International Journal of Pattern Recognition and Artificial Intelligence

Fire is one of the most common serious disasters in human society. It is a kind of burning phenomenon that is out of control in time and space. When a fire occurs, how to detect the fire quickly and remove it in the budding state has become the key content of fire control work. Outdoor fire is very common in our daily life, and once it occurs without effective and timely control, it will cause huge losses. Therefore, it is particularly important to study an intelligent alarm system for outdoor fire. Generally, fire detection technology can be divided into sensor fire detection technology and image fire detection technology. Sensor fire detection technology is low cost and easy to design, but its application field is limited. Under the interference of many factors outside, misjudgement and missed judgement will occur. Image fire detection technology can achieve certain detection function through manual design of features and classifiers, but there are still defects in the application in the actual diversified environment. With the development of neural network technology in recent years, it has made great breakthroughs in the field of image recognition. Its judgment type is obtained through a large number of data training algorithms. Because of its automatic feature extraction and classification characteristics, it can effectively adapt to the external environment. Therefore, this paper proposes an end-to-end two-stream neural network model to detect fires, uses fire video on the network to train the algorithm, and then uses the fire database to test. Compared with the existing fire detection algorithms, it is found that the proposed method has good practicability and versatility, and provides a good reference for the development of fire detection technology.

Many tasks of natural language processing such as information retrieval, intelligent question answering, and machine translation require the calculation of sentence similarity. The traditional calculation methods used in the past could not solve semantic understanding problems well. First, the model structure based on Siamese lack of interaction between sentences; second, it has matching problem which contains lacking position information and only using partial matching factor based on the matching model. In this paper, a combination of word and word’s dependence is proposed to calculate the sentence similarity. This combination can extract the word features and word’s dependency features. To extract more matching features, a bi-directional multi-interaction matching sequence model is proposed by using word2vec and dependency2vec. This model obtains matching features by convolving and pooling the word-granularity (word vector, dependency vector) interaction sequences in two directions. Next, the model aggregates the bi-direction matching features. The paper evaluates the model on two tasks: paraphrase identification and natural language inference. The experimental results show that the combination of word and word’s dependence can enhance the ability of extracting matching features between two sentences. The results also show that the model with dependency can achieve higher accuracy than these models without using dependency.

With the development of the world economy and the accelerating process of urbanization, cars have brought great convenience to people’s lives and activities, and have become an indispensable means of transportation. Intelligent vehicles have the important significance of reducing traffic accidents, improving transportation capacity and broad market prospects, and can lead the future development of the automotive industry, so they have received extensive attention. In the existing intelligent vehicle system, the laser radar is a well-deserved protagonist because of its excellent speed and precision. It is an indispensable part of achieving high-precision positioning, but to some extent, the price hindering its marketization is a major factor. Compared with lidar sensors, vision sensors have the advantages of fast sampling rate, light weight, low energy consumption and low price. Therefore, many domestic and foreign research institutions have listed them as the focus of research. However, the current vision-based intelligent vehicle environment sensing technology is also susceptible to factors such as illumination, climate and road type, resulting in insufficient accuracy and real-time performance of the algorithm. This paper takes the environment perception of intelligent vehicles as the research object, and conducts in-depth research on the existing problems in road recognition and obstacle detection algorithms, including road image vanishing point detection, road image segmentation problem, road scene based on binocular vision. Three-dimensional reconstruction and obstacle detection technology.

Because the shallow neural network has limited ability to represent complex functions with limited samples and calculation units, its generalization ability will be limited when it comes to complex classification problems. The essence of deep learning is to learn a nonlinear network structure, to represent input data distributed representation and demonstrate a powerful ability to learn deeper features of data from a small set of samples. In order to realize the accurate classification of expression images under normal conditions, this paper proposes an expression recognition model of improved Visual Geometry Group (VGG) deep convolutional neural network (CNN). Based on the VGG-19, the model optimizes network structure and network parameters. Most expression databases are unable to train the entire network from the start due to lack of sufficient data. This paper uses migration learning techniques to overcome the shortage of image training samples. Shallow CNN, Alex-Net and improved VGG-19 deep CNN are used to train and analyze the facial expression data on the Extended Cohn–Kanade expression database, and compare the experimental results obtained. The experimental results indicate that the improved VGG-19 network model can achieve 96% accuracy in facial expression recognition, which is obviously superior to the results of other network models.

