Chongqing Public Security Bureau

Chongqing, China

Chongqing Public Security Bureau

Chongqing, China
Time filter
Source Type

Zhang H.,Chongqing Medical University | Luo P.,Chongqing Public Security Bureau | Sui J.,Chongqing Medical University
Bandaoti Guangdian/Semiconductor Optoelectronics | Year: 2017

The detection technique of pupil size has great significance in medical field. The algorithms for pupil size detection was studied by using image processing technology. By mkaing pre-processing on pupil images, acquiring the image processing area, edge detection, recognizing the pupil boundary with encoding chain and coding, and improving Hough circle to fit the pupil size, the detection on pupil size can be realized. Based on the detections on normal pupil images and diseased pupil images, the system can reach a pupil fitting coverage rate of 85%, and the detection only consumes a very short time, making it can be easily applied in practice. © 2017, Editorial Office of Semiconductor Optoelectronics. All right reserved.

Jia C.,Chongqing University of Posts and Telecommunications | Wei L.,Chongqing Public Security Bureau | Wang H.,Beijing Jiaotong University
Mathematical Problems in Engineering | Year: 2015

Wavelet is able to adapt to the requirements of time-frequency signal analysis automatically and can focus on any details of the signal and then decompose the function into the representation of a series of simple basis functions. It is of theoretical and practical significance. Therefore, this paper does subdivision on track irregularity time series based on the idea of wavelet decomposition-reconstruction and tries to find the best fitting forecast model of detail signal and approximate signal obtained through track irregularity time series wavelet decomposition, respectively. On this ideology, piecewise gray-ARMA recursive based on wavelet decomposition and reconstruction (PG-ARMARWDR) and piecewise ANN-ARMA recursive based on wavelet decomposition and reconstruction (PANN-ARMARWDR) models are proposed. Comparison and analysis of two models have shown that both these models can achieve higher accuracy. © 2015 Chaolong Jia et al.

Jia C.,Chongqing University of Posts and Telecommunications | Wei L.,Chongqing Public Security Bureau | Wang H.,Beijing Jiaotong University
Computational Intelligence and Neuroscience | Year: 2014

Focusing on problems existing in track irregularity time series data quality, this paper first presents abnormal data identification, data offset correction algorithm, local outlier data identification, and noise cancellation algorithms. And then proposes track irregularity time series decomposition and reconstruction through the wavelet decomposition and reconstruction approach. Finally, the patterns and features of track irregularity standard deviation data sequence in unit sections are studied, and the changing trend of track irregularity time series is discovered and described. © 2014 Chaolong Jia et al.

Yong F.,Chongqing University | Jichuan G.,Chongqing Public Security Bureau | Junhua C.,Chongqing Public Security Bureau | Lei Z.,CAS Institute of Automation
Proceedings of 2015 IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2015 | Year: 2015

Speaker identification based on Gaussian mixture model (GMM) has achieved good performance in practical applications. However its processing speed is very slow, which greatly affects its practicability. In this paper, we propose two methods to tackle this problem. Firstly, the GMM likelihood score formula is converted into linear, and further converted to the dot product of the model. Based on these conversions, we define the speaker metric space, and three fast speaker identification algorithms are proposed on the speaker metric space. Secondly, we propose a speaker retrieval method based on the high dimensional indexing technology, to further improve the speed of the speaker identification system. Experiments on the NIST2008 dataset show that the speaker identification method based on the speaker metric space can not only greatly improve the recognition speed, but also improve the performance of the speaker identification system. And the proposed speaker retrieval algorithm does facilitate the speed further. © 2015 IEEE.

Xianghui Z.,China Information Technology Security Evaluation Center | Yong P.,China Information Technology Security Evaluation Center | Zan Z.,Chongqing Public Security Bureau | Yi J.,China Information Technology Security Evaluation Center | Yuangang Y.,China Information Technology Security Evaluation Center
Proceedings - 2015 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2015 | Year: 2015

For the question of information security vulnerabilities discovery, the parallel vulnerabilities discovery method is given based on the CAPEC, CWE, CVE and other open source database and text mining. Firstly, we can extract the association vulnerability CWE under the same attack mode, then from CWE associated with CVE based on open source database. That can help us to analyze the potential parallel relationship of the multiple vulnerabilities. Secondly, the vulnerability description information will be vectorized, so that the software system is able to intelligent processing to vulnerability data. That is different from the query based on keyword matching, analyzes the similarity between the multiple vulnerabilities according to the threshold from the training set, and computes the parallel relationship between the multiple vulnerabilities and discovery the parallel vulnerabilities. Finally, this method is correct and effective by the experimental verification and in practice. According to this method, we are able to repair other parallel vulnerabilities when finding a vulnerability is exploited. An advantage of our method is that is applied to network defense. © 2015 IEEE.

