Abstract
Sawn timber is an important component material in furniture manufacturing, decoration, construction and other industries. The mechanical properties, surface colors, textures, use and other properties of sawn timber possesed by different tree species are different. In order to meet the needs of reasonable timber use and product quality of sawn timber products, sawn timber must be identified according to tree species to ensure the best use of materials. In this study, an optimized convolution neural network was proposed to process sawn timber image data to identify the tree species of the sawn timber. The spatial pyramid pooling and attention mechanism were used to improve the convolution layer of ResNet101 to extract the feature vector of sawn timber images. The optimized ResNet (simply called “AM-SPPResNet”) was used to identify the sawn timber image, and the basic recognition model was obtained. Then, the weight parameters of the feature extraction layer of the basic model were frozen, the full connection layer was removed, and using support vector machine (SVM) and XGBoost classifier which were commonly used in machine learning to train and learn the 21 × 1024 dimension feature vectors extracted by feature extraction layer. Through a number of comparative experiments, it is found that the prediction model using linear function as the kernel function of support vector machine learning the feature vectors extracted from the improved convolution layer performed best, and the F1 score and overall accuracy of all kinds of samples were above 99%. Compared with the traditional methods, the accuracy was improved by up to 12%.
Highlights
Sawn timber refers to a type of solid wood board whose size meets the industry standard, and its specifications are unified after a series of processing procedures, such as peeling, cutting, and polishing
This paper first used a linear classifier as the full connection layer of the network, and used the learning rate of 1e−4 and batch-size of 16 and Adam optimizer to train the different convolutional neural network models
It can be found that the identification performance of AM-SPPResNet, which introduces attention mechanism and spatial pyramid pooling strategy, has been greatly improved compared with the original network, and the accuracy rate has been improved by 5.3%, the range of F1 score value between different types was only 0.015, while the range of F1 score value of the original network was 0.108, it shows that the balance of AM-SPPResNet identification performance has been greatly improved
Summary
Sawn timber refers to a type of solid wood board whose size meets the industry standard, and its specifications are unified after a series of processing procedures, such as peeling, cutting, and polishing. Sawn timber has the advantages of small deformation, crack resistance, high bonding strength, diverse colors, and easy splicing. It is a common green and sustainable multifunctional material used in the furniture, decoration, and construction industries. In order to meet the high-quality demands of human beings for timber products under the condition of increasingly tense shortage of log resources, meet the needs of reasonable use of timber products and at the same time ensure the quality of products, it is necessary to identify the sawn timber according to tree species, so as to ensure the best use of materials
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
