International Journal of Pattern Recognition and Artificial Intelligence

Conventional pattern recognition methods employed for differentiating the types of insulation defects in power cables usually rely on the manual extraction of partial discharge features, which is inefficient and easily affected by subjective uncertainty. This work addresses this problem by proposing a new framework based on the automatic features extraction of partial discharge signal. The method first applies a sliding time window to convert partial discharge signals in the time domain into two-dimensional images that serve directly as the input to the convolutional neural networks (CNNs). Then a nonlinear encoder is employed to automatically extract the features of the partial discharge image data as the input of CNNs for classification. In addition, we address the overfitting problem associated with the few-shot by applying a deep convolutional generative adversarial network (DCGAN) to augment the original training dataset. Experimental results demonstrate the validity of the proposed algorithm; it increases the classification accuracy by 4.18% relative to that achieved with manually extracted features; the overall accuracy of the proposed algorithm training with the augmented dataset is 3.175% higher than that with the original experimental dataset.

Aerial image-based target detection has problems such as low accuracy in multiscale target detection situations, slow detection speed, missed targets and falsely detected targets. To solve this problem, this paper proposes a detection algorithm based on the improved You Only Look Once (YOLO)v3 network architecture from the perspective of model efficiency and applies it to multiscale image-based target detection. First, the K-means clustering algorithm is used to cluster an aerial dataset and optimize the anchor frame parameters of the network to improve the effectiveness of target detection. Second, the feature extraction method of the algorithm is improved, and a feature fusion method is used to establish a multiscale (large-, medium-, and small-scale) prediction layer, which mitigates the problem of small target information loss in deep networks and improves the detection accuracy of the algorithm. Finally, label regularization processing is performed on the predicted value, the generalized intersection over union (GIoU) is used as the bounding box regression loss function, and the focal loss function is integrated into the bounding box confidence loss function, which not only improves the target detection accuracy but also effectively reduces the false detection rate and missed target rate of the algorithm. An experimental comparison on the RSOD and NWPU VHR-10 aerial datasets shows that the detection effect of high-efficiency YOLO (HE-YOLO) is significantly improved compared with that of YOLOv3, and the average detection accuracies are increased by 8.92% and 7.79% on the two datasets, respectively. The algorithm not only shows better detection performance for multiscale targets but also reduces the missed target rate and false detection rate and has good robustness and generalizability.

Current state-of-the-art detectors achieved impressive performance in detection accuracy with the use of deep learning. However, most of such detectors cannot detect objects in real time due to heavy computational cost, which limits their wide application. Although some one-stage detectors are designed to accelerate the detection speed, it is still not satisfied for task in high-resolution remote sensing images. To address this problem, a lightweight one-stage approach based on YOLOv3 is proposed in this paper, which is named Squeeze-and-Excitation YOLOv3 (SE-YOLOv3). The proposed algorithm maintains high efficiency and effectiveness simultaneously. With an aim to reduce the number of parameters and increase the ability of feature description, two customized modules, lightweight feature extraction and attention-aware feature augmentation, are embedded by utilizing global information and suppressing redundancy features, respectively. To meet the scale invariance, a spatial pyramid pooling method is used to aggregate local features. The evaluation experiments on two remote sensing image data sets, DOTA and NWPU VHR-10, reveal that the proposed approach achieves more competitive detection effect with less computational consumption.

Imaging sensors with higher resolution and higher frame rates are becoming more popular for wide-area video surveillance (VS) and other applications as technology advances Using Mask-RCNN, we proposed Multiple-Object Detection and Segmentation in High-Resolution Video based on Deep Learning. The ResNet-50 ResNet-101 is used as the backbone in the proposed R-CNN Mask FPN model. The deep residual network’s design overcomes the problem of lower learning efficiency due to the network’s deepening. To reach the objective of the smallest overall error, the deep residual network divided the training series into one training block, minimizing the error of each block. It is roughly divided into five convolutional layer stages. The output scale is cut in half at each point. We used mixed precision FP16 and FP32 for training the model and achieved great speed in training time reduction in inference time for object. The COCO 2014 data set is used to train and validate the proposed model with mixed precision, leading to faster performance. The results of the experiments show that the proposed model can run at 30–48 frames per second with 85% accuracy.

The max–min ant system (MMAS) is a modified ant colony optimization (ACO) algorithm. Its convergence speed is effectively improved by setting the upper and lower bounds of the pheromone and updating it in the optimal path. However, MMAS still has drawbacks, such as long search time and local extremums. In this paper, the hybrid max–min ant system (HMMAS) is proposed to deal with the shortcomings of MMAS. Employing Levy flight strategy, HMMAS can dynamically adjust the parameters to increase the diversity of solutions and expand the search range. Besides, HMMAS uses the OBL strategy to generate opposite solutions in the early stage. In this way, the convergence is accelerated. When HMMAS falls into a local extremum, the path reorganization strategy is utilized. With its help, HMMAS can redistribute the pheromone in each path and achieve global optimum. To verify the effectiveness, HMMAS is first compared with the three conventional ACO algorithms of AS, ACS, and MMAS in 20 sets of experiments. The results indicate that the average results of HMMAS in the 19 sets of TSP instances are better than the other three algorithms, and the standard deviation in the 14 sets of calculation instances is the smallest. Then, HMMAS is compared with some state-of-the-art algorithms, and the results show that HMMAS is better than other comparison algorithms, either by the minimum or the average value.

