Steel Surface Defect Detection Using Texture Segmentation Based on Multifractal Dimension

Yazdchi, Mohammadreza; Yazdi, Mehran; Mahyari, Arash Golibagh

doi:10.1109/icdip.2009.68

Cited by 51 publications

(37 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Karmaşık ve çok olan verilerin bilgisayar destekli incelenmesi hızlı bir şekilde gerçekleştirilebilir [4]. Metal üreticileri kusurlu ürünleri tespit ederek istenmeyen kusurların sebeplerini ortadan kaldırabilir ve müşterilerine devamlı olarak kaliteli ürün temin edebilirler [5]. Haddeleme işlemi sırasında oluşan kusurların tespitine yönelik Wu, K ve arkadaşları, yaptıkları çalışmada FFT özellik çıkarımı metodu kullanılarak genetik algoritma ve yapay sinir ağları yöntemleriyle kusur tespiti yapmışlardır [6].…”

Section: Gi̇ri̇ş (Introduction)unclassified

ÜRETİM AŞAMASINDA RAY ve PROFİLDE OLUŞAN KUSURLARININ TESPİTİNE YÖNELİK BİR PARALEL KUSUR ALGILAMA ALGORİTMASI

Orak¹,

Çelik²

2017

Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

Ray ve profil kusurlarının bilgisayar destekli tespit edilmesi  Paralel görüntü işleme metodu  CUDA kernel yapısının performans kıyaslaması Makale Bilgileri ÖZET Geliş: 17.03.2016 Kabul: 14.11.16 DOI:Otomatik bir sistem tarafından görüntü işlenerek sonuç elde etmek günümüzde pek çok alanda gerekli olabilmektedir. Kusurlu ürün üretimi birçok alanda karşılaşılan üreticiler tarafından da istenmeyen bir durumdur. Görüntü işleyerek görüntü üzerindeki kusurların tespit edilmesi bu alanda kullanılan bir yöntemdir. Görüntü işleme piksel temelli yapıldığından dolayı çok iş yükü oluşturmaktadır. Hızın işlem sürecinde önem arz ettiği durumlarda paralel görüntü işlemenin yapılması bir çözüm olabilmektedir. Bundan dolayı mevcut çok çekirdekli bilgisayarların donanım ya da yazılım yardımıyla paralelleştirilerek görüntülerin işlenmesi performans sağlayacaktır. Paralel görüntü işlemede elde edilen performans, kullanılan algoritmanın paralelliğe uygun olması ve işlemcilere doğru bir şekilde dağıtım yapılması ile ilişkilidir. Kaynakların ortak kullanımı ve veri alışverişinin fazla olması performansı doğrudan etkiler. Bu çalışmada; Kardemir A.Ş. haddehanede, haddeleme işlemi sırasında ray ve profil yüzeylerinde meydana gelen kusurların tespit edilmesine yönelik geliştirilen COLMSTD algoritmasının paralel uygulaması, iki farklı şekilde gerçekleştirilmiştir. 1. Yöntemde GPU'da çalışacak CUDA çekirdek sayıları yazılımsal olarak değiştirilmiş ve 2. Yöntemde ise tek CUDA çekirdeğindeki blok sayıları değiştirilerek, test gerçekleştirilmiştir. GPU (Graphics Processing Unit) üzerinde CUDA (Compute Unified Device Architecture) arayüz desteğinin uygulanması ile elde edilen değerler ile CPU üzerinde elde edilen değerler kıyaslanmıştır.

show abstract

Section: Gi̇ri̇ş (Introduction)unclassified

ÜRETİM AŞAMASINDA RAY ve PROFİLDE OLUŞAN KUSURLARININ TESPİTİNE YÖNELİK BİR PARALEL KUSUR ALGILAMA ALGORİTMASI

Orak¹,

Çelik²

2017

Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

show abstract

“…In reference 2) hough transform is used to classify defects with well defined geometric shapes, and principal component analysis and self-organizing maps are used to classify defects with complex shapes. In reference 3) a new defect detection algorithm is provided with multifractal features and neural work classifier. Reference 4) used local binary pattern algorithm to extract defect features, and employed decision tree and support vector machine as classification method.…”

