Search

Solid wood is renowned as a superior material for construction and furniture applications. However, characteristics such as dead knots, live knots, piths, and cracks are easily formed during timber’s growth and processing stages. These features and defects significantly undermine the mechanical characteristics of sawn timber, rendering it unsuitable for specific applications. This study introduces BDCS-YOLO (Bilateral Defect Cutting Strategy based on You Only Look Once), an artificial intelligence bilateral sawing strategy to advance the automation of timber processing. Grounded on a dual-sided image acquisition platform, BDCS-YOLO achieves a commendable mean average feature detection precision of 0.94 when evaluated on a meticulously curated dataset comprising 450 images. Furthermore, a dual-side processing optimization module is deployed to enhance the accuracy of defect detection bounding boxes and establish refined processing coordinates. This innovative approach yields a notable 12.3% increase in the volume yield of sawn timber compared to present production, signifying a substantial leap toward efficiently utilizing solid wood resources in the lumber processing industry.

In recent years, machine vision has played an important role in product surface quality detection. The promotion and use of this technology have largely avoided the subjectivity caused by human detection and improved detection efficiency and accuracy. Different from the image data commonly used in machine vision, point cloud can describe the spatial structure of an object, provide more information than image data, and have the ability to expand the data to build multi-dimensional data models. Due to the strong anti-interference ability of point cloud sensors and the high accuracy of three-dimensional positioning information point cloud, nondestructive testing technology based on point cloud has received more and more attention. This paper summarizes the research progress of product surface quality detection methods based on 3D point cloud in recent years. According to different data processing methods, the detection research is divided into five categories: based on point cloud contour, based on local geometric feature, based on template matching, based on multimodal point cloud, and based on deep learning. The five methods are reviewed and summarized respectively. Finally, the key problems of each detection method and the future trend of product surface quality detection technology based on 3D point cloud are discussed.

Most green plums need to be processed before consumption, and due to personal subjective factors, manual harvesting and sorting are difficult to achieve using standardized processing. Soluble solid content (SSC) of green plum was taken as the research object in this paper. Visible near-infrared (VIS-NIR) and shortwave near-infrared (SW-NIR) full-spectrum spectral information of green plums were collected, and the spectral data were corrected and pre-processed. Random forest algorithm based on induced random selection (IRS-RF) was proposed to screen four sets of characteristic wavebands. Bayesian optimization CatBoost model (BO-CatBoost) was constructed to predict SSC value of green plums. The experimental results showed that the preprocessing method of multiplicative scatter corrections (MSC) was obviously superior to Savitzky–Golay (S–G), the prediction effect of SSC based on VIS-NIR spectral waveband by partial least squares regression model (PLSR) was obviously superior to SW-NIR spectral waveband, MSC + IRS-RF was obviously superior to corresponding combination of correlation coefficient method (CCM), successive projections algorithm (SPA), competitive adaptive reweighted sampling (CARS), and random forest (RF). With the lowest dimensional selected feature waveband, the lowest VIS-NIR band group was only 53, and the SW-NIR band group was only 100. The model proposed in this paper based on MSC + IRS-RF + BO-CatBoost was superior to PLSR, XGBoost, and CatBoost in predicting SSC, with R2P of 0.957, which was 3.1% higher than the traditional PLSR.

In the open shop scheduling problem, resources and tasks are required to be allocated in an optimized manner, but when the arrival of tasks is dynamic, the problem becomes much more difficult. To solve large scale open shop scheduling problem with release dates, heuristic algorithms are more promising compared with metaheuristic algorithms. In this paper, a framework of general scheduling object is developed, under which open shop scheduling problem is described. Then, a complex scheduling network model for open shop scheduling problem is established, and the problem is transformed into reasonably arranging the node traversal order with the goal of traversing all nodes in the network as quickly as possible, on condition that each node has a traversal time and only disconnected nodes can be traversed simultaneously. Considering that network structural features and local time attributes of nodes can provide heuristic information, six single heuristic rules are raised and a novel complex network based dynamic rule selection approach is proposed to solve dynamic open shop problem by switching dynamically the scheduling rules based on real-time production status. Finally, two experiments are carried out and the experimental results show that the proposed heuristic rules have acceptable performance, and the proposed complex network based dynamic rule selection approach is feasible.

