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

Article Type: Other

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3527-3527, 2020

Authors: Yang, Xinfeng | Hu, Qiping | Li, Shuaihao

Article Type: Research Article

Abstract: With the development of technology, fingerprint identification has become an effective means of personal identification, and has been widely used in the fields of public security, custom, banking, network security and other areas requiring identification. Nowadays, many effective methods have been proposed for fingerprint identification, but these methods are not effective in identifying damaged fingerprints, and the correct recognition rate is low. In order to effectively solve the problem of identification and classification of damaged fingerprints, this paper proposes a method for classification of broken fingerprints based on deep learning fuzzy theory. Firstly, after pre-processing the fingerprint, using the bifurcation …point and the endpoint in the broken fingerprint image as the minutiae, the feature extraction ability of the deep convolutional neural network is utilized to extract the feature of the damaged fingerprint minutiae. Secondly, the fuzzy rough set is used to reduce the feature. Finally, using the reduced feature uses the Softmax classifier to classify the damaged fingerprint image. The simulation results show that, after preprocessing the damaged fingerprint image, using OPTA algorithm to refine the damaged fingerprint image, the features of the fingerprint image can be extracted effectively by deep convolutional neural network, and then the classification accuracy can be improved by using Softmax classifier to reduce the features. Show more

Keywords: Damaged fingerprint recognition, deep learning, OPTA algorithm, deep convolutional neural network, softmax classifier

DOI: 10.3233/JIFS-179575

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3529-3537, 2020

Authors: Yang, Xinfeng | Hu, Qiping | Li, Shuaihao

Article Type: Research Article

Abstract: With the development of society, health has attracted more and more attention. Heart disease is a common and frequently occurring disease, and it is fatal. Rapid and timely diagnosis and treatment of heart disease is very important. Electrocardiogram (ECG) reflects human heart health and is widely used in heart disease examination. Existing methods depending on doctors’ personal experience and diagnostic level are time-consuming and inefficient. Therefore, a classification method that can automatically analyze ECG is required. Aiming at the classification of 12-lead ECG, based on the good performance of convolution neural network, this paper proposes a method of ECG classification …based on lead convolution neural network, which can effectively and accurately detect, recognize and classify ECG. First, the image features are extracted after the ECG is preprocessed, and then using the fuzzy set reduces the extracted ECG image features. Then, residual learning is used to optimize the convolutional neural network, and in order to ensure that the network is easy to train and fast convergence, a random parameter initialization method is introduced to achieve better classification results. The simulation results show that the proposed multi-lead filtering algorithm reduces the loss of useful information while eliminating noise; at the same time, the convolution neural network can effectively and accurately classify ECG images; and the introduction of residual network can improve the classification effect. Show more

Keywords: Electrocardiogram, 12 lead, convolutional neural network, multi lead filter, residual learning

DOI: 10.3233/JIFS-179576

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3539-3548, 2020

Authors: He, Tao | He, Miao

Article Type: Research Article

Abstract:  With the establishment and development of technologies and theories such as computer technology, image processing, pattern recognition and artificial intelligence, image analysis systems have gradually become one of the methods of automatic quantitative analysis and testing in the medical field. However, the current technology is limited to the objectivity and comprehensiveness of blood edge detection, and no detection method with high accuracy can be found. In order to accurately and effectively detect the blood color of the color ultrasound image, this paper classifies the image feature extraction method, and simulates the classical differential algorithm, mathematical morphology algorithm and fuzzy Pal.King. …In the following, the classical canny algorithm and its improved algorithm and the improved fuzzy Pal.King algorithm are introduced in detail. Finally, the simulation results are obtained. Among the indicators tested, DD has the highest accuracy. In the curve analysis, the FI value of FIB was > 0.05, the area under the curve of FDP was 67.9%, the sensitivity was 64%, and the specificity was 59%. In this paper, the quantitative analysis of the image feature extraction effect is given by the calculation results, and the subjective and objective unity is achieved. At the same time, the improved algorithm proposed in this paper is applied to the evaluation system for analysis and summary, and the results obtained are consistent with the theoretical analysis. Show more

