Wood defect detection method with PCA feature fusion and compressed sensing

Abstract

We used principal component analysis (PCA) and compressed sensing to detect wood defects from wood plate images. PCA makes it possible to reduce data redundancy and feature dimensions and compressed sensing, used as a classifier, improves identification accuracy. We extracted 25 features, including geometry and regional features, gray-scale texture features, and invariant moment features, from wood board images and then integrated them using PCA, and selected eight principal components to express defects. After the fusion process, we used the features to construct a data dictionary, and realized the classification of defects by computing the optimal solution of the data dictionary in l ₁ norm using the least square method. We tested 50 Xylosma samples of live knots, dead knots, and cracks. The average detection time with PCA feature fusion and without were 0.2015 and 0.7125 ms, respectively. The original detection accuracy by SOM neural network was 87 %, but after compressed sensing, it was 92 %.

Keywords

Principal component analysis / Compressed sensing / Wood board classification / Defect detection

References

[1] Candes E. Compressive sampling. Proc Int Congr Math, 2006, 3: 1433-14521.

[2] Castellani M, Rowlands H. Evolutionary artificial neural network design and training for wood veneer classification. Eng Appl Artif Intell, 2009, 22: 732-741.

[3] Donoho D. Compressed sensing. IEEE Trans Inf Theory, 2006, 52(4): 1289-1300.

[4] Estévez PA, Perez CA, Goles E. Genetic input selection to a neural classifier for defect classification of radiata pine boards. For Prod J, 2003, 53(7/8): 87-94.

[5] Gu IYH, Andersson H, Vicen R (2008) Automatic classification of wood defects using support vector machines. In: Computer vision and graphics: lecture notes in computer science, vol 5337. Springer, Berlin, pp 356–367

[6] Peck R, Devore JL. The exploration and analysis of data. 2005, Belmont: Duxbury Press, 611 662

[7] Pham DT, Alcock RJ. Automated grading and defect detection: a review. For Prod J, 1998, 48(3): 34-42.

[8] Pham DT, Alcock RJ. Automated visual inspection of wood boards selection of features for defect classification by a neural network. Proc Inst Mech Eng Part E, 1999, 213(4): 231-245.

[9] Ruz GA, Estevez PA, Ramirez PA. Automated visual inspection system for wood defect classification using computational intelligence techniques. Int J Syst Sci, 2009, 40(2): 163-172.

[10] Shi GM, Liu DH, Gao DH, Liu Z, Lin J, Wang LJ. Advances in theory and application of compressed sensing. Acta Electron Sin, 2009, 37(5): 1070-1081.

[11] Zhang YZ, Cao J, Xu L, Yu HL. Wood floor defects segmentation and recognition based on morphological and SOM. Electr Mach Control, 2013, 17(4): 116-120.

[12] Zhang YZ, Liu SJ, Cao J, Li C, Yu HL. A rapid, automated flaw segmentation method using morphological reconstruction to grade wood flooring. J For Res, 2014, 25(4): 959-964.

[13] Zhang YZ, Xu L, Ding L, Cao J. Defects segmentation for wood floor based on image fusion method. Electr Mach Control, 2014, 18(7): 113-118.