JP5823794B2 - Metal plate appearance evaluation method - Google Patents
Metal plate appearance evaluation method Download PDFInfo
- Publication number
- JP5823794B2 JP5823794B2 JP2011209303A JP2011209303A JP5823794B2 JP 5823794 B2 JP5823794 B2 JP 5823794B2 JP 2011209303 A JP2011209303 A JP 2011209303A JP 2011209303 A JP2011209303 A JP 2011209303A JP 5823794 B2 JP5823794 B2 JP 5823794B2
- Authority
- JP
- Japan
- Prior art keywords
- polishing
- polishing unit
- appearance
- value
- metal plate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Description
本発明は、金属板の外観評価方法に関する。 The present invention relates to a method for evaluating the appearance of a metal plate.
従来、例えば鉄道車両の外板として用いられる金属板では、予めベルトグラインダ等による研磨が表面に施されている場合がある。例えば特許文献1に記載の鉄道車両構体では、2枚の板状部材の重ね合わせ部分をレーザビームによって加熱溶融して接合する方法において、レーザビームによる溶接線方向と研磨目とがほぼ平行な方向となるように、予め外板の表面に研磨加工を施すことが記載されている。 Conventionally, for example, in a metal plate used as an outer plate of a railway vehicle, the surface may be previously polished by a belt grinder or the like. For example, in the railway vehicle structure described in Patent Document 1, in a method in which the overlapped portions of two plate-like members are joined by heating and melting with a laser beam, the direction of the welding line by the laser beam and the polishing line are substantially parallel to each other. Thus, it is described that the surface of the outer plate is polished in advance.
金属板における研磨部は、研磨目自体が意匠性を有している。また、そのため、金属板を研磨した後、その一部が著しく変化した場合などに修復のための再研磨を行う場合もある。しかしながら、現状では、研磨部及び再研磨部の仕上がり具合(外観)の評価は目視に頼っており、評価者に熟練した技能が要求されているという問題がある。 As for the grinding | polishing part in a metal plate, the grinding | polishing eyes itself have the designability. For this reason, after polishing the metal plate, re-polishing for repair may be performed when a part of the metal plate changes significantly. However, under the present circumstances, the evaluation of the finish (appearance) of the polishing part and the re-polishing part relies on visual observation, and there is a problem that skill required by the evaluator is required.
本発明は、上記課題の解決のためになされたものであり、研磨部及び再研磨部の仕上がり具合を簡単な手法で定量的に評価できる金属板の外観評価方法を提供することを目的とする。 The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a method for evaluating the appearance of a metal plate that can quantitatively evaluate the finished condition of a polishing part and a re-polishing part by a simple method. .
上記課題の解決のため、本発明に係る金属板の外観評価方法は、所定方向に研磨されてなる研磨部と、研磨部が所定方向に更に研磨されてなり、当該研磨部よりも細かい研磨目を有する再研磨部とを表面に有する金属板の外観評価方法であって、金属板の表面をスキャナで走査し、表面のデジタル画像を取得する画像取得工程と、デジタル画像を構成する各ピクセルのRGB値をグレースケール変換して得られるAc値を行方向及び列方向についてそれぞれ算出し、行方向のピクセル位置に対するAc値の波形パターンと、列方向のピクセル位置に対するAc値の波形パターンとに基づいて、研磨部及び再研磨部の研磨方向がスキャナの走査方向と一致しているデジタル画像を選別する画像選別工程と、研磨部及び再研磨部が含まれるように画像選別工程で選別されたデジタル画像の一部画像を抽出する画像抽出工程と、一部画像を構成する各ピクセルのRGB値をグレースケール変換して得られるAc値を研磨部及び再研磨部における研磨方向に垂直な方向について算出し、研磨方向のピクセル位置に対するAc値の波形パターンに基づいて、研磨部及び再研磨部の外観の可否を判断する外観判断工程と、を備えたことを特徴としている。 In order to solve the above-mentioned problems, a metal plate appearance evaluation method according to the present invention includes a polishing portion polished in a predetermined direction, and the polishing portion is further polished in a predetermined direction, and has a finer mesh than the polishing portion. A method for evaluating the appearance of a metal plate having a re-polished portion on the surface thereof, the image acquisition step of scanning the surface of the metal plate with a scanner to acquire a digital image of the surface, and for each pixel constituting the digital image Ac values obtained by converting the RGB values to gray scale are calculated in the row direction and the column direction, respectively, and based on the waveform pattern of the Ac values for the pixel positions in the row direction and the waveform pattern of the Ac values for the pixel positions in the column direction. An image selection process for selecting a digital image in which the polishing direction of the polishing unit and the re-polishing unit matches the scanning direction of the scanner, and the polishing unit and the re-polishing unit. An image extraction step for extracting a partial image of the digital image selected in the image selection step, and an Ac value obtained by performing gray scale conversion on the RGB value of each pixel constituting the partial image in the polishing unit and the re-polishing unit An appearance determination step for calculating a direction perpendicular to the polishing direction and determining whether or not the appearance of the polishing portion and the re-polishing portion is acceptable based on a waveform pattern of the Ac value with respect to the pixel position in the polishing direction. Yes.
この金属板の外観評価方法では、金属板の表面のデジタル画像をスキャナによって取得し、デジタル画像の一部画像に含まれる各ピクセルのRGB値をグレースケール変換して得られるAc値の波形パターンに基づいて、研磨部及び再研磨部の外観の可否を判断する。このような波形パターンを用いることで、研磨部及び再研磨部の外観を容易かつ定量的に判断することができる。また、この金属板の外観評価方法では、Ac値の波形パターンに基づいて、研磨部及び再研磨部の研磨方向がスキャナの走査方向と一致しているデジタル画像を判断対象として選別する。この前処理により、外観判断工程で用いる波形パターンのS/N比が向上し、判断精度を高めることができる。 In this metal plate appearance evaluation method, a digital image of the surface of the metal plate is obtained by a scanner, and the RGB value of each pixel included in a partial image of the digital image is converted into a waveform pattern of Ac values obtained by gray scale conversion. Based on this, whether or not the appearance of the polishing part and the re-polishing part is acceptable is determined. By using such a waveform pattern, it is possible to easily and quantitatively determine the appearance of the polishing unit and the re-polishing unit. Also, in this metal plate appearance evaluation method, based on the waveform pattern of the Ac value, a digital image in which the polishing direction of the polishing unit and the re-polishing unit matches the scanning direction of the scanner is selected as a determination target. By this preprocessing, the S / N ratio of the waveform pattern used in the appearance determination step can be improved, and the determination accuracy can be increased.