Keywords: Fuzzy algorithm, color ultrasound image

DOI: 10.3233/JIFS-179577

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3549-3556, 2020

Authors: Wang, Shuqiang | Liang, Shuo | Peng, Fei

Article Type: Research Article

Abstract: Edge detection is image processing, analysis and one of the most important areas of research in the field of computer vision; it is the basic tools of pattern recognition and image information extraction. Actual image processing is generally mixed with noise. How to eliminate the false edge caused by noise interference and ensure the accuracy of edge positioning, it becomes an important problem to be solved in edge detection and is also the purpose of this paper. Firstly, a histogram matching image enhancement algorithm based on maximum fuzzy entropy dynamic improvement is proposed. The algorithm first maps gray scale images …from the spatial domain to the fuzzy domain, the target image is divided into several gray layers based on maximum fuzzy entropy. And then, for the characteristics of different gray levels, histogram matching method is used to design corresponding matching function for each gray layer. These matching functions are used to enhance the corresponding gray layer to obtain the enhanced image. Image enhancement method combines fuzzy entropy and histogram matching algorithm, it can effectively suppress noise and improve image contrast ratio. Secondly, an image edge detection algorithm based on improved fuzzy theory is proposed. This algorithm uses the improved fuzzy enhancement algorithm to enhance the original image. The non-maximum suppression algorithm is used to process the enhanced image; the optimal threshold value is obtained by fuzzy extraction and maximum inter-group variance method. This algorithm is used for edge detection of image. Experiments show that the algorithm is feasible and effective, and has some advantages. Show more

Keywords: Edge detection, fuzzy set, histogram algorithm, image processing

DOI: 10.3233/JIFS-179578

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3557-3566, 2020

Authors: Yu, Changqing | Wang, Liguang | Zhao, Jiong | Hao, Li | Shen, Yafeng

Article Type: Research Article

Abstract: With the development of modern remote sensing technology, remote sensing images have become one of the powerful tools for people to understand the Earth and its surroundings. However, there is currently no good classification algorithm that can accurately classify images. In order to accurately classify remote sensing images, this paper studies the content of the article by using fuzzy C-means clustering algorithm and radial basis neural network (RBF). The classification accuracy of SIRI-WHU dataset was analyzed by using the classification accuracy evaluation index such as overall accuracy and Kappa coefficient. The Kappa coefficient of vegetation classification in SIRI-WHU dataset was …0.9678, and the overall accuracy reached 97.18%. According to the classification problem of remote sensing image, according to the characteristics of remote sensing image, the improved model Alex Net-10-FCM is used to classify the remote sensing image dataset, and very high classification accuracy is obtained. Show more

Keywords: Remote sensing image classification, fuzzy C-means clustering algorithm, Kappa coefficient, data set, RBP neural network

DOI: 10.3233/JIFS-179579

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3567-3574, 2020

Authors: Zhang, Le | Wang, Jinsong | An, Zhiyong

Article Type: Research Article

Abstract: The purpose of image segmentation is to select the target region from the existing image, which is the core technology for image understanding, description and analysis. When faced with some complicated problems, the image segmentation effect of the traditional method is often unsatisfactory. As a branch of the swarm intelligence optimization algorithm, Particle Swarm Optimization (PSO) provides a new power and direction for the development of image segmentation. However, the algorithm has a large probability of loss of particle diversity in the late stage, which makes the algorithm converge prematurely. Therefore, the purpose of this paper is to improve the …problem existing in the PSO algorithm and apply the improved algorithm in image segmentation. In this paper, the whole population of PSO algorithm is divided into multiple sub-populations and co-evolution. The mutation operation from the genetic algorithm is introduced at the same time. The worst sub-population is mutated according to the mutation probability. The larger inertia factor is selected to speed the particles. Update, and then carry out simulation experiments on some classical test functions. Finally, combined with the improved PSO algorithm and fuzzy C-means clustering algorithm (FCM), the fuzzy clustering validity index is introduced, and the blood cell image is segmented by the algorithm. The experimental results show that the algorithm can find a reasonable number of cluster center segmentation categories and efficiently perform adaptive segmentation of images. Show more

Keywords: Fractional particle swarm, maximum entropy, c-means clustering algorithm, image segmentation

