Fuzzy c-means （FCM） algorithm is one of the most popular methods for image segmentation. However, the standard FCM algorithm is sensitive to noise because of not taking into account the spatial information in the im...Fuzzy c-means （FCM） algorithm is one of the most popular methods for image segmentation. However, the standard FCM algorithm is sensitive to noise because of not taking into account the spatial information in the image. An improved FCM algorithm is proposed to improve the antinoise performance of FCM algorithm. The new algorithm is formulated by incorporating the spatial neighborhood information into the membership function for clustering. The distribution statistics of the neighborhood pixels and the prior probability are used to form a new membership func- tion. It is not only effective to remove the noise spots but also can reduce the misclassified pixels. Experimental results indicate that the proposed algorithm is more accurate and robust to noise than the standard FCM algorithm.展开更多
The fuzzy C-means clustering algorithm(FCM) to the fuzzy kernel C-means clustering algorithm(FKCM) to effectively perform cluster analysis on the diversiform structures are extended, such as non-hyperspherical data, d...The fuzzy C-means clustering algorithm(FCM) to the fuzzy kernel C-means clustering algorithm(FKCM) to effectively perform cluster analysis on the diversiform structures are extended, such as non-hyperspherical data, data with noise, data with mixture of heterogeneous cluster prototypes, asymmetric data, etc. Based on the Mercer kernel, FKCM clustering algorithm is derived from FCM algorithm united with kernel method. The results of experiments with the synthetic and real data show that the FKCM clustering algorithm is universality and can effectively unsupervised analyze datasets with variform structures in contrast to FCM algorithm. It is can be imagined that kernel-based clustering algorithm is one of important research direction of fuzzy clustering analysis.展开更多
针对模糊C有序均值聚类算法没有考虑图像空间信息,导致难以有效地分割含噪图像的问题,提出一种基于非局部信息和子空间的模糊C有序均值聚类(non-local information and subspace for fuzzy C-ordered means,SFCOM-NLS)算法.首先,利用图...针对模糊C有序均值聚类算法没有考虑图像空间信息,导致难以有效地分割含噪图像的问题,提出一种基于非局部信息和子空间的模糊C有序均值聚类(non-local information and subspace for fuzzy C-ordered means,SFCOM-NLS)算法.首先,利用图像中给定的相似邻域结构的像素提取当前像素的非局部空间信息;其次,计算每个像素的典型性,并对其进行排序,在每次迭代中更新像素的典型性,提高像素聚类的准确性,解决在聚类过程中存在相似类导致的误分类问题;最后,引入子空间聚类概念,为图像不同维度分配适当的权重,提高彩色图像的分割性能.在含噪合成图像和公开数据集BSDS500,MSRA100和AID上实验结果表明,所提算法的模糊划分系数、模糊划分熵、分割精度和标准化互信息平均值分别达到了95.00%,6.66%,98.77%和95.54%,均优于对比的同类算法.展开更多
提出CF-WFCM算法,该算法分为属性权重学习算法和聚类算法两部分.属性权重学习算法,从数据自身的相似性出发,通过梯度递减算法极小化属性评价函数CFuzziness(w),为每个属性赋予一个权重.将属性权重应用于Fuzzy C Mean聚类算法,得到CF-WFC...提出CF-WFCM算法,该算法分为属性权重学习算法和聚类算法两部分.属性权重学习算法,从数据自身的相似性出发,通过梯度递减算法极小化属性评价函数CFuzziness(w),为每个属性赋予一个权重.将属性权重应用于Fuzzy C Mean聚类算法,得到CF-WFCM算法的聚类算法.CF-WFCM算法强化重要属性在聚类过程中的作用,消减冗余属性的作用,从而改善聚类的效果.我们选取了部分UCI数据库进行实验,实验结果证明:CF-WFCM算法的聚类结果优于FCM算法的聚类结果.函数CFuzziness(w)不仅可以评价属性的重要性,而且可以评价属性评价函数的优劣.实验说明了这一问题.最后我们对CF-WFCM算法进行了讨论.展开更多
The fuzzy integration evaluation method （FIEM） is studied in order to select the best orbital elements from the multi-group initial orbits determined by a satellite TT＆C （Tracking, Telemetry and Control） center w...The fuzzy integration evaluation method （FIEM） is studied in order to select the best orbital elements from the multi-group initial orbits determined by a satellite TT＆C （Tracking, Telemetry and Control） center with all kinds of data sources. By employing FIEM together with the experience of TT＆C experts, the index system to evaluate the selection of the best initial orbits is established after the data sources and orbit determination theories are studied. Besides, the concrete steps in employing the method are presented. Moreover, by taking the objects to be evaluated as evaluation experts, the problem of how to generate evaluation matrices is solved. Through practical application, the method to select the best initial orbital elements has been proved to be flexible and effective The originality of the method is to find a new evaluation criterion （comparing the actually tracked orbits） replacing the traditional one （comparing the nominal orbits） for selecting the best orbital elements.展开更多
The main function of electronic support measure system is to detect threating signals in order to take countermeasures against them. To accomplish this objective, a process of associating each interleaved pulse with i...The main function of electronic support measure system is to detect threating signals in order to take countermeasures against them. To accomplish this objective, a process of associating each interleaved pulse with its emitter must be done. This process is termed sorting or de-interleaving. A novel point symmetry based radar sorting (PSBRS) algorithm is addressed. In order to deal with all kinds of radar signals, the symmetry measure distance is used to cluster pulses instead of the conventional Euclidean distance. The reference points of the symmetrical clusters are initialized by the alternative fuzzy c-means (AFCM) algorithm to ameliorate the effects of noise and the false sorting. Besides, the density filtering (DF) algorithm is proposed to discard the noise pulses or clutter. The performance of the algorithm is evaluated under the effects of noise and missing pulses. It has been observed that the PSBRS algorithm can cope with a large number of noise pulses and it is completely independent of missing pulses. Finally, PSBRS is compared with some benchmark algorithms, and the simulation results reveal the feasibility and efficiency of the algorithm.展开更多
基金supported by the National Natural Science Foundation of China(6087403160740430664)
文摘Fuzzy c-means （FCM） algorithm is one of the most popular methods for image segmentation. However, the standard FCM algorithm is sensitive to noise because of not taking into account the spatial information in the image. An improved FCM algorithm is proposed to improve the antinoise performance of FCM algorithm. The new algorithm is formulated by incorporating the spatial neighborhood information into the membership function for clustering. The distribution statistics of the neighborhood pixels and the prior probability are used to form a new membership func- tion. It is not only effective to remove the noise spots but also can reduce the misclassified pixels. Experimental results indicate that the proposed algorithm is more accurate and robust to noise than the standard FCM algorithm.
