Penggunaan Histogram dari Koefisien Aproksimasi Wavelet untuk Deteksi Cacat Tekstil
Main Article Content
Keywords
Abstract
Generally, textile defect inspection at textile industry is still conducted manually by human. This approach is susceptible to errors and tends to be inconsistent due to fatigue and inattentiveness. To guarantee the consistency and inspection quality, an automatic defect detection system is required. This research proposes the use of histograms generated from two-level wavelet’s approximation coefficients as features to detect textile defects. The Euclidian distance that is calculated between feature of reference textile (non-defective textile) and feature of defective one is used as an evaluation parameter. If the Euclidian distances of the features of textile images are higher than a predetermined threshold, the textiles are determined as defective ones, and vice versa. Simulations are conducted using four groups of textile defects. It turns out that the proposed method can achieve 100% detection rate for textile group with ink-spot and textile group with holes.
Keywords : Wavelet coefficient histogram, Euclidean distance, Textile defect, industrial textiles, Image features
Abstrak
Pada industri tekstil, cacat produksi umumnya masih diperiksa secara manual oleh manusia. Pemeriksaan secara manual rentan terhadap kesalahan dan kurang konsisten, karena sifat manusia yang dapat lelah, lupa dan lain sebagainya. Untuk menjamin konsistensi dan kualitas pemeriksaan cacat kain, sebuah sistem deteksi otomatis perlu ada. Penelitian ini mengusulkan penggunaan histogram dari koefisien aproksimasi wavelet dua tingkat sebagai fitur untuk deteksi cacat tekstil. Jarak Euclidian yang dihitung diantara fitur tekstil citra referensi (berasal dari citra tidak cacat) dengan fitur tesktil citra cacat digunakan sebagai parameter evaluasi. Jika jarak Euclidian dari fitur suatu citra tekstil berada di atas nilai ambang yang telah ditentukan sebelumnya, citra tersebut dinyatakan cacat, dan sebaliknya. Penelitian dilaksanakan dengan menjalankan simulasi deteksi cacat tekstil, menggunakan empat kelompok cacat tekstil yang berbeda. Ditemukan bahwa metode usulan mencapai tingkat kebenaran deteksi sebesar 100% untuk citra kelompok cacat tinta dan kelompok cacat lubang.
Kata Kunci : Histogram koefisien wavelet, Jarak Euclidean, Cacat tekstil, Industri tekstil, Fitur citra
References
[2].Tilda Database, diakses di http://lmb.informatik.uni-freiburg.de/resources/datasets/tilda.en.html, Februari 2013.
[3] H. Y. T. Ngan et. al., ”Automated fabric defect detection- A Review”, Image and Vision Computing, Vol. 29, pp. 442-458, 2011.
[4] Chun-Lin, Liu, ”A Tutorial of the wavelet transform”, February 23, 2010, diakses pada 20 Februari 2014.
[5] S. Arivazhagan et. al., ”Fault segmentation in fabric images using Gabor wavelet transform”, Machine Vision Applications, vol. 16, No. 6, pp. 356-363, 2006.
[6] Chi-ho Chan, et. al., ”Fabric defect detection by Fourier analysis”, IEEE Transactions on Industry Applications, Vol. 36, No. 5, pp. 1267-1276, 2000.
[7] A. Serdaroglu, et.al., Defect detection in textile fabric images using wavelet transforms and independent component analysis”, Pattern Recognition and Image Analysis, Vol. 16, No. 1, pp. 61-64, 2004.
[8] H. Y. T. Ngan et. al., ”Wavelet based methodson pattern fabric defect detection”, Pattern recognition, Vol. 38, pp. 559-576, 2005.
[9] Andika, Fitri Arnia dan Khairul Munadi, ”De- teksi cacat tekstil menggunakan fitur koefisien aproksimasi wavelet dan euclidean distance,” karya ilmiah Jurusan Teknik Elektro, Universitas Syiah Kuala, 2013.
[10] V. Murino, et.al., ”Statistical classification of raw textile defects”, Proc. of the 17th International Conference on Pattern Recognition”, 2004.
[11] V. Castelli, ”Image Databases: Search and Retieval of Digital Imagery”, John Wiley and Sons, 2002.
Article Sidebar
Article Details
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
The authors declare that:
1. This paper has not been published in the same form elsewhere.
2. It will not be submitted anywhere else for publication prior to acceptance/rejection by this Journal.
3. A copyright permission is obtained for materials published elsewhere and which require this permission for reproduction.