また、画像選別工程において、波形パターンのAc値及び振幅が所定の閾値以下である場合に、デジタル画像における研磨部及び再研磨部の研磨方向がスキャナの走査方向と一致していると判断することが好ましい。 In the image selection process, when the Ac value and amplitude of the waveform pattern are equal to or less than a predetermined threshold value, it is determined that the polishing direction of the polishing unit and the re-polishing unit in the digital image matches the scanning direction of the scanner. Is preferred.
研磨部及び再研磨部の研磨方向がスキャナの走査方向と交差している場合、研磨部及び再研磨部の研磨方向がスキャナの走査方向と一致している場合に比べて波形パターンの平均値が高くなり、さらに、行方向及び列方向のいずれかの波形パターンにおいて振幅が大きく乱れる傾向を有する。したがって、波形パターンのAc値及び振幅に閾値を設けることで、S/N比が高いデジタル画像を容易に選別できる。 When the polishing direction of the polishing part and the re-polishing part intersects with the scanning direction of the scanner, the average value of the waveform pattern is larger than when the polishing direction of the polishing part and the re-polishing part matches the scanning direction of the scanner. In addition, the amplitude tends to be greatly disturbed in the waveform pattern in either the row direction or the column direction. Therefore, a digital image with a high S / N ratio can be easily selected by providing threshold values for the Ac value and amplitude of the waveform pattern.
また、外観判断工程において、再研磨部の形成方向に沿って一部画像を複数抽出し、これらの一部画像から算出される各波形パターンの立ち上がり部分及び立ち下がり部分の位置が所定の管理限界の範囲内であるか否かに基づいて、再研磨部の位置ずれの可否を判断することが好ましい。 In addition, in the appearance determination process, a plurality of partial images are extracted along the direction of formation of the regrinding part, and the positions of the rising and falling portions of each waveform pattern calculated from these partial images are within a predetermined control limit. It is preferable to determine whether or not the re-polishing part is misaligned based on whether or not it is within the range.
再研磨部の位置ずれは、一部画像から算出される各波形パターンの立ち上がり部分及び立ち下がり部分の位置ずれとして現れる。したがって、これらの波形パターンの立ち上がり部分及び立ち下がり部分の位置に対して管理限界を設けることで、再研磨部の位置ずれの可否を精度良く判断できる。 The positional deviation of the re-polishing part appears as the positional deviation of the rising part and the falling part of each waveform pattern calculated from a partial image. Therefore, it is possible to accurately determine whether or not the re-polishing unit is misaligned by providing control limits for the positions of the rising and falling portions of these waveform patterns.
また、外観判断工程において、再研磨部の形成方向に沿って一部画像を複数抽出し、これらの一部画像から算出される各波形パターンの立ち上がり部分及び立ち下がり部分の傾きが所定の管理限界の範囲内であるか否かに基づいて、研磨部と再研磨部との境界部分のぼかし状態の可否を判断することが好ましい。 In addition, in the appearance determination process, a plurality of partial images are extracted along the direction of formation of the regrinding part, and the slopes of the rising and falling portions of each waveform pattern calculated from these partial images are within a predetermined control limit. It is preferable to determine whether or not the blurring state of the boundary portion between the polishing unit and the re-polishing unit is possible based on whether or not it is within the range.
研磨部と再研磨部との境界部分のぼかし状態は、一部画像から算出される各波形パターンの立ち上がり部分及び立ち下がり部分の傾きとして現れる。したがって、これらの波形パターンの立ち上がり部分及び立ち下がり部分の傾きに対して管理限界を設けることで、研磨部と再研磨部との境界部分のぼかし状態の可否を精度良く判断できる。 The blurring state of the boundary portion between the polishing portion and the re-polishing portion appears as the slope of the rising portion and the falling portion of each waveform pattern calculated from a partial image. Therefore, by providing a control limit for the slopes of the rising and falling portions of these waveform patterns, it is possible to accurately determine whether or not the blurring state of the boundary portion between the polishing portion and the re-polishing portion is possible.
また、外観判断工程において、再研磨部の形成方向に沿って一部画像を複数抽出し、これらの一部画像から算出される各波形パターンの立ち上がり部分及び立ち下がり部分の裾部の高さが所定の管理限界の範囲内であるか否かに基づいて、研磨部及び再研磨部の汚れの可否を判断することが好ましい。 In addition, in the appearance determination step, a plurality of partial images are extracted along the direction of formation of the regrinding portion, and the heights of the skirt portions of the rising and falling portions of each waveform pattern calculated from these partial images are calculated. It is preferable to determine whether or not the polishing unit and the re-polishing unit are soiled based on whether or not they are within a predetermined control limit range.
研磨部及び再研磨部の汚れは、一部画像から算出される各波形パターンの立ち上がり部分及び立ち下がり部分の裾部の高さの違いとして現れる。したがって、これらの波形パターンの立ち上がり部分及び立ち下がり部分の裾部の高さに対して管理限界を設けることで、研磨部及び再研磨部の汚れの可否を精度良く判断できる。 The contamination of the polishing part and the re-polishing part appears as a difference in the height of the skirt of the rising part and the falling part of each waveform pattern calculated from a partial image. Therefore, it is possible to accurately determine whether or not the polishing portion and the re-polishing portion are contaminated by providing a control limit with respect to the height of the bottom portion and the bottom portion of the waveform pattern.
