201233974 六、發明說明： 【發明所屬之技術領域】 本發明係關於一種藉由微細表面凹凸之賦予而進行基 材之防眩處理所使用之模具的檢查裝置，詳言之，係關於 一種於表面具有電鍍層之前述模具的自動檢查裝置。〔「防 眩」與「防炫」在本件之技術領域習慣上已通用，均指 anti-glare，防止反光之意思〕 【先前技術】 以往，為了防止外光映入液晶顯示裝置等圖像顯示裝 置的顯示面，於其表面配置防眩膜。防眩膜係例如使分散 有微粒子之樹脂組成物塗佈於基材薄膜上，藉由該微粒子 而顯現表面凹凸來製作。但，含有微粒子之防眩薄膜係藉 由起因於微粒子之内部散射，使顯示面全體泛白，顯示變 濁色，易產生所謂「白暈」之傾向。 另外，亦已提出一種不含有微粒子，使模具之表面凹 凸形狀轉印至基材薄膜上，而賦予表面凹凸之防眩薄膜。 若依此種防眩薄膜，可防止白暈及實現亮處的高對比度。 又，使用模具賦予表面凹凸之防眩薄膜係不使用微粒子而 只以表面凹凸形狀改變防眩薄膜之光學特性，故有比較容 易依照市場要求之光學特性的設計變更之優點。 在使用模具之防眩薄膜的生產中係模具表面之缺陷管 理極重要。在防眩薄膜之工業生產中，通常使用長尺寸的 基材薄膜連續地生產防眩薄膜，但若模具表面存在缺陷， 長尺寸之防眩薄膜的全長全部會轉印缺陷。因此，可適時 4 323801 201233974 爾 • ,·修補因使用模具所產生（或，從模具製作時存在）之缺陷， • · 要求如下技術：可精度佳地檢測存在於模具表面之缺陷的 及判斷是否為缺陷。 在防眩處理用之模具中，係為防止其使用所造成之擦 傷產生，故一般對其表面實施電鍍，但形成電鍍層時有 時產生例如直徑約100至200 左右深度或高度為數“111 左右的凹狀或凸狀之缺陷，又，亦有時因模具之使用產生 此種缺陷。在模具之檢查中係必須以高精度檢測出此等缺 陷。另一方面，在模具之檢查中，亦必須可精度佳地判斷 疑似為缺陷之處，實際需要修補之缺陷、或僅為附著於表 面之污垢或銹等而不需修補者。 進行模具表面之缺陷檢測的方法係可想到使用顯微光 學系之照相機，取得模具表面之放大圖像的方法。此時， 為了從所取得之放大圖像精度佳地進行缺陷檢測，係必須 於模具表面對準焦點而得到鮮明之放大圖像。以往，顯微 光學系之自動對焦一般係採用對比度檢測方式 ，亦即，可 得到對比度最高的圖像之照相機位置作為焦點位置之方式 {例如日本特開平10-048505號公報（專利文獻1)}。 [先前技術文獻] [專利文獻] [專利文獻1]日本特開平10-048505號公報 【發明内容】 (發明欲解決之課題） 然而’藉由本發明者之檢討係得知在防眩處理用模具 323801 5 201233974 表面之圖像取得中係藉由防眩處理用之表面凹凸的影響， 可得到對比度最高的圖像之照相機位置與實際之焦點位置 · ' 不一致（亦即，在與實際之焦點位置相異的位置對比度變成 · · 最高），在以往一般所使用之對比度檢測方式中係無法得到 鮮明之圖像。 ： 本發明之目的在於提供一種檢查裝置，係對於具有表 面凹凸之防眩處理用模具的表面可取得焦點對準之鮮明放 大圖像，再者，可精度良好地進行檢測缺陷及判斷是否為 缺陷。又，本發明更進一步目的在於提供一種可自動進行 此種缺陷檢查之檢查裝置。 (解決課題之手段） 本發明關於一種用以檢查於表面具有電鍍層之防眩處 理用模具之電鍍表面的檢查裝置，本發明之檢查裝置，係 具備：對電鍍表面之至少一部分照射光之照明；用以取得 被光照射之區域的電鍍表面之圖像且解析度為25〇em以 下之照相機，對於該圖像進行邊緣抽離過處理，以獲得處 理圖像之圖像處理手段；計算該處理圖像之亮度分散值的 第1演算手段；移動照相機而改變照相機與電鍍表面之間 的距離之照相機移動手段；檢測上述亮度分散值成為最大 之照相機位置之對焦位置檢測手段。 上述對焦位置檢測手段較佳係包括：控制照相機之攝 影的第1控制手段；控制上述照相機移動手段之驅動的第 2控制手段；檢測照相機位置之照相機位置檢測手段；記 憶手段，係記憶從照相機位置檢測手段所輸出之照相機攝 323801 t 6 201233974 影時之照相機位置、及從第1演算手段所輸出且從該照相 ' 機攝影所得到之亮度分散值；第2演算手段，係依據記憶 ~ 手段所記憶之照相機位置及亮度分散值，決定亮度分散值 - 成為最大之照相機位置。 , 本發明之檢查裝置係藉由上述對焦位置檢測手段所檢 測出之亮度分散值成為最大之照相機位置移動照相機而取 得圖像，使用該圖像而進行電鍵表面之檢查。 本發明之檢查裝置，可進一步具備：藉由使檢查對象 之防眩處理用模具移動或旋轉，而移動電鍍表面之光照射 的區域之模具移動手段。 上述照明宜為從照相機之光軸與同軸上照射之同軸照 明。又，防眩處理用模具所具有之電鍍層例如為鍍鉻層。 (發明效果） 若依本發明之檢查裝置，藉由採用與習知對比度檢測 方式相異之新的自動對焦機構，而對於具有表面凹凸之防 眩處理用模具的表面而可取得焦點對準之鮮明放大圖像， 故可精度良好地檢測缺陷及判斷是否為缺陷，且自動地進 行。藉此，可提昇模具之持續精度，相同模具因防眩處理 而可經過長期間保持良好的品質。 【實施方式】 〈檢查裝置〉 第1圖係表示本發明之檢查裝置的構成之一例的區塊 圖。本發明之檢查裝置係用以檢查於表面具有電鍍層之防 眩處理用模具之電鍍表面的裝置，更具體而言，使用藉由 7 323801 201233974 照相機取得之放大圖像進行檢測在電鍍表面之凸狀缺陷或 凹狀缺陷，或判斷疑似為缺陷之處實際上是否為缺陷的檢 查裝置。 如第1圖所示般，本發明之檢查裝置係具備：對電鍍 表面之至少一部分照射光之照明101 ;用以取得被光照射 之區域的電鍍表面之圖像之照相機102 ;對於所得到之圖 像進行邊緣抽離過濾處理，得到處理圖像之圖像處理手段 103;計算所得到之處理圖像之亮度分散值的第1演算手段 104 ;移動照相機使得照相機與電鍍表面之間的距離改變之 照相機移動手段105;及檢測亮度分散值成為最大之照相 機位置之對焦位置檢測手段106。以下，更詳細說明有關 本發明之防眩處理用模具的檢查裝置。 (1)照明 對電鍍表面照射光之照明101係可使用白熾燈、螢光 燈、發光二極體（LED)等各種照明，可為面發光型，亦可為 點照明型。照明光之波長係無特別限定，而可使用例如可 見光（波長範圍380至800nm)或白色光。由於不易因電鍍 表面之凹凸而產生影子，又，可得到均一的照明狀態，故， 照明101宜為同軸照明。所謂同軸照明係利用半鏡從與照 相機光軸之同軸上（與照相機之光轴方向平行）照射照明光 之照明。一般而言，同軸照明係與照相機所使用之透鏡系 一體化而使用。可同軸照明之透鏡的市售品可舉例如 Keyence股份有限公司製之「CA-LMA1」、「CA-LMA2」、 「LA-LMA4」等。於此等之透鏡裝載例如Keyence股份有限 8 323801 201233974 Λ ·..公司製之點照明「CA-DPW2」，則可同軸照明。 * (2)照相機 • 帛以取得電鑛表面的圖像之照相機1〇2係'包含透鏡 •彡、從被光照射之區域的⑽表面隔著該透鏡系而入射之 '光轉換成電訊號之攝影元件。攝影元件係宜為⑽、· 等二維攝影元件。從照相機1Q2之攝影元件所輸出之電訊 號係藉由圖像轉換手段（在第i圖中未圖示）而轉換成電鍵 表面之圖像，將該圖像送出至圖像處理手段1〇3。 CCD照相機（透鏡別）之市售品係例如Kw股份有 限公司製之「CV-H500C」、「CV-H500M」、「CMMOC」、 「CV-H200M」、「CV-H100C」、「CV-H100M」、及 0M_ 股份有 限公司製之「FZ-SC5M」、「FZ-S5M」、「FZ-SC2M」、「FZ-S2M」 等。 在本發明之檢查裝置中’照相機1〇2之解析度為25〇 以下。此係如下之理由。亦即如後述般，在本發明之 檢查裝置係使藉由將存在於電鍍表面之溝所形成的網目狀 之圖案解析’並實施強化鱗之邊緣減輯處理，因此， 必須以充分的解析度拍攝網目狀之圖案。就防眩處理用模 具之表面電鍍而言，試作複數次實施一般鍍鉻之模具，並 確認後，使網目狀之圖案的平均線間隔約為25〇以瓜，平均 線寬（溝寬）為約6/zm。因此，判斷為了解析網目狀之圖案， 必須為250以下之解析度。照相機1〇2之解析度係電鍍表 面之攝影區域有鑑於檢查效率等，只要極端小，愈高俞佳。 照相機102之解析度係宜為以下，更宜為 323801 9 201233974 以下。 此處，在本發明令照相機之解析度係對應於在攝影區 電鑛表面上的攝影元件之像素丨邊長度的長度或透鏡 =析度中較大者。透鏡之解析度經常是所使用之 2一透鏡製造商提^值。依製造商刊亦有_解析度 刀析此力等。對於製造商未揭示解析度之透鏡，亦已揭 示開口數時，使用例如稱為「雷立（Rayieigh)^基準」的 下式： 解析度 d 二〇. 61χλ/ΝΑ 昭算出透鏡之解析度5。此處，λ係表示於檢查所使用之 、月光的波長。為了計算解析度5，只要於又代入在測定 =用的波長即可。使用白色光時，可以發絲譜之波峰波 長估計解析度5。_為對物透鏡的開口數。 (3)圖像處理手段 〜圖像處理手段103係對於藉由以照相機102的攝影所 得至ij之雷蚀主 蚩 表面的圖像進行邊緣抽離過濾處理。在本說明 ;，稱邊緣抽離過濾處理後之圖像為處理圖像。如上述 在具有凹凸之防眩處理用模具表面的圖像取得中係受 二、二凹凸的影響，可得到對比度最高之圖像的照相機位 、實際焦點位置不一致（亦即，在與實際之焦點位置相 、、置對比度變成最高）’因此難以藉由習知之對比度檢 3 =到鮮明之圖像。此係因表面凹凸所產生的影子（此 :異=指因表面凹凸而使得照明光朝與照相機之透鏡方向 、、方向散射）與對焦位置相異之位置（未對準焦點的位 323801 10 201233974 . 置）變強，影愈強之處，對比度愈變高，結果出現對焦位置 與顯示最大之對比度的位置相異之照相機位置。 ^ 依本發明，若於所取得之電鍍表面的圖像使用邊緣抽 離過濾時，在該圖像中，存在於電鍍表面之上述圖案依該 ， 過濾的演算法而強化。藉此，起因於該圖案之對比度變大， 另一方面，起因於表面凹凸之影子的對比度相對地變小， 故可得到實質上排除表面凹凸影子之影響且對圖像全體之 對比度的影響之圖像（處理圖像）。此係因在表面凹凸之影 子中，鄰接之像素間的亮度差小而成。有關此種處理圖像 藉由使用以對比度檢測方式之自動對焦，可使顯示最大之 對比度的照相機位置與實際之對焦位置一致，因此，藉由 使照相機移動至顯示最大之對比度的位置，可得到焦點對 準之圖像。 存在於電鍍表面之上述圖案通常為溝狀之圖案，於電 鍍層形成時，為了涵蓋電鍍表面全體而形成網目狀者（此種 溝狀的圖案係例如在鍍鉻中為典型性）。於此種電鍍層形成 時所形成的圖案相較於防眩處理用之表面凹凸之圖案非常 小，不對防眩特性造成影響。藉由照相機102所取得之圖 像係可藉邊緣抽離處理而強化圖案，必須使該圖案以充分 的解析度攝影（因此，如上述般照相機係要求250 #m以下 之解析度）。 此處，說明有關「邊緣抽離過濾」，邊緣抽離過濾係指 強化與鄰接像素之濃度差（亮度差）之處理。若舉出邊緣抽 離過濾之具體例，例如有「邊緣抽離X方向過濾」、「邊緣 11 323801 201233974 抽離Y方向過濾」、「Sobel過濾」、「Prewitt過濾」、「Roberts 過濾」、「Laplacian過濾」等。其中，以1次微分過濾之 「邊緣抽離X方向過滤」、「邊緣抽離γ方向過」、「Sobel 過濾」、「Prewitt過濾」、「Roberts過濾」等為佳。 將Sobel過濾、Prewitt過濾、邊緣抽離X方向過濾、 及邊緣抽離Y方向過濾之演算法分別表示於第2圖（a)至 (d)。任一過濾係使3x3之像素區域之中心像素分別對於9 像素乘上圖所示之係數後，取代成合算之濃度值的處理。 對於全部之像素進行此處理。S〇bel過濾及Prewitt過濾 係合成2個邊緣抽離之結果的邊緣抽離處理。若使用邊緣 才由離X方向過濾’見到此演算法即可知，可得到於橫方向 (X方向）被強化之圖像。在邊緣抽離γ方向過濾中係可得 到進行縱方向（Y方向）被強化之圖像。在S〇bel過濾及 Prewitt過濾中係進行縱方向之強化後，可得到合成進行 橫方向之強化（或進行其反方向）之結果的圖像。 又’在3x3之像素區域中取出濃度值之最大值與最小 值’藉由使用以其差作為中心像素的新濃度值之MAX-MIN 過據的過濾處理或第2圖（e)所示之邊緣強化過濾處理，亦 有時可強化存在於電鍍表面之圖案。 圖像處理手段103係可使用市售之圖像處理裝置、個 人電腦等計算機。市售之圖像處理裴置可例示Keyence股 知有限公司製之「XG-7000系列」、「CV-5000系列」。個人 電月岛係由各公司販售。例如可使用搭載I1 c〇rporat ion 製Core系列、xe〇n系列等CPU的系統組合進行邊緣抽離 12 323801 201233974 • 過渡處理之軟體。又，亦可使用Keyence股份有限公司製 之「CV-5000」等控制器作為圖像處理手段103。 ‘ （4)第1演算手段 • 計算處理圖像的亮度分散值之第1演算手段1〇4係可 ' 同樣地使用作為上述圖像處理手段103所例示者。圖像處 理手段103亦可具備作為第1演算手段1〇4之功能。第i 演算手段104係對於所得到之處理圖像的各像素計算亮 度，藉由算出此等之亮度值的分散而得到亮度分散值。 (5) 照相機移動手段 移動照相機102而改變照相機102與電鍍表面之間的 距離的照相機移動手段105係只要為可對於照相機1〇2之 焦點深度以充分精度定位之驅動器即可，無特別限制，可 使用各種驅動器。較佳之驅動器係可舉例如SUS股份有限 公司製之「XA系列」、Oriental Motor股份有限公司製之 「EZX系列」、THK股份有限公司製之「GLM系列」等附屬 直線運動機構之驅動器。 (6) 對焦位置檢測手段 檢測亮度分散值成為最大之照相機位置之對焦位置檢 ’則手段106如第1圖所示般較佳係包括：控制照相機 之攝影的第1控制手段107 ;控制照相機移動手段1〇5之 驅動的第2控制手段1〇8 ;檢測照相機位置之照相機位置 檢測手段109 ;記憶手段no，其係記憶從照相機位置檢測 手段109所輸出之照相機攝影時之照相機位置、及從第丄 演算手段104所輸出且從照相機攝影所得到之亮度分散 323801 13 201233974 值；第2演算手段111，其係依據記憶手段u〇所記憶之 照相機位置及亮度分散值，決定亮度分散值成為最大之照 相機位置。 第1控制手段107、第2控制手段108、照相機位置檢 測手段109、記憶手段11〇及第2演算手段m係亦可分 別兼具其他任一者一個以上或全部之功能，進一步亦可兼 具圖像處理手段103或第1演算手段1〇4之功能。 對焦位置檢測手段10 6係可使用例如於個人電腦組合 適合之軟體者、或由三菱電機股份有限公司及〇MR〇N股份 有限公司等所製造、販賣之可程式邏輯控制器（PLC)等。 照相機位置檢測手段109係可為組入於第2控制手段 108之脈沖計數器、組入於照相機移動手段1〇5之編碼器、 或與照相機102 —起移動的方式固定於照相機移動手段 105之距離測定器等。組入脈沖計數器之第2控制手段1〇8 係可使用例如OMRON股份有限公司所販賣之位置控制單元 「CJIW-NC 214型」、Cosmotechs股份有限公司之控制器「SP 120」、SUS股份有限公司製之控制器「XA-S4」等。此等之 裝置通常若得到指令，具有將目前之脈沖計算值換算成位 置而輸出之功能。可藉該功能容易地檢測出照相機位置。 組入於照相機移動手段105之編碼器係可使用例如由 Mitutoyo股份有限公司或Renishaw股份有限公司所販賣 之各種線性編碼器。此時，可從編碼器之輸出檢測出照相 機位置。又’距離測定器係可使用Mi tutoyo股份有限公司 所販賣之數位量尺計、Keyence股份有限公司所販賣之雷 323801 14 201233974 Λ . 射移位計「LK-G5000系列」等。 * ' (7)模具移動手段 % 本發明之檢查裝置係可具備使檢查對象之防眩處理用 - 模具移動或旋轉的模具移動手段。模具移動手段係可使用 步進馬達、祠服馬達、DC馬達等各種驅動手段。從移動量 之控制性的觀點，特別宜使用步進馬達。又，藉由必要之 精度、扭力，亦可進一步組合各種齒輪。步進馬達例如有201233974 VI. OBJECTS OF THE INVENTION: The present invention relates to an inspection apparatus for a mold used for anti-glare treatment of a substrate by imparting fine surface irregularities, and more particularly to a surface. An automatic inspection device having the aforementioned mold of the plating layer. ["Anti-glare" and "anti-glare" are common in the technical field of this article, and both refer to anti-glare to prevent reflection. [Prior Art] In the past, in order to prevent external light from being reflected on the image display such as a liquid crystal display device The display surface of the device is provided with an anti-glare film on its surface. The anti-glare film is produced by, for example, applying a resin composition in which fine particles are dispersed to a base film, and revealing surface irregularities by the fine particles. However, the anti-glare film containing fine particles tends to cause whitening of the entire display surface due to internal scattering due to the fine particles, and tends to cause so-called "white halo". Further, an anti-glare film which does not contain fine particles and which transfers the concave shape of the surface of the mold to the base film to impart unevenness to the surface has been proposed. According to this anti-glare film, it can prevent white halo and achieve high contrast in bright places. Further, the antiglare film which imparts unevenness to the surface by using a mold does not use fine particles, and changes the optical characteristics of the antiglare film only by the surface unevenness. Therefore, there is an advantage that the design of the optical characteristics which are more easily required by the market can be changed. Defect management of the mold surface is extremely important in the production of anti-glare films using molds. In the industrial production of an anti-glare film, an anti-glare film is usually continuously produced using a long-sized base film, but if there is a defect on the surface of the mold, all of the long-length anti-glare film will have a transfer defect. Therefore, it is possible to correct the defects caused by the use of the mold (or from the time of mold production), and the following techniques are required: the defects existing on the surface