Home      About Journal      Editorial Board      Instructions      Ethics Statement      Contact Us                   中文

Judgement and Disposal of Academic Misconduct Article Copyright Transfer Agreement Proof of Confidentiality Requirements for Electronic Version More.... Chinese Association of Automation National ResearchCenter for Intelligent Computing System Institute of Intelligent Machines,Chinese Academy of Sciences More....

2017 Vol.30 Issue.11, Published 2017-11-30 Orignal Article        Orignal Article 961 Parallel Surgery: An ACP-Based Approach for Intelligent Operations WANG Fei-Yue, ZHANG Mei, MENG Xiangbing, WANG Rong, WANG Xiao, ZHANG Zhicheng, CHEN Ling, GE Junhua, YANG Tian The ACP theory that comprises artificial societies, computational experiments, and parallel execution is playing an essential role in intelligent technology, especially in modeling and control of complex systems. The ACP theory is introduced into the medical and surgery area, and the basic framework and key techniques for parallel surgery is proposed, in which the artificial surgical systems are used for modeling the doctors and patients to express complex real surgical conditions. The computational experiments are utilized to propose and evaluate a variety of operation plans. Finally, the parallel execution is conducted via interaction to optimize the operation plan and play a key role in real-time operation guidance. This parallel surgery system integrates many technologies including rule extraction, computer graphics, virtual reality/augment reality, machine learning and knowledge automation in order to highly improve the efficiency and accuracy of surgical operations. 2017 Vol. 30 (11): 961-970 [Abstract] ( 806 ) [HTML 1KB] [ PDF 2909KB] ( 629 ) 971 Rough Set Model and Decision Research in Intuitionistic Fuzzy Information System Based on Weighted Multi-granulation JI Xia, ZHAO Peng, YAO Sheng By analyzing the limitation of the current multi-granulation intuitionistic fuzzy rough set(MGIFRS), a MGIFRS based on weighted granulations is presented in this paper. Firstly, the properties of the presented MGIFRS are analyzed. Three intuitionistic fuzzy rough sets, MGIFRS based on weighted granulations, optimistic MGIFRS and pessimistic MGIFRS, are compared to declare their relationships, and the relationship of uncertainty measurements under these three kinds of MGIFRS are also discussed. Then, the certainty factor and support factor of the decision rule are defined. A rule acquisition method is provided to make up for the shortcomings of the existing MGIFRS. Finally, an decision-making example is utilized to verify the validity of the proposed MGIFRS based on weighted granulations. 2017 Vol. 30 (11): 971-982 [Abstract] ( 523 ) [HTML 1KB] [ PDF 581KB] ( 670 ) 983 Target Recognition Algorithm for Maritime Surveillance Radars Based on Clustering and Random Reference Classifier FAN Xueman, HU Shengliang, HE Jingbo To improve the generalization ability of maritime surveillance radars in complicatedly interferential environment, a dynamic ensemble selection algorithm based on k-medoids clustering and random reference classifier(KMRRC) is proposed. Firstly, a pool of base classifiers are generated through Bagging technique. Secondly, k-medoids clustering is used to divided all the base classifiers into several clusters based on pairwise diversity metric. Then, the RRC model for each base classifier is constructed on the basis of validation dataset. Finally, the RRC model is employed to select some of the most competent classifiers from each cluster for ensemble and decision making. The parameters of KMRRC are determined by optimization experiment based on the self-built high resolution range profile(HRRP) dataset, and the performance of KMRRC is compared with nine ensemble methods and the base classification algorithm using the HRRP dataset and other seventeen UCI datasets in Java environment with a Weka stand-alone library. Besides, the influence of the diversity measures on the performance of KMRRC is further studied. The feasibility of KMMRRC in the field of target recognition for maritime surveillance radars is verified by experiments. 2017 Vol. 30 (11): 983-994 [Abstract] ( 887 ) [HTML 1KB] [ PDF 1004KB] ( 443 ) 995 Person Re-identification via Multiple Confidences Re-ranking LI Jiao, ZHANG Xiaohui, ZHU Hong, WANG Jing To improve the low recognition rate caused by the poor similarity measure in person re-identification, a method of person re-identification via multiple confidences re-ranking is proposed, and the accuracy of person re-identification is improved by evaluating the confidences of test samples. Firstly, the target samples and test samples are described by characteristics from deep learning network ResNet50, and the initial ranking is obtained based on the similarity between the target samples and test samples. Secondly, the classified sample sets are formed by samples with similar ranking, and then the cluster center of each category, the minimum, maximum and mean distance between samples and cluster center are acquired to set three confidence intervals with different confidences. Finally, Jaccard distance is used to sort the similarity between the target samples and test samples. The experimental results of three standard test datasets verify the effectiveness of the proposed algorithm. 2017 Vol. 30 (11): 995-1002 [Abstract] ( 788 ) [HTML 1KB] [ PDF 727KB] ( 579 ) 1003 Geography Image Similarity Measurement Method Based on Adaptive Weighting of Similarity Matrix LI Qin, YOU Xiong, LI Ke, TANG Fen Image similarity measurement is crucial to many vision applications. A similarity measurement method based on adaptive weighting of similarity matrix is proposed in this paper. The image is firstly divided into the same-sized patches, and the convolutional neural networks are adopted to construct the descriptor of each patch. The patch similarities are calculated to constitute the similarity matrix. The probability of image pair coming from the same place is evaluated by analyzing the data distribution in similarity matrix. And the similarity weight of each unit is calculated based on the data difference. Ultimately, the overall image similarity is determined. The experimental results show that the proposed method is more robust than the existing ones in image retrieval. Moreover, it effectively solves the loop closure detection in simultaneous localization and mapping. 