また、外観判断工程において、再研磨部の形成方向に沿って一部画像を複数抽出し、これらの一部画像から算出される各波形パターンの立ち上がり部分の傾きの絶対値と立ち下がり部分の傾きの絶対値との差が所定の閾値以下であるか否かに基づいて、研磨部と再研磨部との境界部分の研磨ムラの可否を判断することが好ましい。 In addition, in the appearance determination process, a plurality of partial images are extracted along the formation direction of the regrinding part, and the absolute value of the slope of the rising portion and the slope of the falling portion of each waveform pattern calculated from these partial images It is preferable to determine whether or not the unevenness of polishing at the boundary portion between the polishing portion and the re-polishing portion is possible based on whether or not the difference between the absolute value and the absolute value is equal to or less than a predetermined threshold value.
研磨部と再研磨部との境界部分の研磨ムラは、一部画像から算出される各波形パターンの立ち上がり部分の傾きの絶対値と立ち下がり部分の傾きの絶対値との差として現れる。したがって、これらの波形パターンの立ち上がり部分の傾きの絶対値と立ち下がり部分の傾きの絶対値との差に閾値を設けることで、研磨部と再研磨部との境界部分の研磨ムラの可否を精度良く判断できる。 Polishing unevenness at the boundary portion between the polishing portion and the re-polishing portion appears as a difference between the absolute value of the slope of the rising portion and the absolute value of the slope of the falling portion of each waveform pattern calculated from a partial image. Therefore, by setting a threshold value for the difference between the absolute value of the slope of the rising part and the absolute value of the falling part of these waveform patterns, it is possible to accurately determine whether or not polishing unevenness occurs at the boundary between the polishing part and the re-polishing part. Can judge well.
本発明に係る金属板の外観評価方法によれば、研磨部及び再研磨部の仕上がり具合を簡単な手法で定量的に評価できる。 According to the method for evaluating the appearance of a metal plate according to the present invention, it is possible to quantitatively evaluate the finished condition of the polishing part and the re-polishing part by a simple method.
以下、図面を参照しながら、本発明に係る金属板の外観評価方法の好適な実施形態について詳細に説明する。 Hereinafter, preferred embodiments of a metal plate appearance evaluation method according to the present invention will be described in detail with reference to the drawings.
[外観評価方法の工程]
図1は、本発明に係る金属板の外観評価方法の一実施形態を示すフローチャートである。この金属板の外観評価方法は、例えば鉄道車両用構体の外板に用いられる金属板の評価に適用される方法であり、図1に示すように、画像取得工程(ステップS01)と、画像選別工程(ステップS02)と、画像抽出工程(ステップS03)と、外観判断工程(ステップS04)とを備えて構成されている。
[Process of appearance evaluation method]
FIG. 1 is a flowchart showing an embodiment of a metal plate appearance evaluation method according to the present invention. This metal plate appearance evaluation method is a method applied to, for example, evaluation of a metal plate used for an outer plate of a railway vehicle structure. As shown in FIG. 1, an image acquisition step (step S01) and image selection are performed. A process (step S02), an image extraction process (step S03), and an appearance determination process (step S04) are provided.
[評価対象物]
図2は、外観評価の対象となる金属板の一例を示す図である。図2に示すように、金属板1は、例えばステンレス鋼、アルミニウム合金、マグネシウム合金などによって形成され、厚さ数mm程度の板状をなしている。金属板1の一面側には、例えばベルトグラインダによって形成された研磨部2が全面に形成されている。研磨部2の研磨目は、例えば鉄道車両の長手方向に沿う向きに形成されている。
[Evaluation object]
FIG. 2 is a diagram illustrating an example of a metal plate to be subjected to appearance evaluation. As shown in FIG. 2, the metal plate 1 is formed of, for example, stainless steel, an aluminum alloy, a magnesium alloy, or the like, and has a plate shape with a thickness of about several mm. On one surface side of the metal plate 1, a polishing portion 2 formed by, for example, a belt grinder is formed on the entire surface. The polishing eyes of the polishing unit 2 are formed, for example, in a direction along the longitudinal direction of the railway vehicle.
また、研磨部2の一部は、研磨部2が更に研磨されることにより更に細かい研磨目となった再研磨部3となっている。再研磨部3は、溶接痕や傷などを目立たなくする目的で形成されるものであり、例えば鉄道車両の長手方向に沿って帯状に延びている。再研磨部3の研磨目は、研磨部2の研磨目と同方向に形成されている。研磨部2を形成する研磨材の番手は、例えば40番〜60番であり、再研磨部3を形成する研磨材の番手は、例えば180番〜240番である。 Further, a part of the polishing unit 2 is a re-polishing unit 3 that has become finer as the polishing unit 2 is further polished. The regrind part 3 is formed for the purpose of making welding marks, scratches, etc. inconspicuous, and extends, for example, in a strip shape along the longitudinal direction of the railway vehicle. The polishing marks of the re-polishing unit 3 are formed in the same direction as the polishing marks of the polishing unit 2. The count of the abrasive that forms the polishing unit 2 is, for example, No. 40 to No. 60, and the count of the abrasive that forms the re-polishing unit 3 is, for example, No. 180 to No. 240.
以下、金属板の外観評価方法の各工程について説明する。 Hereinafter, each process of the metal plate appearance evaluation method will be described.
[画像取得工程]
画像取得工程は、金属板1の表面のデジタル画像を取得する工程である。この画像取得工程では、図3に示すように、金属板1の表面をスキャナ4で所定の方向に走査し、例えば2000×500ピクセルで表面のデジタル画像を取得する。スキャナ4の走査は、金属板1の表面において、少なくとも再研磨部3の一部が含まれるように行う。また、スキャナ4の走査は、走査方向を変えて複数回行ってもよい。スキャナ4は、公知の装置を用いることができる。
[Image acquisition process]
The image acquisition step is a step of acquiring a digital image of the surface of the metal plate 1. In this image acquisition process, as shown in FIG. 3, the surface of the metal plate 1 is scanned in a predetermined direction by the scanner 4, and a digital image of the surface is acquired by, for example, 2000 × 500 pixels. The scanning of the scanner 4 is performed so that at least a part of the re-polishing unit 3 is included on the surface of the metal plate 1. The scanning of the scanner 4 may be performed a plurality of times while changing the scanning direction. A known device can be used for the scanner 4.