Aiming at the problem of insufficient detail retention in multimodal medical image fusion (MMIF) based on sparse representation (SR), an MMIF method based on density peak clustering and convolution sparse representation (CSR-DPC) is proposed. First, the base layer is obtained based on the registered input image by the averaging filter, and the original image minus the base layer to obtain the detail layer. Second, for retaining the details of the fused image, the detail layer image is fused by CSR to obtain the fused detail layer image, then the base layer image is segmented into several image blocks, and the blocks are clustered by using DPC to obtain some clusters, and each class cluster is trained to obtain a sub-dictionary, and all the sub-dictionaries are fused to obtain an adaptive dictionary. The sparse coefficient is fused through the learned adaptive dictionary, and the fused base layer image is obtained through reconstruction. Finally, fusing the detail layer and the base layer and reconstructing them forms the ultimate fused image. Experiments show that compared to the state-of-the-art two multi-scale transformation methods and five SR methods, the proposed method(CSR-DPC) outperforms the other methods in terms of the image details, the visual quality and the objective evaluation index, which can be helpful for clinical diagnosis and adjuvant treatment.

In order to identify the disturbance signal in power system and reduce the influence on system security, a voltage disturbance signal classifier based on improved local mean decomposition (ILMD) and BP neural network is proposed. ILMD is used to decompose the disturbance signal in three layers, and the product function (PF) component with amplitude-frequency information of voltage signal is obtained. The signal energy value constructed by PF component is used as the input of BP neural network to identify and classify the voltage disturbance signal. Experiments on four typical voltage disturbance signals show that the signal classifiers based on ILMD and BP neural networks have high accuracy and good working efficiency for the recognition and classification of voltage disturbance signals.

Cognitive radio (CR) is the trending domain in addressing the inadequate bands for communication, and spectrum sensing is the hectic challenge need to be addressed extensively. In the conventional CRs, the communication is restricted to the secondary users (SUs) in the allocated bands causing the underutilization of the available band. Thus, with the aim to afford higher throughput and spectrum efficiency, this paper introduces the hybrid mixture model for spectrum sensing in the multiple-input–multiple-output (MIMO) systems and the effectiveness is evaluated based on the evaluation parameters, such as detection probability and probability of false alarm. The signal received through the orthogonal frequency-division multiplexing (OFDM) antenna is employed for analyzing the spectral availability for which the energy and Eigen statistics of the signal is generated, which forms the input to the Hybrid mixture model. The developed Hybrid mixture model is the integration of the Gaussian Mixture Model (GMM) and Whale Elephant-Herd Optimization (WEHO). The GMM is subjected to the optimal tuning using the WEHO, which is the modification of the standard Whale Optimization Algorithm (WOA) with the Elephant-Herd Optimization (EHO). The analysis reveals that the proposed spectrum sensing model acquired the maximal detection probability and minimal false alarm probability of 99.9% and 46.4%, respectively. The proposed hybrid mixture model derives the spectrum availability and ensures the effective communication in CR without any interference.

A new method about the multi-fault condition monitoring of slurry pump based on principal component analysis (PCA) and sequential probability ratio test (SPRT) is proposed. The method identifies the condition of the slurry pump by analyzing the vibration signal. The experimental model is established using the normal impeller and the faulty impellers where the collected vibration signals were preprocessed using wavelet packet transform (WPT). The characteristic parameters of the vibration signals are extracted by time domain signal analysis and the dimension of data was reduced by PCA. The principal components with the largest contribution rate are chosen as the inputted signal to SPRT to assess the proposed algorithm. The new methodology is reasonable and practical for the multi-fault diagnosis of slurry pump.

