Wood Defects Recognition Research Articles

Review of the current state of application of wood defect recognition technology

Wood utilisation is an important factor affecting production costs, but the combined utilisation rate of wood is generally only 50 to 70%. During the production process, the rejection scheme of wood defects is one of the most important factors affecting the wood yield. This paper provides an overview of the main wood defects affecting wood quality, introduces techniques for detecting and identifying wood defects using different technologies, highlights the more widely used image recognition-based wood surface defect identification methods, and presents three advanced wood defect detection and identification equipment. In view of the relatively fixed wood defect recognition requirements in wood processing production, it is proposed that wood defect recognition technology should be further developed toward deep learning to improve the accuracy and efficiency of wood defect recognition.

RETRACTED: Artificial neural network for defect detection in CT images of wood

• New image processing technique is proposed to increase the volume of sets. • The recognition rate of the network was found to be 98.7% (100 epochs). • Results help to automate and speed up the defect recognition in CT images. • Results can be useful in designing software applications for CT scanners. The timely detection of defects in wood helps optimize operation of sawmills and find effective log processing solutions. This paper aims to develop effective method for automated wood defect detection and recognition in CT images using a multiple-layer convolutional neural network and a reinforcement learning strategy. The data augmentation technique was proposed to increase the volume of training, validation, and test sets. The network can achieve sufficient accuracy up to 98.7% at a total set of 500 images. The study shows a direct non-linear relationship between the dimension of the training set and the recognition rate. The results of calculating performance metrics for the developed method indicate the high accuracy of the ANN prediction model. The study results will be useful in designing software applications for industrial and laboratory CT scanners that are used in lumber production and R&D centers. The proposal can be improved to perform wood defect detection and recognition in color and 3D CT images of logs.

Wood Defect Detection Based on Depth Extreme Learning Machine

The deep learning feature extraction method and extreme learning machine (ELM) classification method are combined to establish a depth extreme learning machine model for wood image defect detection. The convolution neural network (CNN) algorithm alone tends to provide inaccurate defect locations, incomplete defect contour and boundary information, and inaccurate recognition of defect types. The nonsubsampled shearlet transform (NSST) is used here to preprocess the wood images, which reduces the complexity and computation of the image processing. CNN is then applied to manage the deep algorithm design of the wood images. The simple linear iterative clustering algorithm is used to improve the initial model; the obtained image features are used as ELM classification inputs. ELM has faster training speed and stronger generalization ability than other similar neural networks, but the random selection of input weights and thresholds degrades the classification accuracy. A genetic algorithm is used here to optimize the initial parameters of the ELM to stabilize the network classification performance. The depth extreme learning machine can extract high-level abstract information from the data, does not require iterative adjustment of the network weights, has high calculation efficiency, and allows CNN to effectively extract the wood defect contour. The distributed input data feature is automatically expressed in layer form by deep learning pre-training. The wood defect recognition accuracy reached 96.72% in a test time of only 187 ms.

Study on the identification of the wood surface defects based on texture features

Wood surface defect detection technology is the intersect multidisciplinary technology between computer vision and pattern recognition, which has a high value and is widely used in the field of timber production and deep processing. This paper mainly takes three common defects such as dead knots, poles and living knots of wood for the study, it deeply researches on image segmentation and pattern recognition methods of wood. To ensure the reliability of the results of wood defect recognition, the selection of characteristic values is the crucial aspects of pattern recognition. Haipeng Yu extracted texture of the wood based on GLCM, Xuebing Bai also classified texture of the wood based on GLCM. In addition, researchers used wavelets, Markov random, fractal, local binary pattern and histogram to make some useful attempts with the study of wood texture feature extraction. The above study only applied a texture analysis method. As the diversity and complexity of the wood surface defect images, the success rate of using a certain type of feature detection method is still less than ideal from the application point of view. The study proposes a hybrid wood surface texture features based on defect detection method, which combines the integration of Tamura texture and GLCM method advantages of these two methods, so that the accuracy and robustness of the algorithm are effectively protected.

Wood Defects Recognition Based on Fuzzy BP Neural Network

Firstly, we applied the X-ray non-destructive testing technology to detect wood defects for getting the images. After graying the images, we calculated their GLCMS(Gray Level Co-occurrence Matrixes), then we normalized GLCMS to obtain the joint probabilities of GLCMS. The feature vectors of images, which included 13 eigenvalues of images were calculated and extracted by the joint probability of GLCMS. The fuzzy BP neural network(abbreviated as FBP) was designed by combining fuzzy mathematics and BP neural network . And the FBP neural network was regarded as the membership function of feature vectors, the outputs of the network was regarded as the degree of membership to the feature vectors in each category. We use the maximum degree of membership method for the pattern recognition of feature vectors, so the automatic identification and classification for feature vectors were achieved , and then the automatic identification of wood defects was realized. By simulated study and training many times, the results shown that the average recognition success rate of the network was more than 90%, and some FBP networks had an extremely high recognition success rate to training samples and test samples.

Quantitative Identification of Defects in Lumber Based on Modal Frequencies and Artificial Neural Network

This study preliminarily discussed a new method to identify the location and size of internal wood defects using experimental modal analysis (EMA) and artificial neural network. The different defect sizes and locations were simulated by removing mass from intact wood specimens. At room temperature in the laboratory, free vibration testing was conducted to generate the frequency response functions (FRF) of intact and defective Korean Pine (Pinus koraiensis) wood specimens using fast Fourier transform (FFT) analysis system. The first three orders intrinsic frequencies were captured by picking up the location of each order peak of FRF curves. Then, two identification indexes developed by previous research were constructed based on these intrinsic frequencies, and they were used as input parameters to build the networks for localization and size determination of wood defects respectively. These two artificial neural networks were trained and tested for wood defects recognition. The research results showed that: (1) the intrinsic frequencies of defective wood were lower than those of intact wood; and (2) the constructed two identification indexes were capable to effectively detect the location and size of wood defects, which were more sensitive to large size defects than small size defects.

