International Journal of Pattern Recognition and Artificial Intelligence

In this paper, for some typical local slump problems, through the site survey of engineering conditions, we observe the local slumping parts, collect a large number of production data, and adopt the Chiping polar projection method to evaluate the initial stability of the problem area. The unstable structure is combined with the solid scale projection to analyze the sliding direction, the sliding surface, the sliding amount, and the like. The study has a clearer understanding of the local slump problem, and based on the stability judgment results, combined with the site conditions, ultimately serves the engineering examples.

Model accuracy and running speed are the two key issues for flood warning in urban areas. Traditional hydrodynamic models, which have a rigorous physical mechanism for flood routine, have been widely adopted for water level prediction of rainwater pipe network. However, with the amount of pipes increasing, both the running speed and data availability of hydrodynamic models would be decreased rapidly. To achieve a real-time prediction for the water level of the rainwater pipe network, a new framework based on a machine learning method was proposed in this paper. The spatial and temporal autocorrelation of water levels for adjacent manholes was revealed through theoretical analysis, and then a support vector machine (SVM)-based machine learning model was developed, in which the water levels of adjacent manholes and rivers-near-by-outlets at the last time step were chosen as the independent variables, and then the water levels at the current time step can be computed by the proposed machine learning model with calibrated parameters. The proposed framework was applied in Fuzhou city, China. It turns out that the proposed machine learning method can forecast the water level of the rainwater pipe network with good accuracy and running speed.

Deep learning refers to Convolutional Neural Network (CNN). CNN is used for image recognition for this study. The dataset is named Fruits-360 and it is obtained from the Kaggle dataset. Seventy percent of the pictures are selected as training data and the rest of the images are used for testing. In this study, an image size is $100\times 100\times 3$. Training is realized using Stochastic Gradient Descent with Momentum (sgdm), Adaptive Moment Estimation (adam) and Root Mean Square Propogation (rmsprop) techniques. The threshold value is determined as 98% for the training. When the accuracy reaches more than 98%, training is stopped. Calculation of the final validation accuracy is done using trained network. In this study, more than 98% of the predicted labels match the true labels of the validation set. Accuracies are calculated using test data for sgdm, adam and rmsprop techniques. The results are 98.08%, 98.85%, 98.88%, respectively. It is clear that fruits are recognized with good accuracy.

In this study, a natural scene text detection method based on the improved faster region-based convolutional neural network (R-CNN) is proposed. This method extracts image features with the Inception-ResNet architecture, adopts a region proposal network to generate region proposals for the extracted features, merges the fine-tuned features with the region proposals, and finally, uses Fast R-CNN to classify and locate text. The proposed method solves the problems of varying text sizes and the text being obscured in the image. Compared with the original Faster R-CNN, the multilevel Inception-ResNet network model presented in this study can extract deeper text features. The extracted feature map is further sparsely represented by Reduction B, Inception ResNet C and Avg Pool, and then is fused with text regions obtained by the text feature mapping lower layer network to acquire the exact text regions. The text detection method presented in this study is tested on the 2017 dataset of ICDAR2017 Competition on Reading Chinese Text in the Wild (RCTW-17), which contains a large number of distorted, blurry, different scale and size texts. An accuracy of 76.4% is achieved in this platform, thereby proving the efficiency of the proposed method.

Aiming at improving the performance of the nonlinear adaptive filtering under the alpha-stable distribution noise environment, Kernel Affine Projection P-norm (KAPP) algorithm based on the minimum dispersion coefficient criterion and the affine projection is deduced. The accuracy of the gradient estimation is enhanced by using the input signals and the error signals at multiple times. The simulation results on Mackey–Glass chaotic time series prediction show that the KAPP algorithm has faster convergence speed, better steady-state performance and stronger robustness under the Gaussian noise and stable distributed noise environment.

The colorimetric method is usually used to test the concentration of substances. However, this method has a big error since different people have different sensitivities to colors. In this paper, in order to solve the identification problem of the color and the concentration of the test paper, firstly, we found out that the concentration of substance is correlated with the color reading by using the Pearson’s Chi-squared test method. And by the concentration coefficient of Pearson correlation analysis, the concentration of substance and color reading is highly correlated. Secondly, according to the RGB value of the paper image, the color moments of the image are calculated as the characteristics of the image, and the Levenberg–Marquardt (LM) neural network is established to classify the concentration of the substance. The accuracy of the training set model is 94.5%, and the accuracy of the test set model is 87.5%. The model precision is high, and the model has stronger generalization ability. Therefore, according to the RGB value of the test paper image, it is effective to establish the LM neural network model to identify the substance concentration.