of the mold can be accurately detected and judged whether For defects. In the mold for anti-glare treatment, in order to prevent the occurrence of scratches caused by the use thereof, plating is generally performed on the surface thereof, but when the plating layer is formed, for example, a depth of about 100 to 200 or a height of about "111" is sometimes generated. Concave or convex defects, and sometimes such defects are caused by the use of the mold. In the inspection of the mold, these defects must be detected with high precision. On the other hand, in the inspection of the mold, It is necessary to accurately judge the defects that are suspected to be defects, the defects that need to be repaired, or the dirt or rust attached to the surface, etc., without repairing. The method of detecting defects on the surface of the mold is conceivable to use microscopic optics. A camera that acquires an enlarged image of the surface of the mold. In this case, in order to accurately detect the defect from the obtained enlarged image, it is necessary to obtain a sharp enlarged image by focusing the focus on the surface of the mold. The autofocus of the microscopic optical system generally adopts the contrast detection method, that is, the camera position at which the image with the highest contrast is obtained as the focus position. [Patent Document 1] [Patent Document 1] [Patent Document 1] Japanese Laid-Open Patent Publication No. Hei 10-048505 (Summary of the Invention) However, it has been found that the image of the surface of the anti-glare processing mold 323801 5 201233974 is obtained by the surface unevenness of the anti-glare treatment, and the camera position of the image with the highest contrast can be obtained. It does not coincide with the actual focus position ' (that is, the position contrast differs from the actual focus position becomes the highest), and a clear image cannot be obtained in the contrast detection method generally used in the past. An object of the present invention is to provide an inspection apparatus capable of obtaining a sharply magnified image of a focus on a surface of an anti-glare processing mold having surface irregularities, and further capable of accurately detecting a defect and determining whether or not it is a defect. It is still a further object of the present invention to provide an inspection apparatus that can automatically perform such defect inspection. The present invention relates to an inspection apparatus for inspecting a plating surface of an anti-glare treatment mold having an electroplated layer on a surface thereof, the inspection apparatus of the present invention comprising: illumination for irradiating at least a portion of the electroplated surface; a camera having an image of the plated surface and having a resolution of 25 〇 or less, performing edge extraction processing on the image to obtain an image processing means for processing the image; and calculating a brightness dispersion value of the processed image The first calculation means; the camera moving means for changing the distance between the camera and the plating surface by moving the camera; and the focus position detecting means for detecting the camera position where the brightness dispersion value becomes the maximum. The focus position detecting means preferably includes: a first control means for photographing the camera; a second control means for controlling the driving of the camera moving means; a camera position detecting means for detecting the position of the camera; and a memory means for storing the camera taken by the camera position detecting means 323801 t 6 201233974 Camera position and time from the first calculation The brightness dispersion value obtained by the means and obtained from the camera photography; the second calculation means determines the brightness dispersion value based on the camera position and the brightness dispersion value memorized by the memory means - the maximum camera position. In the inspection apparatus of the present invention, the camera is moved by the camera position at which the luminance dispersion value detected by the focus position detecting means is maximized, and an image is obtained, and the surface of the key is inspected using the image. Further, the inspection apparatus of the present invention may further include a mold moving means for moving a region irradiated with light of the plating surface by moving or rotating the mold for preventing glare treatment. Preferably, the illumination is a coaxial illumination that is illuminated from the optical axis of the camera and coaxially. Further, the plating layer of the mold for anti-glare treatment is, for example, a chrome plating layer. (Effect of the Invention) According to the inspection apparatus of the present invention, by using a new autofocus mechanism different from the conventional contrast detection method, focus can be obtained on the surface of the mold for anti-glare treatment having surface irregularities. By clearly magnifying the image, it is possible to accurately detect the defect and determine whether it is a defect and automatically perform it. Thereby, the continuous precision of the mold can be improved, and the same mold can maintain good quality over a long period of time due to anti-glare treatment. [Embodiment] <Inspection apparatus> Fig. 1 is a block diagram showing an example of the configuration of an inspection apparatus according to the present invention. The inspection apparatus of the present invention is for inspecting a plating surface of an anti-glare processing mold having a plating layer on the surface, and more specifically, using a magnified image obtained by a camera of 7 323801 201233974 to detect a convex surface on a plating surface. A defect or a concave defect, or an inspection device that determines whether the defect is actually a defect. As shown in Fig. 1, the inspection apparatus of the present invention includes: an illumination 101 for irradiating at least a part of the plating surface; and a camera 102 for obtaining an image of the plating surface of the region irradiated with light; The image is subjected to edge extraction filtering processing to obtain an image processing means 103 for processing the image; a first calculation means 104 for calculating the luminance dispersion value of the obtained processed image; and moving the camera to change the distance between the camera and the plating surface The camera moving means 105; and the in-focus position detecting means 106 for detecting the camera position in which the luminance dispersion value is the largest. Hereinafter, the inspection apparatus for the mold for anti-glare treatment of the present invention will be described in more detail. (1) Illumination The illumination 101 for illuminating the surface of the plating can be used for various types of illumination such as incandescent lamps, fluorescent lamps, and light-emitting diodes (LEDs), and can be either a surface-emitting type or a point-illuminated type. The wavelength of the illumination light is not particularly limited, and for example, visible light (wavelength range 380 to 800 nm) or white light can be used. Since the shadow is not easily generated by the unevenness of the plating surface, and a uniform illumination state is obtained, the illumination 101 is preferably coaxial illumination. The so-called coaxial illumination uses a half mirror to illuminate the illumination light from the coaxial with the optical axis of the camera (parallel to the optical axis of the camera). In general, a coaxial illumination system is used in combination with a lens used in a camera. Commercial products of the coaxially illuminable lens include, for example, "CA-LMA1", "CA-LMA2", and "LA-LMA4" manufactured by Keyence Corporation. For such lens loading, for example, Keying Co., Ltd. 8 323801 201233974 Λ ·.. company's point lighting "CA-DPW2", can be coaxial lighting. * (2) Camera • The camera that captures the image of the surface of the electric ore is equipped with a lens • 彡, and the light that is incident from the surface of the (10) surface of the light-irradiated area through the lens system is converted into a signal. Photography component. The photographic element is preferably a two-dimensional photographic element such as (10), . The electric signal output from the photographic element of the camera 1Q2 is converted into an image of the surface of the electric key by an image conversion means (not shown in FIG. 1), and the image is sent to the image processing means 1 〇 3 . Commercial products of the CCD camera (lens) are, for example, "CV-H500C", "CV-H500M", "CMMOC", "CV-H200M", "CV-H100C", "CV-H100M" manufactured by Kw Corporation. "FZ-SC5M", "FZ-S5M", "FZ-SC2M", "FZ-S2M", etc., manufactured by 0M_ Corporation. In the inspection apparatus of the present invention, the resolution of the camera 1〇2 is 25 〇 or less. This is the reason for the following. In other words, as described later, in the inspection apparatus of the present invention, the mesh-like pattern formed by the groove existing on the plating