Journal of Intelligent & Fuzzy Systems - Volume 37, issue 4

Authors: Yi, Haibo | Chi, Ruinan | Nie, Zhe

Article Type: Research Article

Abstract: There has been a continuous and increasing interest in research and application of Wireless Sensor Networks (WSNs). Although the future of WSNs is very prospective, the security issues become more and more important because they are used for very critical applications in engineering areas. Furthermore, they are very vulnerable and thus attractive to attacks because of human-unattended deployment and their limited prices. Generally, they are protected by public key cryptosystems. Such systems depends on the difficulty of the elliptic curve discrete logarithms or integer factorizations. However, they can be attacked by Shor’s algorithm on quantum computers. Multivariate Public Key Cryptography …(MPKC) schemes are secure to attacks by quantum computers. Among such schemes, multivariate signatures use various multiplications in a finite field, which are time-consuming operations during signature generations. Thus, we focus on improving multiplications in finite fields for multivariate signatures. We propose a variant of Mastrovito multiplications based on trinomial, Special Trinomial (ST), pentanomial, Special Pentanomial (SP), Equally-spaced-polynomial (ESP), All-One-polynomial (AOP), and successive-one-polynomial (SOP). Our design is implemented on hardware and can be used to improve the implementations of multivariate signatures including UOV, enTTS and Rainbow for protecting data security in WSNs. Show more

Keywords: Multiplication, finite field, wireless sensor networks (WSNs), multivariate signature

DOI: 10.3233/JIFS-179291

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4575-4584, 2019

Authors: Xie, Qubo | Zhou, Ke | Fu, Xiao | Fan, Xiaohu

Article Type: Research Article

Abstract: Recently, many breakthroughs have been achieved in the text detection field; however, printed text detection performance remains unsatisfactory. To address this issue, this paper proposes a refined feature attention based text detection model comprising a feature attention FCNs and text instance segmentation. With the feature attention mechanism, the FCNs model is optimized effectively, which enables the network to learn more precise and accurate features. Therefore, the network can better detect noise-intensive and dense text in medical images. In addition, a centerline-based text region detection algorithm is proposed to process the output of network during text instance segmentation. This algorithm calculates …each text region according to the geometric information of the text instance; thus, it is able to process multi-oriented text instances precisely. The proposed model can be trained end-to-end and does not require post-processing operations, which greatly increases detection efficiency. The proposed model achieved excellent results on a medical text image dataset. Compared to existing text detection models, the proposed model demonstrates significantly better performance in terms of F-meatures and detection speed. Show more

Keywords: Printed text detection, attention mechanism, medical text image, image instance segmentation

DOI: 10.3233/JIFS-179292

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4585-4594, 2019

Authors: Hu, Zhiqiang | Yuan, Xiaoping

Article Type: Research Article

Abstract: A key radiation sources during the laser-cutting process is sparks, the spark jet angle is closely related to the quality of laser-cutting. A new approach to extract the angle of laser-cut spark jets in industrial metal-cutting processes is presented based on an established machine vision platform. This research demonstrated the algorithm and key theory of PCA and its use in image-processing. The laser-cutting experiment carried out on an St12 steel, the spark images captured during real-time operation are analysed by using the proposed algorithm to obtain the angle and shape of spark jets: the cutting kerf is analysed by optical …microscope to acquire the kerf width and associated features. The quantifiable link between sparks angle and kerf width is testified by tests on 1.8-mm thick St12 steel plate, and the results conform to the rules associated with the laser-cutting of metal. Show more

Keywords: Computer vision, laser-cutting, spark jet angle, quality analysis, PCA

DOI: 10.3233/JIFS-179293

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4595-4603, 2019

Authors: Yin, Xiaoqi | Qian, Jiansheng | Guo, Xingge | Fu, Chengfang | Lin, Guohua

Article Type: Research Article

Abstract: Measurement matrix is an important link which has important influence on signal sampling and reconstruction algorithm. Although the traditional random measurement matrix has good effect in reconstruction signal, the hardware implementation is difficult and it requires a lot of storage space. LDPC codes has a sparse check matrix with low density and strong orthogonality. When the RIP conditions are satisfied, the columns and rows are not correlated. In view of the problems existing in the measurement matrix, the deterministic measurement matrix based on the sparsity of photograph LDPC codes is constructed in this paper. Each submatrix is obtained through the …circular shift of other submatrices, which is easy to be implemented on hardware. Experimental results show that the sparse matrix performance of LDPC codes is improved in PSNR and dNMSE compared to the traditional methods by using the same orthogonal matching pursuit (OMP) algorithm for optimization and the same compression ratio. At the same time, it consumes less time in the reconstruction of remote sensing image, and the running speed is greatly improved, which can meet the real-time demand, it also provides an effective measurement matrix construction method for the practical application of compressed sensing theory. Show more

Keywords: Compressed sensing, sparse matrix, LDPC, photograph, orthogonal matching pursuit, wavelet transform

