基于帶鋼表面缺陷檢測(cè)系統(tǒng)的視覺研究進(jìn)展外文文獻(xiàn)翻譯、中英文翻譯,基于,帶鋼,表面,缺陷,缺點(diǎn),檢測(cè),系統(tǒng),視覺,研究進(jìn)展,外文,文獻(xiàn),翻譯,中英文 基于帶鋼表面缺陷檢測(cè)系統(tǒng)的視覺研究進(jìn)展 文 摘 鋼是大量的材料選擇和非常多元化的工業(yè)應(yīng)用。表面質(zhì)量以及其他屬性是最重要的質(zhì)量參數(shù), 特別是對(duì)于扁鋼產(chǎn)品。傳統(tǒng)手工表面檢驗(yàn)程序非常不足,確保保證表面質(zhì)量免費(fèi)。為了確?？蛻舻膰?yán)格要求,自動(dòng)建立鋼鐵表面檢查技術(shù)在過去的二十年被發(fā)現(xiàn)是非常有效和流行的?？紤]到它的重要性,本文試圖通過對(duì)鋼表面缺陷檢測(cè)和分類建立第一個(gè)正式審查技術(shù)發(fā)展水平?？梢钥闯龃蠖鄶?shù)的研究工作一直在進(jìn)行冷鋼帶表面，是客戶需求中最敏感的。對(duì)熱軋帶材和棒材/棒表面缺陷檢測(cè)工作也顯示出增長(zhǎng)在過去的10年。審查涉及總體方面的鋼表面缺陷自動(dòng)檢測(cè)和分類系統(tǒng)使用應(yīng)用技術(shù)。人們的注意也被吸引到報(bào)道成功率以及實(shí)時(shí)操作方面等有關(guān)的問題。? 關(guān)鍵詞:鋼鐵表面檢驗(yàn);缺陷檢測(cè);缺陷分類;自動(dòng)視覺檢測(cè) 審查? 1．簡(jiǎn)介:鋼鐵表面和其自動(dòng)檢查的重要性 鋼鐵可能是最重要的金屬的量子和各種各樣的使用。鋼鐵對(duì)工業(yè)社會(huì)的發(fā)展作出了巨大貢獻(xiàn)。事實(shí)上,鋼鐵消費(fèi)被認(rèn)為是一個(gè)標(biāo)準(zhǔn)來(lái)判斷一個(gè)國(guó)家的發(fā)展?fàn)顩r。根據(jù)世界鋼鐵協(xié)會(huì),在2013年粗鋼產(chǎn)量為15.82億噸(噸),比其他所有的生產(chǎn)圖金屬放在一起。今天,有超過3500等級(jí)的鋼的扁鋼產(chǎn)品貿(mào)易約占50%。?一個(gè)集成的鋼鐵制造工廠生產(chǎn)鐵礦石在高爐鐵水、焦炭、燒結(jié)礦和通量作為輸入。液態(tài)鐵轉(zhuǎn)化為鋼液與指定由中小學(xué)煉鋼流程組成。鋼液不斷鑄石板和坯料。板是典型的矩形截面的尺寸板1600 -毫米寬,250 -毫米厚,12000 -毫米長(zhǎng)。坯料的方形截面通常大約150×150毫米和12000毫米長(zhǎng)。板是隨后條,然后冷卷成熱。坯料軋制成各種維度。一個(gè)簡(jiǎn)化煉鋼過程的流程圖如圖1所示。鋼材的表面質(zhì)量的重要性,冷軋鋼板的下尤其認(rèn)為重要性自1980年代主要是由于要求汽車汽車制造商。在課程的時(shí)候,熱帶材表面質(zhì)量,近年來(lái),結(jié)構(gòu)性產(chǎn)品的表面質(zhì)量如棒/酒吧認(rèn)為重要。傳統(tǒng)上,扁鋼產(chǎn)品的表面質(zhì)量,在線圈形式,判斷手動(dòng)通過削減約30米的無(wú)規(guī)卷曲在一個(gè)批處理和檢查由一個(gè)專家。通常,在手動(dòng)檢查,檢查表面是鋼鐵表面產(chǎn)生約占總數(shù)的0.05%。在冷軋機(jī)復(fù)雜,運(yùn)營(yíng)商有時(shí)駐扎檢查成品的在線缺陷。然而,由于線路速度高、疲勞和其他不利因素,審查過程并不令人滿意。因此,手動(dòng)檢查過程不足以保證鋼材表面沒有缺陷和合理程度的信心,當(dāng)然,需要自動(dòng)表面檢測(cè)做了。?在重大的發(fā)展[1],九個(gè)鋼鐵公司和我們?nèi)齻€(gè)鋁公司在1980年代早期開始一個(gè)研究項(xiàng)目建立鋼鐵表面年檢與兩個(gè)商業(yè)合作組織。一個(gè)原型系統(tǒng)是建立在1987年在幾個(gè)鋼鐵廠和測(cè)試。與此同時(shí),歐洲企業(yè)也開始工作。因此,從1980年代以后的一半,研究工作表面檢查鋼鐵產(chǎn)品開始。今天,建立自動(dòng)表面檢測(cè)系統(tǒng)(網(wǎng)絡(luò))是由許多著名公司。自2006年以來(lái),每年國(guó)際表面檢驗(yàn)峰會(huì)(ISIS)是由組成的一個(gè)財(cái)團(tuán)等等。鋼鐵產(chǎn)品的技術(shù)應(yīng)用自動(dòng)檢查,盡管不是100%準(zhǔn)確已經(jīng)成熟了。? 2．復(fù)雜的鋼鐵表面檢查自動(dòng)化 鋼材表面的實(shí)時(shí)檢測(cè)面臨著一系列的挑戰(zhàn)。困難可以列舉如下： 危險(xiǎn)場(chǎng)所。為檢驗(yàn)設(shè)備安裝地點(diǎn)（照明系統(tǒng)，攝像機(jī)和一些信號(hào)處理設(shè)備），特別是，熱輥的米爾斯是很危險(xiǎn)的。環(huán)境溫度高，粉塵，石油的存在，水液滴和水蒸氣是很常見的。此外，該照明系統(tǒng)和相機(jī)需要防沖擊和振動(dòng)。此外，重型設(shè)備和場(chǎng)地在日常的感動(dòng)，每周和每年的維護(hù)。這些因素都需要適當(dāng)?shù)奈锢砗铜h(huán)境保護(hù)措施，現(xiàn)場(chǎng)設(shè)備的使用。 運(yùn)行速度。在日常生產(chǎn)中，表面的運(yùn)行速度進(jìn)行檢查一般是高。扁平鋼產(chǎn)品，在滾動(dòng)的速度，在檢查設(shè)備的操作，通常是20米/秒長(zhǎng)的產(chǎn)品，特別是線材，速度是225英里/小時(shí)的高（100米/秒）[ 2 ]。在這樣的高速實(shí)時(shí)操作需要特殊的圖像處理設(shè)備和軟件的執(zhí)行時(shí)間小。 在不同的鋼制品表面缺陷的品種有報(bào)道是非常高的[ 3 ]。例如，出版社[ 4 ]已經(jīng)把表面缺陷熱軋產(chǎn)品九大類29個(gè)亞類。這些缺陷都不受任何標(biāo)準(zhǔn)。因此，他們的特點(diǎn)和分類，并從機(jī)廠商有所不同。此外，由于在生產(chǎn)過程中的變化，這些缺陷表現(xiàn)的變化。 大量的攝像機(jī)。扁平鋼產(chǎn)品，兩套檢測(cè)系統(tǒng)-一個(gè)頂部和底部表面需要另一個(gè)。這些反過來(lái)一般由3至4的相機(jī)蓋帶的整個(gè)寬度。長(zhǎng)的鋼產(chǎn)品，多個(gè)攝像機(jī)位于外周以保證整個(gè)表面覆蓋。例如，一個(gè)圓形產(chǎn)品，至少三的相機(jī)同時(shí)使用五臺(tái)攝像機(jī)已在文獻(xiàn)[ 5 ]報(bào)道用。因此，對(duì)圖像采集和實(shí)時(shí)處理是一項(xiàng)艱巨的任務(wù)。 3．現(xiàn)有的文獻(xiàn)綜述 ?多年來(lái),許多審查論文(6 - 12)表面缺陷檢測(cè)的諸多方面的報(bào)告。各個(gè)方面和紋理分析方法一直在審查(13、14)。兩個(gè)相對(duì)最近審查本拍紙簿[6、7]。表面缺陷檢測(cè)使用紋理分析技術(shù)的進(jìn)步已經(jīng)被謝絕[6]覆蓋處理主要應(yīng)用于紡織品、磚和木頭。 [7]了非常全面的研究工作在織物表面缺陷檢測(cè)和提供一些有價(jià)值的結(jié)論。審查論文特別是紋理缺陷和面料也提到鋼鐵表面缺陷分類識(shí)別技術(shù)可以應(yīng)用的地方。值得一提的是,早在1982年,11個(gè)文件是列在“檢驗(yàn)在金屬加工行業(yè)”審查由下巴和哈洛[12]。岡薩雷斯和森林[15]提供了一個(gè)出色的理論背景圖像處理的各個(gè)方面,而理論依據(jù)神經(jīng)網(wǎng)絡(luò)分類由充分浸[16]。然而,作者不能找到任何審查的研究工作領(lǐng)域的鋼表面缺陷檢測(cè)和分類。因此,本文嘗試從學(xué)術(shù)界鞏固已發(fā)表的文獻(xiàn),鋼鐵行業(yè)和制造商的主題自動(dòng)缺陷檢測(cè)和分類的鋼鐵表面。 4．可用性研究的出版物上自動(dòng)建立鋼鐵表面檢查 ?