surface is analyzed and the edge reduction processing of the scale is performed, so that sufficient resolution must be obtained. Take a mesh-like pattern. For the surface plating of the mold for anti-glare treatment, it is tried to perform a general chrome-plated mold in a plurality of times, and after confirming, the average line spacing of the mesh-like patterns is about 25 〇, and the average line width (groove width) is about 6/zm. Therefore, it is judged that in order to analyze the mesh-like pattern, it is necessary to have a resolution of 250 or less. The resolution of the camera 1〇2 is the photographic area of the plating surface. In view of the inspection efficiency, etc., as long as it is extremely small, the higher the Yu Jia. The resolution of the camera 102 is preferably the following, and is preferably 323801 9 201233974 or less. Here, in the present invention, the resolution of the camera corresponds to the length of the pixel edge length of the photographic element on the surface of the illuminating area of the photographic region or the larger of the lens = resolution. The resolution of the lens is often the value of the lens manufacturer used. According to the manufacturer's magazine, there is also a _resolution degree. For lenses that do not reveal the resolution of the manufacturer, when the number of openings is also disclosed, for example, the following formula called "Rayieigh^ benchmark" is used: resolution d 〇. 61χλ/ΝΑ The resolution of the lens is calculated 5 . Here, λ is the wavelength of the moonlight used for inspection. In order to calculate the resolution 5, it is sufficient to substitute the wavelength for measurement = use. When white light is used, the peak length of the hairline spectrum can be estimated by a resolution of 5. _ is the number of openings of the objective lens. (3) Image processing means - The image processing means 103 performs edge extraction filtering processing on the image of the surface of the ruthenium main ridge which is obtained by the imaging of the camera 102. In the present description, the image after the edge extraction filtering process is referred to as a processed image. As described above, in the image acquisition of the surface of the mold for anti-glare treatment having irregularities, the camera position and the actual focus position of the image with the highest contrast are affected by the influence of the second and second unevenness (that is, the actual focus is The position phase and the contrast ratio become the highest) 'It is therefore difficult to check the contrast image by the conventional contrast 3 = to a sharp image. This is the shadow caused by the surface irregularities (this: the difference = the surface of the lens caused by the surface irregularities, the direction of the lens, the direction of scattering) and the focus position is different (the misaligned focus 323801 10 201233974 When the image becomes stronger, the sharper the image, the higher the contrast, and the camera position where the focus position is different from the position where the maximum contrast is displayed. According to the present invention, if the image of the plated surface obtained is subjected to edge extraction filtering, the pattern present on the plated surface in the image is strengthened by the filtering algorithm. As a result, the contrast of the pattern is increased, and on the other hand, the contrast due to the shadow of the surface unevenness is relatively small, so that the influence of the surface unevenness shadow is substantially eliminated and the contrast of the entire image is affected. Image (process image). This is because the difference in luminance between adjacent pixels is small in the shadow of the surface unevenness. With regard to such a processed image, by using the autofocus with the contrast detection method, the camera position at which the maximum contrast is displayed can be made coincident with the actual focus position, and therefore, by moving the camera to the position where the maximum contrast is displayed, The image that is in focus. The pattern existing on the plating surface is usually a groove-like pattern, and when the electroplated layer is formed, a mesh is formed to cover the entire plating surface (such a groove-like pattern is typically chrome-plated, for example). The pattern formed when such a plating layer is formed is very small compared to the pattern of surface irregularities for anti-glare treatment, and does not affect the anti-glare property. The image obtained by the camera 102 can be enhanced by the edge extraction process, and the pattern must be photographed with sufficient resolution (hence, the camera system requires a resolution of 250 #m or less as described above). Here, the description will be given regarding the "edge extraction filter", and the edge extraction filter refers to a process of enhancing the density difference (luminance difference) between adjacent pixels. For specific examples of edge extraction filtering, for example, "edge extraction from X-direction filtering", "edge 11 323801 201233974 extraction from Y-direction filtering", "Sobel filtering", "Prewitt filtering", "Roberts filtering", " Laplacian filtering" and so on. Among them, the "edge extraction X direction filter", "edge extraction γ direction", "Sobel filter", "Prewitt filter", "Roberts filter", etc. are preferred. The algorithms for Sobel filtering, Prewitt filtering, edge-extracting X-direction filtering, and edge-extracting Y-direction filtering are shown in Figures 2(a) through (d), respectively. Any filtering system replaces the central pixel of the pixel region of 3x3 with the factor shown in the figure for 9 pixels, and replaces it with the processing of the concentration value. This processing is performed for all pixels. The S〇bel filter and the Prewitt filter synthesize the edge extraction process of the two edge extraction results. If the edge is used, it is filtered from the X direction. As can be seen from this algorithm, an image enhanced in the horizontal direction (X direction) can be obtained. An image in which the longitudinal direction (Y direction) is reinforced is obtained in the edge extraction γ-direction filtering. In the S〇bel filtration and the Prewitt filtration, the longitudinal direction is strengthened, and an image obtained by synthesizing the transverse direction (or the opposite direction) can be obtained. Further, 'the maximum value and the minimum value of the density values are taken out in the pixel region of 3x3' by the filtering process using the MAX-MIN of the new density value whose center pixel is the difference, or the image shown in Fig. 2(e) Edge-enhanced filtration treatment sometimes also enhances the pattern present on the plated surface. As the image processing means 103, a computer such as a commercially available image processing apparatus or a personal computer can be used. The commercially available image processing apparatus can be exemplified by "XG-7000 Series" and "CV-5000 Series" manufactured by Keyence Corporation. Personal electricity islands are sold by companies. For example, edge extraction can be performed using a system combination of CPUs such as the I1 c〇rporat ion Core series and the xe〇n series. 12 323801 201233974 • Software for transition processing. Further, a controller such as "CV-5000" manufactured by Keyence Co., Ltd. may be used as the image processing means 103. ‘(4) First calculation means • The first calculation means 1 to 4 for calculating the luminance dispersion value of the processed image can be similarly used as the image processing means 103. The image processing means 103 may also have a function as the first calculation means 1〇4. The i-th calculation means 104 calculates the luminance for each pixel of the obtained processed image, and obtains the luminance dispersion value by calculating the dispersion of the luminance values. (5) The camera moving means 105 moves the camera 102 to change the distance between the camera 102 and the plating surface, and the camera moving means 105 is not particularly limited as long as it can be positioned with sufficient precision for the depth of focus of the camera 1〇2. A variety of drives are available. For example, the "XA Series" manufactured by SUS Co., Ltd., the "EZX Series" manufactured by Oriental Motor Co., Ltd., and the "GLM Series" manufactured by THK Co., Ltd., such as the driver of the auxiliary linear motion mechanism. (6) The focus position detecting means detects the focus position detection of the camera position at which the luminance dispersion value becomes the largest. The means 106 as described above preferably includes the first control means 107 for controlling the shooting of the camera; and controls the movement of the camera. The second control means 1〇8 driven by means 1〇5; the camera position detecting means 109 for detecting the position of the camera; and the memory means no, which memorizes the position and the position of the camera at the time of camera shooting output from the camera position detecting means 109 The brightness distribution 323801 