Wood Species Recognition with Small Data: A Deep Learning Approach

Wood species recognition is an important work in the wood trade and wood commercial activities. Although many recognition methods were presented in recent years, the existing wood species recognition methods mainly use shallow recognition models with low accuracy and are still unsatisfying for many real-world applications. Besides, their generalization ability is not strong. In this paper, a novel deep-learning-based wood species recognition method was proposed, which improved the accuracy and generalization greatly. Themethod uses 20X amplifying glass to acquire wood images, extracts the image features with ResNet50 neural network, refines the features with linear discriminant analysis (LDA), and recognizes the wood species with a KNN classifier. Our data was small, but we adopted transfer learning to improve our method. About 3000 wood images were used in our wood species recognition experiments and our method was executed in 25 rare wood species and the results showed our method had better generalization performance and accuracy. Compared with traditional deep learning our results were obtained from a small amount of data, which just confirmed the effectiveness of our method.

[1]  J. Sugiyama,et al.  Neural network for classification of Chinese zither panel wood via near-infrared spectroscopy , 2019, BioResources.

[2]  Peng Zhao,et al.  Wood species identification using feature-level fusion scheme , 2014 .

[3]  Fei Dai,et al.  Application of Deep Learning in Wood Classification , 2019, 2019 IEEE International Conference on Computer Science and Educational Informatization (CSEI).

[4]  Alex C. Wiedenhoeft,et al.  Classification of CITES-listed and other neotropical Meliaceae wood images using convolutional neural networks , 2018, Plant Methods.

[5]  Anisha M. Lal,et al.  Morphology-Combined Gradient Boosting for Recognizing Targets in SAR Images , 2019 .

[6]  Zhi-Hong Mao,et al.  Candidates for Synergies: Linear Discriminants versus Principal Components , 2014, Comput. Intell. Neurosci..

[7]  Jun Wang,et al.  Discrimination of wood borers infested Platycladus orientalis trunks using quartz crystal microbalance gas sensor array , 2020 .

[8]  Rubiyah Yusof,et al.  Statistical feature extraction method for wood species recognition system , 2017, European Journal of Wood and Wood Products.

[10]  C. Fuentealba,et al.  Wood products identification by internal characteristics readings , 2004, 2004 IEEE International Conference on Industrial Technology, 2004. IEEE ICIT '04..

[11]  Georgios Karalis.  Decision Trees and Applications. , 2020, Advances in experimental medicine and biology.

[12]  Avinash C. Kak,et al.  PCA versus LDA , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  C. Kalaitzidis,et al.  Linear Discriminant Analysis of spectral measurements for discrimination between healthy and diseased trees of Olea europaea L. artificially infected by Fomitiporia mediterranea , 2020 .

[14]  Tawani Lorena Naide,et al.  Discrimination of wood and charcoal from six caatinga species by near-infrared spectroscopy , 2018 .

[15]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[16]  Flavio Ruffinatto,et al.  REVIEW OF MACROSCOPIC FEATURES FOR HARDWOOD AND SOFTWOOD IDENTIFICATION AND A PROPOSAL FOR A NEW CHARACTER LIST , 2015 .

[17]  Nikolaos Grammalidis,et al.  Wood species recognition through multidimensional texture analysis , 2018, Comput. Electron. Agric..

[18]  Xiaoguang Yue,et al.  The Wood Slice Cell Image Identification Algorithm Based on Singular Value Decomposition , 2015 .

[19]  Silvana Nisgoski,et al.  An automatic recognition system of Brazilian flora species based on textural features of macroscopic images of wood , 2020, Wood Science and Technology.

[20]  D. Chicco,et al.  The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation , 2020, BMC Genomics.

[21]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Xiaoling Xia,et al.  Inception-v3 for flower classification , 2017, 2017 2nd International Conference on Image, Vision and Computing (ICIVC).

[23]  Mohamed Limam,et al.  Performance Evaluation of One‐Class Classification‐based Control Charts through an Industrial Application , 2013, Qual. Reliab. Eng. Int..

[24]  R. A. Pramunendar,et al.  Wood Types Classification using Back-Propagation Neural Network based on Genetic Algorithm with Gray Level Co-occurrence Matrix for Features Extraction , 2019 .

[25]  Axel Wismüller,et al.  Deep transfer learning for characterizing chondrocyte patterns in phase contrast X-Ray computed tomography images of the human patellar cartilage , 2018, Comput. Biol. Medicine.

[26]  Sameer Singh,et al.  Nearest-neighbour classifiers in natural scene analysis , 2001, Pattern Recognit..

[27]  Hang-jun Wang,et al.  Wood Recognition Using Image Texture Features , 2013, PloS one.

[28]  N. Isabel,et al.  Chemotyping and identification of protected Dalbergiatimber using gas chromatography quadrupole time of flight mass spectrometry. , 2019, Journal of chromatography. A.

[29]  Peng Zhao,et al.  Robust wood species recognition using variable color information , 2013 .

[30]  D. Lee,et al.  Linear Discriminant Analysis for Signatures , 2010, IEEE Transactions on Neural Networks.