JPH03259734A - Automatic evaluating method for variation degree of surface property - Google Patents
Automatic evaluating method for variation degree of surface propertyInfo
- Publication number
- JPH03259734A JPH03259734A JP5947490A JP5947490A JPH03259734A JP H03259734 A JPH03259734 A JP H03259734A JP 5947490 A JP5947490 A JP 5947490A JP 5947490 A JP5947490 A JP 5947490A JP H03259734 A JPH03259734 A JP H03259734A
- Authority
- JP
- Japan
- Prior art keywords
- brightness
- values
- pixels
- group
- value
- 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
Links
- 238000000034 method Methods 0.000 title claims description 16
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000010191 image analysis Methods 0.000 claims description 8
- 238000011109 contamination Methods 0.000 abstract description 10
- 230000000007 visual effect Effects 0.000 abstract description 4
- 238000007689 inspection Methods 0.000 abstract description 2
- 238000012545 processing Methods 0.000 abstract description 2
- 238000000926 separation method Methods 0.000 abstract 1
- 230000001953 sensory effect Effects 0.000 description 13
- 238000000611 regression analysis Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 3
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 239000003973 paint Substances 0.000 description 2
- 238000002834 transmittance Methods 0.000 description 2
- 238000003705 background correction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000005357 flat glass Substances 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000010186 staining Methods 0.000 description 1
Landscapes
- Closed-Circuit Television Systems (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
Abstract
Description
【発明の詳細な説明】
「産業上の利用分野］
本発明は画像解析装置を用いて外壁等の壁面や床面、天
井面あるいは各種部材表面の汚れやキズ、塗料剥落等の
表面性状の変化度を評価する方法に関する。Detailed Description of the Invention "Industrial Field of Application" The present invention uses an image analysis device to detect changes in surface properties such as dirt, scratches, paint peeling, etc. on walls such as exterior walls, floors, ceilings, or the surfaces of various parts. Concerning how to evaluate degree.
［従来の技術］
建築物の床や外壁等の汚れを数値化する方法として、暴
露試料の反射率、色度、分光反射率等を測定したり、窓
ガラスの場合には光透過率を測定することが行なわれて
いる。[Prior art] As a method of quantifying dirt on building floors and exterior walls, etc., there is a method of measuring reflectance, chromaticity, spectral reflectance, etc. of exposed samples, or measuring light transmittance in the case of window glass. things are being done.
［発明が解決しようとする課題］
従来の汚れ測定法は、基準となる標準試料（新品試料又
は非暴露試料）と実際の試料とについて反射率、透過率
等の物性値を比較して汚れの程度を判断するものであり
、常に判断の基準となる標準試料が必要である。[Problem to be solved by the invention] The conventional stain measurement method compares the physical property values such as reflectance and transmittance of a standard sample (new sample or unexposed sample) and an actual sample to determine the degree of contamination. It is used to judge the degree of damage, and a standard sample is always required to serve as a basis for judgment.
ところが、実際の建築物の外壁面の汚れの程度を判定す
る場合、基準となる標準試料が現存せず、対照てきない
ことが多い。However, when determining the level of dirt on the exterior walls of actual buildings, there are often no standard samples available to serve as standards, and therefore no comparison can be made.
また、人がある部材（例えば壁面や床）を視覚した場合
に当該部材が汚れているか否かを判断する場合には、通
常、標準試料との対比により汚れの程度を判断するのて
はなく、視覚した当該試料の表面性状にのみ基いて判断
するのであるから、標準試料との対比に基〈従来の評価
方法では、実際に人か視覚して判断する汚れ度評価との
Ｓ離か大きくなるおそれがある。In addition, when a person looks at a certain part (for example, a wall or floor) and wants to judge whether the part is dirty or not, it is usually not possible to judge the degree of dirt by comparing it with a standard sample. Since judgments are made based only on the visual appearance of the surface of the sample, it is based on comparison with standard samples. There is a risk that
［課題を解決するための手段］
本発明の表面性状の変化度を評価方法は、評価対象面の
撮像を画像解析装置に入力し、各画素の明度値に基づい
て表面性状の変化度を測定する方法に関する。[Means for Solving the Problems] The method for evaluating the degree of change in surface texture of the present invention involves inputting an image of the surface to be evaluated into an image analysis device, and measuring the degree of change in surface texture based on the brightness value of each pixel. Regarding how to.