發(fā)表文獻(xiàn)的可用性鋼表面主要由各學(xué)術(shù)機(jī)構(gòu)的研究工作,鋼鐵廠/鋼鐵廠研究單位和表面檢測(cè)設(shè)備制造商。許多研究工作已經(jīng)聯(lián)合發(fā)表的學(xué)術(shù)/科研院所和鋼鐵廠表明良好的合作伙伴關(guān)系。在過去的10年中,相當(dāng)比例的出版工作在鋼鐵表面系統(tǒng)來(lái)自中國(guó)。這是符合中國(guó)鋼鐵制造業(yè)占主導(dǎo)地位的存在。?已發(fā)表的一些論文報(bào)道的研究工作主要集中在缺陷分類方面實(shí)現(xiàn)商業(yè)采購(gòu)系統(tǒng)。而整體系統(tǒng)和他們的利益被著名致力于良好的文檔記錄,細(xì)節(jié)的缺陷檢測(cè)和分類并不詳細(xì),可能由于知識(shí)產(chǎn)權(quán)問題。 5．鋼表面的類別 ?類型的鋼表面缺陷檢測(cè).研究:板、棒、板、熱地帶,寒冷的地帶,桿/酒吧。它們覆蓋大部分鋼作為材料的應(yīng)用。冷,和后期,桿/酒吧得到更多研究者的關(guān)注。這主要是解釋說(shuō),大比例的這些產(chǎn)品是成品,客戶的質(zhì)量要求越來(lái)越嚴(yán)格。?廣泛、鋼鐵表面可以在平面和長(zhǎng)產(chǎn)品分類(圖2)。 平板產(chǎn)品表面可以進(jìn)一步被分類如下: ——板/坯:都是由連鑄過程鋼液和有一些相似性對(duì)表面和內(nèi)部條件。表面覆蓋規(guī)模越來(lái)越模糊。 ——板是由加熱一塊約為1250°C和隨后滾。表面氧化,甚至相對(duì)對(duì)板。 ——熱條是由加熱一塊約為1250°C和滾動(dòng)滾動(dòng)站在多個(gè)減少厚度所需的值。帶鋼表面氧化。然而,由于軋制力高,大大減少了熱表面粒度帶板。 ——冷帶是由在冷軋機(jī)軋制熱條酸洗過程(去除表面氧化層和清潔)。因此,冷條的表面沒有氧化,表面很光滑由于很高的軋制力用于冷變形過程。-涂帶(鍍鋅、鍍錫)/完成不銹鋼帶表面在本質(zhì)上是高度反光的。 長(zhǎng)的產(chǎn)品表面可以進(jìn)一步被分類如下:? 棒/禁止生產(chǎn)鋼坯熱軋過程,及其表面氧化。進(jìn)一步,表面也不平坦,因此,反射角向外圍從而產(chǎn)生不均勻的圖像強(qiáng)度不同。?等長(zhǎng)的產(chǎn)品角度、通道重型等生產(chǎn)從坯/開花。他們是復(fù)雜的截面和需要特殊照明和相機(jī)的安排。 6．鋼材的表面缺陷列表 ?有一個(gè)大的各種表面缺陷對(duì)不同鋼產(chǎn)品。此外,沒有統(tǒng)一標(biāo)準(zhǔn)的缺陷。也有大型國(guó)際集團(tuán)相似性和內(nèi)部集團(tuán)多元化的[17]各種類的缺陷,使得缺陷分類困難。缺陷目錄發(fā)布的是一家現(xiàn)代化的、德國(guó)[4]作為事實(shí)上的標(biāo)準(zhǔn)。?試圖列出了一些主要已被稱為文學(xué)的表面缺陷檢測(cè)和分類在過去的兩年半。缺陷相對(duì)于上述類別的鋼鐵表面。 7．自動(dòng)表面檢測(cè)系統(tǒng)硬件結(jié)構(gòu)的關(guān)鍵元素 圖3顯示了網(wǎng)絡(luò)多媒體的基本硬件結(jié)構(gòu)。它由一個(gè)或多個(gè)光源,一個(gè)或多個(gè)相機(jī)(亮視場(chǎng)或亮和暗視野),高速圖像處理器、服務(wù)器和操作員界面。? 7.1圖像采集?表面獲得滿意的圖像質(zhì)量,照亮表面充分和統(tǒng)一。事實(shí)上,高質(zhì)量的照明減少圖像處理的計(jì)算負(fù)擔(dān)。兩種類型的照明技術(shù)可用于金屬表面:強(qiáng)度和范圍成像。(在18到22)討論了照明系統(tǒng)的各個(gè)方面,對(duì)金屬表面。研究成像系統(tǒng)的冷帶已經(jīng)被很好地記錄下來(lái)了[23]。?成像范圍提供了高度的信息從而使3 d缺陷突出。成像范圍不是競(jìng)爭(zhēng)強(qiáng)度成像。一般來(lái)說(shuō),使用范圍成像在鋼表面缺陷的研究并不多見。?強(qiáng)度成像的主要是兩種類型:明亮的場(chǎng)和暗場(chǎng)。在明亮的照明領(lǐng)域,傳感器捕捉最直接的反射光。表面看起來(lái)明亮,而缺陷特性顯得更黑。在暗場(chǎng)照明,入射光線的角度表面法向量是非常大的。這個(gè)結(jié)果在一個(gè)黑暗的表面,但有些缺陷圖像中出現(xiàn)明亮。暗視野觀點(diǎn)需要更強(qiáng)烈的照明。約8倍而亮視場(chǎng)照明要求報(bào)道[21]。?不幸的是,所有表面缺陷不會(huì)出現(xiàn)在明亮的領(lǐng)域或僅在暗視野。有很多的例子使用兩套攝像頭覆蓋視圖的字段(24 - 26日)。使用20電荷耦合器件(CCD)區(qū)域掃描相機(jī)用來(lái)捕捉表面圖像的雙方熱軋條使用明視場(chǎng)和暗場(chǎng)模式已報(bào)告在中國(guó)一家鋼鐵工廠[24]。然而,考慮到維護(hù)問題和系統(tǒng)的復(fù)雜性,大多數(shù)的系統(tǒng)將相機(jī)在明視場(chǎng)和暗場(chǎng)之間的位置。 7.2光源?提供所需的光源均勻光盡可能。雖然照明要求特別安排的光電源[27],提供統(tǒng)一的強(qiáng)度是不可能由于使用多個(gè)光源在大多數(shù)情況下。圖4顯示了入射光強(qiáng)度的變化對(duì)鋼的表面使用兩個(gè)至強(qiáng)燈[28]。類型的光源用于一般是:廣泛熒光管、鹵素、至強(qiáng)和領(lǐng)導(dǎo)。? 7.3型攝像機(jī)?一般來(lái)說(shuō),使用高分辨率CCD相機(jī)。使用線掃描和區(qū)域掃描相機(jī)已經(jīng)在文獻(xiàn)報(bào)道。線掃描相機(jī)已被廣泛使用,因?yàn)樗菀滓庾R(shí)到一個(gè)強(qiáng)大的,甚至照明區(qū)域表面進(jìn)行檢查。線掃描相機(jī)的缺點(diǎn)是,他們不能生成一個(gè)完整的形象,需要一個(gè)外部硬件建立圖像從多個(gè)線掃描[7]。大部分的自動(dòng)表面檢測(cè)系統(tǒng)制造商使用線掃描。區(qū)域掃描相機(jī)、運(yùn)輸編碼器的使用是可選的,檢查決議在兩個(gè)方向上獨(dú)立于對(duì)象(web)的速度。然而,盡管使用區(qū)域掃描相機(jī),甚至需要特別注意確保照明面積的掃描盡可能。高分辨率攝像機(jī)也用作免費(fèi)系統(tǒng)[30]。 7.4攝像頭和圖像分辨率?相機(jī)分辨率。線掃描相機(jī)分辨率通常是1024(交叉網(wǎng)絡(luò))×1(網(wǎng)絡(luò))和2048×1像素。[31]報(bào)道使用4096×1像素的相機(jī)。制造商通常使用1024 / 2048/4096×1像素。區(qū)域掃描:已報(bào)告600×400像素的[32]。在[33],4096×1000像素用于板。?圖像分辨率。各種尺寸的圖像的決議已報(bào)告[31日24日,26日,33歲,34)?？鐆eb從0.17毫米到1毫米,而報(bào)道決議從0.25到1.25毫米不等。 7.5圖像處理計(jì)算機(jī)硬件?CCD攝像機(jī)記錄了一個(gè)圖像轉(zhuǎn)移到某種形式的快速、并行處理系統(tǒng)專用的相機(jī)和靠近它[24]。確保實(shí)時(shí)操作的并行處理系統(tǒng)處理大量圖像數(shù)據(jù)并選擇感興趣的和存儲(chǔ)區(qū)域(roi)。并行處理系統(tǒng)可能是相機(jī)本身的一部分,或FPGA與特殊硬件處理器或通用處理器。這一部分系統(tǒng)至關(guān)重要的實(shí)時(shí)操作以及缺陷檢測(cè)和分類的準(zhǔn)確性。此后,與大型備份服務(wù)器內(nèi)存用于進(jìn)一步的處理和操作的接口。 8．缺陷檢測(cè)和分類的方法列表 ?各種方法/技術(shù)用于鋼鐵表面的缺陷檢測(cè)和分類列出。表1顯示了不同的方法的列表.檢測(cè)相對(duì)于獲得本研究的引用。類型的鋼表面也被提到在桌子上。