An effective and accurate lane marking detection algorithm is a fundamental element of the intelligent vehicle system and the advanced driver assistant system, which can provide important information to ensure the vehicle runs in the lane or warn the driver in case of lane departure. However, in the complex urban environment, lane markings are always affected by illumination, shadow, rut, water, other vehicles, abandoned old lane markings and non-lane markings, etc. Meanwhile, the lane markings are weak caused by hard use over time. The dash and curve lane marking detection is also a challenge. In this paper, a new lane marking detection algorithm for urban traffic is proposed. In the low-level phase, an iterative adaptive threshold method is used for image segmentation, which is especially suitable for the blurred and weakened lane markings caused by low illumination or wear. In the middle-level phase, the algorithm clusters the candidate pixels into line segments, and the upper and lower structure is used to cluster the line segments into candidate lanes, which is more suitable for curve and dashed lane markings. In the high-level phase, we compute the highest scores to get the two optimal lane markings. The optimal strategy can exclude interference similar to lane markings. We test our algorithm on Future Challenge TSD-Lane dataset and KITTI UM dataset. The results show our algorithm can effectively detect lane markings under multiple disturbance, occlusions and sharp curves.

Railway fastener recognition and detection is an important task for railway operation safety. However, the current automatic inspection methods based on computer vision can effectively detect the intact or completely missing fasteners, but they have weaker ability to recognize the partially worn ones. In our method, we exploit the EA-HOG feature fastener image, generate two symmetrical images of original test image and turn the detection of the original test image into the detection of two symmetrical images, then integrate the two recognition results of symmetrical image to reach exact recognition of original test image. The potential advantages of the proposed method are as follows: First, we propose a simple yet efficient method to extract the fastener edge, as well as the EA-HOG feature of the fastener image. Second, the symmetry images indeed reflect some possible appearance of the fastener image which are not shown in the original images, these changes are helpful for us to judge the status of the symmetry samples based on the improved sparse representation algorithm and then obtain an exact judgment of the original test image by combining the two corresponding judgments of its symmetry images. The experiment results show that the proposed approach achieves a rather high recognition result and meets the demand of railway fastener detection.

In view of the current consumer trust crisis in food safety, some researchers proposed to build a reliable food safety traceability system solution. Because the food safety traceability system requires storing and processing of massive video data, this paper proposes to build a reliable food safety traceability system by introducing cloud storage technology into video surveillance system. Hadoop platform is built to store and process massive monitoring of video data. In addition, in order to make use of Map Reduce framework for parallel computing, this paper optimizes the traditional particle filter target tracking algorithm, and proposes using the parallel computing framework Map Reduce in Hadoop to achieve the parallel computing of all particles in particle filter. Finally, the simulation results show that when the number of videos is large, the time of parallel processing is obviously shorter than that of single-machine mode. And as the video data becomes larger and larger, the advantages of parallel processing become more and more obvious.

In this paper, we propose two self-adapting patch strategies, which are obtained by employing the integral projection technique on images’ edge images, while the edge images are recovered by the two-dimensional discrete wavelet transform. The patch strategies are equipped with the advantage of considering the single image’s unique properties and maintaining the integrity of some particular local information. Combining the self-adapting patch strategies with local binary pattern feature extraction and the classifier of the forward and backward greedy algorithms under strong sparse constraint, we propose two new face recognition methods. Experiments are run on the Georgia Tech, LFW and AR face databases. The obtained numerical results show that the new methods outperform some related patch-based methods to a larger extent.