DOI: 10.3233/JIFS-179294

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4605-4613, 2019

Authors: Zhang, Yueling | Pu, Geguang | Zhang, Min | Y, William

Article Type: Research Article

Abstract: Deep Neural Network is an application of Big Data, and the robustness of Big Data is one of the most important issues. This paper proposes a new approach named PCD for computing adversarial examples for Deep Neural Network (DNN) and increase the robustness of Big Data. In safety-critical applications, adversarial examples are big threats to the reliability of DNNs. PCD generates adversarial examples by generating different coverage of pooling functions using gradient ascent. Among the 2707 input images, PCD generates 672 adversarial examples with L ∞ distances less than 0.3. Comparing to PGD (state-of-art tool for generating adversarial examples …with distances less than 0.3), PCD finds 1.5 times more adversarial examples than PGD (449) does. Show more

Keywords: Deep neural network, robustness, coverage, big data

DOI: 10.3233/JIFS-179295

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4615-4620, 2019

Authors: Lin, Yu | Yu, Xuejun | Glaubitt, Walther

Article Type: Research Article

Abstract: At present, distributed systems have been widely used, and a lot of research has been done on the reliability of distributed systems. However, trusted 3.0 based distributed system node trust behavior recognition research is still relatively rare. This paper focuses on the identification process and evaluation method of distributed node trust behavior. In this paper, call links and resource consumption of software behavior are obtained by burying points during the development phase. Based on the idea of “sense of words and deeds”, this paper compares the pre-acquired trust call links with the resource consumption sample data, in order to obtain …a credible real-time software behavior model. This paper will focus on the analysis and extend to the distributed environment, and conduct a trusted analysis of software behavior, which provides a new idea for software trusted behavior analysis. Show more

Keywords: Trust 3.0, distributed nodes, trusted behavior, identification model

DOI: 10.3233/JIFS-179296

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4621-4631, 2019

Authors: Li, Yan

Article Type: Research Article

Abstract: As the main tool to monitor remote networking environment, sensor network integrates various functions of data collection, data collection, data processing and information transmission information data encryption in one, which has been gradually applied in all areas of society. However, sensor networks may suffer attacks during the process of data acquisition and transmission. Therefore, data privacy protection becomes the most important problem in wireless sensor networks. In this paper, privacy protection, high precision and data fusion were studied, and a privacy preserving algorithm based on CRAE was proposed on the existing encryption algorithms. The algorithm can not only reduce the …energy consumption of sensor networks, but also have good privacy. Simulation experiments were carried out to compare the proposed algorithm with the existing encryption algorithms and achieve the desired design goal. Show more

Keywords: Privacy protection, sensor networks, data fusion, energy consumption

DOI: 10.3233/JIFS-179297

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4633-4638, 2019

Authors: Li, Yan

Article Type: Research Article

Abstract: With the development and wide application of wireless sensor networks, a data detection method based on time series is proposed to solve the problem that the sampling values of sensors vary greatly in harsh environments and the detection results of events are inaccurate with the increase of fault nodes in wireless sensor networks. The median of the normal data collected by the sensor is used to establish the reference value, and then the confidence interval is constructed. Finally, a method based on calculating the difference degree of the data interval is proposed to determine the source of the anomaly. The …experimental results show that the proposed method keeps the detection rate of abnormal data in sensor networks above 98% and the false alarm rate below 0.6%. It has certain reliability and practicability. Show more

Keywords: Anomaly detection, wireless sensor network, data buffer, confidence interval, difference

DOI: 10.3233/JIFS-179298

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4639-4645, 2019

Authors: ZHOU, Zhou | Xie, Houliang | Li, Fangmin

Article Type: Research Article

Abstract: To deal with the problem of unbalanced load and ignoring task priority in previous task scheduling algorithms, this paper proposes a task scheduling algorithm named P-Min-Min-Max, which combines with priority and greedy strategy. Under the condition of considering the task priority, the algorithm leverages the execution time of tasks for greedy strategy, and binding large task and small task in the task list to form “task pair” to perform scheduling, so as to effectively solve the problem of unbalanced load. To reduce the average response time of tasks, the algorithm priority schedules the small task from the “task pair”. The …experimental results show that, compared with the Min-Min and P-Min-Min algorithm, the proposed algorithm improves the system resource utilization and service quality of user, and saves the total execution time for tasks. Compared with Max-Min and P-Max-Min algorithm, the proposed algorithm improves the system resource utilization and service quality of user, and reduces the total completion time and average response time. Show more

Keywords: Cloud computing, greedy strategy, load balancing, priority, service quality

DOI: 10.3233/JIFS-179299

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4647-4655, 2019

Authors: Cui, Le | Cheng, Libo | Jiang, Xiaoming | Chen, Zhanfang | Albarka,

Article Type: Research Article

Abstract: Outlier detection has always been a more active research topic in statistical diagnosis. Outliers are ubiquitous at data analysis areas in current and may produce erroneous results. In multivariate linear regression model, the existence of outliers will directly affect the modeling, parameter estimation and prediction. A set of data contains abnormal values, which will have a great impact on the estimation of the mean and standard deviation of the data, and also affect the estimation results of the least squares method. In the paper, on the basis of the linear regression model and deleting model and mean shift model, from …the perspective of residual sum of squares and by introducing sample quantile to estimate the overall parameters robustly, a special statistic which is combined with the sample quantile method is used to detected the outliers. Finally, an example is analyzed and compared with the traditional method. The results show that the method is more effective. Show more

Keywords: Regression analysis, outlier, data deletion model, mean shift model, residual sum of squares, sample quantile

DOI: 10.3233/JIFS-179300

Citation: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4657-4664, 2019