[画像選別工程]
画像選別工程は、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向と一致しているデジタル画像を選別する工程である。研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向と交差する場合、研磨目での光の複雑な反射がノイズとなり、後続の外観判断工程での判断精度が低下するおそれがある。そこで、ノイズの少ないデジタル画像を外観判断工程で用いるべく、画像選別工程を実施する。
[Image selection process]
The image selection step is a step of selecting a digital image in which the polishing direction of the polishing unit 2 and the re-polishing unit 3 matches the scanning direction of the scanner 4. When the polishing direction of the polishing unit 2 and the re-polishing unit 3 intersects with the scanning direction of the scanner 4, complicated reflection of light at the polishing eyes becomes noise, and there is a possibility that the determination accuracy in the subsequent appearance determination step is lowered. . Therefore, an image selection process is performed so that a digital image with less noise is used in the appearance determination process.
この画像選別工程では、まず、画像取得工程で取得したデジタル画像を構成する各ピクセルのRGB値を以下の式(1)を用いてグレースケール変換し、Ac値を得る。この式(1)は、テレビ放送の規格NTSC(National Television System Committee)で使用されているYIQカラーモデルに基づく変換式である。
次に、各ピクセルのRGB値をグレースケール変換して得られたAc値を行方向及び列方向についてそれぞれ算出し、行方向のピクセル位置に対するAc値の波形パターンと、列方向のピクセル位置に対するAc値の波形パターンとを算出する。 Next, the Ac values obtained by performing gray scale conversion on the RGB values of each pixel are calculated in the row direction and the column direction, respectively, and the waveform pattern of the Ac value for the pixel position in the row direction and the Ac for the pixel position in the column direction are calculated. The waveform pattern of the value is calculated.
図4に示す例では、画像取得工程で取得したデジタル画像の任意の箇所を所定の範囲で切り出している。この範囲において、1行目〜n行目の各ピクセルのAc値の平均値Ac(x1)〜Ac(Xn)と、1列目〜n列目の各ピクセルのAc値の平均値Ac(y1)〜Ac(Yn)とを算出し、行方向のピクセル位置に対するAc値の波形パターンと、列方向のピクセル位置に対するAc値の波形パターンとをそれぞれ得る。 In the example shown in FIG. 4, an arbitrary portion of the digital image acquired in the image acquisition process is cut out in a predetermined range. In this range, the average value Ac (x 1 ) to Ac (X n ) of the Ac values of the pixels in the first row to the nth row and the average value Ac of the pixels in the first column to the nth column. (Y 1 ) to Ac (Y n ) are calculated, and a waveform pattern of Ac values for pixel positions in the row direction and a waveform pattern of Ac values for pixel positions in the column direction are obtained.
図5は、行方向についてのAc値の波形パターンの一例を示す図であり、図6は、列方向についてのAc値の波形パターンの一例を示す図である。この図5及び図6では、デジタル画像の任意の箇所を250×1500ピクセルの範囲で切りだし、横軸をピクセル位置、縦軸をAc値の平均値としている。 FIG. 5 is a diagram illustrating an example of an Ac value waveform pattern in the row direction, and FIG. 6 is a diagram illustrating an example of an Ac value waveform pattern in the column direction. In FIGS. 5 and 6, an arbitrary portion of the digital image is cut out in a range of 250 × 1500 pixels, the horizontal axis is the pixel position, and the vertical axis is the average value of the Ac values.
この図5及び図6では、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向に対して平行になっている場合の波形パターン(グラフA)、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向に対して斜め(45°程度)になっている場合の波形パターン(グラフB)、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向に対して垂直になっている場合の波形パターン(グラフC)がそれぞれ例示されている。 5 and 6, the waveform pattern (graph A) when the polishing direction of the polishing unit 2 and the re-polishing unit 3 is parallel to the scanning direction of the scanner 4, the polishing unit 2 and the re-polishing unit 3. Waveform pattern (graph B) in the case where the polishing direction is oblique (about 45 °) with respect to the scanning direction of the scanner 4, and the polishing directions of the polishing unit 2 and the re-polishing unit 3 are with respect to the scanning direction of the scanner 4 Each of the waveform patterns (graph C) in the case of being vertical is illustrated.
同図に示す結果から、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向に対して垂直になっている場合には、Ac値の波形パターンの振幅に大きな変動が現れることがわかる(図5のグラフC参照)。また、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向に対して平行になっている場合には、垂直及び斜めの場合に比べてAc値が3分の1程度になっていることがわかる(図5及び図6のグラフA参照)。 From the results shown in the figure, when the polishing direction of the polishing unit 2 and the re-polishing unit 3 is perpendicular to the scanning direction of the scanner 4, a large variation appears in the amplitude of the waveform pattern of the Ac value. Yes (see graph C in FIG. 5). Further, when the polishing direction of the polishing unit 2 and the re-polishing unit 3 is parallel to the scanning direction of the scanner 4, the Ac value is about one-third compared to the vertical and oblique directions. (See graph A in FIGS. 5 and 6).
そこで、画像選別工程では、波形パターンのAc値及び振幅に対して閾値を設け、波形パターンのAc値及び振幅がいずれも閾値以下となるデジタル画像を、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向と一致しているデジタル画像であると判断し、画像取得工程で取得したデジタル画像の中から選別する。 Therefore, in the image selection step, a threshold is provided for the Ac value and amplitude of the waveform pattern, and a digital image in which both the Ac value and amplitude of the waveform pattern are equal to or less than the threshold is used for polishing directions of the polishing unit 2 and the re-polishing unit 3 Is a digital image that matches the scanning direction of the scanner 4, and is selected from the digital images acquired in the image acquisition process.
