。ちなみに、画像メモリー５の仕様は濃淡階調
64、画素数 512×512 、画面数４である。そして、ＣＰ
Ｕにおいて各画素の濃度の差値を全画面にわたって測定
し、測定される差値を、比較した画素のうち濃度の濃い
側の濃度値ごとに和算し、この和算した値を各濃度ごと
の画素数で割った値を各濃度値の頻度と成す微分ヒスト
グラムを形成する。この微分ヒストグラムＱＳ(i) を求
める式を下記に示す。Next, the surface of the polished and etched base is captured by the television camera 4, and the captured image is recorded in the image memory 5 as a discrimination target screen. At this time, in the image memory 5, the discrimination target screen is divided into a plurality of pixels to form pixels, and the gray scale of each pixel is recorded as a density value set in a plurality of stages. By the way, the specification of the image memory 5 is gray scale.
It has 64 pixels, 512 x 512 pixels, and 4 screens. And CP
In U, the difference value of the densities of each pixel is measured over the entire screen, and the measured difference value is summed for each density value on the darker side of the compared pixels, and the summed value is calculated for each density. A differential histogram in which the value divided by the number of pixels of is the frequency of each density value is formed. The formula for obtaining this differential histogram QS (i) is shown below.
ＱＳ(i) ，ｊ，ｉ濃度値（ｉ＝０〜63） ａ（ｉ，ｊ）（各画素の濃度値） Δａ＝ａ（ｉ＋１，ｊ）−ａ（ｉ，ｊ） ＤＳ（qa）＝ＤＳ（qa）＋Δａ そして、ＤＳ（qa）を全画面について計測し、 ＱＳ(i) ＝ＤＳ(i) ／ＨＳ(i) を求める。ただし、ＨＳ(i) は濃度ヒストグラムであ
る。このとき、各組織間の濃度差は同組織の間ににおい
て小さく、異組織の間において大きくなる。このため
に、比較によって得られる濃度の差値を、比較した画素
のうち濃度の濃い側の濃度値ごとに和算することによ
り、濃度の濃い組織が有するもっとも多い濃度値の頻度
が、全画面に分布する濃度値のうちでもっとも大きくな
る。したがって、この頻度がもっとも大きい濃度値が、
濃い組織すなわち黒鉛を判別する閾値となる。そして、
頻度のピークは、各組織の境界ごとにあらわれるので、
各ピークの頻度を与える濃度値が、残りの組織パーライ
トとフェライトを判別する閾値となる。ちなみに、第３
図は微分ヒストクラムの一例を示すものであり、ＰＬは
黒鉛を判別するための閾値ＰＬ＝27を与える頻度のピー
クである。また、パーライトとフェライトを判別するた
めの閾値ＰＵを与えるピークが明確でない場合には、Ｐ
Ｌ＜について再度微分ヒストグラム（圧間微分ヒストグ
ラム）を求める。第４図は圧間微分ヒストグラムであ
り、ＰＵ＝42が明示されている。なお、第２図におい
て、濃度はｉ＝０でまっ黒、ｉ＝63でまっ白となる。そ
して、Ａ部が黒鉛の画素量を示し、Ｂ部がパーライトの
画素量を示している。またＰＬ＜について微分ヒストグ
ラムを求めることは、黒鉛を除いた組織を対象とするこ
とになり、このことによってパーライトとフェライトの
境界が求められることとなる。したがって、各組織を閾
値に基づいて判別し、各々の組織を２値化によって抽出
し、黒鉛の球状化率、パーライト量フェライト量を計測
することができる。したがって、微分ヒストグラムから
導いた閾値に基づいて判別対象画面を二値化することに
より、エッチングの程度や光量の変動に左右されずに、
基地を構成する組織の判別を行える。QS (i), j, i density value (i = 0 to 63) a (i, j) (density value of each pixel) Δa = a (i + 1, j) -a (i, j) DS (qa) = DS (qa) + Δa Then, DS (qa) is measured for all screens, and QS (i) = DS (i) / HS (i) is obtained. However, HS (i) is a density histogram. At this time, the concentration difference between the tissues is small between the same tissues and is large between the different tissues. For this reason, the difference value of the density obtained by the comparison is summed for each density value on the darker density side of the compared pixels, so that the frequency of the most dense density value that the dense tissue has is It is the largest of the concentration values distributed in. Therefore, the density value with the highest frequency is
It is a threshold value for distinguishing a dense structure, that is, graphite. And
