基于PSO优选特征的实木板材缺陷的压缩感知分选方法

李超; 刘思佳; 曹军; 于慧伶; 张怡卓

doi:10.13332/j.1000-1522.20140385

基于PSO优选特征的实木板材缺陷的压缩感知分选方法

李超^1,2,
刘思佳¹,
曹军¹,
于慧伶¹,
张怡卓^1, ,

1.
1 东北林业大学机电工程学院;
2.
2 东北林业大学机械工程流动站

基金项目:

林业公益性行业科研专项(201304510)、黑龙江省自然基金项目(C201405)、中央高校基本科研业务费专项(DL13CB02、DL13BB21)

详细信息

作者简介:
李超,博士,讲师。主要研究方向:信号处理与图像处理。Email:lchao820225@163.com 地址:150040黑龙江省哈尔滨市和兴路26号东北林业大学机电工程学院。

责任作者:
张怡卓,博士,副教授。主要研究方向:图像处理与模式识别。Email:nefuzyz@163.com 地址:同上。

计量
- 文章访问数: 1560
- HTML全文浏览量: 212
- PDF下载量: 21
出版历程
- 收稿日期: 2014-10-22
- 修回日期: 2014-10-22
- 发布日期: 2015-07-30

The method of wood defect recognition based on PSO feature selection and compressed sensing

1.
1 College of Mechanical and Electrical Engineering, Northeast Forestry University, Harbin, Heilongjiang, 150040, P. R. China;
2.
2 Mechanical Engineering Station, Northeast Forestry University, Harbin, Heilongjiang, 150040, P. R. China

摘要

摘要: 针对实木板材表面缺陷的复杂性与随机性,提出了一种快速、准确的识别方法。首先,对实木板材表面图像进行3级双树复小波分解,提取低频子带、高频子带、原图像的均值、标准差和熵,共40维特征向量;然后,运用粒子群算法(PSO)优选出20个关键特征;最后,采用压缩感知理论将优选后的特征向量作为样本矩阵列,构建训练样本数据字典,通过最小残差完成缺陷识别。对4类柞木样本进行了仿真实验,活结、死结、虫眼、裂纹的分类正确率分别为93.3%、86.7%、100%和93.3%,结果表明:双树复小波良好的方向性能够表达实木板材表面复杂的信息;基于粒子群算法的特征选择能够提高分类效率;压缩感知分类器与传统分类器相比,具有结构简单、分类精度高的特点。
- 缺陷识别 /
- 双树复小波 /
- 粒子群算法 /
- 压缩感知
Abstract: Aimed at the complexity and randomness of the wood board defects, we propose a novel and efficient method in this paper. Firstly, three-level dual-tree complex wavelet decomposition was used to extract 40 features, including average value, standard deviation and entropy from low-frequency, high-frequency sub-bands and the original image. Then, the particle swarm optimization (PSO) algorithm was applied and 20 key features obtained. Finally, a data dictionary of training samples was constructed based on compressed sensing, and classification of defects was completed by the minimal reconstruction error. Four types of Xylosma racemosum wood samples, i.e., live knot, dead knot, pinhole and crack, were used for the experiment. The recognition rates of the four types were 93.3%, 86.7%, 100% and 93.3%, respectively. Experimental results showed that the good directionality of dual-tree complex wavelets can reflect the complex information of wood board, the PSO can improve the efficiency of classification, and the compressed sensing has the advantages of simple structure and high classification accuracy.
- defect recognition /
- dual-tree complex wavelet /
- particle swarm optimization /
- compressed sensing

HTML全文

参考文献(28)

