[1]刘瑞明,黄佳炜,刘勇,等.基于相关向量机回归的涂泥木线条表面点状小缺陷检测[J].林业工程学报,2019,4(05):115-120.[doi:10.13360/j.issn.2096-1359.2019.05.016]
 LIU Ruiming,HUANG Jiawei,LIU Yong,et al.Detection of small point surface defects of gesso primed wood moldings based on relevant vector machine[J].Journal of Forestry Engineering,2019,4(05):115-120.[doi:10.13360/j.issn.2096-1359.2019.05.016]
点击复制

基于相关向量机回归的涂泥木线条表面点状小缺陷检测()
分享到:

《林业工程学报》[ISSN:1001-8081/CN:32-1160/S]

卷:
4
期数:
2019年05期
页码:
115-120
栏目:
装备与信息化
出版日期:
2019-09-16

文章信息/Info

Title:
Detection of small point surface defects of gesso primed wood moldings based on relevant vector machine
文章编号:
2096-1359(2019)05-0115-06
作者:
刘瑞明1黄佳炜2刘勇2孙帅成2王双永3周海宾3
1.江苏海洋大学电子工程学院,江苏 连云港 222005; 2.江苏海洋大学机械与海洋工程学院, 江苏 连云港 222005; 3.中国林业科学研究院木材工业研究所,北京 100091
Author(s):
LIU Ruiming1 HUANG Jiawei2 LIU Yong2 SUN Shuaicheng2 WANG Shuangyong3ZHOU Haibin3
1.School of Electronic Engineering, Jiangsu Ocean University, Lianyungang 222005, Jiangsu, China; 2.School of Mechanical and Ocean Engineering, Jiangsu Ocean University, Lianyungang 222005, Jiangsu, China; 3.Research Institute of Wood Industry, Chinese Academy of Forestry, Beijing 100091, China
关键词:
木线条 相关向量机(RVR) 核相关系数 表面缺陷检测 信杂比 背景抑制因子
Keywords:
wood molding relevance vector regression kernel correlation coefficients surface defect detection signal-to-clutter ratio background suppression factor
分类号:
S781.5
DOI:
10.13360/j.issn.2096-1359.2019.05.016
文献标志码:
A
摘要:
木线条是一种用途广泛的建筑材料,主要用于装饰、装修和家具制造等行业。涂泥木线条具有美观、强度高、耐潮湿等优点,深受用户的青睐。在生产过程中,为了提高产品质量,需要对涂泥木线条的表面缺陷进行检测,基于数字图像处理技术的缺陷检测方法已成为主要技术手段。在图像中,面积较大的缺陷相对容易检测,但小的点状缺陷由于包含像素少、缺乏纹理特征等特点,检测非常困难。使用传统的、基于滤波的方法检测小缺陷,主要利用了小缺陷的高频特性,但高频图像噪声会造成误检,导致检测效果并不理想。本研究提出了基于相关向量回归结合后处理的方法对小缺陷进行检测,相关向量回归与支持向量回归相比,具有超参数少、表达更稀疏、核函数不需要满足梅西定理等优点。该方法通过以下步骤实现小缺陷检测:首先利用相关向量回归算法对图像进行处理,将回归值作为像素灰度值构建回归图像,然后求原始图像与回归图像的差图像,求取差图像的核相关系数后,再经过取反、二值化和局部平均等后处理,最终得到信杂比较大的检测图像。通过检测指标的比较,与基于Top-hat滤波的检测算法相比,本研究提出的方法具有更好的检测效果。
Abstract:
Molding is a kind of widely used building material, which is utilized primarily for decorating and furniture manufacturing.The gesso primed wood molding is favored by users because it has beautiful appearance, high strength and moisture resistance.In the molding production, surface defects of gesso primed wood moldings must be checked and eliminated in order to improve the quality of products.The defect detection methods based on digital image processing technology have been widely applied in the wood industry.It is relatively easy to detect those defects with a large area in images.However, detecting tiny point defects is very difficult due to fewer pixels and lack of texture features, which is a current technical challenge.The traditional defect detection methods based on filtering technology mainly utilize the high-frequency characteristics of tiny defects.Unfortunately, image noise has also high-frequency characteristics, which may cause false detection and the detection performance is unsatisfactory.In this paper, a method based on relevance vector regression combined with some post-processing was applied to detect tiny point defects on the gesso primed wood molding.Compared with support vector regression, the relevance vector regression does not need to set many hyper-parameters.Its representation was sparser than the support vector regression, which means less computation required by the relevance vector regression.Another advantage was that its kernel function was not necessary to satisfy the Mercer's condition.This method executed the detection of tiny point defects with the following steps.Firstly, the image was processed by the relevance vector regression algorithm, and the regression value of every pixel was obtained.The regression value was used as the pixel gray value to construct a regression image.The pixels of tiny defects were smoothed in the regression image.The background pixels were unchanged.Then, the difference between the original image and the regression image was calculated.In the different images, the different values of tiny defects were larger than those of background.The kernel correlation coefficient of the difference image was computed.The post-processing methods, such as inversion, binarization and local averaging, were executed in order to improve the detection performance further.Finally, the detection image with high signal-to-clutter ratio and large background suppression factor was constructed.By comparing the detection indicators, the proposed method had better detection performance than that of the Top-hat filter-based detection algorithm.