Frequency peaks appear at the boundaries of each tissue, so
The concentration value giving the frequency of each peak becomes the threshold value for distinguishing the remaining tissue pearlite and ferrite. By the way, the third
The figure shows an example of the differential histocrum, and P L is the peak of the frequency that gives the threshold value P L = 27 for discriminating graphite. If the peak giving the threshold P U for distinguishing between pearlite and ferrite is not clear, P
The differential histogram (pressure differential histogram) is obtained again for L <. FIG. 4 is a pressure differential histogram, which clearly shows P U = 42. In FIG. 2, the density is pure black when i = 0 and pure white when i = 63. Then, the portion A shows the pixel amount of graphite, and the portion B shows the pixel amount of pearlite. Further, obtaining the differential histogram for P L <is intended for the structure excluding graphite, and thus the boundary between pearlite and ferrite is obtained. Therefore, it is possible to determine each structure based on the threshold value, extract each structure by binarization, and measure the spheroidization rate of graphite and the amount of pearlite and ferrite. Therefore, by binarizing the discrimination target screen based on the threshold value derived from the differential histogram, regardless of the degree of etching or the fluctuation of the light amount,
The organization of the base can be identified.
ちなみに、第１表は光量変化試験の結果を示すものであ
り、第１表より明らかなように、光量に応じて適切な閾
値が決定されている。Incidentally, Table 1 shows the results of the light quantity change test, and as is clear from Table 1, an appropriate threshold value is determined according to the light quantity.
第２表にエッチング時間を変化させた場合の結果を示す
が、エッチング時間に応じて11〜18秒の範囲ならば適切
な閾値が決定されている。 Table 2 shows the results when the etching time was changed, and an appropriate threshold value was determined within the range of 11 to 18 seconds according to the etching time.
発明の効果 以上述べたごとく、本発明によれば、各画素の濃度の差
値を頻度として形成される微分ヒストグラムから閾値を
決定し、この閾値により各組織を判別することによっ
て、エッチングの程度や光量の変動に左右されずに、基
地を構成する組織を判別することができる。 EFFECTS OF THE INVENTION As described above, according to the present invention, the threshold value is determined from the differential histogram formed by using the difference value of the density of each pixel as the frequency, and each tissue is discriminated by the threshold value, so that the degree of etching and It is possible to discriminate the tissues that make up the base without being affected by the fluctuation of the light amount.
第１図は本発明の一実施例を示す全体構成図、第２図は
基地組織を示す図、第３図は微分ヒストグラム、第４図
は圧間微分ヒストグラムである。 １……管、４……テレビカメラ、５……画像メモリー、
６……ＣＰＵ、７……モニターＴＶ。FIG. 1 is an overall configuration diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing a base organization, FIG. 3 is a differential histogram, and FIG. 4 is a pressure differential histogram. 1 ... Tube, 4 ... TV camera, 5 ... Image memory,
6 ... CPU, 7 ... monitor TV.