技術(shù)后可能廣泛統(tǒng)計(jì),形態(tài),空間域?yàn)V波、頻域分析、聯(lián)合空間/局部分析和分形模型?？臻g域?yàn)V波、形態(tài)學(xué)操作和關(guān)節(jié)空間y域過濾被發(fā)現(xiàn)廣泛用于所有類型的表面。?表面檢查的最終目標(biāo)是使用分類歸類指定類缺陷。作為一個(gè)過程、分類開始后缺陷局部分割。通常在這個(gè)階段,很多功能是提取的區(qū)域。理想情況下,不同的組合匹配所需的這些特性是獨(dú)特和不同類型的缺陷。匹配通常是使用學(xué)習(xí)方法如神經(jīng)網(wǎng)絡(luò)反向傳播(NN-BP)、支持向量機(jī)(SVM)等。自適應(yīng)學(xué)習(xí)的兩種類型:1)監(jiān)督的網(wǎng)絡(luò)提供了大量已知的典型輸入。此后,網(wǎng)絡(luò)產(chǎn)生已知輸出盡可能基于培訓(xùn)。b)在無(wú)監(jiān)督學(xué)習(xí),網(wǎng)絡(luò)需要各種輸入之間的關(guān)系沒有被告知。?然而,鋼表面缺陷展覽大型國(guó)際集團(tuán)相似性和內(nèi)部多樣性。因此,找到合適的特性和識(shí)別分類器計(jì)算成本較低是主要的研究領(lǐng)域。表2顯示了分類方法的列表引用和類型的表面。 結(jié)論 本文處理的自動(dòng)化檢測(cè)方法對(duì)鋼鐵表面使用圖像處理技術(shù)。審查出版物在兩年半的提供了一個(gè)了解發(fā)生在這一領(lǐng)域的最新進(jìn)展。主要觀察如下: a)由于惡劣的環(huán)境,需要特別注意照明和成像系統(tǒng)的設(shè)計(jì)。鋼鐵表面圖像據(jù)報(bào)道,由于表面氧化皮含有大量的噪聲,振動(dòng),異常/變量照明,存在偽缺陷等表面缺陷的不規(guī)則形狀和他們的類型和特征發(fā)生顯著的變化從一個(gè)工廠到另一個(gè)。特征的缺陷也依賴生產(chǎn)條件。 b)已發(fā)表的文獻(xiàn)表明,相對(duì)重視為冷軋帶鋼表面缺陷的檢測(cè)。最近,注意力也集中在表面的熱條和酒吧/棒。多種技術(shù),無(wú)論是在空間和頻率域,已經(jīng)申請(qǐng)了缺陷檢測(cè)。通常,組合的幾個(gè)技術(shù)提供了有用的結(jié)果。關(guān)于缺陷分類, 某種形式的神經(jīng)網(wǎng)絡(luò)或基于支持向量機(jī)技術(shù)找到的使用。實(shí)時(shí)操作的自動(dòng)化檢查系統(tǒng)通常需要非?？斓奶幚韴D像的軋機(jī)速度通常是非常高的平面和鋼產(chǎn)品。這需要每個(gè)攝像機(jī)的專用硬件系統(tǒng)具有并行處理能力。 c)不謹(jǐn)慎的比較不同技術(shù)的結(jié)果是由于缺乏共同的標(biāo)準(zhǔn)對(duì)圖像和實(shí)驗(yàn)方法。這個(gè)問題是進(jìn)一步復(fù)雜由于缺乏標(biāo)準(zhǔn)定義的缺陷類型。 d)商業(yè)化生產(chǎn)的自動(dòng)化應(yīng)用檢查系統(tǒng)網(wǎng)絡(luò)材料已達(dá)到高水平的成熟。然而,他們需要得到適當(dāng)?shù)恼{(diào)整為特定的應(yīng)用程序。也連續(xù)設(shè)計(jì)師和用戶之間的協(xié)作是必要的安裝系統(tǒng)適應(yīng)新品種/特征的缺陷在同一安裝位置。 引用 1。杜邦F、C Odet、米箱、優(yōu)化的缺陷識(shí)別的扁鋼產(chǎn)品成本矩陣?yán)碚摗ＨQ于國(guó)際30(1),3 - 10(1997)。7月的12日訪問 2。賈慶林,YL Murphey,J施正榮,T,表面缺陷檢測(cè)的智能實(shí)時(shí)視覺系統(tǒng)(IEEE-Proceedings第17屆國(guó)際會(huì)議在模式識(shí)別,2004),頁(yè)2 – 5 3。Sharifzadeh Alirezaee,R Amirfattahi,距首都普里什蒂納,檢測(cè)鋼缺陷使用圖像處理算法(IEEE國(guó)際會(huì)議,2008),頁(yè)125 – 127 4。C公園,SC贏了,一個(gè)自動(dòng)化web表面熱線材使用非抽取小波變換和支持向量機(jī)(工業(yè)電子、IECON 09年,IEEE的35年會(huì)上,2009),頁(yè)2411 – 2415 5。X謝,審查使用紋理表面缺陷檢測(cè)的最新進(jìn)展分析技術(shù)。電子。列托人。視覺形象肛門。7(3),1-22(2008) 6。庫(kù)馬爾,織物疵點(diǎn)檢測(cè):一項(xiàng)調(diào)查。IEEE反式。印第安納州。電子。55(1),348 - 363(2008) 7。M Shirvaikar,自動(dòng)視覺檢測(cè)的趨勢(shì)。j . Proc實(shí)時(shí)圖像。1(1),41-43(2006) 8。Y李,來(lái)自G培華學(xué)院、自由表面檢查技術(shù)最先進(jìn)的審查。愛思唯爾、計(jì)算機(jī)輔助Des。36歲,1395 - 1417(2004) E V I E W Open Access Review of vision-based steel surface inspection systems Abstract Steel is the material of choice for a large number and very diverse industrial applications. Surface qualities along with other properties are the most important quality parameters, particularly for flat-rolled steel products. Traditional manual surface inspection procedures are awfully inadequate to ensure guaranteed quality-free surface. To ensure stringent requirements of customers, automated vision-based steel surface inspection techniques have been found to be very effective and popular during the last two decades. Considering its importance, this paper attempts to make the first formal review of state-of-art of vision-based defect detection and classification of steel surfaces as they are produced from steel mills. It is observed that majority of research work has been undertaken for cold steel strip surfaces which is most sensitive to customers' requirements. Work on surface defect detection of hot strips and bars/rods has also shown signs of increase during the last 10 years. The review covers overall aspects of automatic steel surface defect detection and classification systems using vision-based techniques. Attentions have also been drawn to reported success rates along with issues related to real-time operational aspects. Keywords: Steel surface inspection; Defect