Section: Introductionmentioning

confidence: 99%

“…References [1][2][3][4][5] can effectively extract features and realize classification for the strip steel surface defect. However, samples with the same type of defect may be different in scale and rotation in real problems, which requires the algorithm of defect feature extraction to be of scale and rotation invariance.…”

Section: Introductionmentioning

confidence: 99%

“…Defect feature extraction and defect classification are the crucial steps and the study of them is becoming more and more widely and deeply. [1][2][3][4][5] In reference 1) extended haar rectangle features are extracted and weak classifier adaptive enhancement classification method is used. In reference 2) hough transform is used to classify defects with well defined geometric shapes, and principal component analysis and self-organizing maps are used to classify defects with complex shapes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Strip Steel Surface Defect Recognition Based on Novel Feature Extraction and Enhanced Least Squares Twin Support Vector Machine

2014

View full text Add to dashboard Cite

Feature extraction and classification are two important steps in the process of strip steel surface defect recognition. Traditional methods of defect feature extraction are not of scale and rotation invariance. Moreover, traditional methods of defect classification have a conflict between efficiency and accuracy in. In order to solve the above two problems, a novel recognition method is proposed in this paper. On one hand, the novel defect feature extraction scheme is realized by building sampling benchmark scale (SBS) information for training dataset and using gradient magnitude and gradient orientation co-occurrence matrix (GMGOCM), gray level and gradient orientation co-occurrence matrix (GLGOCM), and moment invariant features. On the other hand, K-nearest neighbor and R-nearest neighbor algorithms are used to prune training dataset, and amplification factors of pruned samples are used to improve least squares twin support vector machine (LSTWSVM) classifier in efficiency and accuracy. The experimental results show that the novel recognition method can not only realize defect feature extraction with scale and rotation invariance but also realize defect classification with high efficiency and accuracy.

show abstract

“…Image segmentation is an indispensable step for most popular defect detection methods, including statistical methods, 2-9 model-based methods, [10][11][12] and frequency spectral methods. [13][14][15][16][17][18][19][20][21] In all of these methods, the determination and location of defects depends on image segmentation.…”

Section: Introductionmentioning

confidence: 99%

Rapid surface defect detection based on singular value decomposition using steel strips as an example

2018

View full text Add to dashboard Cite

On the radiated EMI current extraction of dc transmission line based on corona current statistical measurements AIP Advances 8, 055001 (2018) For most surface defect detection methods based on image processing, image segmentation is a prerequisite for determining and locating the defect. In our previous work, a method based on singular value decomposition (SVD) was used to determine and approximately locate surface defects on steel strips without image segmentation. For the SVD-based method, the image to be inspected was projected onto its first left and right singular vectors respectively. If there were defects in the image, there would be sharp changes in the projections. Then the defects may be determined and located according sharp changes in the projections of each image to be inspected. This method was simple and practical but the SVD should be performed for each image to be inspected. Owing to the high time complexity of SVD itself, it did not have a significant advantage in terms of time consumption over image segmentation-based methods. Here, we present an improved SVD-based method. In the improved method, a defect-free image is considered as the reference image which is acquired under the same environment as the image to be inspected. The singular vectors of each image to be inspected are replaced by the singular vectors of the reference image, and SVD is performed only once for the reference image off-line before detecting of the defects, thus greatly reducing the time required. The improved method is more conducive to real-time defect detection.

show abstract

Steel Surface Defect Detection Using Texture Segmentation Based on Multifractal Dimension

Cited by 51 publications

References 11 publications

ÜRETİM AŞAMASINDA RAY ve PROFİLDE OLUŞAN KUSURLARININ TESPİTİNE YÖNELİK BİR PARALEL KUSUR ALGILAMA ALGORİTMASI

ÜRETİM AŞAMASINDA RAY ve PROFİLDE OLUŞAN KUSURLARININ TESPİTİNE YÖNELİK BİR PARALEL KUSUR ALGILAMA ALGORİTMASI

Strip Steel Surface Defect Recognition Based on Novel Feature Extraction and Enhanced Least Squares Twin Support Vector Machine

Rapid surface defect detection based on singular value decomposition using steel strips as an example

Contact Info

Product

Resources

About