[画像抽出工程]
画像抽出工程は、研磨部2及び再研磨部3が含まれるように画像選別工程で選別されたデジタル画像の一部画像を抽出する工程である。より具体的には、例えば図7に示すように、デジタル画像10の全領域の中から、再研磨部3が中央部分で略直交するような範囲で複数(本実施形態では3箇所)の一部画像11(11a,11b,11c)を抽出する。一部画像11a,11b,11cは、例えば2000×500ピクセルの範囲で、再研磨部3の延在方向に沿って互いに離間して抽出することが好ましい。
[Image extraction process]
The image extraction step is a step of extracting a partial image of the digital image selected in the image selection step so that the polishing unit 2 and the re-polishing unit 3 are included. More specifically, for example, as shown in FIG. 7, a plurality (three in this embodiment) of re-polished portions 3 are within a range where the re-polished portions 3 are substantially orthogonal to each other in the central portion. The partial image 11 (11a, 11b, 11c) is extracted. The partial images 11a, 11b, and 11c are preferably extracted in a range of 2000 × 500 pixels, for example, separated from each other along the extending direction of the re-polishing unit 3.
[外観判断工程]
外観判断工程は、研磨部2及び再研磨部3の外観の可否を判断する工程である。この外観判断工程では、まず、画像選別工程と同様の手法で、一部画像11a,11b,11cに含まれる各ピクセルのRGB値をグレースケール変換し、Ac値を得る。次に、行方向(再研磨部3に垂直な方向)について、1行目〜n行目の各ピクセルのAc値の平均値Ac(x1)〜Ac(Xn)を算出し、行方向のピクセル位置に対するAc値の波形パターンを一部画像11a,11b,11cごとに算出する。
[Appearance judgment process]
The appearance determination step is a step of determining whether or not the appearance of the polishing unit 2 and the re-polishing unit 3 is acceptable. In this appearance determination step, first, the RGB value of each pixel included in the partial images 11a, 11b, and 11c is subjected to gray scale conversion by the same method as in the image selection step to obtain an Ac value. Next, average values Ac (x 1 ) to Ac (X n ) of the Ac values of the pixels in the first row to the n-th row are calculated in the row direction (direction perpendicular to the re-polishing unit 3). The waveform pattern of the Ac value for each pixel position is calculated for each of the images 11a, 11b, and 11c.
本実施形態では、金属板1の外観評価項目として、再研磨部3の位置ずれ、研磨部2と再研磨部3との境界部分のぼかし状態、研磨部2及び再研磨部3の汚れ、及び研磨部2と再研磨部3との境界部分の研磨ムラの4項目を例示する。 In the present embodiment, the appearance evaluation items of the metal plate 1 include positional deviation of the re-polishing unit 3, blurring of the boundary portion between the polishing unit 2 and the re-polishing unit 3, dirt on the polishing unit 2 and the re-polishing unit 3, and Four items of polishing unevenness at the boundary portion between the polishing unit 2 and the re-polishing unit 3 are illustrated.
図8は、再研磨部3の位置ずれの可否の判断例を示す図である。同図では、横軸をピクセル位置、縦軸をAc値とし、一部画像11a,11b,11cからそれぞれ算出される3つの波形パターンA,B,Cを示している。波形パターンA,B,Cは、基本的な形として、再研磨部3に対応する凸部A1,B1,C1と、研磨部2に対応する裾部A2,B2,C2とを有している。また、波形パターンA,B,Cは、一方側の研磨部2と再研磨部3との境界部分に対応する立ち上がり部分A3,B3,C3と、他方側の研磨部2と再研磨部3との境界部分に対応する立ち下がり部分A4,B4,C4とを有している。 FIG. 8 is a diagram illustrating an example of determining whether the re-polishing unit 3 is misaligned. In the figure, the horizontal axis represents the pixel position and the vertical axis represents the Ac value, and three waveform patterns A, B, and C calculated from the partial images 11a, 11b, and 11c are shown. The waveform patterns A, B, and C have, as basic shapes, convex portions A 1 , B 1 , and C 1 corresponding to the re-polishing portion 3 and skirt portions A 2 , B 2 , and C 2 corresponding to the polishing portion 2. And have. The waveform patterns A, B, and C include rising portions A 3 , B 3 , and C 3 corresponding to the boundary portion between the polishing unit 2 and the re-polishing unit 3 on one side, and the polishing unit 2 and the re-polishing on the other side. It has falling portions A 4 , B 4 , and C 4 corresponding to the boundary portion with the portion 3.
再研磨部3の位置ずれは、例えば管理限界を用いた波形解析によって評価する。この波形解析では、予め基準用(位置ずれに異常がないと判断されるもの)の波形パターンのAc値の移動平均を求め、移動平均化後の波形パターンに含まれる各Ac値を、その標準偏差σに基づいて正規化する。次に、標準偏差σを3倍して得られる正規化値を移動平均値に加算したものを上管理限界(+3σ)、標準偏差を3倍して得られる正規化値を移動平均値から減算したものを下管理限界(−3σ)とする。 The positional deviation of the regrinding unit 3 is evaluated by, for example, waveform analysis using a control limit. In this waveform analysis, a moving average of Ac values of a waveform pattern for reference (which is determined to be normal in positional deviation) is obtained in advance, and each Ac value included in the waveform pattern after moving average is used as its standard. Normalization is performed based on the deviation σ. Next, the standardized value obtained by multiplying the standard deviation σ by 3 is added to the moving average value, the upper control limit (+ 3σ), and the normalized value obtained by multiplying the standard deviation by 3 is subtracted from the moving average value This is taken as the lower control limit (−3σ).