Applying multifractal spectrum combined with fractal discrete Brownian motion model to wood defects recognition

Wood nondestructive testing technology is a new and multidisciplinary industry scientific research. It has attained fast development and achievements in recent years. X-ray computed tomography (CT) scanning technology is a kind of wood nondestructive testing technology in practice. CT scanning technology has been applied to the detection of internal defects in the logs for the purpose of obtaining prior information, which can be used to reach better wood sawing decision. Fractal geometry and its extension multifractal are used for describing, modeling, analyzing, and processing of different complex shapes and images. A method in CT image edge detection using multifractal theory combined with fractal Brownian motion is applied in the paper. First, its multifractal spectrum is estimated. Then, different types of pixels are classified by the spectrum; they are smoothing edge points and singular edge points. From the images processed by multifractal spectrum theory and compared with each image by different spectrum values, it can be seen that the larger the range of threshold is set, the more exact the edge can be detected. The paper provides a new method to recognize the defect information and to saw it in the condition of nondestructive wood.

Effective semi-supervised nonlinear dimensionality reduction for wood defects recognition

Dimensionality reduction is an important preprocessing step in high-dimensional data analysis without losing intrinsic information. The problem of semi-supervised nonlinear dimensionality reduction called KNDR is considered for wood defects recognition. In this setting, domain knowledge in forms of pairs constraints are used to specify whether pairs of instances belong to the same class or different classes. KNDR can project the data onto a set of 'useful' features and preserve the structure of labeled and unlabeled data as well as the constraints defined in the embedding space, under which the projections of the original data can be effectively partitioned from each other. We demonstrate the practical usefulness of KNDR for data visualization and wood defects recognition through extensive experiments. Experimental results show it achieves similar or even higher performances than some existing methods.

Pattern recognition and size determination of internal wood defects based on wavelet neural networks

This paper presented the results of a preliminary study on detecting internal wood defects using an ultrasound technique coupled with wavelet transform and artificial neural networks analysis. At room temperature in the laboratory, the type and size of the wood defects in 275 Elm specimens were detected using a RSM-SY5 ultrasonic instrument. The original signals of the Elm specimens were decomposed using wavelet packets, the energy variation of each node in the fifth layer was calculated, and back-propagation artificial neural networks (BP ANN) were trained and employed for wood defect recognition. The energy variation caused by wood defects mostly depends on the degree of the defect's deterioration (i.e., the more serious the wood defect's deterioration, the larger the energy variation). By comparing the energy variation of all 32 node signals in the fifth layer wavelet packet, the variation of the node (5,0) was the largest and contained the most defect information. The node (5,0) was used as the input in back-propagation artificial neural networks in order to detect the type of defects. The accuracy rate for Elm specimens was at least 90%. The same method was used to test the size and position of the hole-defects in Elm specimens with an accuracy rate of at least 80%.

Contribution of fuzzy reasoning method to knowledge integration in a defect recognition system

This article presents the improvement of a defect recognition system for wooden boards by using knowledge integration from two expert fields. These two kinds of knowledge to integrate respectively concern wood expertise and industrial vision expertise. First of all, extraction, modelling and integration of knowledge use the Natural Language Information Analysis method (NIAM) to be formalized from their natural language expression. Then, to improve a classical industrial vision system , we propose to use the resulting symbolic model of knowledge to partially build a numeric model of wood defect recognition. This model is created according to a tree structure where each inference engine is a fuzzy rule based inference system. The expert knowledge model previously obtained is used to configure each node of the resulting hierarchical structure. The practical results we obtained in industrial conditions show the efficiency of such an approach.

Robust Spatial Autoregressive Modeling for Hardwood Log Inspection

We explore the application of a stochastic texture modeling method toward a machine vision system for log inspection in the forest products industry. This machine vision system uses computerized tomography (CT) imaging to locate and identify internal defects in hardwood logs. The application of CT to such industrial vision problems requires efficient and robust image analysis methods. This paper addresses one particular aspect of the problem of creating such a computer vision system, namely, the use of image texture modeling for wood defect recognition. In particular, we contribute the first application of spatial autoregressive (SAR) modeling to wood-grain texture analysis of CT images of hardwood logs. Thereto a circularly shifted correlation approach is developed to discriminate the circular texture patterns on the cross-sectional CT images of logs. A robust algorithm for parameter estimation is applied to obtain model parameters associated with individual defects occurring inside a log. Based on the estimated model features, a simple minimum distance correlation-classifier is constructed which classifies an unknown defect into one of the prototypical defects. Experimental results from the proposed method, applied to CT images from different red oak wood, are given and show the efficacy of our approach.

Automatic Image Recognition and Optimal Control in the Woodworking Industry

Abstract The problem of the automatic wood defects recognition is analyzed. The theoretical background of the defect detection best systems, namely the vision systems, is Introduced. Recognition methods useful in the woodworking industry are described. The methods solve the problem of model matching to the image and problem of the wood defects recognition. The latter are based on the spatial tone dependence matrices proposed by Haralick and applied in wood by Conners and co-workers. Further the consequences of the automatic defects recognition on optimal sawing of dimensional wood materials is studied in the paper. The methods for optimal edging and sawing of boards are presented.