With the global population surge, the consumption of nonrenewable resources and pollution emissions have reached an alarming level. Engineered bamboo is widely used in construction, mechanical and electrical product packaging, and other industries. Its main damage is the material fracture caused by the expansion of initial cracks. In order to accurately detect the length of crack propagation, digital image correlation technology can be used for calculation. At present, the traditional interpolation method is still used in the reconstruction of engineered bamboo speckle images for digital correlation technology, and the performance is relatively lagging. Therefore, this paper proposes a super-resolution reconstruction method of engineering-bamboo speckle images based on an attention-dense residual network. In this study, the residual network is improved by removing the BN layer, using the L1 loss function, introducing the attention model, and designing an attention-intensive residual block. An image super-resolution model based on the attention-dense residual network is proposed. Finally, the objective evaluation indexes PSNR and SSIM and subjective evaluation index MOS were used to evaluate the performance of the model. The ADRN method was 29.19 dB, 0.938, and 3.19 points in PSNR, SSIM, and MOS values. Compared to the traditional BICUBIC B-spline interpolation method, the speckle images reconstructed by this model increased by 8.55 dB, 0.323, and 1.43 points, respectively. Compared to the SRResNet method, the speckle images reconstructed by this model were increased by 4.53 dB, 0.111, and 0.14 points, respectively. The reconstructed speckle images of engineered bamboo were clearer, and the image features were more obvious, which could better identify the tip crack position of the engineered bamboo. The results show that the super-resolution reconstruction effect of engineered-bamboo speckle images can be effectively improved by adding the attention mechanism to the residual network. This method has great application value.

In the real industrial scenario, the setup times and delivery times are two non-negligible factors, but only few studies have considered the open shop scheduling problem with sequence-dependent setup times and delivery times (OSSP-STDT). In this paper, a mixed integer linear programming model is formulated firstly to accurately solve small-size problems, but it will fail when the size of the problem increases. Then, a complex scheduling network model is developed to characterize OSSP-STDT. After comprehensively considering the local topological features and time attributes in the complex network model, an effective heuristic rule based on complex network is established for solving large-size problems. Finally, an actual warehouse scheduling problem is converted into the aforementioned problem and used as one typical application scenario. Experiments have been conducted and computational results show that compared with exact solutions and meta-heuristics, the proposed algorithm can solve the large-size open shop scheduling problem with sequence-dependent setup times and delivery times more effectively and efficiently.

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%.

Bolted connections are widely used in timber structures. Bolt looseness is one of the most important factors leading to structural failure. At present, most of the detection methods for bolt looseness do not achieve a good balance between cost and accuracy. In this paper, the detection method of small angle of bolt loosening in a timber structure is studied using deep learning and machine vision technology. Firstly, three schemes are designed, and the recognition targets are the nut’s own specification number, rectangular mark, and circular mark, respectively. The Single Shot MultiBox Detector (SSD) algorithm is adopted to train the image datasets. The scheme with the smallest identification angle error is the one identifying round objects, of which the identification angle error is 0.38°. Then, the identification accuracy was further improved, and the minimum recognition angle reached 1°. Finally, the looseness in a four-bolted connection and an eight-bolted connection are tested, confirming the feasibility of this method when applied on multi-bolted connection, and realizing a low operating costing and high accuracy.

Solid wood flooring has good esthetic properties and is an excellent material for interior decoration. To meet the artistic effects of specific interior decoration requirements, the color of solid wood flooring needs to be coordinated. Thus, the color of the produced solid wood flooring needs to be sorted to meet the individual needs of customers. In this work, machine vision, deep learning methods, and ensemble learning methods are introduced to reduce the cost of manual sorting and improve production efficiency. The color CCD camera was used to collect 108 solid wood floors of three color grades provided by the company and obtained 108 18,000 × 2048 pixel wood images. A total of 432 images were obtained after data expansion. Deep learning methods, such as VGG16, DenseNet121, and XGBoost, were compared. After using XGBoost to filter the features, the accuracy of solid wood flooring color classification was 97.22%, the training model time was 5.27 s, the average test time for each picture was 51 ms, and a good result was achieved.

The acidity of green plum has an important influence on the fruit’s deep processing. Traditional physical and chemical analysis methods for green plum acidity detection are destructive, time-consuming, and unable to achieve online detection. In response, a rapid and non-destructive detection method based on hyperspectral imaging technology was studied in this paper. Research on prediction performance comparisons between supervised learning methods and unsupervised learning methods is currently popular. To further improve the accuracy of component prediction, a new hyperspectral imaging system was developed, and the kernel principle component analysis—linear discriminant analysis—extreme gradient boosting algorithm (KPCA-LDA-XGB) model was proposed to predict the acidity of green plum. The KPCA-LDA-XGB model is a supervised learning model combined with the extreme gradient boosting algorithm (XGBoost), kernel principal component analysis (KPCA), and linear discriminant analysis (LDA). The experimental results proved that the KPCA-LDA-XGB model offers good acidity predictions for green plum, with a correlation coefficient (R) of 0.829 and a root mean squared error (RMSE) of 0.107 for the prediction set. Compared with the basic XGBoost model, the KPCA-LDA-XGB model showed a 79.4% increase in R and a 31.2% decrease in RMSE. The use of linear, radial basis function (RBF), and polynomial (Poly) kernel functions were also compared and analyzed in this paper to further optimize the KPCA-LDA-XGB model.

