Journal of Intelligent & Fuzzy Systems - Volume 38, issue 4

Authors: Guo, Rui | Shen, Xuanjing | Kang, Hui

Article Type: Research Article

Abstract: Image segmentation is very important for various fields. With the development of computer technology, computer technology has become more and more effective for image segmentation, and it is studied on the basis of partial differential equations. The curve representation method in plane differential geometry is expounded, with the SegNet-v2 segmentation model analyzed and tested in medical image segmentation. The test results show that the partial differential equation image segmentation algorithm can achieve more accurate segmentation, especially in medical image segmentation, which can achieve good results, and it is worth in practice to further promote.

Keywords: Partial differential equation, image segmentation, algorithm analysis

DOI: 10.3233/JIFS-179614

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3903-3909, 2020

Authors: Kung, Wang

Article Type: Research Article

Abstract: In order to improve the signal processing technology of contact image sensor, the optimization of image processing technology is carried out from the aspects of feature extraction, fine-grained semantic feature generation and semantic analysis matching optimization. To solve the problem of inaccurate feature extraction, a multi-scale feature representation algorithm for food images is proposed. By extracting the features of multi-scale convolution layer, and according to the food image, the feature extraction process is dispersed into each convolution process. By comparing the features of the layers with the image library, the most accurate features are selected for transmission. To solve the …problem of conservative vocabulary and poor generalization performance of generated sentences, a fine-grained image semantic analysis algorithm based on subword segmentation is proposed. The results show that compared with the mainstream methods, the proposed method has improved in varying degrees on the four evaluation indicators. The research provides a reference for the optimization of image sensor signal processing technology and the wide application of BP neural network algorithm. Show more

Keywords: BP neural network algorithms, contact image, sensors, signal processing technology

DOI: 10.3233/JIFS-179616

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3911-3919, 2020

Authors: Yi, Bo | Cao, Yuan Ping | Song, Ying

Article Type: Research Article

Abstract: With the rapid development of information science and technology, network security has occupied a very important position in people’s lives. Since the network security situation problem does not form a unified optimal solution in the model and algorithm, it is still necessary for researchers to continue to explore. In order to better evaluate the network security risk, based on fuzzy theory, particle swarm optimization and RBF neural network, this paper proposes a network security risk assessment model based on fuzzy theory. By mining the rules in the historical data of the network security situation and combining with the current network …status, the assessment of the current network security situation is realized, and the objectivity and comprehensibility of the evaluation results are improved. The experimental comparison shows that the fuzzy theory prediction model with PSO-RBF neural network has more rapid and effective evaluation and prediction results than the fuzzy theory prediction model with RBF neural network only. Show more

Keywords: Cyber security risk assessment, fuzzy theory

DOI: 10.3233/JIFS-179617

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3921-3928, 2020

Authors: Wang, Jing

Article Type: Research Article

Abstract: In recent years, technology of face recognition has developed rapidly, more and more face recognition technologies have been integrated into our work and life. In practical applications, due to influence of various factors, the resolution of the face image is low, the noise interference is large, and the illumination changes sharply during the imaging process, which brings difficulties to the face recognition, which seriously affects the accuracy of the face recognition method. This paper aims to introduce two-type fuzzy theory into face recognition and study its extraction and recognition methods of face feature. Firstly, it introduce the face recognition technology …simply. Face recognition is a technique that uses a computer to analyze a face image and extract valid identification information to identify the identity. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two methods for extracting features from face recognition. Principal Component Analysis (PCA) is a data analysis method that uses a small number of characterizations to reduce the number of dimensions, which reduces computational complexity greatly. The purpose of linear discriminant analysis is to extract data from high-dimensional feature spaces. Extracted the low-dimensional features with recognition ability, and studied the two-type fuzzy system based on fuzzy sets deeply. Obtained the output function of the two-type fuzzy system by studying the structure of each layer of the two-type fuzzy system. Introduce two types of fuzzy ideas into linear discriminant analysis. Discussed the construction of fuzzy membership functions, the selection of kernel functions and the determination of clustering rules. Finally, the ORL face database of the trained fuzzy face recognition model. As a result, the face recognition method based on the type 2 fuzzy has certain feasibility. The experimental results show that face recognition based on interval two-type fuzzy neural network has good recognition rate and anti-noise ability. Show more

Keywords: Type 2 fuzzy rules, linear identification, face recognition, feature extraction