In the concrete implementation of the fuzzy vault algorithm, the geometric hash method is a common technique for automatic calibration of biometric templates. For the fuzzy problem of parameter acquisition, the matching accuracy of fuzzy vault template is affected in the three parameters: the pixel size, hash table and hash table quantization parameters ($\alpha$ and $\beta$). The single factor experiment method obtains the optimal range of these three parameters, and the extraction range of the fuzzy point and the selection rule of the base point distance are improved for the fuzzy vault algorithm. Finally, based on the FVC fingerprint database, their matching precision is compared for the algorithm before and after optimization. The experimental results show that the false rejection rate (FRR) of the optimized algorithm is reduced by at least 9.84%, and the false acceptance rate (FAR) is reduced by at least 7.12%, indicating that the optimization scheme improves the matching accuracy of the algorithm. The algorithm has certain robustness and practicability.

This paper explores the performance and obtains a reasonable cleaning effect of the cleaning system of combine harvester and studies the relationship between the cleaning effect of the combine harvester cleaning system and its influencing factors. We established a neural network model between the cleaning loss rate and the clean system parameters. First, we tested the results of the cleaning performance of each group under different combinations of conditions, and analyzed the direct or indirect relationship between the cleaning loss rate and the parameters in the experiment under each working condition. Then, according to the experimental data obtained in the experiment, we predict the clearance loss rate for several sets of conditions by this model. The experimental results show that the prediction results of the model can meet the experimental requirements under the condition that the accuracy is not very high.

In today’s rapid development of science and technology, the manual work in the factory has been gradually replaced by machines, and the process of industrial intelligence has been further deepened. Workpiece recognition is the use of machine learning, computer vision and other technologies to identify the target workpiece, and the robot intelligent capture is a higher level of operation of the workpiece recognition, which is the key to realize the intelligentization of industrial robots. Due to the complex environmental factors and the diversity of the shape and size of the objects to be grasped, the accuracy and efficiency of the workpiece recognition are not ideal at this stage, and intelligent crawling is even more difficult to talk about. Aiming at the above problems, this paper builds a crawling planning model based on the convolutional neural network and the grasping pose mapping rules. Based on the established crawling planning model, the sampling candidate grabbing algorithm is designed and the migration learning method is adopted. The pretrained convolutional neural network for image recognition on the ImageNet dataset was migrated to the capture detection task of the Carnegie Mellon dataset. Experiments show that the network model proposed in this paper performs well, and its correct crawl rate is as high as 81.27%. This is to achieve a more stable and reliable identification and intelligent crawling work.

Aiming at the problem of portrait of members in shopping malls, this paper analyzes the similarities and differences of consumption behaviors between member groups and nonmember groups, and constructs the LRFMC model with $k$-means algorithm to analyze the value of membership. Second, active states of members are divided according to the consumption time interval, and KNN algorithm model is established to predict member states and used to predict the membership status. Finally, it discusses which types of goods are more suitable for promotional activities and can bring more profits to the shopping mall.

High dynamic range image (HDRI) which is combined with low dynamic range image (LDRI) needs to be mapped to a low dynamic area to display. In the process of mapping, it is impossible to determine the contribution of low dynamic image sequences in the display images, so that it results in a problem that the low dynamic images cannot be accurately selected. In this paper, for the first time, a contribution algorithm from LDRI to HDRI according to the corresponding response curve of the camera is proposed.

According to the analysis and prediction of engineering cost, a BIM-aided analysis method based on GA network model is proposed. First, we improve the neural network by genetic algorithm; second, according to the engineering feature vector, BIM software is used to train the GA network model; finally, the GA network model reaches a steady state, given prediction of Engineering cost. According to the experimental study of 20 high-rise residential buildings in YJW area, the experimental results show that the proposed GA model combined with BIM auxiliary analysis method can accurately and easily complete the project cost prediction.