detection; Defect classification; Automated visual inspection Review 1. Introduction: importance of steel surface and its automated inspection Steel is probably the most important of all metals in terms of its quantum and variety of use. Steel has contributed immensely towards the development of industrial society. In fact, consumption of steel is considered to be one of the yardsticks to judge the developmental status of a country. As per World Steel Association, production of crude steel during 2013 was 1,582 million tons (Mt), which is more than production figure of all other metals put together. Today, there are more than 3,500 grades of steel available out of which trade in flat steel products accounts for about 50%. An integrated iron and steel making plant produces liquid iron in blast furnace with iron ore, coke, sinter and flux as input. Liquid iron is converted to liquid steel with specified constituent by primary and secondary steel making processes. Liquid steel is continuously cast into slabs and billets. Slabs are of rectangular cross-section  with dimension of a typical slab being 1,600-mm-wide, 250-mm-thick and 12,000-mm-long. Billets are normally of square cross-section of about 150 × 150 mm and about 12,000-mm-long. Slabs are subsequently rolled into hot strips and then to cold strips. Billets are rolled into struc-tural of various dimensions. A simplified flow chart of steel making processes is shown in Figure 1. Importance of surface quality of steel products, par-ticularly that of cold-rolled steel assumed importance since 1980s primarily due to demands from automotive car makers. In course of time, hot strip surface quality, and in recent times, surface quality of structural products like rods/bars have assumed significant importance. Traditionally, surface quality of flat steel products, which are in coil form, is judged manually by cutting about 30 m of a random coil in a batch and inspected by an expert. Typically, in manual inspection, the inspected surface is about 0.05% of the total steel surface produced. In cold rolling mill complex, operators are sometimes stationed to inspect the finished product online for any defect. However, due to high line speed, fatigue and other adverse factors, inspection process is hardly satisfactory. Thus, the manual inspection process is not sufficient to guarantee defect-free surface of steel products with reasonable degree of confidence and naturally, need for automated surface inspection grew. In a significant development [1], nine steel companies and three aluminium companies in US started a research project in early 1980s on vision-based steel surface inspec-tion in collaboration with two commercial organisations. A prototype system was built and tested in several steel plants during 1987. At the same time, European companies also started working. Thus, from later half of 1980s, system-atic research work on surface inspection of steel products started. Today, vision-based automated surface inspection systems (ASIS) are produced by many reputed companies. Since 2006, an annual International Surface Inspection Summit (ISIS) is organised by a consortium of manufac-turers and others. Technology of vision-based automatic inspection of steel products, even though not 100% accurate has matured. This paper attempts to find out the status of development of vision-based ASIS for steel surfaces through review of published literature during the last two and a half decades. 