本発明では、各画素の明度値の集合を閾値ＬｔによりＬ
ｔよりも小なる明度値のグループとＬ ｔよりも犬なる
明度値のグループとに分割した場合に、次式で示される
２グループの分散０１σＴ ＝（ＬＩＳ （ＬＳ−Ｌａ
） ２＋ωｂ（Ｌｂ−Ｌａ）２
ωｓ：低明度値グループに属する画素比率ωｂ：高明度
値グループに属する画素比率Ｌｓ：低明度値グループに
属する画素の平均明度
■、ｂ：高明度値グループに属する画素の平均明度
Ｌａ・全画素の平均明度
（明度はＪＩＳ Ｚ ８７２９のＬ１値である。）が最
大となる閾値Ｌｔを求め、このＬｔ値により低明度値の
グループの画素と高明度値のグループの画素とを確定す
ると共に、各画素のａ′値及びｂ０値（ＪＩＳ Ｚ ８
７２９）ヲ演算’１− ル。In the present invention, a set of brightness values of each pixel is set to L by a threshold Lt.
When divided into a group with brightness values smaller than t and a group with brightness values smaller than Lt, the variance of the two groups is expressed by the following formula: 01σT = (LIS (LS - La
) 2+ωb(Lb-La)2 ωs: Pixel ratio belonging to the low brightness value group ωb: Pixel ratio belonging to the high brightness value group Ls: Average brightness of pixels belonging to the low brightness value group ■, b: Belonging to the high brightness value group The threshold value Lt at which the average brightness La of the pixel and the average brightness of all pixels (lightness is the L1 value of JIS Z 8729) is determined is determined, and based on this Lt value, pixels in a group of low brightness values and a group of high brightness values are determined. At the same time, determine the a' value and b0 value (JIS Z 8
729) Operation '1- rule.
請求項（１）の方法では、この確定されたグループのう
ち一方のグループに属する画素数と全画素数との比率な
Ａｒとし、確定された両グルブの平均明度差Ｌｂ−Ｌｓ
をコントラストΔし、平均ａ′値の差ａ ’ｂ−ａ ＊
ｓをΔａ、平均ｂ′値の差ｂ”、、−ａ″５をΔｂとし
、これらから色差ΔＥをΔＥ＝ ΔＬ ＋ Δａ ＋
Δｂ）２として求め、一方のグループに属する画素の
平均明度Ｌｂ又はＬｓとぅｒＷ］７Ｔ−１丁との重回帰
により評価対象面の変化度を特徴する
請求項（２）の方法では、上記のようにして評価対象面
の変化度を評価する請求項１の方法において、評価対象
面に目地があるときに、目地の明度値の範囲を設定し、
設定された範囲にある明度値を有した画素を評価対象画
素から除外して表面性状の変化度を評価する。In the method of claim (1), Ar is the ratio between the number of pixels belonging to one of the determined groups and the total number of pixels, and the average brightness difference between the two determined groups is Lb - Ls.
is the contrast Δ, and the difference in the average a′ value is a′ba−a*
Let s be Δa, and the difference b", -a"5 in average b' value be Δb, and from these, the color difference ΔE can be calculated as ΔE= ΔL + Δa +
In the method of claim (2), the degree of change of the evaluation target surface is characterized by multiple regression with the average brightness Lb or Ls of pixels belonging to one group, calculated as Δb)2, and the average brightness of pixels belonging to one group. In the method according to claim 1, in which the degree of change of the surface to be evaluated is evaluated as follows, when there is a joint in the surface to be evaluated, a range of brightness values of the joint is set,
Pixels having brightness values within a set range are excluded from evaluation target pixels to evaluate the degree of change in surface texture.
［作用］
本発明に従い評価対象面の汚れの程度を評価する場合を
例にとりて本発明の詳細な説明する。[Operation] The present invention will be described in detail by taking as an example the case where the degree of contamination of the surface to be evaluated is evaluated according to the present invention.
評価対象面に汚れが付着していると、通常の場合、この
汚れは評価対象面全面に均一に付着するのではなく、あ
る程度のムラを伴なうように付着する。When dirt adheres to the surface to be evaluated, normally the dirt does not adhere uniformly to the entire surface to be evaluated, but rather adheres unevenly to some extent.
本発明方法では、評価対象面の撮像の各画素の明度値に
基づいて各画素が高明度値グループに属するか、低明度
値グループに属するかを分類する。In the method of the present invention, each pixel is classified as to whether it belongs to a high brightness value group or a low brightness value group based on the brightness value of each pixel in the captured image of the evaluation target surface.
本発明では、この確定されたグループのうち方のグルー
プ（多くの場合は低明度値グループ）に属する画素の集
合が汚染部分として扱われる。In the present invention, a set of pixels belonging to one of the determined groups (often a low brightness value group) is treated as a contaminated portion.
なお、本発明では、判別分析法により、高明度値グルー
プと低明度値グループとを区画する閾値が決定されてい
る。Note that, in the present invention, a threshold value that separates a high brightness value group and a low brightness value group is determined by a discriminant analysis method.