Automatic music genre classification based on distance metric learning (DML) is proposed in this paper. Three types of timbral descriptors, namely, mel-frequency cepstral coefficient (MFCC) features, modified group delay features (MODGDF) and low-level timbral feature sets are combined at the feature level. We experimented with k nearest neighbor (kNN) and support vector machine (SVM)-based classifiers for standard and DML kernels (DMLK) using GTZAN and Folk music dataset. Standard kernel-based kNN and SVM-based classifiers report classification accuracy (in%) of 79.03 and 90.16, respectively, on GTZAN dataset and 86.60 and 92.26, respectively, for Folk music dataset, with the best performing RBF kernel. A further improvement was observed when DML kernels were used in place of standard kernels in the kernel kNN and SVM-based classifiers with an accuracy of 84.46%, 92.74% (GTZAN), 90.00 and 96.23 (Folk music dataset) for DMLK-kNN and DMLK-SVM, respectively. The results demonstrate the potential of DML kernels in music genre classification task.

Existing dimensionality reduction (DR) techniques such as principal component analysis (PCA) and its variants are not suitable for target class mining due to the negligence of unique statistical properties of class-of-interest (CoI) samples. Conventionally, these approaches utilize higher or lower eigenvalued principal components (PCs) for data transformation; but the higher eigenvalued PCs may split the target class, whereas lower eigenvalued PCs do not contribute significant information and wrong selection of PCs leads to performance degradation. Considering these facts, the present research offers a novel target class-guided feature extraction method. In this approach, initially, the eigendecomposition is performed on variance–covariance matrix of only the target class samples, where the higher- and lower-valued eigenvectors are rejected via statistical analysis, and the selected eigenvectors are utilized to extract the most promising feature subspace. The extracted feature-subset gives a more tighter description of the CoI with enhanced associativity among target class samples and ensures the strong separation from nontarget class samples. One-class support vector machine (OCSVM) is evaluated to validate the performance of learned features. To obtain optimized values of hyperparameters of OCSVM a novel $N$-ary search-based autonomous method is also proposed. Exhaustive experiments with a wide variety of datasets are performed in feature-space (original and reduced) and eigenspace (obtained from original and reduced features) to validate the performance of the proposed approach in terms of accuracy, precision, specificity and sensitivity.

Scene text detection is an important research branch of artificial intelligence technology whose goal is to locate text in scene images. In the Tibetan areas of China, scene images containing both Tibetan and Chinese texts are ubiquitous. Thus, detecting bilingual Tibetan-Chinese scene texts is important in promoting intelligent applications for minority languages. In this study, a scene text detection database for bilingual Tibetan-Chinese is constructed using a manually labeled method and an automatic synthesis method, and a text detection method is proposed. First, we predict a text rectangular region and the text center region for each text instance and simultaneously learned the expansion distance of the text center region. Second, based on the classification score of the text center region and the text rectangular region, we obtain the final confidence of each text instance and then filter out the text center region with lower confidence. Third, through the learned expansion distance, the full-text instance from the remaining text center region is recovered. The results show that our method obtains good detection performance; it achieves an accuracy of up to 75.47% during the text detection phase, laying the foundation for scene text recognition in the subsequent step.

The ellipsoidal magnetization model has a wide range of application scenarios. For example, in aviation magnetic field prospecting, mineral prospecting, seabed prospecting, and UXO (unexploded ordnance) detection. However, because the existing ellipsoid magnetization formula is relatively complicated, the detection model is usually replaced by a dipole. Such a model increases the error probability and poses a significant challenge for subsequent imaging and pattern recognition. Based on the distribution of ellipsoid gravity potential and magnetic potential, the magnetic anomaly field distribution equation generated by the ellipsoid is deduced by changing the aspect ratio, making the ellipsoid equivalent to a sphere. The result of formula derivation shows that the two magnetic anomaly fields are consistent. This paper uses COMSOL finite element software to model UXO, ellipsoids, and spheres and analyzes magnetic anomalies. The conclusion shows that the ellipsoid model can completely replace the UXO model when the error range of 1nT is satisfied. Finally, we established two sets of ellipsoids and calculated the magnetic anomalous field distributions on different planes using deduction formulas and finite element software. We compared the experimental results and found that the relative error of the two sets of data was within $\pm 0.25$‰. Error analysis found that the error distribution is standardized and conforms to the normal distribution. The above mathematical analysis and finite element simulation prove that the calculation method is simple and reliable and provides a magnetic field distribution equation for subsequent UXO inversion.