2. Complexities of steel surface inspection automation Real-time inspection of steel surfaces faces a number of challenges. The difficulties may be enumerated as follows: Hazardous site. The place for installation of inspec-tion equipment (illumination system, camera and some signal processing equipment), particularly, for hot roll-ing mills is very hazardous. Presence of high ambient temperature, dust, oil, water droplet and vapour is very common. Additionally, the illumination system and the cameras require protection against shock and vibration. Further, heavy equipment is moved in and out of site during daily, weekly and annual maintenance. All above factors necessitate the use of appropriate physical and environmental protective measures for site equipment. Operating speed. During regular production, operating speed of the surface to be inspected is generally high. For flat steel products, speed at the end of rolling, where the inspection equipment has to operate, is typically 20 m/s. For long products, particularly wire rods, speed  could be as high as 225 miles/h (100 m/s) [2]. Real-time operation at such high speed requires special image pro-cessing equipment and software with small execution time. Varieties of surface defects in different steel products are reported to be very high [3]. For example, Verlag Stahleisen [4] have categorised surface defects of hot-rolled products in nine main classes and 29 subclasses. These defects are not governed by any standard. Thus, their characteristics and classification vary from mill to mill and from operator to operator. Further, manifestation of these defects changes due to variations in production process. Large number of cameras. For flat steel products, two sets of inspection systems - one for top and another for bottom surface - are needed. Each of these sets in turn gen-erally consists of 3 to 4 cameras to cover the entire width of the strip. For long steel products, multiple cameras are to be located peripherally to ensure coverage of entire surface. For example, for a round product, at least three cameras are used while use of five cameras has been reported in the literature [5]. Thus, gathering of images and their real-time processing is a daunting task. 3. Prior literature review Over the years, a number of review papers [6-12] on vari-ous aspects of surface defect detection have been reported. Various aspects and methods for texture analysis have been reviewed in [13,14]. Two comparatively recent review pa-pers are [6,7]. Advances in surface defect detection using texture analysis techniques have been dealt with by Xie [6] covering applications in mainly textiles, tiles and wood. Kumar [7] has covered very comprehensively research work done in fabric surface defect detection and provided some valuable conclusions. Review papers particularly on texture defects and defects in fabrics also mention steel surface as a category where identified techniques can be applied. It is worth mentioning that as early as in 1982, 11 papers were listed under ‘Inspection in Metal Processing Industry’ cat-egory in a review by Chin and Harlow [12]. Gonzalez and Woods [15] provide an excellent theoretical background to all aspects of image processing, whereas theoretical basis for neural network-based classification is adequately cov-ered by Haykins [16]. However, the authors could not locate any review of research work done in the field of steel surface defect detection and classification. Therefore, in this paper, attempt has been made to consolidate the published literature from academia, steel industry and manufacturers on the topic of automatic defect detection and classification of steel surfaces. 