基于AlexNet优化的板材材种识别方法

doi:10.11707/j.1001-7488.20220316

Abstract

Abstract:

Objective: Based on the cell characteristics of the processed image of cell on wood end faces, the search for corresponding machine learning methods can significantly increase the accuracy of identifying and realize efficient utilization and processing of wood, with the aim to provide an important basis for the identification and protection of rare wood species. Method: Using the cells on end faces of five kinds of wood(Abies nephrolepis, Larix olgensis, Picea jezoensis var.microsperma, Liriodendron chinense, Magnolia officinalis subsp. biloba) as research samples, a variety of differential images are extracted as data sets. Feature information is extracted by image processing, and support vector machine classifier(SVM) and AlexNet neural network are used for classification and recognition. According to the difference of wood end face cell distinction, BN (batch normalization) algorithm is added to the AlexNet neural network architecture for optimization, and a more efficient wood recognition method is designed. Result: The 29 680 enhanced images are divided into a 7∶3 ratio and saved in the training set and the test set folders. The test samples are labeled and put into the same folder, then conduct an overall batch test of the three classification algorithms. The results show that the overall recognition accuracy of the test set of the support vector machine classifier is 84.67%, the overall recognition accuracy of the test set of the AlexNet neural network is 88.76%, the overall recognition accuracy of the AlexNet neural network based on the BN algorithm is 91.15%. It can be seen that the recognition accuracy of the AlexNet neural network based on the BN algorithm is better. Conclusion: When the sample size is sufficient, the classification effect of the AlexNet neural network on images of cell on wood end faces is significantly better than that of the SVM classifier. The optimized AlexNet neural network based on the BN algorithm architecture is more sensitive to the linear features of the image, retains the fitting optimization of the AlexNet neural network and speeds up the optimization rate, which can effectively improve the classification accuracy and achieve high-precision wood classification.

Key words: wood species identification, machine learning, AlexNet, SVM

CLC Number:

S781.1

Nannan Yang,Yan Bai,Suyi Jiang,Chunmei Yang,Kaihong Xu. Recognition Method of Plate and Wood Based on ALexNet Optimaization[J]. Scientia Silvae Sinicae, 2022, 58(3): 149-158.