The green plum is rich in amino acids, lipids, inorganic salts, vitamins, and trace elements. It has high nutritional value and medicinal value and is very popular among Chinese people. However, green plums are susceptible to collisions and pests during growth, picking, storage, and transportation, causing surface defects, affecting the quality of green plums and their products and reducing their economic value. In China, defect detection and grading of green plum products are still performed manually. Traditional manual classification has low accuracy and high cost, which is far from meeting the production needs of green plum products. In order to improve the economic value of green plums and their products and improve the automation and intelligence level of the product production process, this study adopted deep learning methods based on a convolutional neural network and cost-effective computer vision technology to achieve efficient classification of green plum defects. First, a camera and LEDs were used to collect 1240 green plum images of RGB, and the green plum experimental classification standard was formulated and divided into five categories, namely, rot, spot, scar, crack, and normal. Images were randomly divided into a training set and test set, and the number of images of the training set was expanded. Then, the stochastic weight averaging (SWA) optimizer and w-softmax loss function were used to improve the VGG network, which was trained and tested to generate a green plum defect detection network model. The average recognition accuracy of green plum defects was 93.8%, the test time for each picture was 84.69 ms, the recognition rate of decay defect was 99.25%, and the recognition rate of normal green plum was 95.65%. The results were compared with the source VGG network, resnet18 network, and green lemon network. The results show that for the classification of green plum defects, the recognition accuracy of the green plum defect detection network increased by 9.8% and 16.6%, and the test speed is increased by 1.87 and 6.21 ms, respectively, which has certain advantages.

Considering the difficult problem of robot recognition and grasping in the scenario of disorderly stacked wooden planks, a recognition and positioning method based on local image features and point pair geometric features is proposed here and we define a local patch point pair feature. First, we used self-developed scanning equipment to collect images of wood boards and a robot to drive a RGB-D camera to collect images of disorderly stacked wooden planks. The image patches cut from these images were input to a convolutional autoencoder to train and obtain a local texture feature descriptor that is robust to changes in perspective. Then, the small image patches around the point pairs of the plank model are extracted, and input into the trained encoder to obtain the feature vector of the image patch, combining the point pair geometric feature information to form a feature description code expressing the characteristics of the plank. After that, the robot drives the RGB-D camera to collect the local image patches of the point pairs in the area to be grasped in the scene of the stacked wooden planks, also obtaining the feature description code of the wooden planks to be grasped. Finally, through the process of point pair feature matching, pose voting and clustering, the pose of the plank to be grasped is determined. The robot grasping experiment here shows that both the recognition rate and grasping success rate of planks are high, reaching 95.3% and 93.8%, respectively. Compared with the traditional point pair feature method (PPF) and other methods, the method present here has obvious advantages and can be applied to stacked wood plank grasping environments.

Wood is widely used in construction, the home, and art applications all over the world because of its good mechanical properties and aesthetic value. However, because the growth and preservation of wood are greatly affected by the environment, it often contains different types of defects that affect its performance and ornamental value. To solve the issues of high labor costs and low efficiency in the detection of wood defects, we used machine vision and deep learning methods in this work. A color charge-coupled device camera was used to collect the surface images of two types of wood from Akagi and Pinus sylvestris trees. A total of 500 images with a size of 200 × 200 pixels containing wood knots, dead knots, and checking defects were obtained. The transfer learning method was used to apply the single-shot multibox detector (SSD), a target detection algorithm and the DenseNet network was introduced to improve the algorithm. The mean average precision for detecting the three types of defects, live knots, dead knots and checking was 96.1%.

Real-time monitoring and fault diagnosis of bearings are of great significance to improve production safety, prevent major accidents, and reduce production costs. However, there are three primary concerns in the current research, namely real-time performance, effectiveness, and generalization performance. In this paper, a deep learning method based on stacked residual dilated convolutional neural network (SRDCNN) is proposed for real-time bearing fault diagnosis, which is subtly combined by the dilated convolution, the input gate structure of long short-term memory network (LSTM) and the residual network. In the SRDCNN model, the dilated convolution is used to exponentially increase the receptive field of convolution kernel and extract features from the sample with more points, alleviating the influence of randomness. The input gate structure of LSTM could effectively remove noise and control the entry of information contained in the input sample. Meanwhile, the residual network is introduced to overcome the problem of vanishing gradients caused by the deeper structure of the neural network, hence improving the overall classification accuracy. The experimental results indicate that compared with three excellent models, the proposed SRDCNN model has higher denoising ability and better workload adaptability.