13 201233974 value which is output by the first calculation means 104 and obtained from camera photography; the second calculation means 111 determines the brightness dispersion value to be the maximum according to the camera position and the brightness dispersion value memorized by the memory means u Camera position. The first control means 107, the second control means 108, the camera position detecting means 109, the memory means 11 and the second arithmetic means m may each have one or more or all of the functions of the other, and may further have both The function of the image processing means 103 or the first calculation means 1〇4. For the focus position detecting means 106, for example, a suitable software for a personal computer combination or a programmable logic controller (PLC) manufactured or sold by Mitsubishi Electric Corporation and 〇MR〇N Co., Ltd. can be used. The camera position detecting means 109 may be a pulse counter incorporated in the second control means 108, an encoder incorporated in the camera moving means 1〇5, or fixed to the camera moving means 105 in such a manner as to move together with the camera 102. Tester, etc. For the second control means 1〇8 of the pulse counter, for example, the position control unit "CJIW-NC 214" sold by OMRON Co., Ltd., the controller "SP 120" of Cosmotechs Co., Ltd., and SUS Co., Ltd. can be used. Controller "XA-S4" and so on. Such devices typically have the function of converting the current pulse calculation value into a position and outputting it if commanded. The camera position can be easily detected by this function. The encoder incorporated in the camera moving means 105 can use various linear encoders such as those sold by Mitutoyo Co., Ltd. or Renishaw Co., Ltd. At this point, the camera position can be detected from the output of the encoder. Further, the distance measuring device can use a digital scale sold by Mi Tutoyo Co., Ltd., a mine sold by Keyence Co., Ltd. 323801 14 201233974 Λ . A shift meter "LK-G5000 series". * (7) Mold moving means % The inspection apparatus of the present invention may be provided with a mold moving means for moving or rotating the mold for preventing the anti-glare treatment of the object to be inspected. The mold moving means can use various driving means such as a stepping motor, a squat motor, and a DC motor. It is particularly preferable to use a stepping motor from the viewpoint of controllability of the amount of movement. Further, various gears can be further combined by the necessary precision and torque. Stepper motors are for example
Oriental Motor股份有限公司製之「AS系列」、「AR系列」，"AS Series" and "AR Series" made by Oriental Motor Co., Ltd.
伺服馬達例如有Oriental Motor股份有限公司製之「NX 系列」’可取得分別與齒輪組合之製品。藉由具備模具移動 手#又’可對模具之電鍵表面全體自動進行檢查。 〈使用檢查裝置之防眩處理用模具的檢查〉 其次，一邊參照第3圖，一邊說明有關使用上述本發 明之檢查裝置之防眩處理用模具的檢查方法。首先，於檢 查對象之防眩處理用模具的電鍍表面，使用照明1〇1照設 照明光’使照設照明光之區域的圖像依據第1控制手段1〇7 之攝影指令（快門觸發（Shutter trigger)發送訊號），以照 相機102攝影（步驟3〇1)。從照相機102之攝影元件所輸 出的電訊號係可藉由圖像轉換手段而轉換成圖像，發送訊 號至圖像處理手段1Q3。另—方面，攝影時之照相機位置（位 置座標或照相機與電锻表面之距離等）藉由照相機位置檢 測手& 109而檢測’於記憶手段11〇記憶此照相機位置的 資訊。 繼而’藉由圖像處理手段1〇3 ’對於所得到之圖像而 323801 201233974 進行邊緣抽離過濾處理（步驟302)。如上述般，藉由使用For the servo motor, for example, the "NX series" manufactured by Oriental Motor Co., Ltd. can be combined with the gears. The entire surface of the key surface of the mold can be automatically inspected by having the mold moving hand #又'. <Inspection of mold for anti-glare treatment using inspection apparatus> Next, an inspection method of the mold for anti-glare treatment using the inspection apparatus of the present invention will be described with reference to Fig. 3 . First, in the plating surface of the anti-glare processing mold for the inspection object, the illumination light is used to illuminate the image of the region in which the illumination light is applied, according to the imaging command of the first control means 1〇7 (shutter trigger ( The shutter signal is transmitted by the camera 102 (step 3〇1). The electric signal output from the photographing element of the camera 102 can be converted into an image by an image converting means, and the signal is sent to the image processing means 1Q3. On the other hand, the position of the camera at the time of photographing (the position coordinate or the distance between the camera and the electric forging surface, etc.) is detected by the camera position detecting hand & 109, and the information of the position of the camera is memorized by the memory means 11. Then, the edge extraction filtering processing is performed on the obtained image by the image processing means 1?3' (step 302). As above, by using
邊緣抽離過濾，強化在圖像令之電鍍表面的圖案，可得到 實質上排除表面凹凸之影子影響以及作為圖像全體之對比 度的影響之處理輯。然後，藉由第丨演算手段⑽，計 算處理圖像之亮度分散值（步驟3〇3)。所得到之亮度分散 值已記憶於記憶手段110。 U又 藉由以第2控制手段108控制照相機移動手段1〇5 , 使照相機102與電錢表面之間的距離改變，重複上述步驟 301至303,照相機位置與亮度分散值之相關的資訊儲^於 記憶手段m。第2演算手段U1係依據此資訊，決定亮 度分散值成為最大之照相機位置(步驟3〇4)产 ： 二=:機位置係例如亦可由所取到之處二像的 冗度：散值之t，選擇最大之亮度分散值，從所得到 數=度分散值製作近㈣線，亦可從此近㈣線求得顯 不最大的免度分散值之照相機位置。藉由以上之了 亮度分散值成為最大且焦點對準之照相機位置^ 由觀察或解析在賴相機位置之攝影_像， 藉 有無缺陷或姻缺陷之_(需要修補之缺陷 ，值成為最大之位置再度可=機表;= 在步驟30丨所攝影之圖像收藏於記憶手段ιΐ(^面錢 用此收藏之圖像而進行缺陷檢查。 ’、可使 有關模具一部分的電鍍表面進 用模具移動手段而變更檢查位置，即使於其=分= 323801The edge extraction filter enhances the pattern on the plating surface of the image, and the treatment effect that substantially obscures the shadow effect of the surface unevenness and the influence of the contrast of the entire image can be obtained. Then, the luminance dispersion value of the processed image is calculated by the third calculation means (10) (step 3〇3). The resulting luminance dispersion value has been memorized in the memory means 110. U further changes the distance between the camera 102 and the surface of the money by controlling the camera moving means 1 〇 5 by the second control means 108, and repeats the above steps 301 to 303, and the information relating to the camera position and the luminance dispersion value is stored. In memory means m. Based on this information, the second calculation means U1 determines the camera position at which the luminance dispersion value becomes the maximum (step 3〇4). Production: 2 =: The machine position is, for example, the redundancy of the two images taken: the scatter value t, select the maximum brightness dispersion value, and make the near (four) line from the obtained number = degree dispersion value, and also obtain the camera position of the most disproportionate dispersion value from the near (four) line. By the above, the brightness dispersion value becomes the maximum and the focus of the camera position is observed or analyzed by the camera image at the position of the camera, and the defect is the defect. Once again = the machine table; = the image captured in step 30 is stored in the memory means ιΐ (^ face money to use this collection of images for defect inspection. ', can be part of the mold on the plating surface into the mold movement Means to change the inspection position, even if it = = 323801
16 201233974 ， 表面亦可進行同樣的檢查。 * · 〈成為檢查對象之防眩處理用模具〉 * ' 本發明之檢查裝置較佳係使用於例如具有鍍鉻層作為 - 表面鍍層之防眩處理用模具的缺陷檢查。以下，說明有關 ， 具有鍍鉻層之防眩處理用模具的較佳之一例。 第4圖及第5圖係分別模式地表示具有鍍鉻層之防眩 處理用模具較佳的製造方法之一例中前半步驟及後半步驟 的圖。第4圖及第5圖係模式地表示在各步驟之模具的截 面。該模具之製造方法係基本上包含：[1]第1電鍍步驟、 [2]研磨步驟、[3]感光性樹脂膜形成步驟、[4]曝光步驟、 [5]顯影步驟、[6]第1蝕刻步驟、[7]感光性樹脂膜剝離 步驟、[8]第2電鍍步驟。 [1]第1電鍍步驟 本步驟中，首先對用於模具之基材的表面實施鍍銅或 鍍鎳。如此地，藉由於模具用基材的表面實施鑛銅或鍍錄， 可提昇後面之第2電鍍步驟中之鍍鉻的密著性或光澤性。 在第1電鍍步驟中所使用的銅或鎳，除了可為各別之 純金屬外，亦可為以銅作為主體之合金、或以鎳作為主體 之合金，因此，在本說明書中所謂之「銅」係包含銅及銅 合金之意，又，「鎳」係包含鎳及鎳合金之意。鍍銅或鍍鎳 係可分別進行電解電鍍，亦可進行無電解電鍍，但通常採 用電解電鍍。 實施鍍銅或鍍鎳時，若電鍍層太薄，無法排除基底表 面之影響，故其厚度宜為50/zm以上。電鍍層厚度之上限 17 323801 201233974 宜為500 // m左右。 適宜使用於模具用基材之金屬材料，從成本之觀點， 可舉例如鋁、鐵等。從操作之便利性，更宜為使用輕量之 鋁。此處所謂之鋁或鐵，亦可分別為純金屬外，亦可為以 鋁或鐵作為主體之合金。又，模具用基材之形狀係在該領 域中可為以往所採用之適當形狀，例如平板狀之外，亦可 為圓柱狀或圓筒狀之輥。 [2]研磨步驟 繼續研磨步驟中係在上述之第1電鍍步驟實施鍍銅或 鍍鎳的基材表面研磨。經過該步驟，基材表面係宜研磨至 接近鏡面的狀態。此係作為基材之金屬板或金屬輥為了達 到期望的精度，經常實施切削或研削等機械加工，藉此， 於基材表面殘留加工痕，即使實施鍍銅或鍍鎳之狀態，亦 殘留有此等加工痕，又，電鍍之狀態，不限於表面完全平 滑。於第4圖（a)中係對平板狀之模具用基材7之表面實施 鍍銅或鍍鎳（未圖示），進一步，模式地表示具有因研磨步 驟而被鏡面研磨之表面8的狀態。 有關實施鍍銅或鍍鎳之表面研磨的方法係無特別限 制，亦可使用機械研磨法、電解研磨法、化學研磨法之任 一者。機械研磨法係可例示超精加工法、摩擦法、流體研 磨法、緩衝液研磨法等。研磨後之表面粗度係依據JIS B 0601之規定為基準的中心線平均粗度Ra為0. 1 /z m以下， 更宜為0. 05 // m以下。若研磨後之中心線平均粗度Ra大於 0. 1/zm，有可能於最終模具表面之凹凸形狀殘留表面粗度 18 323801 201233974 , 之影響。 •舞 [3] 感光性樹脂膜形成步驟 繼續感光性樹脂膜形成步驟中係於已實施鏡面研磨之 • 模具用基材7被研磨的表面8，塗佈感光性樹脂，藉由加 ， 熱、乾燥’形成感光性樹脂膜。於第4圖（b)中模式地表示 於模具用基材7被研磨的表面8形成感光性樹脂膜9之狀 態。 於感光性樹脂係可使用以往公知者。具有感光部分硬 化之性質的負型感光性樹脂，可使用例如於分子中具有兩 烯酸基或甲基丙烯酸基之丙烯酸酯單體或預聚物、雙疊氮 與二烯橡膠之混合物、聚乙烯基肉桂酸酯系化合物等。又， 具有藉由顯影溶出感光部分，僅殘留未感光部分之性質的 正型感光性樹脂，可使用例如紛（pheno 1)樹脂系或酴酸清 漆樹脂系等。又，感光性樹脂中必要時亦可調配增感劑、 顯影促進劑、密著性改質劑、塗佈性改良劑等各種添加劑。 感光性樹脂為了形成良好之塗膜，宜稀釋於適當的溶劑而 塗佈於模具用基材7被研磨之表面8。 塗佈感光性樹脂溶液的方法，可使用彎月式塗佈 (meniscus coat)、喷泉塗佈、浸潰塗佈、旋轉塗佈、輥塗 佈、線棒塗佈、氣刀塗佈、刮刀塗佈或簾幕塗佈等公知的 方法。塗佈膜之厚度宜為乾燥後1至6/zm的範圍。 [4] 曝光步驟 繼而，在曝光步驟中使對應期望之表面凹凸形狀的圖 案曝光於感光性樹脂膜9上。使用於曝光步驟之光源只要 19 323801 201233974 符合所塗佈之感光性樹脂的感光波長或感度等而適當選擇 即可，可使用例如高壓水銀燈之忌線（波長·· 436nm)、高壓 水銀燈之h線（波長：4〇5nm)、高壓水銀燈之i線（波長： 365nm)、半導體雷射（波長：830nm、532nm、488nm、405nm 等）、YAG雷射（波長：1064nm)、KrF準分子雷射（波長： 248nm)、ArF準分子雷射（波長：193nm)、F2準分子雷射（波 長：157nm)等。使圖案曝光之方法可舉例如雷射描繪。 於第4圖（c)係模式性表示於感光性樹脂膜9上使圖 案曝光之狀態。使感光性樹脂膜以負型的感光性樹脂形 時，被曝光之區域1〇係藉由曝光而進行樹脂之交聯 、 對於後述之顯影液的溶解性會降低。因而，在噸$弗％， 未被曝光之區域11可藉顯影液而溶解，僅被曝^驟中 10在基材表面上殘留成為遮罩。另外，使感 區域 正型的感光性W脂形成時 ’被曝光之區域1 〇係藉. 、从 進行樹脂之鍵結被切斷，對於後述之顯影液的，光而 加。