Figures/Tables 16

Table 1

Fig.1

Fig.2

Fig.3

Fig.4

Table 2

Fig.5

Fig.6

Fig.7

Table 3

Fig.8

Table 4

Table 5

Fig.9

Table 6

Fig.10

References 0

	白雪冰, 王克奇, 王辉. 基于灰度共生矩阵的木材纹理分类方法的研究. 哈尔滨工业大学学报, 2005, 37 (12): 1667- 1670. doi: 10.3321/j.issn:0367-6234.2005.12.022
	Bai X B , Wang K Q , Wang H . Research on the classification of wood texture based on gray level co-occurrence matrix. Journal of Harbin Institute of Technology, 2005, 37 (12): 1667- 1670. doi: 10.3321/j.issn:0367-6234.2005.12.022
	高洁, 任洪娥, 马岩. 基于细胞数字特征的板材材种识别技术. 林业机械与木工设备, 2006, 34 (1): 39- 41. doi: 10.3969/j.issn.2095-2953.2006.01.013
	Gao J , Ren H E , Ma Y . The techniquel on distinguishing wood base on the digital characteristic of cell. Forestry Machinery & Woodworking Equipment, 2006, 34 (1): 39- 41. doi: 10.3969/j.issn.2095-2953.2006.01.013
	耿帅, 王希常. 一种基于中值滤波的图像去噪算法. 计算机与现代化, 2011, (11): 90- 92. doi: 10.3969/j.issn.1006-2475.2011.11.024
	Geng S , Wang X C . An image denoising algorithm based on median filter. Computer and Modernization, 2011, (11): 90- 92. doi: 10.3969/j.issn.1006-2475.2011.11.024
	简丽琼. 基于二维直方图均衡化的图像增强算法. 信息与电脑(理论版), 2015, (22): 49- 54.
	Jian L Q . Image enhancement algorithm based on two-dimensional histogram equalization. China Computer & Communication(Theoretical Edition), 2015, (22): 49- 54.
	孔祥溢, 王任直. 人工智能及在医疗领域的应用. 医学信息学杂志, 2016, 37 (11): 2- 5.
	Kong X Y , Wang R Z . Artificial intelligence and its application in medical field. Journal of Medical Informatics, 2016, 37 (11): 2- 5.
	李好. 废弃木质材料回收利用方式的分析. 林产工业, 2012, 39 (4): 48- 50. doi: 10.3969/j.issn.1001-5299.2012.04.013
	Li H . Analysis of recycling measures of waste wood-based materials. China Forest Products Industry, 2012, 39 (4): 48- 50. doi: 10.3969/j.issn.1001-5299.2012.04.013
	刘栋, 李素, 曹志冬. 深度学习及其在图像物体分类与检测中的应用综述. 计算机科学, 2016, 43 (12): 13- 23. doi: 10.11896/j.issn.1002-137X.2016.12.003
	Liu D , Li S , Cao Z D . State-of-the-art on deep learning and its application in image object classification and detection. Computer Science, 2016, 43 (12): 13- 23. doi: 10.11896/j.issn.1002-137X.2016.12.003
	林泽文. 深度学习及其在图像分类识别中的运用探讨. 信息与电脑(理论版), 2018, (12): 146- 147.
	Lin Z W . Deep learning and its application in image classification and recognition. China Computer & Communication(Theory Edition), 2018, (12): 146- 147.
	吕鸿蒙, 赵地, 迟学斌. 基于增强AlexNet的深度学习的阿尔茨海默病的早期诊断. 计算机科学, 2017, 44 (S1): 50- 60.
	Lü H M , Zhao D , Chi X B . Deep learning for early diagnosis of Alzheimer's disease based on intensive AlexNet. Computer Science, 2017, 44 (S1): 50- 60.
	孙芳利, ProsperNayebare Kokwara, 吴华平, 等. 木竹材防腐技术研究概述. 林业工程学报, 2017, 2 (5): 1- 8.
	Sun F L , Kakwara P , Wu H P , et al. A review on the development of wood and bamboo preservation. Journal of Forestry Engineering, 2017, 2 (5): 1- 8.
	孙娅楠, 林文斌. 梯度下降法在机器学习中的应用. 苏州科技大学学报(自然科学版), 2018, 35 (2): 26- 31.
	Sun Y N , Lin W B . Application of gradient descent method in machine learning. Journal of Suzhou University of Science and Technology(Natural Science Edition), 2018, 35 (2): 26- 31.
	于海鹏, 刘一星, 刘镇波. 基于图像纹理特征的木材树种识别. 林业科学, 2007, 43 (4): 77- 81. 77-81, 146
	Yu H P , Liu Y X , Liu Z B . Wood species retrieval on base of image textural features. Scientia Silvae Sinicae, 2007, 43 (4): 77- 81. 77-81, 146
	曾志强, 吴群, 廖备水, 等. 一种基于核SMOTE的非平衡数据集分类方法. 电子学报, 2019, 37 (11): 2489- 2495.
	Zeng Z Q , Wu Q , Liao B S , et al. A classification method for imbalance data set based on kernel SMOTE. Acta Electronica Sinica, 2009, 37 (11): 2489- 2495.
	Cortes C , Vapnik V . Support-vector networks. Machine Learning, 1995, 20 (3): 273- 297.
	Ioffe S, Szegedy C. 2015. Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv: 1502. 03167[cs. LG]. https://arxiv.org/abs/1502.03167.
	Jia Y, Shelhamer E, Donahue J, et al. 2014. Caffe: convolutional architecture for fast feature embedding, ACM International Conference on Multimedia, 675-678.
	Krizhevsky A , Sutskever I , Hinton G E . ImageNet classification with deep convolutional neural networks. Communications of the ACM, 2017, 60 (6): 84- 90. doi: 10.1145/3065386
	Marr D , Hildreth E . Theory of edge detection. Proceedings of the Royal Society of London Series B, Biological Sciences, 1980, 207 (1167): 187- 217.
	Muntarina K , Shorif S B , Uddin M S . Notes on edge detection approaches. Evolving Systems, 2021, 13 (1): 169- 182.
	Zhao P , Dou G , Chen G S . Wood species identification using feature-level fusion scheme. Optik-International Journal of Light and Electron Optics, 2014, 125 (3): 1144- 1148. doi: 10.1016/j.ijleo.2013.07.124

序号No.	树种Tree species	科Aceae	属Genus	产地Producing area
1	臭冷杉A. nephrolepis	松科Pinaceae	冷杉属Abies	黑龙江省Heilongjiang Province
2	长白落叶松L. olgensis	松科Pinaceae	落叶松属Larix	吉林省长白山Changbai Mountain，Jilin Province
3	鱼鳞云杉P. jezoensis var. microsperma	松科Pinaceae	云杉属Picea	黑龙江省Heilongjiang Province
4	鹅掌楸L. chinense	木兰科Magnoliaceae	鹅掌楸属Liriodendron	湖南省Hunan Province
5	凹叶厚朴M. officinalis subsp. biloba	木兰科Magnoliaceae	木兰属Magnolia	安徽省Anhui Province

	臭冷杉 A. nephrolepis	长白落叶松 L. olgensis	鱼鳞云杉 P. jezoensis var. microsperma	鹅掌楸 L. chinense	凹叶厚朴 M. officinalis subsp. biloba
臭冷杉A. nephrolepis	0.867	0.000	0.133	0.000	0.000
长白落叶松L. olgensis	0.000	0.883	0.117	0.000	0.000
鱼鳞云杉 P. jezoensis var. microsperma	0.240	0.021	0.739	0.000	0.000
鹅掌楸L.chinense	0.000	0.000	0.000	0.909	0.091
凹叶厚朴 M. officinalis subsp. biloba	0.000	0.000	0.000	0.107	0.893