2017 Vol. 30 (11): 1003-1011 [Abstract] ( 780 ) [HTML 1KB] [ PDF 1939KB] ( 647 ) 1012 Hesitant Fuzzy Graph and Its Application to Multi-attribute Decision Making ZHANG Chao, LI Deyu As an effective tool for describing indecisiveness quantitatively, hesitant fuzzy sets deal with the hesitation and the fuzziness in uncertain information simultaneously to solve multi-attribute decision making problems under the background of indecisiveness. Aiming at multi-attribute decision making problems with hesitant fuzzy attribute values, the related model and the multi-attribute decision making approach based on the fuzzy graph theory are studied. Firstly, the concept of hesitant fuzzy graph and some common operational laws are presented. Then, a general hesitant fuzzy graph-based multi-attribute decision making method is established. Finally, an illustrative example and the comparative analysis are conducted to verify the feasibility of the proposed method. 2017 Vol. 30 (11): 1012-1018 [Abstract] ( 616 ) [HTML 1KB] [ PDF 578KB] ( 471 ) 1019 Gene Markers Identification Algorithm for Detecting Colon Cancer Patients XIE Juanying, FAN Wen To detect those few informative genes with strong classification information and identify colon cancer patients as correctly as possible, an algorithm is proposed in this paper to identify the gene markers for detecting colon cancer patients. The densities and distances are defined for genes firstly. All genes are scattered in a 2D space with gene density and distance as X-axis and Y-axis, respectively. Those genes at high density peaks are selected to construct the optimal gene subset. Then, those samples only with genes in the optimal gene subset of colon dataset are clustered by DP_K-medoids clustering algorithm. The distances between genes or samples are calculated via Euclidean distance, Manhattan distance, Chebyshev distance and the cosine distance, respectively. The experimental results demonstrate that the proposed algorithm can find the optimal gene subset of colon cancer with high accuracy, sensitivity, specificity and MCC, and with a very few number of genes as well. 2017 Vol. 30 (11): 1019-1029 [Abstract] ( 532 ) [HTML 1KB] [ PDF 1217KB] ( 472 ) 1030 Radio Fingerprint Extraction Based on Marginal Fisher Deep Autoencoder HUANG Jianhang, LEI Yingke Aiming at the difficulty in radio fingerprint extraction caused by insufficient traditional training methods with small labeled samples, a deep autoencoder regularized by marginal Fisher analysis algorithm for radio fingerprint extraction is proposed. Based on deep autoencoder, the training procedure is divided into two parts, unsupervised pre-training and supervised finetuning based on marginal Fisher analysis. Firstly, the radio individual class information contained in the large amount of unlabeled samples is extracted. And the information is sent to the deep autoencoder for parameters optimization. Then, the trainable parameters are analyzed on the basis of marginal Fisher method with the assistant of labeled samples to improve the discriminant capability of fingerprint feature between radio individuals of the same model. The classification experiment is conducted on several communication radio signal datasets. The results show that the difference of radio individuals of the same model can be represented effectively by the proposed algorithm. 2017 Vol. 30 (11): 1030-1038 [Abstract] ( 568 ) [HTML 1KB] [ PDF 1173KB] ( 605 ) 1039 Personalized Query Expansion Method Based on Multiple Semantic Relationships WU Xuan, ZHOU Dong Due to the ever-increasing amount of digital contents in the internet, the traditional information retrieval technology is unable to meet the demands for high precision information of different users. In this paper, a personalized query expansion method based on multiple semantic relationships is proposed. It is used for personalized search based on social tagging systems. An tag-topic model is utilized to generate the user interesting model. Therefore, more precise semantics can be captured. The performance of the search can also be improved. Based on the user model, a personalized search method based on multiple semantic relationships from social data is further presented to select suitable expansion terms. Experiments conducted on a large social tagging dataset show that the proposed method outperforms several non-personalized methods as well as the existing personalized search methods based on social tagging systems. 2017 Vol. 30 (11): 1039-1047 [Abstract] ( 530 ) [HTML 1KB] [ PDF 686KB] ( 394 ) 1048 Rule Acquisition Algorithm for Neighborhood Multi-granularity Rough Sets Based on Maximal Granule CHEN Jingwen, MA Fumin, ZHANG Tengfei, ZENG Yonggang Granular computing based rule acquisition algorithms remedy the defects of rule acquisition algorithms to some extent. However, most of these algorithms can merely deal with categorical data. To further process the numerical or mixed data from the perspective of multi-granularity and multi-level, the neighborhood multi-granularity rough set model is adopted. Through calculating neighborhood multi-granularity condition granules and decision granules, the redundancy relation of condition granules in the process of rule acquisition is analyzed, and thus the redundant condition granules are further pruned. A rule acquisition algorithm for neighborhood multi-granularity rough set based on maximal granule is developed. The validity and superiority of the proposed algorithm are demonstrated by theoretical analysis and comparable experiments. 2017 Vol. 30 (11): 1048-1056 [Abstract] ( 490 ) [HTML 1KB] [ PDF 602KB] ( 537 )

Supervised by China Association for Science and Technology Sponsored by Chinese Association of Automation NationalResearchCenter for Intelligent Computing System Institute of Intelligent Machines, Chinese Academy of Sciences Published by Science Press

Copyright © 2010 Editorial Office of Pattern Recognition and Artificial Intelligence Address: No.350 Shushanhu Road, Hefei, Anhui Province, P.R. China Tel: 0551-65591176 Fax:0551-65591176 Email: bjb@iim.ac.cn Supported by Beijing Magtech 　Email:support@magtech.com.cn