参考文献/References:

[1] 马旭, 刘应安, 业宁, 等.基于核PCA与SVM算法的木材缺陷识别[J].常州大学学报(自然科学版), 2017, 29(3): 60-68.DOI: 10.3969/j.issn.2095-0411.2017.03.009.
MA X, LIU Y A, YE N, et al.Application of KPCA and SVM to wood defect recognition[J].Journal of Changzhou University(Natural Science Edition), 2017, 29(3): 60-68.
[2] 贾壮, 戴天虹, 李昊, 等.木材缺陷的图像分割提取技术综述[J].西部林业科学, 2015, 44(5): 148-152.DOI: 10.16473/j.cnki.xblykx1972.2015.05.029.
JIA Z, DAI T H, LI H, et al.A review of extraction technology of image segmentation of wood defect[J].Journal of West China Forestry Science, 2015,44(5): 148-152.
[3] 刘传泽, 葛浙东, 张连滨, 等.数字图像处理技术在木材科学中的应用[J].林业机械与木工设备, 2018, 46(2): 13-17.DOI: 10.13279/j.cnki.fmwe.2018.0014.
LIU C Z, GE Z D, ZHANG L B, et al.Application of digital image processing technology to wood science[J].Forestry Machinery & Woodworking Equipment, 2018, 46(2): 13-17.
[4] 张未, 吕志娟, 徐兆军, 等.基于激光三角测距的锯材表面缺陷检测方法[J].林业工程学报, 2017, 2(6): 116-120.DOI:10.13360 /j.issn.2096-1359.2017.06.020.
ZHANG W, LYU Z J, XU Z J, et al.Detecting method for surface defect of swan timber based on laser triangulation[J].Journal of Forestry Engineering, 2017, 2(6): 116-120.
[5] 白雪冰,李润佳,许景涛,等.基于图割算法的木材表面缺陷图像分割[J].林业工程学报, 2018, 3(2): 116-122.DOI: 10.13360/j.issn.2096-1359.2018.02.019.
BAI X B, LI R J, XU J T, et al.Wood surface defect image segmentation based on Graph Cut algorithm[J].Journal of Forestry Engineering, 2018, 3(2): 116-122.
[6] 任重昕, 毕剑华, 谢琳, 等.采用图像分割方法进行木材表面缺陷的定量检测[J].液晶与显示, 2014, 29(5): 785-792.DOI: 10.3788/YJYXS20142905.0785.
REN C X, BI J H, XIE L, et al.Quantitative detection of wood surface defects by image segmentation method[J].Chinese Journal of Liquid Crystals and Displays, 2014, 29(5): 785-792.
[7] 王泽润, 方益明, 冯海林, 等.木材节子缺陷检测与定位方法[J].激光与光电子学进展, 2018, 55(5): 317-324.DOI: 10.3788/LOP55.051501.
WANG Z R, FANG Y M, FENG H L, et al.Method for wooden knot detection and localization[J].Laser & Optoelectronics Progress, 2018, 55(5): 317-324.
[8] 杨建华, 于淼, 范磊, 等.中密度纤维板表面缺陷检测软件设计[J].木材加工机械, 2018,29(1): 10-12.DOI:10.13594/j.cnki.mcjgjx.2018.01.003.
YANG J H, YU M, FAN L, et al.Design of surface defect detection software for medium density fiberboard[J].Wood Processing Machinery, 2018,29(1): 10-12.
[9] 程玉柱, 蔡云飞.基于分数阶CV模型的木材缺陷图像分割算法[J].林业机械与木工设备, 2018, 46(4): 44-47.DOI: 10.13279/j.cnki.fmwe.2018.0043.
CHENG Y Z, CAI Y F.Wood defect image segmentation algorithm based on fractional order CV model[J].Forestry Machinery & Woodworking Equipment, 2018, 46(4): 44-47.