因而’在顯影步驟中未被曝光之區域u t會i曾 二解’僅被曝光之區域11在基材表面上殘留成為 [5]顯影步騍 蕈。 樹脂膜9時Γ步驟係使用負_感光性樹脂作I 、 未破曝光之區域11係以顯影液溶解，’"走个 =區域10殘軸制基材，繼而，儀姨年 性榭㈣Q 士另—方面，使用正塑的感光性樹脂作 光之I 時，僅被曝光之區域10被顯影液溶解，.、織3 “品域1殘存於模具用基材上，繼而，在 灰效与 麵刳- 201233974 驟中作用為遮軍。 驟之顯影液係可使用以往公知者。 可舉例如氲氧化鈉、㈣备& 虱軋化鉀、碳酸鈉、矽酸鈉、偏矽酸 鈉、氨水等無機鹼類、？胶 战咕 貝乙胺、正丙胺等第一胺類、二乙胺、 ，正丁胺專第二胺類、三乙胺、甲基二乙胺等第三胺類、 =曱基=胺、三乙醇胺等醇胺類、氫氧化四甲基録、氣 氧化四乙J'氫氧化三曱基經基乙基錢等第 四級錄鹽、 比各$定等*狀胺類等驗性水溶液，及二甲苯、甲笨等 有機溶劑等。 顯私步驟中之顯影方法係無特別限制，可使用浸潰顯 影、喷塗顯影、刷_影、超音波顯 於第4圖⑷中係模式地表示於感光性樹脂膜9使用 負型之感光崎I ’進行顯影處理之狀態。在第4圖^ 中未曝光之區域11被顯影液溶解’僅被曝光之區域10幾 留於基材表面上’成為遮罩12。於第4圖⑷中係模式地 表不於感，樹月日冑9使用正型之感光性樹脂，進行顯影 處理之狀先、S第4圖（c)中曝光之區域1〇被顯影液溶解， 未被曝从區域11殘留於基材表面上，成為遮罩12。 [6]第1蝕刻步驟 繼而’在第1㈣步驟中使關存於模具用基材表面 上之感光性能9料料，主要_無鮮處的模具用 基材’被研磨之電錢面形成凹凸。於第5圖⑷中係模式 地表示於藉㈣步驟，主要㈣無遮罩處13的模具 用基材7之狀態。遮罩12之下部的模具用基材7係從模具 3238〇1 201233974 用基材表面未被蝕刻，但進行蝕刻同時並進行源自無遮罩 處13的钱刻。因而，在遮罩12與無遮罩處13的邊界附近， 係遮罩12之下部的模具用基材7亦被敍刻（側餘刻）。第6 圖模式地表示側钱刻之進行的情形。第6圖之虛線14係階 段性表示與蝕刻之進行同時變化之模具用基材的表面。 在第1钱刻步驟中之蝕刻處理通常使用氯化鐵（FeCl3) 液、氣化銅（CuCh)液、鹼蝕刻液（Cu(NH3)4C1〇等，藉由腐 姓金屬表面來進行’但亦可使用鹽酸或硫酸等強酸，亦可 使用電解電鍍時與施加逆電位所產生的逆電解蝕刻。於實 施#刻處理時之模具用基材所形成的凹形狀係依基底金屬 之種類、感光性樹脂膜之種類及蝕刻方法等而異，故不能 一概而論’但蝕刻量為l〇/zm以下時，從接觸於蝕刻液之 金屬表面略等方向性地被蝕刻。此處所謂之蝕刻量係藉由 餘刻削去之基材的厚度。 在第1蝕刻步驟中的蝕刻量宜為1至5〇μιη。蝕刻量 未達1//in時，於金屬表面幾乎不會形成凹凸形狀，而成為 幾乎平坦的模具’故不適宜作為防眩處理用模具。又，蝕 刻量超過50 y m時，於金屬表面所形成之凹凸形狀的高低 差變大，使用所得到之模具製作且使用防眩薄膜之圖像顯 示裝置中有產生白暈之虞。在第1蝕刻步驟中之蝕刻處理 可藉由1次蝕刻處理來進行，亦可將蝕刻處理分成2次以 上而實施。將蝕刻處理分成2次以上而實施時，宜使2次 以上之蝕刻處理中的蝕刻量之合計為1至。 [7]感光性樹脂膜剝離步驟16 201233974 , the same inspection can be performed on the surface. * "The mold for anti-glare treatment to be inspected" * The inspection apparatus of the present invention is preferably used for defect inspection of, for example, an anti-glare treatment mold having a chromium plating layer as a surface plating layer. Hereinafter, a preferred example of the mold for antiglare treatment having a chrome plating layer will be described. Fig. 4 and Fig. 5 are views schematically showing the first half step and the second half step in an example of a preferred manufacturing method of the mold for preventing glare treatment having a chrome plating layer. Fig. 4 and Fig. 5 schematically show the cross section of the mold at each step. The manufacturing method of the mold basically comprises: [1] a first plating step, [2] a polishing step, [3] a photosensitive resin film forming step, [4] an exposure step, [5] a developing step, [6] 1 etching step, [7] photosensitive resin film peeling step, and [8] second plating step. [1] First plating step In this step, first, the surface of the substrate for the mold is subjected to copper plating or nickel plating. In this way, by performing ore plating or plating on the surface of the substrate for a mold, the adhesion or gloss of chrome plating in the second plating step in the subsequent step can be improved. The copper or nickel used in the first electroplating step may be an alloy mainly composed of copper or an alloy mainly composed of nickel, in addition to being a pure metal, and therefore, it is referred to in the present specification. "Copper" means copper and copper alloy, and "nickel" means nickel and nickel alloy. The copper plating or the nickel plating may be separately performed by electrolytic plating or electroless plating, but usually electrolytic plating is used. When copper plating or nickel plating is performed, if the plating layer is too thin to exclude the influence of the surface of the substrate, the thickness thereof is preferably 50/zm or more. The upper limit of the thickness of the plating layer 17 323801 201233974 should be about 500 / m. A metal material suitable for use in a substrate for a mold is, for example, aluminum, iron, or the like from the viewpoint of cost. From the convenience of operation, it is more suitable to use lightweight aluminum. The aluminum or iron referred to herein may be a pure metal or an alloy mainly composed of aluminum or iron. Further, the shape of the base material for the mold may be an appropriate shape conventionally used in the field, and may be, for example, a flat plate shape or a cylindrical or cylindrical roll. [2] Polishing step In the polishing step, the surface of the substrate subjected to copper plating or nickel plating is polished in the first plating step described above. Through this step, the surface of the substrate is preferably ground to a state close to the mirror surface. In order to achieve the desired precision, the metal plate or the metal roll as the base material is often subjected to machining such as cutting or grinding, whereby the machining marks remain on the surface of the substrate, and even if copper plating or nickel plating is performed, the residual remains. These processing marks, in addition, the state of plating, are not limited to the surface being completely smooth. In the fourth drawing (a), the surface of the flat-shaped mold substrate 7 is plated with copper or nickel (not shown), and the state of the surface 8 which has been mirror-polished by the polishing step is schematically shown. . The method for performing surface polishing of copper plating or nickel plating is not particularly limited, and any of mechanical grinding, electrolytic grinding, and chemical grinding may be used. The mechanical polishing method may, for example, be a superfinishing method, a rubbing method, a fluid grinding method or a buffer grinding method. The thickness of the surface roughness after the grinding is 0. 1 /z m or less, more preferably 0. 05 // m or less, based on the JIS B 0601. If the average thickness Ra of the center line after grinding is greater than 0.1/zm, there is a possibility that the surface roughness of the surface of the final mold surface remains 18 323801 201233974. • Dance [3] The photosensitive resin film forming step continues in the photosensitive resin film forming step in the surface 8 where the mold substrate 7 has been polished, and the photosensitive resin is applied by adding heat, heat, Drying 'forms a photosensitive resin film. The state in which the photosensitive resin film 9 is formed on the surface 8 on which the substrate 7 for a mold is polished is schematically shown in Fig. 4(b). A conventionally known one can be used for the photosensitive resin. A negative photosensitive resin having a photohardenable partial curing property, for example, an acrylate monomer or prepolymer having a dienoic acid group or a methacrylic group in a molecule, a mixture of a diazide and a diene rubber, and a poly A vinyl cinnamate compound or the like. Further, a positive photosensitive resin having a property of eluting a photosensitive portion by development and leaving only an unexposed portion can be used, for example, a pheno 1 resin or a phthalic acid resin. Further, various additives such as a sensitizer, a development accelerator, an adhesion modifier, and a coatability improver may be blended in the photosensitive resin. In order to form a good coating film, the photosensitive resin is preferably applied to the surface 8 to be polished by the substrate 7 for a mold by diluting it in an appropriate solvent. The method of applying the photosensitive resin solution may be a meniscus coat, a fountain coating, a dip coating, a spin coating, a roll coating, a wire bar coating, an air knife coating, or a knife coating. A known method such as cloth or curtain coating. The thickness of the coating film is preferably in the range of 1 to 6/zm after drying. [4] Exposure step Next, in the exposure step, a pattern corresponding to the desired surface unevenness shape is exposed on the photosensitive resin film 9. The light source used in the exposure step may be appropriately selected as long as 19 323801 201233974 conforms to the photosensitive wavelength or sensitivity of the applied photosensitive resin, and for example, a high-pressure mercury lamp (wavelength·· 436 nm) and a high-pressure mercury lamp h-line can be used. (wavelength: 4〇5nm), i-line of high-pressure mercury lamp (wavelength: 365nm), semiconductor laser (wavelength: 830nm, 532nm, 488nm, 405nm, etc.), YAG laser (wavelength: 1064nm), KrF excimer laser ( Wavelength: 248 nm), ArF excimer laser (wavelength: 193 nm), F2 excimer laser (wavelength: 157 nm), and the like. A method of exposing the pattern may be, for example, laser drawing. Fig. 4(c) is a view schematically showing a state in which the pattern is exposed on the photosensitive resin film 9. When the photosensitive resin film is formed into a negative photosensitive resin, the exposed region 1 is subjected to crosslinking of the resin by exposure, and the solubility of the developer to be described later is lowered. Therefore, in the range of ton %, the unexposed area 11 can be dissolved by the developing solution, and only the exposed portion 10 remains on the surface of the substrate as a mask. In addition, when the photosensitive W-fat which is positive in the sensitive region is formed, the exposed region 1 is removed, and the bond of the resin is cut, and light is applied to the developing solution to be described later. Thus, the region which is not exposed in the developing step will have a second solution. Only the exposed region 11 remains on the surface of the substrate as [5] developing