ここで、再研磨部3の位置ずれは、図8に示すように、一部画像11a,11b,11cから算出される各波形パターンA,B,Cの立ち上がり部分A3,B3,C3及び立ち下がり部分A4,B4,C4の横軸方向の位置ずれとして現れる。したがって、立ち上がり部分A3,B3,C3及び立ち下がり部分A4,B4,C4の全てが上管理限界と下管理限界との間に収まっている場合には、位置ずれに異常がないと判断し、立ち上がり部分A3,B3,C3及び立ち下がり部分A4,B4,C4のいずれかが上管理限界と下管理限界との間に収まっていない場合には、位置ずれに異常があると判断する。 Here, as shown in FIG. 8, the position shift of the re-polishing unit 3 is caused by rising portions A 3 , B 3 , C 3 of the waveform patterns A, B, C calculated from the partial images 11a, 11b, 11c. And it appears as a position shift in the horizontal axis direction of the falling portions A 4 , B 4 , C 4 . Therefore, if all of the rising portions A 3 , B 3 , C 3 and the falling portions A 4 , B 4 , C 4 are within the upper control limit and the lower control limit, the positional deviation is abnormal. If any of the rising portions A 3 , B 3 , C 3 and the falling portions A 4 , B 4 , C 4 is not within the upper control limit and the lower control limit, Judge that the deviation is abnormal.
図9は、研磨部2と再研磨部3との境界部分のぼかし状態の可否の判断例を示す図である。ここでのぼかし状態とは、研磨部2と再研磨部3との相似性を指している。研磨部2と再研磨部3との境界部分のぼかし状態は、一部画像11a,11b,11cから算出される各波形パターンA,B,Cの立ち上がり部分A3,B3,C3及び立ち下がり部分A4,B4,C4の傾きとして現れる。 FIG. 9 is a diagram illustrating an example of determining whether or not the blurring state of the boundary portion between the polishing unit 2 and the re-polishing unit 3 is possible. Here, the blurred state refers to the similarity between the polishing unit 2 and the re-polishing unit 3. The blurring state of the boundary portion between the polishing unit 2 and the re-polishing unit 3 includes rising portions A 3 , B 3 , C 3 and rising portions of the waveform patterns A, B, C calculated from the partial images 11a, 11b, 11c. Appears as the slopes of the descending portions A 4 , B 4 , C 4 .
したがって、図8の場合と同様の手法で、予め基準用(ぼかし状態が設計どおりであると判断されるもの)の波形パターンから上管理限界(+3σ)及び下管理限界(−3σ)を設定し、立ち上がり部分A3,B3,C3及び立ち下がり部分A4,B4,C4の全てが上管理限界と下管理限界との間に収まっている場合には、ぼかし状態が設計どおりであると判断し、立ち上がり部分A3,B3,C3及び立ち下がり部分A4,B4,C4のいずれかが上管理限界と下管理限界との間に収まっていない場合には、ぼかし状態が設計どおりでないと判断する。 Therefore, the upper management limit (+ 3σ) and the lower management limit (−3σ) are set in advance from the waveform pattern for reference (those in which the blurring state is determined to be as designed) in the same manner as in FIG. When the rising portions A 3 , B 3 , C 3 and the falling portions A 4 , B 4 , C 4 are all within the upper control limit and the lower control limit, the blurring state is as designed. If any of the rising portions A 3 , B 3 , C 3 and the falling portions A 4 , B 4 , C 4 does not fall between the upper control limit and the lower control limit, it is blurred. Judge that the condition is not as designed.
図10は、研磨部2及び再研磨部3の汚れの可否の判断例を示す図である。ここでの汚れとは、例えばけがき跡、ペン跡、油汚れ、指紋、シンナー汚れなどを指している。研磨部2及び再研磨部3の汚れは、一部画像11a,11b,11cから算出される各波形パターンA,B,Cの裾部A2,B2,C2の高さの違いとして現れる。 FIG. 10 is a diagram illustrating an example of determining whether or not the polishing unit 2 and the re-polishing unit 3 are dirty. The dirt here refers to, for example, scratch marks, pen marks, oil stains, fingerprints, thinner stains, and the like. The contamination of the polishing unit 2 and the re-polishing unit 3 appears as a difference in height between the skirt portions A 2 , B 2 , and C 2 of the waveform patterns A, B, and C calculated from the partial images 11a, 11b, and 11c. .
したがって、図8の場合と同様の手法で、予め基準用(汚れに異常がないと判断されるもの)の波形パターンから上管理限界(+3σ)及び下管理限界(−3σ)を設定し、裾部A2,B2,C2の全てが上管理限界と下管理限界との間に収まっている場合には、汚れの異常がないと判断し、裾部A2,B2,C2のいずれかが上管理限界と下管理限界との間に収まっていない場合には、汚れの異常があると判断する。 Accordingly, the upper management limit (+ 3σ) and the lower management limit (−3σ) are set in advance from the waveform pattern for reference (those that are judged to have no abnormality in dirt) by the same method as in FIG. When all of the parts A 2 , B 2 , and C 2 are within the upper management limit and the lower management limit, it is determined that there is no abnormality in dirt, and the skirts A 2 , B 2 , and C 2 If any of them does not fall between the upper control limit and the lower control limit, it is determined that there is an abnormality in dirt.
図11は、研磨部2と再研磨部3との境界部分の研磨ムラの可否の判断例を示す図である。ここでの研磨ムラとは、研磨部2と再研磨部3との境界部分の一部で研磨不足(かすれ)が生じていることを指している。研磨部2と再研磨部3との境界部分の研磨ムラは、一部画像11a,11b,11cから算出される各波形パターンA,B,Cの立ち上がり部分A3,B3,C3の傾きの絶対値と立ち下がり部分A4,B4,C4の傾きの絶対値との差として現れる。 FIG. 11 is a diagram illustrating an example of determining whether or not polishing unevenness is present at the boundary portion between the polishing unit 2 and the re-polishing unit 3. Here, the uneven polishing refers to the occurrence of insufficient polishing (fading) at a part of the boundary portion between the polishing unit 2 and the re-polishing unit 3. The unevenness of polishing at the boundary between the polishing unit 2 and the re-polishing unit 3 is the inclination of the rising portions A 3 , B 3 , C 3 of the waveform patterns A, B, C calculated from the partial images 11a, 11b, 11c. And the absolute value of the slope of the falling portions A 4 , B 4 , C 4 appear.