4. Availability of research publications on automated vision-based steel surface inspection Availability of the published literature on steel surface in-spection mostly consists of research work done at various academic institutions, steel plants/steel plant research units and surface inspection equipment manufacturers. A number of research works have been published jointly by academic/research institutes and steel plants indicating good collaborative partnership. During the last 10 years, a significant percentage of published work on steel surface in-spection systems came from China. This is commensurate with China's dominant presence in steel manufacturing. Some papers have been published with reported research work mainly on defect classification aspects implemented in commercially procured systems. While overall systems and their benefits are well documented by reputed manu-facturers, details of defect detection and classification tech-niques are not elaborated, probably due to issues regarding intellectual property rights. 5. Categories of steel surfaces Types of steel surfaces studied for defect detection/classi-fication are: slab, billet, plate, hot strip, cold strip, rod/bar. They cover a large proportion of applications of steel as a material. cold strips, and off late, rod/bars have received more attention of researchers. This is mainly explained by the fact that large proportions of these products are finished product and quality requirements of customers have become more stringent over time. Broadly, steel surfaces can be categorised in flat and long products (Figure 2). Flat product surfaces can further be classified as follows: – Slab/billet: both are produced by continuous casting process from liquid steel and have some similarity with respect to surface and internal conditions. Surface is scale covered and more grainy. – Plates are produced by reheating a slab at about 1,250°C and rolled subsequently. The surface is oxidised and comparatively even with respect to that of slab. – Hot strips are produced by reheating a slab at about 1,250°C and rolling in multiple rolling stands to reduce the thickness to desired value. The strip surface is oxidised. However, due to high rolling force, the surface granularity of hot strip is considerably reduced compared to slab. – Cold strips are produced by rolling hot strips in cold rolling mill after pickling process (which removes the oxide layer and cleans the surface). Thus, the surface of cold strips is not oxidised, and the surface is quite smooth due to very high rolling forces used in cold deformation process. – Coated strip (galvanised, tinned)/finished stainless strip surfaces are highly reflective in nature. Long product surfaces can further be classified as follows: Rods/bars are produced from billet by hot rolling process, and their surface is fairly oxidised. Further, the surface is also not flat, and therefore, angle of reflection varies towards the periphery thus producing nonuniform image intensity. Other long products like angles, channels, heavy struc-tural, rails etc. are produced from billet/bloom. They are of complex cross-section and require special lighting and camera arrangements. 6. List of surface defects for steel products There is a large variety of surface defects for different steel products. Further, there is no agreed standard for defects. There is also large ‘inter group similarity and intra group diversity’ [17] for various classes of defects, which makes defect classification difficult. Defect catalogues published by Verlag Stahleisen GmbH, Germany [4] act as defacto standards for this purpose. An attempt has been made to list some of the main de-fects which have been referred in the literature for surface defect detection and classification during the last two and a half decades. Defects have been listed vis-à-vis the categories of steel surfaces mentioned above. 