この低明度値グループに属する画素の集合域を汚染部分
として扱う場合を例にとって説明する。An example will be explained in which a collection area of pixels belonging to this low brightness value group is treated as a contaminated area.
汚染部分の面積、平均明度等の特徴を抽出するためには
、汚染部分を特定しなければならない。In order to extract features such as the area and average brightness of a contaminated area, the contaminated area must be identified.
しかしながら、汚染部分の境界線は明確でない場合が多
く、そのような濃淡画像の２値化は困難が伴なう。通常
、人が汚染部分を認識する状況を考えると、色の濃い部
分と薄い部分との差が最も明確となる境界線を想定し、
前者を汚染部分として認識している。このような人間の
判断に近い２値化の方法として、本発明では判別分析法
が適用されている。具体的には、画像解析装置に取り込
んた全画素の明度分布において、明度値の集合を閾値Ｌ
ｔで２つのグループに分割した場合に、２つのグループ
の分１！ｌｆｔ （分散）が最も大きくなるようにＬｔ
値を決定する。低明度値グループ（汚染部、Ｌ ＜ Ｌ
ｔ、このＬは任意の画素の明度値である。）及び高明
度値グループ（汚染かない又は少ない下地部、Ｌ２：Ｌ
ｔ）の画素比率と平均明度とをそれぞれωｓ、ωｂ、Ｌ
ｓ、Ｌｂ、全体の平均明度をＬａとすると、グループ間
の分散σ１は（１，ｓ （ＬＳ−Ｌａ）２＋ωｂ （Ｌ
ｂ−Ｌａ）２となり、これを最大にするＬｔが求められ
る。そして、この求められた、上記分散を最大にするＬ
ｔ値をもって低明度値（汚染部）グループと高明度値グ
ループ（非汚染部）との区画を確定させる。However, the boundaries of contaminated areas are often not clear, and it is difficult to binarize such grayscale images. Normally, when considering the situation in which people recognize contaminated areas, we assume a boundary line where the difference between dark and light areas is the clearest.
The former is recognized as a contaminated part. In the present invention, a discriminant analysis method is applied as a binarization method that is similar to human judgment. Specifically, in the brightness distribution of all pixels taken into the image analysis device, a set of brightness values is set to a threshold L.
When divided into two groups by t, 1 of the two groups! Lt so that lft (variance) is the largest
Determine the value. Low brightness value group (contaminated area, L < L
t, and this L is the brightness value of an arbitrary pixel. ) and high brightness value group (underlying area with no or little contamination, L2:L
t) pixel ratio and average brightness as ωs, ωb, and L, respectively.
s, Lb, and the overall average brightness is La, the variance σ1 between groups is (1, s (LS-La)2+ωb (L
b-La)2, and Lt that maximizes this is found. Then, the obtained L that maximizes the above variance is
The division into a low brightness value group (contaminated area) and a high brightness value group (uncontaminated area) is determined using the t value.
この確定された閾値により分類されたグループのうちの
一方（以下、この［作用コの項では低明度値をこの一方
のグループとした場合を例にして説明する。）のグルー
プに属する画素数と撮像されて入力された全画素数との
比率をＡｒとする。The number of pixels belonging to one of the groups classified by this determined threshold value (hereinafter, in the section on effects, we will explain the case where low brightness values are set to one of these groups as an example). Let Ar be the ratio to the total number of pixels imaged and input.
また、確定された両グループの平均明度差Ｌ ｂ −Ｌ
ｓをコントラストΔＬとする。さらに、両グループの
平均ａ″値の差ａ″ｂ−ａ″８をＬａ、平均ｂ″値の差
ｂ″、−ｂ□をΔｂとする。そして、色差ΔＥを
ΔＥ−ΔＬ ＋ Ｌａ ＋ Δｂ として求める。Also, the determined average brightness difference between both groups L b −L
Let s be the contrast ΔL. Further, the difference a″b−a″8 in the average a″ values of both groups is set as La, and the difference b″, −b□ in the average b″ values of both groups is set as Δb.Then, the color difference ΔE is set as ΔE−ΔL + La + Δb. Find it as.
しかして、本発明者が種々研究を重ねた結果、汚染部分
の平均明度値ＬｂとＪ”ＥＶ ｒ ７丁丁との重回帰は
、検査員の視覚」二の汚染度合の判断結果と良好な相関
関係を有することが認められた。請求項（１）はかかる
知見に基く。As a result of various studies conducted by the present inventor, the multiple regression between the average brightness value Lb of the contaminated area and J"EV r 7-cho has a good correlation with the judgment result of the degree of contamination based on the inspector's visual sense. It was recognized that they had a relationship. Claim (1) is based on this knowledge.