step 蕈. When the resin film 9 is used, the negative-photosensitive resin is used as the I, and the unexposed region 11 is dissolved in the developer solution, and the substrate is made up of the residual region of the region 10, and then the 榭 榭 榭 (4) Q In the other aspect, when the photosensitive resin is used as the light I, only the exposed region 10 is dissolved by the developer, and the woven 3 "product 1 remains on the substrate for the mold, and then, the effect is grayed out. It is known that it is used in the case of 刳 刳 - 201233974. The developing solution can be used by conventionally known ones. For example, sodium bismuth oxide, (iv) preparation & strontium potassium, sodium carbonate, sodium citrate, sodium metasilicate Inorganic bases such as ammonia, etc., first amines such as diethylamine, n-propylamine, diethylamine, n-butylamine, second amine, triethylamine, methyldiethylamine, etc. Class, = fluorenyl = amine, triethanolamine and other alcohol amines, tetramethyl hydride, gas oxidized tetraethyl J' oxyhydroxide, hydrazine, etc., fourth grade salt, etc. An aqueous solution such as an amine, an organic solvent such as xylene or methyl bromide, etc. The development method in the smear step is not particularly limited, and a dip can be used. Development, spray development, brush-shadow, and ultrasonic sound are shown in Fig. 4 (4) in a state in which the photosensitive resin film 9 is subjected to development processing using a negative-type photosensitive image I. In Fig. 4 The exposed region 11 is dissolved by the developer 'only the exposed region 10 remains on the surface of the substrate' to become the mask 12. In the fourth figure (4), the pattern is not sensed, and the tree is used for the positive type. The photosensitive resin is subjected to development processing, and the exposed region 1 in FIG. 4(c) is dissolved by the developer, and remains on the surface of the substrate without being exposed from the region 11 to form the mask 12. [6] 1 etching step and then in the first (fourth) step, the photosensitive property 9 material stored on the surface of the substrate for the mold is used, and the surface of the mold for the mold which is mainly used for the mold is not formed. The figure (4) is schematically shown in the state of the (four) step, mainly (iv) the substrate 7 for the mold without the mask portion 13. The substrate 7 for the mold at the lower portion of the mask 12 is from the surface of the substrate for the mold 3238〇1 201233974 Not etched, but etched while carrying out the money from the unmasked portion 13. Thus, In the vicinity of the boundary between the cover 12 and the unmasked portion 13, the base material 7 for the mold which is the lower portion of the mask 12 is also scribed (side engraved). Fig. 6 schematically shows the progress of the side money. The dotted line 14 of the figure indicates the surface of the substrate for the mold which changes simultaneously with the etching. The etching treatment in the first etching step usually uses ferric chloride (FeCl3) liquid or vaporized copper (CuCh) liquid. An alkali etching solution (Cu(NH3)4C1〇, etc., is carried out by the surface of the rotted metal), but a strong acid such as hydrochloric acid or sulfuric acid may be used, and reverse electrolytic etching by applying a reverse potential during electrolytic plating may be used. The concave shape formed by the base material for the mold at the time of the engraving treatment varies depending on the type of the base metal, the type of the photosensitive resin film, the etching method, and the like, so that it cannot be generalized, but when the etching amount is l〇/zm or less, The metal surface in contact with the etchant is etched in a substantially directional manner. The amount of etching referred to herein is the thickness of the substrate which is removed by remnant. The etching amount in the first etching step is preferably from 1 to 5 μm. When the etching amount is less than 1/in, the uneven shape is hardly formed on the metal surface, and the mold is almost flat. Therefore, it is not suitable as an anti-glare treatment mold. Further, when the etching amount exceeds 50 μm, the difference in height between the concavo-convex shapes formed on the metal surface becomes large, and an image display device using the obtained mold and using an anti-glare film has a halo. The etching treatment in the first etching step can be performed by one etching treatment, or the etching treatment can be carried out by dividing the etching treatment twice or more. When the etching treatment is carried out in two or more steps, the total amount of etching in the etching treatment of two or more times is preferably 1 to. [7] Photosensitive resin film peeling step
22 323801 C 201233974 繼而，在感光性樹脂膜剝離步驟中，將在第丨蝕刻步 驟作為遮罩使用之殘存的感光性樹脂膜完全溶解除去。在 感光性樹脂_離步驟中使關離液溶解感紐樹脂膜。 剝離液可使用與上述之顯影液相同者。在感光㈣㈣剝 離步驟巾麟方法並無特觀制，可制浸潰㈣、喷塗 顯影、刷塗顯影、超音波顯影等方法。第5祕）係模式地 表示藉由感光性樹脂_離步射，在第丨_步驟作為 遮罩12使社感光性樹賴完全溶解除去之狀態。利用由 感光性樹脂朗構成之遮罩12藉由㈣使第丨表面凹凸形 狀15形成於模具用基材表面。 [8]第2電鍍步驟 繼而’藉由對所形成之凹凸面（第！表面凹凸形狀15) 實施錢鉻’而使表面之凹凸形狀鈍化。於第5圖⑹中表示 藉由於第1表面凹凸形狀15形成鑛鉻層16，形成凹凸較 第1表面凹凸形狀15更鈍之表面（鍵鉻的表面17)的狀態。 鍍鉻的種類並無特別限制，但宜使用稱為所謂光澤鍵 絡或裝飾峡鉻等顯示良好的紐之鍍絡。鍍鉻—般藉由 電解進行，其電鍍浴係可使用含有⑽）與少^硫 酸的水溶液。藉由調節電流密度與電解時間可控制鍵絡之 厚度。 ，在上述模具的製造方法中，藉由對形成有微細表面凹 凸形狀之表面實施鍍鉻’使凹凸形狀鈍化，同時可得到其 表面硬度提高之模具。此時之凹凸鈍化情形係依基底金屬 之種類、藉由第1_步驟得到的凹凸尺寸與深度、或電 323801 23 201233974 錄種類或厚度等而異，故無法一概而論，但控制鈍化情形 上最大的因素為電鍍厚度。若鍍鉻之厚度薄時，使凹凸形·.， 狀鈍化之效果不足，使其凹凸形狀轉印於透明膜等之透明 · 基板上所得到的實施有防眩處理之透明基材（防眩膜等）的 光學特性有不佳之傾向。另一方面，若電鑛厚度厚時，生， 產ί1生憂差§產生稱為節（n〇duie)之突起狀的電錄缺陷。 因此，鑛鉻之厚度宜為1至之範圍内，更宜為3至 6 y m之範圍内。 在第2電鍍步驟所形成之鍍鉻層宜為以維氏 硬度成為800以上的方式形成，更宜為以成為以上的 方式形成。 又’於上述之[7]感光性樹脂膜剝離步驟與[8]第2電 鍛步驟之間’且包含將藉由第】姓刻步驟所形成之凹凸面 以儀刻處理而鈍化之第2關步驟。在第2㈣步驟中， 將使用感光性樹脂膜作為遮罩之第i钱刻步驟所形成之第 1表面凹凸形狀15藉由钱刻處理而鈍化。藉由第 理使藉㈣刻處理所形成之第丨表面凹凸形二= 吻表面傾斜部分消失，將使用所得到之模具所製造之防眩 膜等實施防眩處理之透明基材的光學特性改變成較佳的方 向。第7圖中藉由第2㈣處理，使模具用基材7之第1 表面凹凸形狀15鈍化’表面傾斜陡崎部分被純化，顯示形 成具有平緩表面傾斜之第2表面凹凸形狀18的狀態。 在第2银刻步驟之钱刻處理亦與第！银刻步驟相同地 通常使用氯化鐵（FeCla)液、氯化銅（CuCh)液、驗飯刻液 323801 24 201233974 (Cu(NH3)4C12)等’藉由腐餘表面來進行，但亦可使用鹽酸 或硫酸等強酸’亦可使用施加與電解電艘時逆電位所產生 的逆電解蝕刻。實施蝕刻處理後之凹凸鈍化情形，係依基 底金屬之種類、钱刻方法、及藉由第1飯刻步驟所得到的 凹凸尺寸與深度等而異，故無法一概而論，但控制鈍化情 形上最大的因素為蝕刻量。此處所謂之蝕刻量，亦與第i 钱刻步驟相同地為被敍刻削去之基材的厚度。若姓刻量小 時，藉由第1蝕刻步驟所得到之凹凸表面形狀鈍化之效果 不足’使其凹凸形狀轉印至透明膜等之透明基材上所得到 的實施有防眩處理之透明基材（防眩膜等）的光學特性有不 佳之傾向。另外’若蝕刻量大時，凹凸形狀幾乎消失，成 為幾乎平坦的模具。因此，蝕刻量宜為1至50 am之範圍 内’更宜為4至20/zm之範圍内。在第2蝕刻步驟中，钱 刻處理與第1蝕刻步驟相同地可藉由1次蝕刻處理來進 行，亦可使蝕刻處理分成2次以上而實施。使蝕刻處理分 成2次以上而實施時，宜使2次以上之蝕刻處理中的姓刻 量之合計為1至50/zm。 [實施例] 以下’例示實施例及比較例而更具體地說明本發明， 但本發明係不限定於此等例。 〈實施例1> (檢查裝置之製作） 於Keyence股份有限公司製CCD照相機「CV-H200C」[芦 載I92萬像素（1600xl200)之CCD攝影元件]裴載〖灯⑼⑺ 323801 25 201233974 股份有限公司製同軸微距透鏡「CA-LMA4」（解析度：4·5 、焦點深度：中心值±〇. 〇86mm)，連接圖像處理手段及 作為第1演算手段之Keyence股份有限公司製控制器 「CV-5700」。於微距透鏡「CA-LMA4」裝載Keyence股份有 限公司製之點照明CA-DPW2，形成可同軸照明。上述CCD 照相機「CV-H200C」係使用同軸微距透鏡「CA-LMA4」時’ 視野為約1_正方，對應於電鍍表面上之攝影元件的像素 1邊的長度之長度約成為lem。因此，在本實施例之檢查 裝置中的照相機之解析度為4. 5" m(因透鏡之解析度大）。 「CV-5700」係隔著Ethernet(註冊商標），連接第1控制 手段、記憶手段及作為第2演算手段之Dell股份有限公司 製之個人電腦「Vostro 220」。裝載有上述透鏡之ccd照相 機係以透鏡對向於作為檢查對象之模具電鍍表面的方式裝 載於照相機移動手段之SUS股份有限公司製之電動驅動器 XA-42L-250EBW」。可藉由電動驅動器調整照相機與模具 電錄表面之距離。電動驅動器連接於第2控制手段及作為 照相機位置檢測手段之SUS股份有限公司製控制器 「XA-S4」，控制器隔著RS232C連接於上述個^電腦 「Vostro 220」。又，在本檢查裝置中，個人電腦「v〇str〇 220」對控制器「CV-5700」發送快門觸發（Shuttertrigger) 訊號’依據該訊號照相機進行攝影。 (模具電鍍表面之缺陷檢查） 使用上述檢查裝置’對藉由上述方法所製作之具有鐘 鉻表面的模具（於側面全體具有表面凹凸之 323801 26 201233974 缺陷檢查。亦即’藉由個人電腦「v〇stro 220」對控制器 「CV-5700」發送快門觸發訊號，同時從電動驅動器之控制 器「XA-S4」取得快門觸發訊號發送時之電動驅動器的目前 座標之操作，改變照相機與模具電鍍表面之距離並進行複 數次。所取得之電動驅動器之位置座標記憶於個人電腦 「Vostro 220」。 另一方面，藉由控制器「CV-5700」對於所攝影之圖像 進行邊緣抽離過濾處理，強化在鍍鉻表面所形成之網目狀 圖案。在該過濾處理中，分別使用1次於控制器「Cv_57〇〇」 所搭載之「邊緣抽離X方向過濾」及「邊緣抽離γ方向過 濾」。第8圖係使用上述檢查裝置所攝影之鍍鉻表面的焦點 位置之攝影圖像（未邊緣抽離過濾處理），第9圖係對於第 8圖所示之圖像使用上述邊緣抽離過遽後的處理圖像。從 此等之圖的比較可清楚得知，藉由邊緣抽離過濾處理，可 得到強化電鍍表面之網目圖案的圖像。 藉由控制器「CV-5700」，對於邊緣抽離過濾處理後之 圖像進行亮度分散值的計算。具體上係選擇於控制器 「CV-5700」所搭載之計測模式「濃淡檢查」而計算亮度分 散值。所片算之免度分散值係藉由(註冊商標）發 送至個人電腦「Vostro 220」。 ^ 第10圖係表示如上述般所得到之電動驅動器的位置 座標、與邊緣抽離過滤處理圖像的亮度分散值之關係圖。 橫軸之「移動距離」係相當於電動驅動器之位置座標，具 體上係意指從最初攝影時之電動驅動器之移動距離。縱輪 323801 27 201233974 係相當於邊緣抽離過濾、處理圖像之亮度分散值，使亮度分 散值最高之處理圖像的亮度分散值為1，而使亮度分散值 標準化者。繼而，一併表示共計6點的移動距離中之電錄 表面的攝影圖像（未邊緣抽離過渡處理者）。如圖所示，於 亮度分散值最尚的照相機位置（移動距離+ 〇. 〇50mm之位置） 可得到對焦性最高的圖像。 藉由使電動驅動器之位置再度調整至移動距離+〇. 〇5〇 mm之位置，取得鮮明之圖像，可良好地確認缺陷之狀態。 又’進一步驗證本實施例之檢查裝置（但，使用驅動 器’使照相機沿著輥狀模具的側面（凹凸表面）的方式朝輥 狀模具之長方向移動）的自動對焦效果。具體上係使具有鍍 絡表面之概狀模具’在照相機透鏡與模具之電鍍表面的距 離相關於模具長方向非固定的狀況中，確認照相機朝模具 長方向移動時之圖像的對焦性。將結果表示於第n圖。第 11圖之橫軸「位置（p〇siti〇n)」係表示從電鍍表面之照相 機的距離（mm)。又’縱軸rAF偏移〇ffset)」係從焦 點深度之中心值的照相機透鏡之距離（111]11)。所使用之透鏡 「CA-LMA4」之焦點深度係中心值±〇 〇86_。圖中記載為 「AF-ON」之圖像係使用本實施例之裝置的自動對焦功能時 的圖像’記載為「AF-OFF」之圖像係未使用自動對焦功能 時的圖像。「AF-OFF」之圖像係從位置 0 mm之焦點位置， 未改變照相機-電鍍表面間之距離而攝影者。又，圖中之 點’係藉由自動對焦功能所檢測出之其位置中可得到焦點 對準之圖像的照相機透鏡位置（從焦點深度之中心值的位 28 323801 201233974 置），表示若干的圖像作為「AF-0N」的圖像。 如圖所示，未使用自動對焦功能時，在位置〇 _附近 雖焦點對準，但照相機透鏡與模具之電鍍表面的距離相關 於模具長方向未固定，故在不同位置中可得到焦點未對準 之圖像（例如’右邊3個之「AF-0FF」圖像）。另一方面， 使用自動對焦功能時’照相機透鏡與模具之電鍍表面的距 離超出焦點深度範圍之位置中，亦可得到焦點對準之鮮明 圖像（例如，右邊之「AF-0N」圖像）。 〈比較例1〉 除了未進行邊緣抽離處理，利用從藉由照相機攝影取 得之圖像所得到之亮度分散值（亦即，未具有圖像處理手段） 以外’其餘與實施例1同樣做法，檢測出亮度分散值為最 大時之照相機位置。第12圖係表示藉此作法所得到之電動 驅動器之位置座標、與圖像（未邊緣抽離過濾處理）之亮度 分散值的關係圖。橫轴及縱轴之意義係與第10圖相同。如 圖所示’可得到電動驅動器之移動距離在+0. 050mm之地點 焦點最對準的鮮明圖像，但可得到最高的亮度分散值係移 動距離~〇. 〇5〇_之地點，有〇. 100mm之差距。所使用之透 鏡「CA-LMA4」的焦點深度係中心值±0. 086mm，故此非可容 許之誤差。 焦點最對準之照相機位置與可得到最高亮度分散值之 照相機位置不一致，係如上述般，在模具之表面凹凸焦點 以外的位置產生最多的影子。