7. Key elements of automatic surface inspection system hardware structure Figure 3 shows the basic hardware structure of ASIS. It consists of one or more light source, one or more camera (bright field or both bright and dark field), fast image processor, server and the operator interface. 7.1 Image acquisition To obtain satisfactory surface image quality, it is import-ant to illuminate the surface adequately and uniformly. In fact, high quality of illumination reduces computational burden of image processing. Two types of illumination techniques can be used for metallic surfaces: intensity im-aging and range imaging. [18-22] have discussed various aspects of illumination systems for metallic surfaces. Research on imaging systems for cold strips has been well documented in [23]. Range imaging provides height information thereby making 3D defects prominent. Range imaging is not competitive to intensity imaging. In general, use of range imaging is not common in steel surface defect studies. Intensity imaging is primarily of two types: bright field and dark field. In bright field illumination, the sensor captures most of the directly reflected light. The surface appears bright, whereas the defect features appear darker. In dark field illumination, the angle of the incident light rays to the surface normal vector is very large. This results in a dark appearance of the surface, but some defects appear bright in the image. Dark field view requires more intense lighting. Requirement of about eight times compared to bright field lighting has been reported [21]. Unfortunately, all surface defects do not show up either in bright field or in dark field alone. There are many examples of the use of two sets of cameras covering both the fields of view [24-26]. Use of 20 charge-coupled device (CCD) area scan cameras which are used to capture surface image of both sides of hot-rolled strips using both bright field and dark field modes have been reported in an iron and steel plant of China [24]. However, considering maintenance issues and system complexity, most of the systems place the cameras in between the bright field and dark field locations. 7.2 Source of light The light source is required to provide uniform ripple-free light as far as possible. While ripple-free illumination calls for special arrangement of light power supply [27], provid-ing uniform intensity is not possible due to the use of more than one light source in majority of the cases. Figure 4 shows the variation of incident light intensity on to a steel surface using two xeon lights [28]. Types of light source which are used in general are: wide spectrum tung-sten, fluorescent tubes, halogen, xeon and LED. 7.3 Type of camera In general, high-resolution CCD cameras are used. Use of both line scan and area scan cameras has been reported in the literature. Line scan cameras have been widely used as it is easier to realise a strong and even illumination to the surface area to be inspected. The disadvantage with the line scan cameras is that they do not generate a complete image at once and requires an external hardware to build up images from multiple line scans [7]. Most of the automatic surface inspection system manufacturers use line scan cam-era. For area scan cameras, the usage of transport encoder is optional and the inspection resolution in both directions is independent of the object (web) speed. However, while using area scan camera, special attention is needed to ensure even illumination of the total area under scan to the extent possible. High-resolution video cameras are also used as complimentary systems [30]. 