なお、評価対象面に目地が存在すると、この目地部分は
例え汚染されていなくても周囲とは明度が著しく相違す
る。（通常の場合、目地は著しい暗色を呈する。）
そこて、請求項（２）では、この目地部か有するてあろ
うと推測される明度範囲を指定しておき、この指定範囲
に該当する画素を評価対象から除外することにより、評
価精度を向上させるようにした。Note that if a joint exists on the surface to be evaluated, the brightness of this joint will be significantly different from the surrounding area even if it is not contaminated. (Normally, the joint exhibits an extremely dark color.) Therefore, in claim (2), the brightness range that the joint is presumed to have is specified, and the pixels corresponding to this specified range are The evaluation accuracy was improved by excluding it from the evaluation target.
［実施例］
実施例１（外壁面の汚染程度を評価）
以下の如くして外壁面の表面の撮像を画像解析して諸物
埋置を計測し、汚染度を自動評価した。[Example] Example 1 (Evaluation of the degree of contamination of the outer wall surface) As described below, images of the surface of the outer wall surface were analyzed to measure the amount of buried items, and the degree of contamination was automatically evaluated.
なお、後述の通り、汚れに対する感覚量も測定し、実壁
面に対し感覚量と物理量の評価を行なった。In addition, as described later, the sensory amount of dirt was also measured, and the sensory amount and physical amount were evaluated for the actual wall surface.
（１） 物理量の計測方法
物理量の計測は画像解析装置（Ｎｅｘｕｓ ６５１０）
およびマイクロコンピュータ−によフた。試料（本実施
例では試料を撮影した写真）の色彩情報はビデオカメラ
によりＲＧＢ （赤、緑、青）各２５６諧調のデータと
して画像解析装置にシェーディング補正後、取り込まれ
る。(1) Method of measuring physical quantities Physical quantities were measured using an image analysis device (Nexus 6510)
and a microcomputer. The color information of the sample (in this embodiment, a photograph taken of the sample) is captured by a video camera into an image analysis device after shading correction as data in each of 256 tones of RGB (red, green, and blue).
ここで光量を一定に保つため、カラーチャートの白の諧
調を２５５にした状態で写真をカメラに取込むことにし
た。In order to keep the amount of light constant, I decided to set the white tone on the color chart to 255 and capture the photo into the camera.
なお、色の値はＸＹＺ表色系とＬ” ａ’ ｂ”表色系
で表わす。画像解析装置のカラー表示であるＲＧＢは、
ＣＩＥ （国際照明委員会）の定めるＲＧＢ表色系と同
一のものではなく、この装置特有のものである。ＲＧＢ
の値を正式な表色系に変換するための変換式を以下の方
法により導いた。Note that color values are expressed in the XYZ color system and the L"a'b" color system. RGB, which is the color display of image analysis equipment, is
It is not the same as the RGB color system defined by CIE (Commission Internationale de Illumination), and is unique to this device. RGB
The conversion formula for converting the value of into the formal color system was derived using the following method.
まず、標準色標ＪＩＳ Ｚ ８７２１の色を画像解析装
置て測定しそれぞれのＲＧＢの値を出す。そして色表に
記載されているＸＹＺ（７）値（ＪＩＳ Ｚ ８７０１
）ＸＹＺ表色系による）と照合し、ＲＧＢとＸＹＺとの
関係式をつくった。変換式を以下に示す。First, the colors of the standard color standard JIS Z 8721 are measured using an image analysis device and the respective RGB values are obtained. Then, the XYZ (7) values listed in the color table (JIS Z 8701
) based on the XYZ color system), and created a relational expression between RGB and XYZ. The conversion formula is shown below.
Ｘ−０，０６６５Ｒ＋０．０７０４Ｇ−０，０３２３Ｂ
＋０．０００７５［ｉＲ２＋０．００００５０Ｇ２４０
．０００３５５８２＋３．１６８６Ｙ−０，０６７８Ｒ
＋０．３３３１Ｇ−０，０１８０Ｂ＋０．００００１０
４３Ｇ３ｉ、００１２６９Ｇ２＋０． ００５１４ＲＧ
＋４．４７０３Ｚ−０，０１６５Ｒ＋０．００８８Ｇ＋
０．０８６１Ｂ−０，００００６４Ｒ２０，０００１５
７Ｇ’＋０．００１４３８８２＋０．７２４にのｘ、ｙ
、ｚ値からＪＩＳ Ｚ ８７２９の明度Ｌ″値と色相及
び彩度に係る色度ａ″値及びｂ″値とを次式で算出した
。X-0,0665R+0.0704G-0,0323B
+0.00075[iR2+0.000050G240
．． 00035582+3.1686Y-0,0678R
+0.3331G-0,0180B+0.000010
43G3i, 001269G2+0. 00514RG
+4.4703Z-0,0165R+0.0088G+
0.0861B-0,000064R20,00015
x, y of 7G'+0.00143882+0.724
, z value, JIS Z 8729 lightness L'' value and chromaticity a'' value and b'' value related to hue and saturation were calculated using the following formula.