[10] 程玉柱, 顾权, 王众辉, 等.基于深度学习的木材缺陷图像检测方法[J].林业机械与木工设备, 2018, 46(8): 33-36.DOI: 10.13279/j.cnki.fmwe.2018.0087.
CHENG Y Z, GU Q, WANG Z H, et al.Wood defect image segmentation based on deep learning[J].Forestry Machinery & Woodworking Equipment, 2018, 46(8): 33-36.
[11] 胡忠康,刘英,周晓林,等.基于深度置信网络的实木板材缺陷及纹理识别研究[J/OL].计算机应用研究,2019,36(12).http://kns.cnki.net/kcms/detail/51.1196.TP.20181011.1330.052.html.DOI: 10.3969/j.issn.1001-3695.2018.07.0438.
HU Z K, LIU Y, ZHOU X L, et al.Research on defects and textures recognition of solid wood lumbers based on deep belief network[J/OL].Application Research of Computers, 2019,36(12).http://kns.cnki.net/kcms/detail/51.1196.TP.20181011.1330.052.html.
[12] 马旭, 刘应安, 业宁, 等.基于核PCA与SVM算法的木材缺陷识别[J].常州大学学报(自然科学版), 2017, 29(3): 60-68.DOI: 10.3969/j.issn.2095-0411.2017.03.009.
MA X, LIU Y A, YE N, et al.Application of KPCA and SVM to wood defect recognition[J].Journal of Changzhou University(Natural Science Edition), 2017, 29(3): 60-68.
[13] 梁浩, 曹军, 林雪, 等.基于贝叶斯神经网络的近红外光谱实木地板表面缺陷检测[J].光谱学与光谱分析, 2017, 37(7): 2041-2045.DOI: 10.3964/j.issn.1000-0593(2017)07-2041-05.
LIANG H, CAO J, LIN X, et al.Surface defects detection of solid wood board using near-infrared spectroscopy based on bayesian neural network[J].Spectroscopy and Spectral Analysis, 2017, 37(7): 2041-2045.
[14] 谢永华, 赵丽花, 王金聪.基于颜色和数学形态学的木材缺陷检测研究[J].现代科学仪器, 2013(2): 75-77.
XIE Y H, ZHAO L H, WANG J C.Detection of timber defects based on color and mathematical morphology[J].Modern Scientific Instruments, 2013(2): 75-77.
[15] 王程程,刘亚秋.基于动态目标检测与跟踪算法(TLD)的刨花板表面缺陷检测技术[J].木材工业,2018,32(4): 44-47.DOI: 10.19455/j.mcgy.20180412.
WANG C C, LIU Y Q.Detecting surface defects of particleboard based on tracking-learning detection technology[J].China Wood Industry, 2018, 32(4): 44-47.
[16] 吕凭乐, 赵丹新, 孙胜利,等.天基微弱运动点目标检测研究综述[J].红外, 2017, 38(1): 1-5.DOI: 10.3969/j.issn.1672-8785.2017.01.001.
LYU P Y, ZHAO D X, SUN S L, et al.Overview of space weak moving point target detection[J].Infrared, 2017, 38(1): 1-5.
[17] DENG L Z, ZHU H, ZHOU Q, et al.Adaptive top-hat filter based on quantum genetic algorithm for infrared small target detection[J].Multimedia Tools and Applications, 2018, 77(9): 10539-10551.DOI:10.1007/s11042-017-4592-2.
[18] WANG X L, JIANG B, LU N Y.Adaptive relevant vector machine based RUL prediction under uncertain conditions[J].ISA Transactions, 2019(87): 217-224.DOI:10.1016/j.isatra.2018.11.024.

备注/Memo

备注/Memo:
收稿日期:2019-01-12 修回日期:2019-04-22
基金项目:国家自然科学基金(31770603)。
作者简介:刘瑞明,男,副教授,研究方向为模式识别与智能系统。E-mail:liurm@hhit.edu.cn
更新日期/Last Update: 2019-09-10