[1]	BAI X B, ZOU L H. Segmentation method of timber surface defects based on gray level-gradient co-occurrence matrix [J]. Forest Engineering, 2007, 23(2): 16-18.
[1]	RUZ G A, ESTEVEZ P A, RAMIREZ P A. Automated visual inspection system for wood defect classification using computational intelligence techniques [J]. International Journal of Systems Science 2009, 40(2): 163-172.
[2]	WANG L, BAI X B. Image segmentation method of surface defects of wood based on Gabor transform [J]. Computer Engineering and Design, 2010, 31(5): 1066-1069.
[2]	白雪冰, 邹丽晖. 基于灰度-梯度共生矩阵的木材表面缺陷分割方法[J]. 森林工程, 2007, 23(2): 16-18.
[3]	王林, 白雪冰. 基于Gabor变换的木材表面缺陷图像分割方法[J]. 计算机工程与设计, 2010, 31 (5): 1066-1069.
[3]	XIAO B J, SHU W Q. Research on feature extraction of defects on wood surfaces based on image fusion [J]. Computer Science, 2011, 38(4): 282-285.
[4]	ZHANG Y Z, CAO J, XU L, et al. Wood floor defects segmentation and recognition based on morphological and SOM [J]. Electric Machines and Control, 2013, 17(4):117-120.
[4]	PHAM D T, ALCOCK R J. Automated grading and defect detection: a review [J]. Forest Products Journal, 1998, 48(3): 34-42.
[5]	ZHANG Y Z, XU L, CAO J. Defects segmentation for wood floor based on image fusion method [J]. Electric Machines and Control, 2014, 17(4): 117-120.
[5]	PHAM D T, ALCOCK R J. Automated visual inspection of wood boards selection of features for defect classification by a neural network [J]. Proceedings of the Institution of Mechanical Engineering, 1999, 213(4): 231-245.
[6]	CASTELLANI M, ROWLANDS H. Evolutionary artificial neural network design and training for wood veneer classification [J]. Engineering Applications of Artificial Intelligence, 2009, 22:732-741.
[6]	SUN Y K. Wavelet transform and image processing technology [M]. Beijing: Tsinghua University Press, 2012.
[7]	WU Q T, CAO J B, ZHENG R J, et al. Intrusion feature selection algorithm based on Particle Swarm Optimization [J]. Computer Engineering and Applications, 2013, 49(7): 89-92.
[7]	肖宾杰, 殳伟群. 基于图像融合的木板表面缺陷特征提取方法研究[J]. 计算机科学, 2011, 38(4): 282-285.
[8]	MAHRAM A. Classification of wood surface defects with hybrid usage of statistical and textural features[C]. 35th International Conference on Telecommunications and Signal Processing. New York: IEEE, 2012:749-752.
[8]	WANG K Q, BAI X B, WANG H. Classification of wood surface texture based on wavelets transform [J]. Journal of Harbin Institute of Technology, 2009, 41(9): 232-234.
[9]	CHEN L J, WANG K Q, WANG H, et al. The confirmation of wavelet base and decomposition progression in wood texture analysis [J]. Forestry Machinery & Woodworking Equipment, 2007, 35(5): 25-27.
[9]	张怡卓, 曹军, 许雷, 等. 实木地板缺陷形态学分割与SOM识别[J]. 电机与控制学报, 2013, 17(4):117-120.
[10]	张怡卓, 许雷, 曹军. 基于图像融合分割的实木地板表面缺陷检测方法[J]. 电机与控制学报, 2014, 17(4): 117-120.
[11]	TSAI D M, HSIAO B. Automatic surface inspection using wavelet reconstruction [J]. Pattern Recognition, 2001, 34(6): 1285-1305.
[12]	HAN Y F, SHI P F. An adaptive level-selecting wavelet transform for texture defect detection [J]. Image and Vision Computing, 2007, 25(8) :1239-1248.
[13]	KINGSBURY N. Complex wavelets for shift invariant analysis and filtering of signals [J]. Applied and Computational Harmonic Analysis, 2001, 10(3): 234-253.
[14]	孙延奎. 小波变换与图像、图形处理技术[M]. 北京: 清华大学出版社, 2012.
[15]	吴庆涛, 曹继邦, 郑瑞娟, 等. 基于粒子群优化的入侵特征选择算法[J]. 计算机工程与应用, 2013, 49 (7): 89-92.
[16]	EBERHART R C,KENNEDY J. A discrete binary version of the particle swarm algorithm [C]∥Proceedings of the IEEE Conference on Systems, Man, and Cybernetics. Orlando, FL: IEEE, 1997: 4104-4109.
[17]	DONOHO D L. Compressed sensing [J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
[18]	王克奇, 白雪冰, 王辉. 基于小波变换的木材表面纹理分类[J]. 哈尔滨工业大学学报, 2009, 41(9): 232-234.
[19]	陈立君, 王克奇, 王辉, 等. 木材纹理分析中小波基的选择和分解级数的确定[J]. 林业机械与木工设备, 2007, 35(5): 25-27.