Claims (1)
の画面と成し、この判別対象の両面を複数に分割して画
素と成し、隣り合う画素の濃淡の比較により得られる濃
度の差値を全画面にわたって測定し、測定される差値
を、比較した画素のうち濃度の濃い側の濃度値ごとに和
算し、この和算した値を各濃度値ごとの画素数で割った
値を各濃度値の頻度と成す微分ヒストグラムを形成し、
頻度の大きい濃度値を閾値として各画素の組織を判別す
ることを特徴とする基地組織判別方法。1. A density obtained by dividing the surface of a base to be tested into a screen to be discriminated, dividing both surfaces of the discrimination target into a plurality of pixels, and comparing the shades of adjacent pixels. The difference value of is measured over the entire screen, the measured difference value is summed for each density value on the darker side of the compared pixels, and the summed value is divided by the number of pixels for each density value. Form a differential histogram of the measured value and the frequency of each density value,
A base tissue discriminating method characterized by discriminating a tissue of each pixel using a density value having a high frequency as a threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP63024252A JPH0623751B2 (en) | 1988-02-03 | 1988-02-03 | Base organization identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP63024252A JPH0623751B2 (en) | 1988-02-03 | 1988-02-03 | Base organization identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPH01197656A JPH01197656A (en) | 1989-08-09 |
| JPH0623751B2 true JPH0623751B2 (en) | 1994-03-30 |
Family
ID=12133056
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP63024252A Expired - Fee Related JPH0623751B2 (en) | 1988-02-03 | 1988-02-03 | Base organization identification method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0623751B2 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7622534B2 (en) * | 2021-04-21 | 2025-01-28 | 株式会社ジェイテクト | Polishing condition determination device, polishing condition determination method, and quality evaluation device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS59166863A (en) * | 1983-03-14 | 1984-09-20 | Komatsu Ltd | Fracture analyzing method of cast iron |
| JPS60104258A (en) * | 1983-11-10 | 1985-06-08 | Daihatsu Motor Co Ltd | Device for measuring spheroidization rate of graphite particle in spheroidal graphite cast iron |
| JPS60143769A (en) * | 1983-12-29 | 1985-07-30 | Kawasaki Steel Corp | Particle size and second phase fraction measuring apparatus |
-
1988
- 1988-02-03 JP JP63024252A patent/JPH0623751B2/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| JPH01197656A (en) | 1989-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7158674B2 (en) | Scene change detection apparatus | |
| CN1981514B (en) | A Method for Dynamically Identifying and Correcting Bad Pixels | |
| CN100582753C (en) | Methods of Detecting Defects | |
| WO2000039329A1 (en) | Method of counting microorganisms and device for accomplishing the counting | |
| JP3279868B2 (en) | Defect inspection method and device for inspected pattern | |
| JP3494429B2 (en) | Crack measurement method for structure surface | |
| KR100543467B1 (en) | Fine pattern line width measuring method and system | |
| JP3333568B2 (en) | Surface defect inspection equipment | |
| KR20010098460A (en) | Appearance inspection method and appearance inspection apparatus having high inspection processing speed | |
| JPH0623751B2 (en) | Base organization identification method | |
| US6888957B1 (en) | System and method for detecting blobs on an end surface of an optical fiber | |
| JP2004279239A (en) | Display defect detection method and display defect inspection device for liquid crystal panel | |
| Wang et al. | Illumination normalization of retinal images using sampling and interpolation | |
| JPH06116914A (en) | Method and apparatus for diagnosing coating film deterioration | |
| US6455832B1 (en) | Defect discriminating method and device for solid-state detector | |
| JPS60205683A (en) | Method and device for discrminating external appearance | |
| JP2743509B2 (en) | Method for determining binarization threshold of floc measuring device | |
| JP3246375B2 (en) | Method and apparatus for detecting gas distribution in blast furnace | |
| JP3216669B2 (en) | Optical disk defect inspection method | |
| JPH11119684A (en) | Display panel defect inspection equipment | |
| JPH08235543A (en) | Method and apparatus for detecting defects such as dirt | |
| Wu | Advanced image analysis techniques for quantifying carpet appearance | |
| JPH0781962B2 (en) | Foreign object detection method | |
| Yuting et al. | Blind Detection of Partial-Color-Manipulation Based on Self-PRNU Estimation | |
| JPS63200278A (en) | Image quality inspection apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| LAPS | Cancellation because of no payment of annual fees |