DOI: 10.3233/JIFS-179580

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3575-3584, 2020

Authors: Lu, Chunfeng

Article Type: Research Article

Abstract: The moving target tracking is a very important in computer vision research topic and is widely used, it is of great research significance, has been widely used, and involved in image recognition, production automation, intelligent pattern recognition, artificial intelligence, weather information, and other fields, but in some high speed movement, such as target under complex background target trajectory when not much target tracking is still more difficult. This paper mainly studies the Kalman tracking algorithm of ping-pong robot based on fuzzy real-time image. For table tennis high-speed motion blurred images, air resistance and the camera imaging distortion caused by factors …such as the error problem, puts forward an adaptive measurement covariance discrete Kalman trajectory estimation algorithm. The algorithm with dynamic adjustment of measuring the size of the covariance, has realized the accurate tracking of the target motion trajectory, and further laid the groundwork for table tennis balls prediction and hitting arm. The experimental results show that the algorithm can effectively overcome the interference of measurement noise and data loss and give excellent tracking results when the image acquisition rate is higher than 70 frames /s and the table tennis speed is higher than 5 m/s. Show more

Keywords: Fuzzy ecognition, ping-pong robot, real-time tracking, image recognition, Kalman filtering

DOI: 10.3233/JIFS-179581

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3585-3594, 2020

Authors: Yin, Deshuai | Hou, Rui | Du, Junchao | Chang, Liang | Yue, Hongxuan | Wang, Liusheng | Liu, Jiayue

Article Type: Research Article

Abstract: OBJECTIVE: The purpose of this study is to realize the precise detection of Synthetic Aperture Radar (SAR) image changes. METHODS: In this study, an intuitionistic fuzzy C-means clustering algorithm is used to accurately detect the target changes in SAR images. The change of SAR image is detected by the constructed intuitionistic fuzzy C-means clustering algorithm. Then, the effect of intuitionistic fuzzy C-means clustering algorithm, block principal component analysis (PCA) and logarithmic ratio method is compared and analyzed in the aspects of stability, accuracy, image extraction, restoration, error and work efficiency of the algorithm. RESULTS: Compared with …block PCA and logarithmic ratio methods, intuitionistic fuzzy C-means clustering algorithm has obvious advantages in stability, with standard deviation of 0.010 and other two algorithms of 0.014 and 0.017. In terms of detection accuracy and error, the algorithm in this study also has a good performance, and the detection accuracy can reach 92.4%. In addition, the intuitionistic fuzzy C-means clustering algorithm is clear and efficient for SAR image target extraction and restoration. Compared with the other two algorithms, the algorithm in this study improves by at least 20% in operation speed. There is no significant difference in the detection results of the proposed algorithm for SAR images with different targets, such as objects, people, geographical environment, etc. CONCLUSION: In this study, based on intuitionistic fuzzy C-means clustering algorithm, target changes in SAR images are detected, and the operation of the algorithm is studied. The algorithm used in this study shows a relatively comprehensive and good result, and also shows that the algorithm is a comprehensive result, which requires a good operation at many levels. This research greatly improves the recognition of intuitionistic fuzzy C-means clustering algorithm and SAR image. Show more

Keywords: Algorithms, models, images, SAR, change

DOI: 10.3233/JIFS-179582

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3595-3604, 2020

Authors: Xu, Huan Chun | Hou, Rui | Liu, Lan | Cai, Jiao Yong | Chen, Ji Gang | Liu, Jia Yue

Article Type: Research Article

Abstract: The conventional image segmentation algorithm of the colorimetric sensor array is inefficient and vulnerable to the interferences of the environment. Therefore, in order to improve the conventional algorithm, an image segmentation algorithm based on fuzzy C-means clustering (FCM) algorithm is proposed in this study. Through the information of the gray-scale distribution histogram, the proposed algorithm divides the different wave-peak regions, where the pixels are relatively concentrated, into different clusters to determine the number of clusters. In addition, the gray values of these clusters are calculated to determine the initial cluster center. Next, the calculation results are used as the input …of the FCM algorithm to complete the clustering segmentation of FCM. The research results show that the algorithm proposed in this study avoids the human participations of the traditional FCM algorithm. Also, based on the original algorithm, the proposed algorithm can reduce the calculation iterations, thereby improving the computational efficiency and obtaining the number of clusters with reference significance. As the results indicate, the proposed algorithm can better describe the fuzzy information in the image, thereby avoiding the problem of classifying the pixels into one category. Besides, the exponential function is used to control the influence weight of the neighboring pixels, and the adaptive weighting of the pixel grayscale is realized to improve the calculation accuracy of pixel grayscale and realize the image segmentation. Show more

Keywords: Fuzzy C-means clustering, image segmentation, colorimetric sensor array

DOI: 10.3233/JIFS-179583

Citation: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 4, pp. 3605-3613, 2020