文摘The fuzzy C-means clustering algorithm(FCM) to the fuzzy kernel C-means clustering algorithm(FKCM) to effectively perform cluster analysis on the diversiform structures are extended, such as non-hyperspherical data, data with noise, data with mixture of heterogeneous cluster prototypes, asymmetric data, etc. Based on the Mercer kernel, FKCM clustering algorithm is derived from FCM algorithm united with kernel method. The results of experiments with the synthetic and real data show that the FKCM clustering algorithm is universality and can effectively unsupervised analyze datasets with variform structures in contrast to FCM algorithm. It is can be imagined that kernel-based clustering algorithm is one of important research direction of fuzzy clustering analysis.
文摘针对模糊C有序均值聚类算法没有考虑图像空间信息,导致难以有效地分割含噪图像的问题,提出一种基于非局部信息和子空间的模糊C有序均值聚类(non-local information and subspace for fuzzy C-ordered means,SFCOM-NLS)算法.首先,利用图像中给定的相似邻域结构的像素提取当前像素的非局部空间信息;其次,计算每个像素的典型性,并对其进行排序,在每次迭代中更新像素的典型性,提高像素聚类的准确性,解决在聚类过程中存在相似类导致的误分类问题;最后,引入子空间聚类概念,为图像不同维度分配适当的权重,提高彩色图像的分割性能.在含噪合成图像和公开数据集BSDS500,MSRA100和AID上实验结果表明,所提算法的模糊划分系数、模糊划分熵、分割精度和标准化互信息平均值分别达到了95.00%,6.66%,98.77%和95.54%,均优于对比的同类算法.
文摘提出CF-WFCM算法,该算法分为属性权重学习算法和聚类算法两部分.属性权重学习算法,从数据自身的相似性出发,通过梯度递减算法极小化属性评价函数CFuzziness(w),为每个属性赋予一个权重.将属性权重应用于Fuzzy C Mean聚类算法,得到CF-WFCM算法的聚类算法.CF-WFCM算法强化重要属性在聚类过程中的作用,消减冗余属性的作用,从而改善聚类的效果.我们选取了部分UCI数据库进行实验,实验结果证明:CF-WFCM算法的聚类结果优于FCM算法的聚类结果.函数CFuzziness(w)不仅可以评价属性的重要性,而且可以评价属性评价函数的优劣.实验说明了这一问题.最后我们对CF-WFCM算法进行了讨论.
基金This project was supported by the Evaluate Quality of Satellite TT&C Mission(C0112)
文摘The fuzzy integration evaluation method （FIEM） is studied in order to select the best orbital elements from the multi-group initial orbits determined by a satellite TT＆C （Tracking, Telemetry and Control） center with all kinds of data sources. By employing FIEM together with the experience of TT＆C experts, the index system to evaluate the selection of the best initial orbits is established after the data sources and orbit determination theories are studied. Besides, the concrete steps in employing the method are presented. Moreover, by taking the objects to be evaluated as evaluation experts, the problem of how to generate evaluation matrices is solved. Through practical application, the method to select the best initial orbital elements has been proved to be flexible and effective The originality of the method is to find a new evaluation criterion （comparing the actually tracked orbits） replacing the traditional one （comparing the nominal orbits） for selecting the best orbital elements.
基金supported by the National Natural Science Foundation of China(61172116)
文摘The main function of electronic support measure system is to detect threating signals in order to take countermeasures against them. To accomplish this objective, a process of associating each interleaved pulse with its emitter must be done. This process is termed sorting or de-interleaving. A novel point symmetry based radar sorting (PSBRS) algorithm is addressed. In order to deal with all kinds of radar signals, the symmetry measure distance is used to cluster pulses instead of the conventional Euclidean distance. The reference points of the symmetrical clusters are initialized by the alternative fuzzy c-means (AFCM) algorithm to ameliorate the effects of noise and the false sorting. Besides, the density filtering (DF) algorithm is proposed to discard the noise pulses or clutter. The performance of the algorithm is evaluated under the effects of noise and missing pulses. It has been observed that the PSBRS algorithm can cope with a large number of noise pulses and it is completely independent of missing pulses. Finally, PSBRS is compared with some benchmark algorithms, and the simulation results reveal the feasibility and efficiency of the algorithm.