如此地，在習知對比度檢測 方式中，無法得到進行缺陷檢查之充分鮮明之模具表面圖 29 323801 201233974 像。 【圖式簡單說明】 第1圖係表示本發明之檢查裝置的構成之一例的區塊 圖。 第2圖係表示若干邊緣抽離過濾的演算法之圖。 第3圖係表示使用本發明之檢查裝置的防眩處理用模 具之檢查方法之一例的流程圖。 第4圖係模式地表示具有鍍鉻層之防眩處理用模具較 佳的製造方法之一例中前半部分的圖。 第5圖係模式地表示具有鍍鉻層之防眩處理用模具較 佳的製造方法之一例中前半部分的圖。 第6圖係模式地表示在第1蚀刻步驟中進行側融刻情 形之圖。 第7圖係模式地表示因第1钱刻步驟中所形成之凹凸 面藉由第2钮刻步驟而純化之狀態圖。 第8圖係使用在實施例1所製作之檢查裝置而攝影之 鍍鉻表面的焦點位置之攝影圖像（未邊緣抽離過濾處理）。 第9圖係對於第8圖所示之圖像使用邊緣抽離過濾後 的處理圖像。 第10圖係表示實施例1所得到之電動驅動器之位置座 標（照相機位置）、與邊緣抽離過濾處理圖像之亮度分散值 的關係圖。 第11圖係表示實施例1所製作之檢查裝置的自動對焦 效果之檢驗結果之圖。 323801 30 201233974 , 第12圖係表示比較例1所得到之電動驅動器之位置座 標（照相機位置）、與未處理圖像之亮度分散值的關係圖。 ^ 【主要元件符號說明】 - 7 模具用基材 . 8 藉由研磨步驟所研磨的表面 9 感光性樹脂膜 10 在曝光步驟被曝光之感光性樹脂膜 11 在曝光步驟未被曝光之感光性樹脂膜 12 遮罩 13 無遮罩處 14 藉由蝕刻階段性所形成之表面 15 第1蝕刻步驟後之基材表面（第1表面凹凸形狀） 16 鍍鉻層 17 錢鉻之表面 18 第2蝕刻步驟後之基材表面（第2表面凹凸形狀） 101 照明 102 照相機 103 圖像處理手段 104 第1演算手段 105 照相機移動手段 106 對焦位置檢測手段 107 第1控制手段 108 第2控制手段 109 照相機位置檢測手段 31 323801 201233974 110 記憶手段 111 第2演算手段 32 32380122 323 801 C 201233974 Then, in the photosensitive resin film peeling step, the photosensitive resin film remaining as a mask in the second etching step is completely dissolved and removed. In the photosensitive resin _ leaving step, the kinetic resin film is dissolved. The stripping solution can be the same as the developer described above. In the photosensitive (four) (four) peeling step, the towel method has no special system, and can be formed by dipping (four), spray development, brush development, ultrasonic development, and the like. The fifth mode is a state in which the photosensitive resin is completely dissolved and removed as a mask 12 by the photosensitive resin_off-step. The second surface unevenness 15 of the second surface is formed on the surface of the substrate for a mold by the mask 12 made of a photosensitive resin. [8] Second plating step Next, the surface irregularities are passivated by applying money chromium to the formed uneven surface (the first surface uneven shape 15). In the fifth graph (6), the ore-plated layer 16 is formed by the first surface uneven shape 15 to form a surface having a surface which is more blunt than the first surface uneven shape 15 (the surface 17 of the bonded chromium). The type of chrome plating is not particularly limited, but it is preferable to use a plating which is well-known as a so-called gloss bond or decorative gorge. The chrome plating is generally carried out by electrolysis, and an electroplating bath may use an aqueous solution containing (10)) and less sulfuric acid. The thickness of the bond can be controlled by adjusting the current density and the electrolysis time. In the method for producing a mold described above, the surface of the finely formed concave-convex shape is chrome-plated to passivate the uneven shape, and a mold having improved surface hardness can be obtained. The concave-convex passivation at this time differs depending on the type of the base metal, the unevenness and depth obtained by the first step, or the type or thickness of the electric 323801 23 201233974, so it cannot be generalized, but the maximum control passivation is achieved. The factor is the plating thickness. When the thickness of the chrome plating is thin, the effect of the uneven shape is not sufficient, and the uneven shape is transferred to a transparent substrate such as a transparent film to obtain an antiglare-treated transparent substrate (anti-glare film). The optical properties of etc. have a tendency to be poor. On the other hand, if the thickness of the electric ore is thick, the production and production of the electric slag will produce a projection-like electro-acoustic defect called a knot (n〇duie). Therefore, the thickness of the ore chromium is preferably in the range of 1 to more preferably in the range of 3 to 6 y m. The chromium plating layer formed in the second plating step is preferably formed so that the Vickers hardness is 800 or more, and more preferably formed in the above manner. Further, 'between the above [7] photosensitive resin film peeling step and [8] second electric forging step" and include the second step of passivating the uneven surface formed by the first surname step by engraving Close the steps. In the second step (4), the first surface uneven shape 15 formed by using the photosensitive resin film as the masking step is passivated by the etching process. By the third reason, the third surface of the kiss surface formed by the processing of the fourth surface is eliminated, and the optical characteristics of the transparent substrate subjected to the anti-glare treatment using the obtained anti-glare film of the mold are changed. In a better direction. In the second (fourth) process, the first surface uneven shape 15 of the mold base material 7 is passivated. The surface inclined steep portion is purified, and the second surface uneven shape 18 having a gentle surface inclination is formed. The money engraving process in the second silver engraving step is also the same! The silver engraving step is generally carried out using ferric chloride (FeCla) solution, copper chloride (CuCh) solution, and rice grind 323801 24 201233974 (Cu(NH3)4C12), etc., by using a residual surface, but The use of a strong acid such as hydrochloric acid or sulfuric acid can also be used for reverse electrolytic etching generated by applying a reverse potential to an electrolytic electric boat. The case of the concave-convex passivation after the etching treatment differs depending on the type of the base metal, the method of engraving, and the size and depth of the unevenness obtained by the first meal step, so that it cannot be generalized, but the maximum passivation is controlled. The factor is the amount of etching. The amount of etching referred to herein is also the thickness of the substrate which is cut and cut in the same manner as the step of engraving. When the last name is small, the effect of passivating the surface shape of the uneven surface obtained by the first etching step is insufficient, and the transparent substrate having the anti-glare treatment obtained by transferring the uneven shape onto the transparent substrate such as a transparent film is obtained. The optical characteristics (anti-glare film, etc.) tend to be poor. Further, when the etching amount is large, the uneven shape almost disappears, and it becomes a nearly flat mold. Therefore, the etching amount is preferably in the range of from 1 to 50 am, more preferably in the range of from 4 to 20/zm. In the second etching step, the etching process can be performed by one etching process in the same manner as the first etching step, or the etching process can be carried out by dividing the etching process into two or more times. When the etching treatment is carried out in two or more steps, it is preferable to make the total of the surnames in the etching treatment of two or more times from 1 to 50/zm. [Examples] Hereinafter, the present invention will be specifically described by way of examples and comparative examples, but the present invention is not limited thereto. <Example 1> (Production of inspection device) CCD camera "CV-H200C" manufactured by Keyence Co., Ltd. [CCD camera component of I92 megapixel (1600xl200)] 灯 Lamp (9) (7) 323801 25 201233974 Coaxial macro lens "CA-LMA4" (resolution: 4·5, depth of focus: center value ± 〇. 〇 86mm), connected image processing means and controller of Keyence Co., Ltd. as the first calculation means "CV -5700". The macro lens "CA-LMA4" is loaded with the point-of-light CA-DPW2 manufactured by Keyence Co., Ltd. to form coaxial illumination. When the coaxial CCD camera "CV-H200C" is used as the coaxial macro lens "CA-LMA4", the field of view is about 1_square, and the length of the length of the side of the pixel 1 corresponding to the photographic element on the plated surface is about lem. Therefore, the resolution of the camera in the inspection apparatus of the present embodiment is 4.5 " m (due to the large resolution of the lens). The "CV-5700" is connected to the first control unit, the memory means, and the personal computer "Vostro 220" manufactured by Dell Corporation, which is the second calculation method, via the Ethernet (registered trademark). The ccd camera equipped with the above-mentioned lens is mounted on the electric drive XA-42L-250EBW manufactured by SUS Co., Ltd., which is a camera moving means, in such a manner that the lens is opposed to the surface of the mold to be inspected. The distance between the camera and the surface of the die can be adjusted by an electric drive. The electric drive is connected to the second control means and the "XA-S4" controller of SUS Co., Ltd., which is a camera position detecting means, and the controller is connected to the above-mentioned computer "Vostro 220" via RS232C. Further, in the inspection apparatus, the personal computer "v〇str〇 220" transmits a shutter trigger (Shuttertrigger signal) to the controller "CV-5700" to shoot according to the signal camera. (Inspection of defects on the surface of the mold plating) Using the above-mentioned inspection apparatus 'the mold having the clock-chromium surface produced by the above method (the defect inspection of the surface of the entire surface is 