下記式において、Ｘｏ 、ｙｏ、ｚｏは照明光源の三刺
激値であり、値は次の通りである。In the formula below, Xo, yo, and zo are the tristimulus values of the illumination light source, and the values are as follows.
ｘｏ ＝９８，０７２
Ｙｏ＝１００
Ｚｏ ＝１１８． ２２５
ａ ’ −５００（（Ｘ／Ｘｏ）”３−（Ｙ／Ｙｏ）”
３）ｂ” −２００（（ｙ／ｙｏ）”３（ｚ／ｚｏ）”
３）Ｌ” ＝１１６（Ｙ／Ｙ、、）”３−１６次に、次
式で定義される分散０１を最大にする明度閾値Ｌｔを求
めた。xo =98,072 Yo=100 Zo =118. 225 a'-500((X/Xo)"3-(Y/Yo)"
3)b"-200((y/yo)"3(z/zo)"
3) L"=116(Y/Y,,)"3-16 Next, the brightness threshold Lt that maximizes the variance 01 defined by the following equation was determined.
σアーωｓ （Ｌｓ−Ｌａ）２
＋ωｂ （Ｌｂ−Ｌａ）２
なお、（ＬＩＳ、 ωｂ、Ｌｓ、Ｌａ、Ｌｂは次の通り
である。σArωs (Ls-La)2 +ωb (Lb-La)2 Note that (LIS, ωb, Ls, La, and Lb are as follows.
ωｓ、低明度値グループに属する画素比率ωｂ＝高明度
値グループに属する画素比率ＬＳ：低明度値グループに
属する画素の平均明度
Ｌｂ：高明度値グループに属する画素の平均明度
１
２
Ｌａ：全画素の平均明度
この閾値Ｌｔを境界として低明度値グループの画素と高
明度値のグループの画素とを確定した。ωs, pixel ratio belonging to the low brightness value group ωb = pixel ratio belonging to the high brightness value group LS: average brightness of pixels belonging to the low brightness value group Lb: average brightness of pixels belonging to the high brightness value group 1 2 La: all pixels The pixels in the low brightness value group and the pixels in the high brightness value group were determined using this threshold value Lt as a boundary.
この確定された低明度値のグループに属する画素数と全
画素数との比率をＡｒとし、確定された両グループの平
均明度差Ｌｂ−ＬｓをコントラストΔＬヒした。また、
両グループの平均ａ１値の差ａ ｍ ｂａ ＋＝８をΔ
ａ、平均ｂ゛りの差ｂ ”、−ｂ ’″をΔｂとし、
ΔＬ ＋ ｂａ）＋（Δｂ）２を△Ｅとした。The ratio between the number of pixels belonging to the determined low brightness value group and the total number of pixels is set as Ar, and the determined average brightness difference Lb-Ls of both groups is calculated as the contrast ΔL. Also,
The difference between the average a1 values of both groups a m ba +=8 is expressed as Δ
a, the difference b '', -b '' of the average b ゛ is Δb,
ΔL + ba) + (Δb)2 was defined as ΔE.
また、次式のに値を２値化後自動計測し、汚染部形状の
複雑さを表わす物理量とした。In addition, the value of the following equation was automatically measured after being binarized, and was used as a physical quantity representing the complexity of the contaminated part shape.
Ｋ＝ １−２ＦＦ丁丁７丁Ｔ
ＳＳ：汚染部面積 Ｐｓ・汚染部周長（２） 感覚量
の測定方法
汚れに対する感覚量の測定は官能検査（系列範ちゅう法
）によった。写真試料に対する評価の妥当性の検討を行
った。K = 1-2FF 7-cho T SS: Area of contaminated area Ps/Perimeter of contaminated area (2) Method for measuring sensory amount The sensory amount for staining was measured by a sensory test (series category method). We examined the validity of the evaluation for photographic samples.
宇都宮大学構内において３４の壁面に対して感覚量と物
理量の評価を行った。実験は、実物の壁面と写真試料に
ついて、その汚れの程度を５段階において検査員により
評化してもらった。実物壁面と写真試料を尺度値により
比較したグラフを第１図に示す。Sensory and physical quantities were evaluated on 34 walls within the campus of Utsunomiya University. In the experiment, inspectors were asked to rate the degree of dirt on an actual wall surface and a photographic sample on a five-point scale. Figure 1 shows a graph comparing the actual wall surface and photographic sample using scale values.