DOI: 10.3233/JIFS-179618

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3929-3938, 2020

Authors: Lyu, Yi | Jiang, YiJie

Article Type: Research Article

Abstract: The purpose of this paper is to accurately locate the fault prediction and diagnosis technology, to have a high degree of automation, and to handle it quickly, for the large aircraft avionics system failure presents the feature of multiple coupling, complex impact and rapid spread. At the same time, the fault prediction diagnosis technology is one of the most important contents of the avionics system equipment prediction, so how to quickly and effectively predict the failure of key system parts of avionics is the core essential to ensure the complete operation of the whole system. This paper through establishing the …gray neural network model, combining the advantages of gray model to deal with poor information and the characteristics of artificial neural network processing nonlinear data, to realize the fault prediction of avionics system, At the same time, At the same time, through the fuzzy recognition method based on the deterioration degree, established the bridge between the two, in turn, to achieve the health prediction management of system. The method mainly includes: Firstly, by combining gray theory and artificial neural network algorithm with fuzzy recognition to establish a network model that contains gray neural network models and can reflect the excellent characteristics of fuzzy recognition and conduct experimental analysis; Second, on this basis, improve the weight update strategy of the gray neural network by using additional learning rate method which based on momentum and improve the accuracy of the algorithm. Therefore, it can be concluded that the predictions presented in this paper should not be directly imitated when the system disturbance factor is too large or the system is abnormally caused by a serious disturbance suddenly appearing at a certain point in time, but should properly processed the data firstly according to the actual situation. According to the time series of the actual situation, several models are established, and the data correction is explained from the model prediction effect, and the gray model and description are improved. The improved combination of gray neural network and gray neural network can not only improve the prediction accuracy, but also provide a feasible method for such time series prediction, which provides a practical and effective technical method for avionics system fault prediction. Show more

Keywords: Ashy neural network, avionics system, fuzzy recognition, fault prediction, combined forecast

DOI: 10.3233/JIFS-179619

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3939-3947, 2020

Authors: Liu, Yanbing | Dhakal, Sanjev | Hao, Binyao

Article Type: Research Article

Abstract: The coal-rock interface identification function enables the shearer to automatically identify the coal-rock interface and demonstrates outstanding advantages in improving economic efficiency and safe operation. It can improve the recovery rate of coal seam, reduce the content of rock, ash and sulfur in coal, improve the efficiency of coal mining operation and reduce equipment wear. It is one of the key equipments to realize coal mining automation. At present, there are more and more researchers on the research of coal rock interface identification technology. A common method is to use a single sensor to establish a coal rock identification system, …and use the neural network algorithm as the core algorithm of the system. Therefore, this paper proposes a recognition system based on wavelet packet decomposition and fuzzy neural network. A variety of sensors are used to collect the response signal of the shearer, and then the multi-signal feature extraction and data fusion of the coal-rock interface identification method are realized, thereby improving the recognition rate. On the basis of the physical simulation system of coal and rock interface, a large number of tests were carried out, and a large amount of test data was collected through experiments. In view of the many advantages of wavelet analysis, this paper uses wavelet packet technology to extract signal features. An energy allocation method based on wavelet packet decomposition can determine the sensitive frequency band of each sensor signal and extract each feature value. The wavelet packet energy method is used for feature extraction, which completes the conversion from mode space to feature space, and provides reliable and accurate feature level data for data fusion. The results show that neural networks and genetic neural networks can be trained and simulated using experimental data. Data fusion based on genetic neural network can perform state recognition and has high recognition accuracy. Multi-sensor data fusion technology based on genetic neural network is feasible in coal-rock interface identification. Show more

Keywords: Multi-sensor, coal-rock interface identification, fuzzy neural network, wavelet packet decomposition

DOI: 10.3233/JIFS-179620

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3949-3959, 2020

Authors: Fan, Linyuan

Article Type: Research Article

Abstract: Multiple sensor information fusion technology originated from the military. It has developed into a very active and popular field of defense research. It is also a high-level universal key technology, which has attracted attention in many disciplines and fields. Many countries, including China, have listed them on the research list. The purpose of this paper is to study multiple sensor data fusion through fuzzy sets and statistical theory. According to the scholars’ research results at home and abroad, proposed a new evidence synthesis algorithm. The algorithm combines the advantages of modifying the original evidence and modifying the comprehensive rules. By …comparing the accuracy of 100 sensor data in multiple sensors and single sensors, the consistency information and conflict information between the evidences are, mined and comprehensively considered the consistency information and conflict information between the evidences,, and analyzed the evidence empirically. Considered ENCE in the weight distribution of conflict evidence fully. By comparing the results of multiple sets of experiments, established multiple sensor data fusion algorithm based on fuzzy sets and statistical theory. Record experimental data and analyze the experimental results. The experimental results show that compared with other methods, the method can reflect the credibility of the evidence more objectively, the convergence speed is faster, and the fusion result is more in line with the actual situation. The experimental results show that the fusion algorithm based on the best fusion set and the new integrated method of conflict evidence can be used as the core algorithm of local fusion center and global fusion center in the fusion system respectively, and can also be used as the secondary fusion model of the fusion system. Show more