したがって、立ち上がり部分A3,B3,C3の傾きの絶対値と立ち下がり部分A4,B4,C4の傾きの絶対値との差に閾値を設けておき、傾きの絶対値の差が閾値以下である場合には、研磨ムラの異常がないと判断し、傾きの絶対値の差が閾値を超える場合には、研磨ムラの異常があると判断する。 Therefore, a threshold is provided for the difference between the absolute values of the slopes of the rising portions A 3 , B 3 , and C 3 and the absolute values of the slopes of the falling portions A 4 , B 4 , and C 4 , and the difference between the absolute values of the slopes is set. Is equal to or less than the threshold value, it is determined that there is no abnormality in polishing unevenness, and when the difference between the absolute values of the slopes exceeds the threshold value, it is determined that there is abnormality in polishing unevenness.
なお、傾きの絶対値の差は、波形パターンごとに算出すればよい。すなわち、立ち上がり部分A3の傾きの絶対値と立ち下がり部分A4の傾きの絶対値との差、立ち上がり部分B3の傾きの絶対値と立ち下がり部分B4の傾きの絶対値との差、立ち上がり部分C3の傾きの絶対値と立ち下がり部分C4の傾きの絶対値との差をそれぞれ算出し、いずれかの傾きの絶対値の差が閾値を超える場合に、研磨ムラの異常があると判断すればよい。 Note that the difference between the absolute values of the slopes may be calculated for each waveform pattern. That is, the difference between the absolute value of the slope of portion A 4 downward absolute value and falling slope of the rising portion A 3, the difference between the absolute value of the slope of the portion B 4 decreases the absolute value and falling slope of the rising portion B 3, the difference in the absolute value of the slope of the falling portion C 4 of the slope of the rising portion C 3 are calculated respectively, if the difference absolute value of any gradient exceeds a threshold value, an abnormality of the polishing nonuniformity It can be judged.
以上説明したように、この金属板の外観評価方法では、金属板1の表面のデジタル画像をスキャナ4によって取得し、デジタル画像10の一部画像11に含まれる各ピクセルのRGB値をグレースケール変換して得られるAc値の波形パターンに基づいて、研磨部2及び再研磨部3の外観の可否を判断する。このような波形パターンを用いることで、研磨部2及び再研磨部3の外観を容易かつ定量的に判断することができる。 As described above, in this metal plate appearance evaluation method, a digital image of the surface of the metal plate 1 is acquired by the scanner 4, and the RGB value of each pixel included in the partial image 11 of the digital image 10 is converted to grayscale. On the basis of the waveform pattern of the Ac value obtained in this manner, whether or not the appearance of the polishing unit 2 and the re-polishing unit 3 is acceptable is determined. By using such a waveform pattern, the appearance of the polishing unit 2 and the re-polishing unit 3 can be easily and quantitatively determined.
また、この金属板の外観評価方法では、Ac値の波形パターンに基づいて、研磨部2及び再研磨部3の研磨方向がスキャナ4の走査方向と一致しているデジタル画像10を判断対象として選別する。この前処理により、外観判断工程で用いる波形パターンのS/N比が向上し、判断精度を高めることができる。 Further, in this metal plate appearance evaluation method, based on the waveform pattern of the Ac value, the digital image 10 in which the polishing direction of the polishing unit 2 and the re-polishing unit 3 coincides with the scanning direction of the scanner 4 is selected as a determination target. To do. By this preprocessing, the S / N ratio of the waveform pattern used in the appearance determination step can be improved, and the determination accuracy can be increased.
1…金属板、2…研磨部、3…再研磨部、4…スキャナ、10…デジタル画像、11(11a,11b,11c)…一部画像、A,B,C…波形パターン、A2,B2,C2…裾部、A3,B3,C3…立ち上がり部分、A4,B4,C4…立ち下がり部分。 1 ... metal plate, 2 ... polishing unit, 3 ... regrinding unit, 4 ... scanner, 10 ... digital image, 11 (11a, 11b, 11c ) ... Some images, A, B, C ... wave pattern, A 2, B 2 , C 2 ... bottom, A 3 , B 3 , C 3 ... rising part, A 4 , B 4 , C 4 ... falling part.