この結果、実物の壁面と写試料の価値では、尺度値の値
が水平４５度の直線上にあり、官能検査を行う上で写真
試料により、壁面と同様のデータを得られると言える。As a result, it can be said that the values of the real wall surface and the photographic sample have scale values on a straight line at 45 degrees horizontally, and that data similar to that of the wall surface can be obtained from the photographic sample when conducting sensory tests.
また、部位別に変動係数を表したグラフを第２図に示す
。これにより、全体的に変動係数の値は屋外が屋内より
大きくなっているのが分かる。Furthermore, a graph showing the coefficient of variation for each region is shown in FIG. As a result, it can be seen that the value of the coefficient of variation is generally larger outdoors than indoors.
これは、写真試料の方が、検査のバラつきが少ないため
と思われる。なお、ばらつきが大きい壁面は判断に嗜好
がまざったものだと考えられるため以降の分析から取り
除いた。This seems to be because photographic samples have less variation in inspection. Note that walls with large variations were removed from the subsequent analysis because it was thought that the judgments were based on preferences.
（３） 物理量と感覚量の評価の対応結果様々な壁面に
ついて写真試料のみの官能検査を行った。壁面について
分類わけをすると表−１のように２つに分類できた。重
回帰の結果を表−２、表−３に示す。重回帰分析の結果
を表４、表−５、表−６に示す。(3) Corresponding results of evaluation of physical quantities and sensory quantities We conducted sensory tests using only photographic samples on various wall surfaces. When we categorized the wall surfaces, we were able to classify them into two categories as shown in Table 1. The results of multiple regression are shown in Tables 2 and 3. The results of the multiple regression analysis are shown in Table 4, Table-5, and Table-6.
それぞれの試料について重回帰分析を行った結果、色差
ΔＥ、汚染部明度、汚染部面積比、複雑さ係数にのとき
相関が非常によく色差のＴ値は、４１て相関係数は約０
．８となった。As a result of performing multiple regression analysis on each sample, it was found that the color difference ΔE, the brightness of the contaminated area, the area ratio of the contaminated area, and the complexity coefficient had a very good correlation.The T value of the color difference was 41, and the correlation coefficient was approximately 0.
．． It became 8.
この結果、平均明度や汚染部面積比は重回帰ではそれほ
ど相関がよくないが他の物理量と組み合わずことにより
相関かたかくなっているのがわかる。また、色差の方が
明度差より相関が高くなりでいる。これは、明度たりで
は色彩の情報を取入れられないためであると考えられる
。官能尺度の対数をとると感覚尺度のときよりも一層、
各物理量との相関が高くなることが分かる。As a result, it can be seen that although the correlation between the average brightness and the contaminated area ratio is not so good in multiple regression, the correlation becomes stronger because it is not combined with other physical quantities. Further, the correlation is higher for color difference than for brightness difference. This is thought to be because color information cannot be taken in based on brightness. When we take the logarithm of the sensory scale, even more so than when we take the sensory scale.
It can be seen that the correlation with each physical quantity becomes high.
以上のように、汚染度に関する物理測定量と感覚量での
対応結果を比較・検討してみたところ汚染度に対する人
間の感覚量は、明度差よりも色差か相関が高くそれに汚
染部面積・複雑さ係数などを加えることにより、よりよ
い相関が得られることか認められた。また、壁面に目地
が入ったりしていた場合であっても、相関性はある程度
落ちるものの、
れた。As mentioned above, when we compared and examined the corresponding results of physically measured quantities and sensory quantities regarding the degree of contamination, we found that the human sense of the degree of contamination has a higher correlation with the color difference than with the brightness difference, as well as the area and complexity of the contaminated area. It was found that a better correlation could be obtained by adding factors such as coefficients. Furthermore, even when there were joints in the wall surface, the correlation remained, although the correlation decreased to some extent.
表
十分に精度良く判定できることも認めら表
写真試料の分類
目地を含む写真試料での
重回帰分析（Ｔ値）
５
表−３
目地を除いた写真試料での
重回帰分析（Ｔ値）
表−６感覚尺度の対数をとった値と
相関が高い物理量の重回帰分析
表−４
表−５
目地を含む写真試料での重回帰分析
目地を除いた写真試料での重回帰分析
［効果］
以上の実施例からも明らかな通り本発明方法によると、
標準試料と対比することなく、評価対象面の撮像を画像
処理することによるだけで評価対象面の汚染の程度を判
別することができる。この判別は、汚染部の大きさ、形
状、明度、色度、目地の有無等に基いて総合的に判断す
るものであり、人の目視判断結果に極めて近似した結果
か得られる。It was also confirmed that the classification of photographic samples could be determined with sufficient accuracy. Multiple regression analysis (T value) on photographic samples including joints 5 Table 3 Multiple regression analysis on photographic samples excluding joints (T value) Table - Multiple regression analysis of physical quantities that are highly correlated with the logarithm values of the six sensory scales Table-4 Table-5 Multiple regression analysis on photographic samples that include joints Multiple regression analysis on photographic samples that exclude joints [Effects] As is clear from the examples, according to the method of the present invention,
The degree of contamination of the evaluation target surface can be determined simply by image processing the captured image of the evaluation target surface without comparing it with a standard sample. This determination is a comprehensive judgment based on the size, shape, brightness, chromaticity, presence or absence of joints, etc. of the contaminated area, and results are obtained that are very similar to human visual judgment results.
