基于深度学习的木材表面缺陷图像检测

陈献明; 王阿川; 王春艳

doi:10.3788/YJYXS20193409.0879

您当前的位置：

首页 >

文章列表页 >

基于深度学习的木材表面缺陷图像检测

图像处理 | 更新时间：2020-07-05

- 基于深度学习的木材表面缺陷图像检测
- Image detection of wood surface defects based on deep learning
- 液晶与显示 2019年34卷第9期页码：879-887
- 作者机构：
  
  1.东北林业大学信息与计算机工程学院, 黑龙江哈尔滨 150040
  2.扎兰屯职业学院, 内蒙古扎兰屯 162650
- 作者简介：
  
  [ "陈献明(1995—), 男, 黑龙江鹤岗人, 硕士研究生, 2017年于东北林业大学获得学士学位, 主要从事机器学习、图像处理等方面的研究, E-mail:chenxianming@nefu.edu.cn" ]
  [ "王阿川(1964—), 男, 四川达州人, 博士, 教授, 博士生导师, 2011年于东北林业大学获得博士学位, 主要从事模式识别、计算机视觉、3S技术等方面的研究。E-mail:wangcal1964@126.com" ]
- 基金信息：
  
  黑龙江省自然科学基金资助项目(C201414);Harbin Foundation for Good Leaders of Disciplines(2014RFXXJ040);the Natural Science Foundation of Heilongjiang Province(C201414)
- DOI：10.3788/YJYXS20193409.0879
  中图分类号： TP391.41
- 收稿日期：2019-04-12，
  
  录用日期：2019-6-19，
  
  纸质出版日期：2019-09-05
- 稿件说明：
移动端阅览
陈献明, 王阿川, 王春艳. 基于深度学习的木材表面缺陷图像检测[J]. 液晶与显示, 2019,34(9):879-887.

Xian-ming CHEN, A-chuan WANG, Chun-yan WANG. Image detection of wood surface defects based on deep learning[J]. Chinese journal of liquid crystals and displays, 2019, 34(9): 879-887.
陈献明, 王阿川, 王春艳. 基于深度学习的木材表面缺陷图像检测[J]. 液晶与显示, 2019,34(9):879-887. DOI： 10.3788/YJYXS20193409.0879.

Xian-ming CHEN, A-chuan WANG, Chun-yan WANG. Image detection of wood surface defects based on deep learning[J]. Chinese journal of liquid crystals and displays, 2019, 34(9): 879-887. DOI： 10.3788/YJYXS20193409.0879.

摘要

针对木材活节、死节、虫眼等缺陷图像的检测问题，本文提出了一种基于深度学习的木材缺陷图像检测方法。首先，通过对Faster-RCNN网络进行训练，得到了可以对木材缺陷定位和识别的检测模型；然后，应用NL-Means方法对图像进行去噪，通过线性滤波、调整对比度和亮度实现图像增强；再对图像进行二值化处理，根据像素值差异提取缺陷边缘特征点集，实现了对木材缺陷的精细分割；最后，对椭圆拟合方法进行了改进，实现了对木材缺陷边缘点集的椭圆拟合，提供了新的木材缺陷加工方案。实验结果表明，该算法具有较好的木材缺陷定位和分类能力，得到了较好的分割及拟合效果，可在缺陷修补这一环节减少约10%的木材填充量。

Abstract

Aiming at the defect image detection of live knots

dead knots

wormhole and so on

a wood defect image detection method based on depth learning was proposed. Firstly

by training Faster-RCNN network

a detection model for locating and recognizing wood defects is obtained. Secondly

the image is denoised by NL-Means method

and image enhancement is achieved by linear filtering

adjusting contrast and brightness. Thirdly

the image is processed by binarization

and the edge feature points of defects are extracted according to the difference of pixel values to realize wood defects fine segmentation. Finally

the ellipse fitting method is improved to realize the ellipse fitting of wood defect edge point set

and a new wood defect processing scheme is provided. The experimental results show that the algorithm has better wood defect location and classification ability

and gets better segmentation and fitting effect. The filling volume of wood can be reduced by about 10% in the process of defect repair.

关键词

Keywords

references

房友盼.基于图像识别的实木板材优选系统研究[D].南京: 南京林业大学, 2016.

FANG Y P. Research on optimization system of solid wood based on image recognition[D]. Nanjing: Nanjing Forestry University, 2016. (in Chinese)

赵鹏, 赵匀, 陈广胜.基于3D扫描技术的木材缺陷定量化分析[J].农业工程学报, 2017, 33(7):171-176.