Keywords: Multiple sensor data fusion, weight distribution, fusion precision, convergence speed

DOI: 10.3233/JIFS-179621

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3961-3970, 2020

Authors: Yuan, Yana | Chai, Huaqi

Article Type: Research Article

Abstract: The discussion of knowledge management and technological innovation has never stopped, and the discussion of the relationship between the two has not only important practical significance but also profound theoretical significance. The purpose of this paper is to study a new method of knowledge fusion from the perspective of science and technology philosophy. From a new perspective, this paper analyzes defects of subjective tendencies, decision-dependent partial attributes of knowledge source existed in practical applications of management field and proposed a knowledge fusion method based on fuzzy set theory. This paper firstly explains the characteristics of knowledge sources, and transforms knowledge …into a new knowledge layer through combining multi-source knowledge, then improves the connotation, level and self-confidence of knowledge, finally improves the ability of the system to accomplish tasks and goals. Then combine the fuzzy set theory with the knowledge fusion algorithm reasonably and effectively, and obtain the results of knowledge fusion by using evidence synthesis and decision rules, so as to make up for the lack and defects in the knowledge fusion process and solve the uncertainty problem in knowledge reasoning. Finally, through the practical example, merged the fuzzy set theory proposed in this paper into knowledge fusion to deal it, obtain a kind of processing of fuzzy set theory, forming a knowledge fusion method based on fuzzy set theory. Based on fuzzy set theory, obtain the observation results of knowledge fusion algorithm combined with the various warning models, then to discuss and analyze the enterprise warning problem deeply. Therefore, the examples and simulation results show that the advantages in practicality and versatility of knowledge fusion method proposed through fuzzy set theory is higher than the common knowledge fusion method. The method used in the production of manufacturing products can help manufacturing companies improve development quality of product and shorten development cycle of product. Moreover, in the product design industry, it has verified that knowledge fusion can promote the dissemination of knowledge in the field of knowledge management, which helps to share and reuse design knowledge, reduce difficulty of development and improve efficiency of development. Show more

Keywords: Knowledge management, information fusion, fuzzy set theory, fuzzy petri net, knowledge update

DOI: 10.3233/JIFS-179622

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3971-3979, 2020

Authors: Wang, Renqiang | Miao, Keyin | Sun, Jianming

Article Type: Research Article

Abstract: In view of the objectively ambiguous feature of infrared image of unmanned autonomous ship, this paper presents a quantitative method to deal with the ambiguity problem in infrared image by using the fuzzy mathematical model to realize the purpose of intelligent recognition of infrared imaging target. In order to simplify the computation of target recognition and improve the response time and accuracy in the selection of target features in infrared images, three features of target location, radiation distribution and shape are selected for analysis in this paper. The membership functions of these three features are weighted to calculate the confidence, …and the classification and recognition are realized according to the confidence. Finally, the simulation results show that the recognition method proposed in this paper can effectively identify the target, and the recognition rate is very high. Compared with the recognition methods based on neural network and SVM, the recognition distance of this method is longer than that of the latter two methods. Show more

Keywords: Fuzzy mathematical model, unmanned autonomous ship, infrared imaging, intelligent target recognition

DOI: 10.3233/JIFS-179623

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3981-3989, 2020

Authors: Wang, Wei | Hu, Xiaohui | Wang, Mingye

Article Type: Research Article

Abstract: With the development of Internet technology, the growth of network services is accelerating. For more and more network service requests, how to ensure the response speed and query accuracy required by users is a huge challenge. In order to realize fast clustering of large data business request data and improve the accuracy of clustering. This paper presents a data fuzzy clustering algorithm based on Adaptive Incremental learning time series. The algorithm defines large data clustering in time series, and the incremental time series clustering method is used. Firstly, the complexity of network data is reduced by data compression, and then …time series data clustering based on service time similarity is carried out. In this paper, the time series fuzzy clustering algorithm based on Adaptive Incremental Learning inherits the clustering structure information obtained by previous clustering. Initialize the current clustering process, and then search the outlier samples in the current data block adaptively without setting parameters. Automatically create new clusters from outlier samples, and finally check empty cluster recognition. Identification determines whether certain clusters need to be deleted to ensure the efficiency of subsequent cluster processes. The experimental results show that the algorithm has good clustering accuracy and efficiency for isochronous and unequal time series. Show more

Keywords: Network data, adaptive incremental learning, time series, fuzzy clustering algorithm

DOI: 10.3233/JIFS-179624

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3991-3998, 2020