本発明方法は壁面等の各種部材の汚染たけてなく、塗料
の剥れ、ひび割れなど、視覚上、閾値により２値化でき
る部分の外観の判断を行なえる。The method of the present invention can visually judge the external appearance of various parts such as walls and other parts that are contaminated, peeling paint, cracks, etc. that can be visually converted into binarization based on a threshold value.
第１図は写真試料と実際壁面との相関を示すグラ乙 第２図は変動係数を示すグラフである。 Figure 1 is a graph showing the correlation between the photographic sample and the actual wall surface. FIG. 2 is a graph showing the coefficient of variation.
Claims (2)
素の明度、色相及び彩度に基づいて表面性状の変化度を
測定する方法であって、 各画素の明度値の集合を閾値ＬｔによりＬｔよりも小な
る明度値のグループとＬｔよりも大なる明度値のグルー
プとに分割した場合に、次式で示される２グループの分
散σ＿Ｔ σ＿Ｔ＝ωｓ（Ｌｓ−Ｌａ）＾２＋ ωｂ（Ｌｂ−Ｌａ）＾２ ωｓ：低明度値グループに属する画素比率 ωｂ：高明度値グループに属する画素比率 Ｌｓ：低明度値グループに属する画素の 平均明度 Ｌｂ：高明度値グループに属する画素の 平均明度 Ｌａ：全画素の平均明度 （明度はＪＩＳＺ８７２９のＬ＾＊値である。）が最大
となる閾値Ｌｔを求め、このＬｔ値により低明度値のグ
ループの画素と高明度値のグループの画素とを確定する
と共に、 各画素のａ＾＊値及びｂ＾＊値（ＪＩＳＺ８７２９）を
演算し、 この確定されたグループのうち一方のグループに属する
画素数と全画素数との比率をＡｒとし、確定された両グ
ループの平均明度差＠Ｌｂ＠−＠Ｌｓ＠をコントラスト
ΔＬ、平均ａ＾＊値の差＠ａ＾＊＿ｂ＠−＠ｂ＾＊＿ｓ
＠をΔａ、平均＠ｂ＾＊＿ｂ−＠ｂ＾＊＿ｓ＠をΔｂと
し、これらから色差ΔＥを ΔＥ＝√［（ΔＬ）＾２＋（Δａ）＾２＋（Δｂ）＾２
］として求め、 一方のグループに属する画素の平均明度＠Ｌｂ＠又は＠
Ｌｓ＠と√（Ａｒ・ΔＥ）との重回帰により評価対象面
の変化度を評価することを特徴とする表面性状の変化度
の自動評価方法。(1) A method in which a captured image of the surface to be evaluated is input to an image analysis device and the degree of change in surface texture is measured based on the brightness, hue, and saturation of each pixel, and the set of brightness values of each pixel is set as a threshold. When Lt is divided into a group with brightness values smaller than Lt and a group with brightness values larger than Lt, the variance σ_T of the two groups is expressed by the following formula: σ_T=ωs(Ls-La)^2+ωb( Lb-La)^2 ωs: Pixel ratio belonging to the low brightness value group ωb: Pixel ratio belonging to the high brightness value group Ls: Average brightness of pixels belonging to the low brightness value group Lb: Average brightness of pixels belonging to the high brightness value group La: Find the threshold Lt at which the average brightness of all pixels (lightness is the L^* value of JIS Z8729) is maximum, and use this Lt value to distinguish between pixels in the low brightness value group and pixels in the high brightness value group. At the same time, calculate the a^* value and b^* value (JIS Z8729) of each pixel, set the ratio of the number of pixels belonging to one of the determined groups to the total number of pixels as Ar, and calculate the determined value. The average brightness difference between both groups @Lb@-@Ls@ is the contrast ΔL, and the difference between the average a^* values @a^*_b@-@b^*_s
@ is Δa, average @b^*_b-@b^*_s@ is Δb, and from these, the color difference ΔE is ΔE=√[(ΔL)^2+(Δa)^2+(Δb)^2
], and the average brightness of pixels belonging to one group @Lb@ or @
An automatic evaluation method for the degree of change in surface texture, characterized in that the degree of change in the surface to be evaluated is evaluated by multiple regression between Ls@ and √(Ar・ΔE).