Claims (6)
前記金属板の表面をスキャナで走査し、前記表面のデジタル画像を取得する画像取得工程と、
前記デジタル画像を構成する各ピクセルのRGB値をグレースケール変換して得られるAc値を行方向及び列方向についてそれぞれ算出し、前記行方向のピクセル位置に対する前記Ac値の波形パターンと、前記列方向のピクセル位置に対する前記Ac値の波形パターンとに基づいて、前記研磨部及び前記再研磨部の研磨方向が前記スキャナの走査方向と一致しているデジタル画像を選別する画像選別工程と、
前記研磨部及び前記再研磨部が含まれるように前記画像選別工程で選別された前記デジタル画像の一部画像を抽出する画像抽出工程と、
前記一部画像を構成する各ピクセルのRGB値をグレースケール変換して得られるAc値を前記研磨部及び前記再研磨部における研磨方向に垂直な方向について算出し、前記研磨方向に垂直な方向のピクセル位置に対する前記Ac値の波形パターンに基づいて、前記研磨部及び前記再研磨部の外観の可否を判断する外観判断工程と、を備えたことを特徴とする金属板の外観評価方法。 A method for evaluating the appearance of a metal plate having a polishing portion polished in a predetermined direction and a re-polishing portion on the surface, the polishing portion being further polished in the predetermined direction and having finer polishing eyes than the polishing portion. There,
An image acquisition step of scanning the surface of the metal plate with a scanner and acquiring a digital image of the surface;
Ac values obtained by performing gray scale conversion on the RGB values of each pixel constituting the digital image are calculated in the row direction and the column direction, respectively, and the waveform pattern of the Ac value with respect to the pixel position in the row direction, and the column direction An image selection step of selecting a digital image in which the polishing direction of the polishing unit and the re-polishing unit matches the scanning direction of the scanner based on the waveform pattern of the Ac value with respect to the pixel position of
An image extraction step of extracting a part of the digital image selected in the image selection step so as to include the polishing unit and the re-polishing unit;
An Ac value obtained by performing gray scale conversion on the RGB value of each pixel constituting the partial image is calculated in a direction perpendicular to the polishing direction in the polishing unit and the re-polishing unit, and the direction perpendicular to the polishing direction is calculated . An appearance evaluation step of determining whether or not the appearance of the polishing part and the re-polishing part is acceptable based on a waveform pattern of the Ac value with respect to a pixel position.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011209303A JP5823794B2 (en) | 2011-09-26 | 2011-09-26 | Metal plate appearance evaluation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011209303A JP5823794B2 (en) | 2011-09-26 | 2011-09-26 | Metal plate appearance evaluation method |
Publications (2)
Publication Number | Publication Date |
---|---|
JP2013068583A JP2013068583A (en) | 2013-04-18 |
JP5823794B2 true JP5823794B2 (en) | 2015-11-25 |
Family
ID=48474428
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2011209303A Active JP5823794B2 (en) | 2011-09-26 | 2011-09-26 | Metal plate appearance evaluation method |
Country Status (1)
Country | Link |
---|---|
JP (1) | JP5823794B2 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108257193A (en) * | 2018-01-05 | 2018-07-06 | 上海爱优威软件开发有限公司 | A kind of terminal data form conversion method and system |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1125254A (en) * | 1997-06-27 | 1999-01-29 | Misawa Homes Co Ltd | Methods for picture analysis and picture edition |
JP2001266142A (en) * | 2000-01-13 | 2001-09-28 | Nikon Corp | Method/device for sorting data, method/device for processing signal, method/device for detecting position, method/device for processing image, method/device for exposure and method for manufacturing device |
JP3481605B2 (en) * | 2001-04-26 | 2003-12-22 | アジアエレクトロニクス株式会社 | Direction determination device |
JP3927054B2 (en) * | 2002-03-14 | 2007-06-06 | 株式会社精工技研 | Optical fiber end face polishing method and ferrule used therefor |
JP4363953B2 (en) * | 2003-10-23 | 2009-11-11 | 大日本印刷株式会社 | Optical characteristic non-uniformity measuring method and apparatus, and product quality judgment method and apparatus using the same |
JP5435904B2 (en) * | 2008-07-11 | 2014-03-05 | 株式会社メガトレード | How to detect fatal wounds |
JP5217841B2 (en) * | 2008-09-25 | 2013-06-19 | オムロン株式会社 | Image processing apparatus and method for supporting parameter setting relating to defect detection on image |
JP5472575B2 (en) * | 2009-02-26 | 2014-04-16 | 独立行政法人 宇宙航空研究開発機構 | Apparatus and method for measuring distance between opposed surfaces at predetermined positions in opposed surface, and high flatness machining method using them |
JP5202442B2 (en) * | 2009-05-29 | 2013-06-05 | 本田技研工業株式会社 | Surface inspection device |
-
2011
- 2011-09-26 JP JP2011209303A patent/JP5823794B2/en active Active
Also Published As
Publication number | Publication date |
---|---|
JP2013068583A (en) | 2013-04-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109141232B (en) | Online detection method for disc castings based on machine vision | |
CN108629775B (en) | Thermal state high-speed wire rod surface image processing method | |
CN108416766B (en) | Double-side light-entering type light guide plate defect visual detection method | |
CN109507192B (en) | Magnetic core surface defect detection method based on machine vision | |
CN111667462B (en) | Automatic detection method and system for small and weak targets in welding defects | |
JP5421192B2 (en) | Crack detection method | |
JP4282643B2 (en) | Strain evaluation apparatus and strain evaluation method | |
JP2017053819A (en) | Crack detection method and detection program of concrete | |
JP5812705B2 (en) | Crack detection method | |
US8610770B2 (en) | Image sensing apparatus and method for sensing target that has defective portion region | |
JP5864005B1 (en) | Image processing method, object life evaluation method, and image processing system | |
JP2015161575A (en) | Tire deterioration evaluation device, system thereof, method thereof, and program thereof | |
JP5705711B2 (en) | Crack detection method | |
CN110610516A (en) | Railway fastener nut center positioning method | |
JP5823794B2 (en) | Metal plate appearance evaluation method | |
JP2018048979A (en) | Surface property inspection apparatus, surface property insection method, and program | |
CN106482635A (en) | Image measuring apparatuss | |
JP6213783B2 (en) | Inspection method and apparatus for vehicle wind glass | |
CN114972317B (en) | Square grain placement planning method | |
TWI447606B (en) | Method for detecting sliversof a billet | |
JP2020085526A (en) | Method for evaluating quality of weld zone | |
JP4980739B2 (en) | Crack detection method | |
JP4978215B2 (en) | Machining surface defect judgment method | |
JP2003329428A (en) | Device and method for surface inspection | |
CN113092486B (en) | Method suitable for detecting water ripples on surface of high-gloss surface product |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20140820 |
|
A977 | Report on retrieval |
Free format text: JAPANESE INTERMEDIATE CODE: A971007 Effective date: 20150417 |
|
A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20150428 |
|
A521 | Written amendment |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20150529 |
|
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: 20151006 |
|
A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20151008 |
|
R150 | Certificate of patent or registration of utility model |
Ref document number: 5823794 Country of ref document: JP Free format text: JAPANESE INTERMEDIATE CODE: R150 |