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. The method 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.
The objective of this study was to develop a computer-aided method to quantify the obvious degree of growth ring boundaries of softwood species, based on data analysis with some image processing technologies. For this purpose, a 5× magnified cross-section color microimage of softwood was cropped into 20 sub-images, and then every image was binarized as a gray image according to an automatic threshold value. After that, the number of black pixels in the gray image was counted row by row and the number of black pixels was binarized to 0 or 100. Finally, a transition band from earlywood to latewood on the sub-image was identified. If everything goes as planned, the growth ring boundaries of the sub-image would be distinct. Otherwise would be indistinct or absent. If more than 50% sub-images are distinct, with the majority voting method, the growth ring boundaries of softwood would be distinct, otherwise would be indistinct or absent. The proposed method has been visualized as a growth-ring-boundary detecting system based on the .NET Framework. A sample of 100 micro-images (see S1 Fig via https://github.com/senly2019/Lin-Qizhao/) of softwood crosssections were selected for evaluation purposes. In short, this detecting system computes the obvious degree of growth ring boundaries of softwood species by image processing involving image importing, image cropping, image reading, image grayscale, image binarization, data analysis. The results showed that the method used avoided mistakes made by the manual comparison method of identifying the presence of growth ring boundaries, and it has a high accuracy of 98%.
In order to achieve rapid acquisition, identification and measurement of the average ray height of softwood based on tangential section photographs, a new method is proposed. Firstly, labels the digital image of the softwood tangential section with the 100 magnification, that is, mark the rays and scales on the image, and establish the dataset; Secondly, the dataset is randomly divided into training set and validation set. YOLOv5s is used for model training to obtain the best target recognition model of rays and scale. The experimental results show that the model trained with YOLOv5s can achieve 93.5% accuracy, 95.6% recall and 96.7% average accuracy in the validation set; Thirdly, using the YOLOv5s trained model, a visual program for automatically calculating the ray height and obtaining the ray characteristics of softwood is designed, which lowered the threshold for wood identification workers to use such software.
In order to achieve rapid acquisition, identification and measurement of the average ray height of softwood based on tangential section photographs, a new method is proposed. Firstly, labels the digital image of the softwood tangential section with the 100 magnification, that is, mark the rays and scales on the image, and establish the dataset; Secondly, the dataset is randomly divided into training set and validation set. YOLOv5s is used for model training to obtain the best target recognition model of rays and scale. The experimental results show that the model trained with YOLOv5s can achieve 93.5% accuracy, 95.6% recall and 96.7% average accuracy in the validation set; Thirdly, using the YOLOv5s trained model, a visual program for automatically calculating the ray height and obtaining the ray characteristics of softwood is designed, which lowered the threshold for wood identification workers to use such software.
The tracking of the wood product is an important technology in the trade activity of rare plants. Normally, factories use Quick Response (QR) and Radio-Frequency Identification (RFID) to identify the individual wood product, but these technologies are not safe enough because they can be easily falsified. The traditional methods are hard to catch the detail of the slim wood texture from the wood product. In this study, a novel method is employed to resolve these problems using a biometric feature on the surface of the real wood product to distinguish the individual wood product. AKAZE is used to extract the keypoint of wood texture. A sub-area detection technique along with a serialization method is then developed to improve the rate of identification. The sub-area detection technique deals with picking out a sub-region in which there are enough AKAZE points as small as possible. The serialization method is also utilized to reduce the redundant process of feature extraction. The experimental results demonstrate that the values of accuracy, recall, and F 1 reach 0.98, 0.96, and 0.96, respectively. The match time that uses serialized function is reduced to 1/3 of which has no application in the original image. Validated results also reveal that our proposed methodology improve the robustness of the wood product identification, and it can be used in Wood Traceability System (WTS) with the blockchain to resolve the digital trust problem and the fast distinction issues of the real wood product.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.