に、目地の明度値の範囲を設定し、設定された範囲にあ
る明度値を有した画素を評価対象画素から除外すること
を特徴とする表面性状の変化度の自動評価方法。(2) Claim 1 is characterized in that when there is a joint on the surface to be evaluated, a range of brightness values of the joint is set, and pixels having brightness values within the set range are excluded from the pixels to be evaluated. An automatic evaluation method for the degree of change in surface texture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5947490A JPH0786471B2 (en) | 1990-03-09 | 1990-03-09 | Automatic evaluation method of degree of change of surface texture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5947490A JPH0786471B2 (en) | 1990-03-09 | 1990-03-09 | Automatic evaluation method of degree of change of surface texture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPH03259734A true JPH03259734A (en) | 1991-11-19 |
| JPH0786471B2 JPH0786471B2 (en) | 1995-09-20 |
Family
ID=13114342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP5947490A Expired - Fee Related JPH0786471B2 (en) | 1990-03-09 | 1990-03-09 | Automatic evaluation method of degree of change of surface texture |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0786471B2 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009068983A (en) * | 2007-09-13 | 2009-04-02 | Inoue Mfg Inc | Roll surface inspection device of roll mill |
| JP2014196927A (en) * | 2013-03-29 | 2014-10-16 | 大和ハウス工業株式会社 | Method and apparatus for assessing deterioration of exterior material |
| CN111999209A (en) * | 2020-08-28 | 2020-11-27 | 刘翡琼 | Anti-boiling sample thermogravimetric analysis device |
-
1990
- 1990-03-09 JP JP5947490A patent/JPH0786471B2/en not_active Expired - Fee Related
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009068983A (en) * | 2007-09-13 | 2009-04-02 | Inoue Mfg Inc | Roll surface inspection device of roll mill |
| JP2014196927A (en) * | 2013-03-29 | 2014-10-16 | 大和ハウス工業株式会社 | Method and apparatus for assessing deterioration of exterior material |
| CN111999209A (en) * | 2020-08-28 | 2020-11-27 | 刘翡琼 | Anti-boiling sample thermogravimetric analysis device |
Also Published As
| Publication number | Publication date |
|---|---|
| JPH0786471B2 (en) | 1995-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Briones et al. | Image analysis of changes in surface color of chocolate | |
| US6668078B1 (en) | System and method for segmentation of images of objects that are occluded by a semi-transparent material | |
| Lemaire et al. | Evaluating concrete surfaces using an image analysis process | |
| Leon et al. | Color measurement in L∗ a∗ b∗ units from RGB digital images | |
| WO2003029811A1 (en) | Assessing colour fastiness | |
| US9582899B2 (en) | Real-time digitally enhanced imaging for the prediction, application, and inspection of coatings | |
| JPH07231883A (en) | Skin surface analyzing system, and skin surface analyzing method | |
| JP2000350702A (en) | Measurement method and apparatus for cutaneous constituents and characteristics | |
| US20220381615A1 (en) | Method and Device for Identification of Effect Pigments in a Target Coating | |
| TWI550526B (en) | Method of setting the check condition of the sprocket and the method of checking the sprocket | |
| JP2018004421A (en) | Apparatus and method for evaluating feel of metal surface | |
| JPH03259734A (en) | Automatic evaluating method for variation degree of surface property | |
| JP3396960B2 (en) | Automatic coating deterioration diagnosis method | |
| JP3140226B2 (en) | Automatic color tone judgment method | |
| WO2003031956A1 (en) | System and method for classifying workpieces according to tonal variations | |
| JP7411155B2 (en) | Color unevenness inspection device and color unevenness inspection method | |
| JP2707743B2 (en) | Automatic evaluation method of degree of change of surface texture | |
| CN111192263B (en) | Intelligent energy-saving indoor people counting method based on machine vision | |
| JP2011179987A (en) | Method and device for measurement of residue | |
| JPH09101265A (en) | Method for automatically inspecting al plate for surface defect | |
| Tzu et al. | Non-uniformity evaluation of flat panel display by automatic optical detection | |
| JPH08297098A (en) | Method and apparatus for measuring gloss irregularity of coated paper board | |
| Boukouvalas et al. | Automatic grading of textured ceramic tiles | |
| van Dalen et al. | Colour analysis of inhomogeneous stains on textile using flatbed scanning and image analysis. | |
| JPH04147002A (en) | Inspecting apparatus of surface of fruit and vegetable |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |
|
| LAPS | Cancellation because of no payment of annual fees |