ZHAO P, ZHAO Y, CHEN G S. Quantitative analysis of wood defect based on 3D scanning technique[J]. Transactions of the Chinese Society of Agricultural Engineering , 2017, 33(7): 171-176. (in Chinese)

李佳, 王阿川, 马欣然.基于多模型融合的木材表面缺陷图像快速识别[J].东北林业大学学报, 2014, 42(12):114-118, 140.

LI J, WANG A C, MA X R. Fast recognition for wood surface defect image based on multi-model fusion[J]. Journal of Northeast Forestry University , 2014, 42(12): 114-118, 140. (in Chinese)

张富文, 许清风, 张治宇, 等.钻入阻抗法检测木材缺陷[J].无损检测, 2016, 38(1):6-9, 74.

ZHANG F W, XU Q F, ZHANG Z Y, et al . Wood defect inspection by drilling resistance method[J]. Nondestructive Testing , 2016, 38(1): 6-9, 74. (in Chinese)

刘嘉新, 吴彤, 王克奇.基于C-V模型的木材缺陷重建图像特征提取[J].东北林业大学学报, 2015, 43(12):78-81.

LIU J X, WU T, WANG K Q. Feature extraction of wood defect reconstructed images based on the C-V model[J]. Journal of Northeast Forestry University , 2015, 43(12): 78-81. (in Chinese)

PACKIANATHER M S, DRAKE P R. Neural networks for classifying images of wood veneer. Part 2[J]. The International Journal of Advanced Manufacturing Technology , 2000, 16(6): 424-433.

熊伟俊, 杨绪兵, 云挺, 等.基于快速 l 1 算法和LBP算法的木材缺陷识别[J].数据采集与处理, 2017, 32(6):1223-1231.

XIONG W J, YANG X B, YUN T, et al . Wood defect recognition based on fast l 1 algorithm and LBP algorithm[J]. Journal of Data Acquisition and Processing , 2017, 32(6): 1223-1231. (in Chinese)

王阿川, 仇逊超.木材缺陷识别新方法——改进C-V模型与小波变换[J].计算机工程与应用, 2011, 47(8):211-214, 235.

WANG A C, QIU X C. New method of lumber recognition using improved C-V model and wavelet transform[J]. Computer Engineering and Applications , 2011, 47(8): 211-214, 235. (in Chinese)

REN S Q, HE K M, GIRSHICK R, et al . Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2017, 39(6): 1137-1149.

BUADES A, COLL B, MORELJ M. A non-local algorithm for image denoising[C]// Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition . San Diego, CA, USA: IEEE, 2005.

贾迪, 孟祥福, 孟琭, 等.RGB空间下结合高斯曼哈顿距离图的彩色图像边缘检测[J].电子学报, 2014, 42(2):257-263.

JIA D, MENG X F, MENG L, et al . Color image edge detection combining with gauss Manhattan distance map in RGB space[J]. Acta Electronica Sinica , 2014, 42(2): 257-263. (in Chinese)

GIRSHICK R. Fast R-CNN[C]// Proceedings of 2015 IEEE International Conference on Computer Vision . Santiago, Chile: IEEE, 2015: 1440-1448.

王健博, 杨航, 吴笑天.结合图像分割的改进导引滤波[J].液晶与显示, 2017, 32(5):380-386.

WANG J B, YANG H, WU X T. Segmentation based improved guided filter[J]. Chinese Journal of Liquid Crystals and Displays , 2017, 32(5): 380-386. (in Chinese)

吴晨睿, 张树有, 何再兴.基于梯度分类的复杂背景椭圆快速检测方法[J].浙江大学学报(工学版), 2018, 52(5):943-950.

WU C R, ZHANG S Y, HE Z X. Fast detection method for ellipse in complex background based on gradient grouping[J] . Journal of Zhejiang University ( Engineering Science ), 2018, 52(5): 943-950. (in Chinese)

熊风光, 李希, 韩燮.基于整体最小二乘的椭圆拟合方法[J].微电子学与计算机, 2017, 34(1):102-105.

XIONG F G, LI X, HANG X. A method of ellipse fitting based on total least squares[J]. Microelectronics & Computer , 2017, 34(1): 102-105. (in Chinese)

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.

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. (in Chinese)

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

RGB深度图像显著性目标检测方法设计

无人机绝对视觉定位的研究进展

基于自适应遮挡恢复与拓扑姿态双向感知的三维手部重建方法

基于Mamba-2的视频快照压缩成像重构方法

基于光流预测的三维光场显示透视关系校正