The directional logging-while-drilling measurements enable to monitor the distance to formation boundaries and their orientation and facilitate proactive well placement, and thus can provide directionality information useful in detecting physical parameters such as bed boundaries, formation dip and formation azimuth, so the oil and gas recovery can be greatly improved. The characteristics of a kind of electromagnetic wave resistivity logging while drilling (LWD) tool with tilted antennas are computed via the magnetic-current-source dyadic Green’s function for horizontally stratified anisotropic media. The current characteristic at the formation interface of the tilted antennas is compared with that of the axial antenna, and how the LWD tool with tilted antennas to detect the formation interface position and orientation is revealed. The amplitude-attenuation and phase-shift characteristics of LWD tool with tilted antennas are analyzed, and how to design the angle of tilted antenna is presented. The relationship between the emitting frequency and emitting-receiving antenna space parameters is expounded, and the emitting-receiving antenna space parameter suitable for different frequencies is presented. The detection characteristic of the electromagnetic wave measuring instrument is studied. For the highly deviated well formation model, the directional response characteristics of the different relative dip angle isotropic and anisotropic formation are numerically simulated and analyzed. The directional response characteristics of the two kinds of coil arrangement (unilateral arrangement and symmetrical arrangement) are analyzed by numerical simulation, and the solution to reduce or eliminate the influence of the directional response of the isotropy on the directional response is obtained. The conclusions obtained by theoretical analysis and numerical simulation provide a guide for the instrument principle research and application.

To overcome the disadvantages of the traditional power inversion algorithm, which has obvious suppression effect on strong interference signals but unobvious effect on suppression of weak interference signals, a matrix decomposition method is proposed to extract the interference sub-matrix and noise sub-matrix, so that the power inversion algorithm can be adopted to weaken and filter out the signals in the case of weak interference. Through the optimized power inversion algorithm, the interference signal can be weakened obviously at the antenna receiving end, and the quality of the effective signal received by the antenna can be improved significantly. Afterwards, the algorithm was used to carry out MATLAB simulation in space based on the anti-jamming performance of different array models, which determined the anti-jamming antenna array structure as a uniform matrix and optimized the array structure. The simulation results show that the optimized power inversion algorithm can filter strong and weak interference better, and the anti-jamming performance is better under different array models and two kinds of interferences. Through comparative analysis, it is concluded that the four elements have the best interference suppression effect in uniform circular array structure, which provides an important reference for the selection of antenna array in adaptive interference suppression hardware structure. At the same time, the BeiDou four-element adaptive circular array antenna, as an important part of intelligent geological disaster monitoring and early warning system, has good anti-jamming ability.

In this paper, we propose a novel ensemble gene selection method to obtain a gene subset. Then we provide a reverse construction method of gene network derived from expression profile data of the gene subset. The uncertainty coefficient based on information entropy are used to define the existence of logical relations among these genes. If the uncertainty coefficient between some genes exceeds predefined thresholds, the gene nodes will be connected by directed edges. Thus, a gene network is generated, which we define as gene logical network. This method is applied to the breast cancer data including control group and experimental group, with comparisons of the 2nd-order logic type distribution, average degree as well as average path length of the networks. It is found that these structures with different networks are quite distinct. By the comparison of the degree difference between control group and experimental group, the key genes are picked up. By defining the dynamics evolution rules of state transition based on the logical regulation among the key genes in the network, the dynamic behaviors for normal breast cells and cells with cancer of different stages are simulated numerically. Some of them are highly related to the development of breast cancer through literature inquiry. The study may provide a useful revelation to the biological mechanism in the formation and development of cancer.

Association rule (AR) mining in complex scene has attracted extensive attention of researchers in recent years. Typically, many researchers focused on an algorithm itself and ignored a generalization method to improve the performance of AR mining. Tuna et al., presented a general data structure Speeding-Up AR Structure with Inverted Index Compression (SAII) which could be utilized in most of the existing algorithms to improve their performance IEEE Trans. Cybern. 46(12) (2016) 3059–3072. However, we found that this algorithm consumes a lot of time in re-ordering data because a one-to-one comparison method is used in this process, which is the main reason that the speeding-up structure is difficult to establish when coping with much more large amount of data. To overcome these problems, this paper aims to propose an improved speeding-up AR algorithm based on group similarity and Apache Spark framework to further reduce the memory requirements and runtime. Our simulation results on the police business big dataset make clear that our improved approach performs well and is more suitable for a big data environment.

With the rapid development of the Internet, people are confronted with information overload. Many recommendation methods are designed to solve this problem. The main contributions of recommendation methods proposed in this paper are as follows: (1) An improved collaborative filtering recommendation algorithm based on user clustering is proposed. Clustering is performed according to user similarity based on the user-item rating matrix. So the search space of recommendation algorithm is reduced. (2) Considering the factor that user’s interests may dynamically change over time, a time decay function is introduced. (3) A method of real-time recommendation based on topic for microblogs is designed to realize real-time recommendation effectively by preserving intermediate variables of user similarity. Experiments show that the proposed algorithms have been improved in terms of running time and accuracy.