	第1次1st	第2次2nd	第3次3rd	第4次4th	第5次5th	平均Mean
识别准确率 Recognition accuracy(%)	87.92	89.06	89.42	88.63	88.75	88.76

参数项Parameter item	参数取值Parameter value
batch_size	32
num_output	5
test_iter	100
test_interval	1 000
base_lr	0.000 1
max_iter	10 000

	臭冷杉 A. nephrolepis	长白落叶松 L. olgensis	鱼鳞云杉 P. jezoensis var. microsperma	鹅掌楸 L.chinense	凹叶厚朴 M. officinalis subsp. biloba
臭冷杉A. nephrolepis	89.42	1.04	9.53	0.01	0.00
长白落叶松L. olgensis	1.77	92.39	5.84	0.00	0.00
鱼鳞云杉 P. jezoensis var. microsperma	10.65	2.81	86.54	0.00	0.00
鹅掌楸L.chinense	0.00	0.00	0.00	93.66	6.34
凹叶厚朴 M. officinalis subsp. biloba	0.00	0.00	0.00	4.93	95.07

Recognition Method of Plate and Wood Based on ALexNet Optimaization

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 16

References 0

Related Articles 12

Recommended Articles

Metrics

Comments

[1]	Jiaqi You,Mingze Li,Wenyi Fan,Ying Quan,Bin Wang,Zhukun Mo,Zixiao Zhu. Stand Type Identification Based on Hyperspectral and LiDAR Data [J]. Scientia Silvae Sinicae, 2021, 57(5): 119-129.
[2]	Fang Chen,Xianbao Cheng,Anmin Huang,Xueshun Wang. Elastic Modulus Prediction of Cunninghamia lanceolata Based on Artificial Bee Colony Algorithm SVM and NIR [J]. Scientia Silvae Sinicae, 2021, 57(1): 161-168.
[3]	Mao Xuegang, Zhu Liang, Liu Yitong, Yao Yao, Fan Wenyi. Object-Oriented Classification for Tree Species Based on High Spatial Resolution Images and Spaceborne Polarimetric SAR Cooperation with Feature [J]. Scientia Silvae Sinicae, 2019, 55(9): 92-102.
[4]	Mao Xuegang, Xing Xiuli, Li Jiarui, Tan Liangquan, Fan Wenyi. Object-Oriented Recognition of Forest Gap Based on Aerial Orthophoto [J]. Scientia Silvae Sinicae, 2019, 55(2): 87-96.
[5]	Zhang Ying, Zhang Xiaoli, Li Hongzhi, Li Liangcai. A Comparison of Landcover Classification Based on the Improved Transformed Divergence Analysis [J]. Scientia Silvae Sinicae, 2018, 54(8): 88-98.
[6]	Tan Nian, Wang Xueshun, Huang Anmin, Wang Chen. Wood Density Prediction of Cunninghamia lanceolata Based on Gray Wolf Algorithm SVM and NIR [J]. Scientia Silvae Sinicae, 2018, 54(12): 137-141.
[7]	Cao Zhengbin, Liu Xiaoping, Du Guangyue, Chu Xin, Liu Dawei, Zhou Yucheng. Temperature Prediction Model of Heat Storage of Wooden Flooring for Ground with Heating System Based on TWSVM and Fuzzy [J]. Scientia Silvae Sinicae, 2018, 54(11): 45-52.
[8]	Mao Xuegang, Chen Wenqu, Wei Jingyu, Fan Wenyi. Effect and Evaluation of Segmentation Scale on Object-Based Forest Species Classification [J]. Scientia Silvae Sinicae, 2017, 53(12): 73-83.
[9]	Mao Xuegang, Hou Jiyu, Fan Wenyi. Object-Based Automatic Recognition for Forest Gaps Using Aerial Image and LiDAR Data [J]. Scientia Silvae Sinicae, 2017, 53(11): 94-103.
[10]	Xu Huadong, Wang Lihai, Song Shiquan. Two Dimensional Image Construction of Ultrasonic Wave for [J]. Scientia Silvae Sinicae, 2014, 50(8): 126-130.
[11]	Zhang Yan, Song Kuiyan, Tong Da. Prediction of Mature Wood Anatomical Properties of Pinus banksiana Plantation Based on Support Vector Machines (SVM) [J]. Scientia Silvae Sinicae, 2013, 49(9): 119-125.
[12]	Huang Yuxia;Xu Dongbei;Pu Su. Application of SVM Method on the Prediction of Forest Fire Danger [J]. Scientia Silvae Sinicae, 2007, 43(10): 77-82.