Soft measurement of wood defects based on LDA feature fusion and compressed sensor images

Abstract

We proposed a detection method for wood defects based on linear discriminant analysis (LDA) and the use of compressed sensor images. Wood surface images were captured, using a camera Oscar F810C IRF camera, and then the image segmentation was performed, and the defect features were extracted from wood board images. To reduce the processing time, LDA algorithm was used to integrate these features and reduce their dimensions. Features after fusion were used to construct a data dictionary and a compressed sensor was designed to recognize the wood defects types. Of the three major defect types, 50 images live knots, dead knots, and cracks were used to test the effects of this method. The average time for feature fusion and classification was 0.446 ms with the classification accuracy of 94%.

Keywords

Compressed sensing / Defect detection / Linear discriminant analysis / Wood-board classification

References

[1] Candes E (2006) Compressive sampling. In: Proceedings of the international congress of mathematicians, Madrid, Spain, vol 3, pp 1433–1452

[2] Donoho D. Compressed sensing. IEEE Trans Inf Theory, 2006, 52(4): 1289-1306.

[3] Kwon BK, Won JS, Kang DJ. Fast defect detection for various types of surfaces using random forest with VOV features. Int J Precis Eng Manuf, 2015, 16: 965-970.

[4] Lampinen J, Smolander S, Korhonen M (1995) Wood surface inspection system based on generic visual features. In: International conference on artificial neural networks ICANN, vol 95. pp 9–13

[5] Li C, Huang JY, Chen CM. Soft computing approach to feature extraction. Fuzzy Set Syst, 2004, 147(1): 119-140. in Chinese)

[6] Li C, Su YW, Zhang YZ, Yang HM. Root imaging from ground penetrating radar data by CPSO-OMP compressed sensing. J For Res, 2017, 28(1): 155-162.

[7] Niskanen M, Silvén O, Kauppinen H (2001) Color and texture based wood inspection with non-supervised clustering. In: Proceedings of the Scandinavian conference on image, pp 336–342

[8] Peck R, Devore JL. Statistics: the exploration & analysis of data, 2011, Belmont: Duxbury Press 611 662

[9] Pham DT, Alcock RJ. Automated grading and defect detection: a review. For Prod J, 1998, 48(4): 34-42.

[10] Schütt C, Aschoff T, Winterhalder D, Thies M, Kretschmer U, Spiecker H (2004) Approaches for recognition of wood quality of standing trees based on terrestrial laserscanner data. In: Thies M, Koch B, Spiecker H (eds)

[11] Silvén O, Niskanen M, Kauppinen H. Wood inspection with non-supervised clustering. Mach Vis Appl, 2003, 13(5–6): 275-285.

[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.

[14] Zhang YZ, Xu C, Li C, Yu HL, Cao J. Wood defect detection method with PCA feature fusion and compressed sensing. J For Res, 2015, 26(3): 745-751.