323801 26 201233974. That is, by the personal computer "v 〇stro 220" sends a shutter trigger signal to the controller "CV-5700", and at the same time, the current coordinate operation of the electric drive when the shutter trigger signal is sent is obtained from the controller "XA-S4" of the electric drive controller, and the camera and mold plating surface are changed. The distance of the obtained electric drive is stored in the personal computer "Vostro 220". On the other hand, the controller "CV-5700" performs edge extraction filtering on the captured image. The mesh pattern formed on the chrome-plated surface is reinforced. In the filtering process, "edge extraction X-direction filtration" and "edge extraction γ-direction filtration" mounted on the controller "Cv_57〇〇" are used once. Figure 8 is a photographic image of the focus position of the chrome-plated surface photographed using the above inspection apparatus (without edge extraction filter) Fig. 9 is a processed image obtained by using the above-mentioned edge extraction after the image shown in Fig. 8. From the comparison of the figures, it is clear that the edge extraction filtering process can be enhanced. The image of the mesh pattern on the plated surface. The controller "CV-5700" is used to calculate the brightness dispersion value of the image after the edge extraction filtering process. Specifically, it is selected from the controller "CV-5700". In the measurement mode "darkness check", the luminance dispersion value is calculated. The exempted value of the calculation is transmitted to the personal computer "Vostro 220" by (registered trademark). ^ Figure 10 shows the electric motor obtained as described above. The relationship between the position coordinates of the driver and the brightness dispersion value of the edge-extracted filtered image. The "moving distance" of the horizontal axis corresponds to the position coordinate of the electric drive, specifically means the electric drive from the initial shooting. Moving distance. The vertical wheel 323801 27 201233974 is equivalent to the edge dispersion filtering, processing the brightness dispersion value of the image, and the brightness dispersion value of the processed image with the highest brightness dispersion value is 1 and bright. The degree-divided value standardizer. Then, the photographic image of the transcribed surface in the total moving distance of 6 points (the edge-extraction transition processor) is shown together. As shown in the figure, the camera position where the luminance dispersion value is the most. (Moving distance + 〇. 〇50mm position) The image with the highest focus can be obtained. By adjusting the position of the electric drive to the moving distance + 〇. 〇5〇mm position, a clear image can be obtained. The state of the defect is confirmed. Further, the autofocus effect of the inspection apparatus of the present embodiment (however, the actuator is moved toward the longitudinal direction of the roll mold by the side surface (concave surface) of the roll mold using the actuator ') is further verified. Specifically, in the case where the distance between the camera lens and the plating surface of the mold is not fixed in the longitudinal direction of the mold, the shape of the mold having the plating surface is confirmed to be in focus when the camera moves in the mold length direction. The result is shown in the nth picture. The horizontal axis "position (p〇siti〇n)" in Fig. 11 indicates the distance (mm) from the camera on the plated surface. Further, the 'vertical axis rAF offset 〇 ffset) is the distance (111) 11 from the camera lens of the center value of the focal depth. The focal length of the lens "CA-LMA4" used is the center value ± 〇 〇 86_. In the figure, the image of "AF-ON" is an image when the autofocus function of the apparatus of the present embodiment is used, and the image described as "AF-OFF" is an image when the autofocus function is not used. The image of "AF-OFF" is from the focus position of 0 mm, and the photographer is not changed by the distance between the camera and the plating surface. In addition, the point in the figure is a camera lens position (position 28 323801 201233974 from the center value of the depth of focus) detected by the autofocus function in its position, indicating a number of The image is an image of "AF-0N". As shown in the figure, when the autofocus function is not used, the focus is in the vicinity of the position 〇_, but the distance between the camera lens and the plated surface of the mold is not fixed in the longitudinal direction of the mold, so the focus is not correct in different positions. A quasi-image (for example, 'AF-0FF' on the right side). On the other hand, when the autofocus function is used, the distance between the camera lens and the plated surface of the mold is out of the depth of focus range, and a sharp image with the focus can be obtained (for example, the "AF-0N" image on the right side). . <Comparative Example 1> The same procedure as in the first embodiment was carried out except that the edge dispersion processing was not performed, and the luminance dispersion value obtained from the image obtained by camera photography (that is, the image processing means was not provided) was used. The position of the camera when the luminance dispersion value is the maximum is detected. Fig. 12 is a graph showing the relationship between the position coordinates of the electric actuator obtained by this method and the luminance dispersion value of the image (non-edge extraction filtering processing). The meanings of the horizontal axis and the vertical axis are the same as those in Fig. 10. As shown in the figure, the sharpest image with the closest focus of the moving distance of the electric drive at +0. 050mm can be obtained, but the highest brightness dispersion value can be obtained. The moving distance is ~〇. 〇5〇_ 〇. The gap of 100mm. The depth of focus of the lens "CA-LMA4" used is ±0. 086 mm, so this is not an allowable error. The position of the camera whose focus is most aligned does not coincide with the position of the camera which can obtain the highest brightness dispersion value. As described above, the most shadow is generated at a position other than the uneven focus on the surface of the mold. Thus, in the conventional contrast detection method, it is not possible to obtain a sufficiently clear mold surface image for defect inspection 29 323801 201233974. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram showing an example of the configuration of an inspection apparatus according to the present invention. Figure 2 is a diagram showing the algorithm of several edge-extraction filtering. Fig. 3 is a flow chart showing an example of an inspection method of an anti-glare treatment mold using the inspection apparatus of the present invention. Fig. 4 is a view schematically showing the first half of an example of a manufacturing method of a mold for preventing glare treatment having a chrome plating layer. Fig. 5 is a view schematically showing the first half of an example of a manufacturing method of a mold for preventing glare treatment having a chrome plating layer. Fig. 6 is a view schematically showing a side melting situation in the first etching step. Fig. 7 is a view schematically showing a state in which the uneven surface formed in the first step is purified by the second button etching step. Fig. 8 is a photographic image of a focus position of a chrome-plated surface photographed using the inspection apparatus manufactured in Example 1 (unedge detachment filtration processing). Fig. 9 is a processed image after edge filtering using the image shown in Fig. 8. Fig. 10 is a view showing the relationship between the position coordinates (camera position) of the electric actuator obtained in the first embodiment and the luminance dispersion value of the edge extraction filtering processed image. Fig. 11 is a view showing the results of inspection of the autofocus effect of the inspection apparatus produced in the first embodiment. 323801 30 201233974, Fig. 12 is a view showing the relationship between the position coordinates (camera position) of the electric actuator obtained in Comparative Example 1 and the luminance dispersion value of the unprocessed image. ^ [Main component symbol description] - 7 Mold base material. 8 Surface to be polished by the polishing step 9 Photosensitive resin film 10 Photosensitive resin film 11 exposed in the exposure step Photosensitive resin not exposed in the exposure step Membrane 12 Mask 13 Unmasked portion 14 Surface formed by etching gradation 15 Substrate surface after first etching step (first surface uneven shape) 16 Chrome-plated layer 17 Surface of chrome 18 After the second etching step Substrate surface (second surface uneven shape) 101 Illumination 102 Camera 103 Image processing means 104 First calculation means 105 Camera moving means 106 Focus position detecting means 107 First control means 108 Second control means 109 Camera position detecting means 31 323801 201233974 110 Memory means 111 Second calculation means 32 323801