Earthquake prediction based on extreme imbalanced precursor data is a challenging task for standard algorithms. Since even if an area is in an earthquake-prone zone, the proportion of days with earthquakes per year is still a minority. The general method is to generate more artificial data for the minority class that is the earthquake occurrence data. But the most popular oversampling methods generate synthetic samples along line segments that join minority class instances, which is not suitable for earthquake precursor data. In this paper, we propose a Safe Zone Synthetic Minority Oversampling Technique (SZ-SMOTE) oversampling method as an enhancement of the SMOTE data generation mechanism. SZ-SMOTE generates synthetic samples with a concentration mechanism in the hyper-sphere area around each selected minority instances. The performance of SZ-SMOTE is compared against no oversampling, SMOTE and its popular modifications adaptive synthetic sampling (ADASYN) and borderline SMOTE (B-SMOTE) on six different classifiers. The experiment results show that the quality of earthquake prediction using SZ-SMOTE as oversampling algorithm significantly outperforms that of using the other oversampling algorithms.

This paper takes a project comprehensive office building project of Chaoyang financial center as an example, based on the construction visualization requirements of BIM technology, this paper analyzes the modeling requirements and Suggestions applicable to the project. With the help of Revit modeling software, the architecture and structure model of the whole drawing and three floors underground and two floors above ground electromechanical model are completed. There is a collision detection between the structural and electromechanical subject model that be established by Navisworks Manage 2016 software, also, the paper analyzes the causes of the collision and puts forward corresponding solutions, and find out the problems in the model and make changes in time, which prepare the model for the construction visualization simulation application. On this basis, the construction progress plan set up by Microsoft Project 2013 was carried out and related to the three-dimensional (3D) parameterized model components, which implement the visual simulation of the construction process of research and the 4D simulation of construction process of the project. What’s more, the application of visualization simulation based on BIM technology can greatly reduce the occurrence of changes and rework in actual construction, thus greatly improving the efficiency of construction work.¹¹

As intelligence technology develops, there is a higher requirement for computing speed and accuracy of water injection system simulation. In this paper, aiming at the tree-shaped water injection pipe network system of large-scale oilfields, based on the energy equation for calculating the pressure drop $\mathrm{\Delta }H$ of pipe section, a mathematical model of the pipeline unit and the node unit is established, and finally, a mathematical model of pipe network for the entire water injection system is established; then, the improved iterative algorithm is used to solve the simulation model of water injection system. In this way, we determine the boundary calculation conditions, take the water injection station as reference node, and use the maximum pressure of water injection well as the initial value of the reference node for calculation, which reduces the number of iterations in model calculation; by comparing the simulation results of different iteration steps, 0.01 is selected as the iteration step size due to its higher calculation accuracy; and the calculation process has also been optimized. The process of solving the characteristic matrix $K$ is combined with the process of calculating the pressure drop $\mathrm{\Delta }H$ of pipe section, and placed outside the algorithm loop, thereby shortening the calculation time of a single cycle and reducing the calculation amount of the algorithm. The application cases show that the proposed optimization algorithm for water injection system pipe network simulation can be used as an effective method to improve the solution speed and calculation accuracy of the simulation algorithm of tree-shaped water injection system in large-scale oilfields.

Optimal disassembly sequencing is an NP-hard problem and has always been an ambition for industry production. In the context of increasing public concerns over environmental impacts, in addition to the feasibility of a disassembly sequence, dismantling enterprises have to consider the relationship between potential profits and the impacts. Thus, an ideal disassembly sequence should weight these three factors comprehensively. Up to now, an appropriate ELV disassembly sequence still mainly relies on people’s intuitive experience and seeking an optimal disassembly sequencing method assumes enormous importance. This paper aims to address the optimal disassembly sequencing problem of ELVs by means of an improved genetic algorithm, in which a matrix coding mechanism and an elite strategy are employed. The weight of different factors can be adjusted according to the actual conditions of factories. The paper gives a case and a series of Pareto fronts are obtained. The effects of population size and maximum evolutionary time on the Pareto solutions were investigated. Ultimately, the optimal Pareto disassembly sequence corresponding to balanced profit and environmental impact is achieved, thereby providing an appropriate disassembly depth defined by the aforementioned disassembly sequence. This can contribute to timely disassembly decisions for end-of-life vehicle (ELV) dismantling enterprises, achieving a cost-effective disassembly process for survival in the context of growing environmental concerns. This paper seeks to offer a viable decision-making approach prior to real disassembly of ELVs by detailing a Pareto disassembly depth and sequence.

This paper proposed a pulse-coupled synchronization protocol for wireless sensor networks. The time information of the node can be encoded at the Mac layer and then sent and received in the form of pulse waves. During the exchange of time information between a pair of nodes, one node will adjust its own clock information according to certain rules after receiving the pulse time signal of the other node. This process is repeated in the entire wireless sensor network. Under certain environmental and estimated parameter conditions, all nodes in the network can finally converge to a coherent frequency and phase, thus realizing time synchronization. The relationship between time synchronization and coupling coefficient was proved theoretically, and the optimal coupling coefficient was derived. An optimized algorithm was proposed after the optimization of the protocol. Finally, the correctness of the proposed protocol and its optimized algorithm was verified by data simulation.