7.4 Camera and image resolution Camera resolution. Line scan camera resolution is generally 1,024(cross web) × 1(down web) and 2,048 × 1 pixels. Yazdchi et al. [31] reported the use of 4,096 × 1 pixel camera. Manufacturers normally use 1,024/2,048/4,096 × 1 pixels. For area scan: 600 × 400 pixels have been reported by [32]. In [33], 4,096 × 1,000 pixels have been used for slab. Image resolution. Various dimensions of image resolu-tions have been reported [24,26,31,33,34]. Cross web reso-lutions vary from 0.17 mm to about 1 mm while reported down-web resolutions vary from 0.25 to 1.25 mm. 7.5 Image processing computer hardware Images captured by a CCD camera are transferred to some form of fast, parallel processing system dedicated to the camera and located close to it [24]. The parallel processing system ensures real-time operation by processing bulk image data and selecting and storing regions of interest (RoIs). The parallel processing system could be a part of the camera itself, or a FPGA processor or a general purpose processor with special hardware. This part of the system is vitally important both from real-time operation as well as accuracy of defect detection and classification. Thereafter, a server with a large backup memory is used for further processing and for operator's interface. 8. List of defect detection and classification methods Various methods/techniques used for defect detection and classification of steel surfaces are listed in the litera-ture. Table 1 shows the list of different methods of de-fect detection vis-à-vis references obtained for this study. Types of steel surfaces have also been mentioned in the table. Techniques followed may broadly be cate-gorised as statistical, morphological, spatial domain filtering, frequency domain analysis, joint spatial/spatial-frequency analysis and fractal models. Spatial domain filtering, morphological operations and joint spatial/fre-quency domain filtering are found to be used extensively for all types of surfaces. Ultimate objective of surface inspection is to categorise defects in specified classes using classification tech-niques. As a process, classification starts after defects are localised by segmentation. At this stage, generally a number of features are extracted from regions of inter-est. Ideally, different combinations of these features are required to match uniquely with that of different types of defects. The matching is normally done using adap-tive learning methods such as neural network with back propagation (NN-BP), support vector machine (SVM)  etc. Adaptive learning is of two types: a) supervised where the network is provided with a large number of known in-puts. Thereafter, the network produces the known outputs as closely as possible based on training. b) In unsupervised learning, the network is required to work out relationships between various inputs without being told. However, steel surface defects exhibit large ‘inter group similarity and intra group diversity ’. Thus, finding suitable features and identifying classifiers with low computational cost are the major areas of research activ-ity. Table 2 shows the list of classification methods with respect to references and types of surface. Conclusions This paper dealt with review of automated inspection?methods for steel surfaces using image processing techniques. Review of publications over two and a half decades?has provided an idea of recent advances that have taken?place in this field.?Main observations are as follows: a) Due to harsh environment of a steel