Jia C.,Beijing Jiaotong University | Xu W.,Beijing Jiaotong University | Wei L.,Chongqing Public Security Bureau | Wang H.,Beijing Jiaotong University
Mathematical Problems in Engineering | Year: 2013

Good track geometry state ensures the safe operation of the railway passenger service and freight service. Railway transportation plays an important role in the Chinese economic and social development. This paper studies track irregularity standard deviation time series data and focuses on the characteristics and trend changes of track state by applying clustering analysis. Linear recursive model and linear-ARMA model based on wavelet decomposition reconstruction are proposed, and all they offer supports for the safe management of railway transportation. © 2013 Jia Chaolong et al.

Du C.,Chongqing Public Security Bureau
Proceedings - International Conference on Natural Computation | Year: 2016

Owing to the shortcoming of local convergence of the particle swarm optimization algorithm, presenting relative distances between particles to enhance probability selection formula, an improved particle swarm optimization with immune theory is introduced. A particle updates its velocity and position not only by individual and global optima,but also by individual optima of a specific particle chosen by roulette method according to certain probability, to maintain population diversity and prevent precocity and stagnation. This method is used to solve the maximum entropy model, estimating OD matrix from traffic flows. By an experiment on a junction in Chongqing City, the results demonstrate that the particle swarm algorithm overcomes the defect of Newton method that strictly relies on initial values, and the improved particle swarm algorithm has much higher optimization capability than the basic particle swarm algorithm and the basic genetic algorithm. © 2015 IEEE.

Guo Y.D.,Central South University | Cai J.F.,Central South University | Meng F.M.,Central South University | Chang Y.F.,Central South University | And 4 more authors.
Medical and Veterinary Entomology | Year: 2012

With the development of molecular identification, there has been a great deal of discussion about the feature of the mitochondrial DNA (mtDNA) fragments. Although longer fragments may minimize stochastic variation across taxa and be more likely to reflect broader patterns of nucleotide divergence, shorter fragments have many advantages, such as quick, easy and economical. Extensive application of long mtDNA segments for species identification cannot always be achieved as a result of constraints in time and money. In the present study, a molecular identification method involving the sequencing of a 272-bp 'barcode' fragment of the mitochondrial cytochrome oxidase subunit I (COI) gene from 55 specimens, representing 7 Chinese sarcophagid species from varying populations, was evaluated. Phylogenetic analysis of the sequenced segments showed that all sarcophagid specimens were properly assigned into seven species, which indicated the possibility of separation congeneric species with the short fragments. The results of this research will be instrumental for the implementation of the Chinese Sarcophagidae database. © 2011 The Authors. Medical and Veterinary Entomology © 2011 The Royal Entomological Society.

Zhang J.,Chongqing Public Security Bureau | Tao D.-Q.,Chongqing Public Security Bureau | Zhao H.,Chongqing Medical University | Yin Z.-Y.,Chongqing Medical University
Chinese Journal of Traumatology - English Edition | Year: 2012

Objective: To investigate the expression of Caspase-3 and Hsp70 in rabbits after severe traumatic brain injury (TBI) and to explore the feasibility of its application in estimation of injury time in forensic medicine. Methods: A rabbit model of heavy TBI was developed by high velocity impact on the parietal bone with an iron stick. Totally 8 healthy adult New Zealand white rabbits were randomly divided into control group (n=2) and injury group (n=6). Four hours after injury, tissue specimens from the parietal lobe, temporal lobe, occipital lobe, cerebellum and brainstem were harvested to detect the expression of Hsp70 and Caspase-3 by immunohistochemistry. Besides, the gray values of cells positive for Hsp70 and Caspase-3 were analyzed with an image analyzer. Results: Immunohistochemistry staining demonstrated a low level of Caspase-3 and Hsp70 expression in normal control group. While in injury group, both the Caspase-3 and Hsp70 expression was significantly elevated (P<0.05). Positive cells gathered around the lesion focus. Occipital lobe and cerebellum had fewer positive cells while temporal and brainstem had the fewest. Conclusion: The expression of Caspase-3 and Hsp70 at an early stage following severe TBI is characteristic and can be applied to estimate the time of injury.

Feng Y.,Chongqing University | Xiong Q.,Chongqing University | Shi W.,Chongqing University | Cao J.,Chongqing Public Security Bureau
Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument | Year: 2016

The identity vector (i-vector) representation based on Total Variability Model (TVM) of speech utterances is widely utilized in speaker recognition task. In this paper, a speaker feature extraction algorithm based on restricted Boltzmann machines (RBM) called RBM-i-vector is presented for speaker verification, which can be viewed as an alternative for the i-vector based on TVM. The method obtains the model parameters of RBM through training, then the linear output of the hidden layer is used to represent the speaker information conveyed in the Gaussian mixture model (GMM) supervector. The influences of different bipartite structures and modules, including Bernoulli or Gaussian distributions in RBM training, linear discriminative analysis (LDA) and etc. on the speaker verification are compared. As a novel i-vector feature extractor, comparative evaluations were conducted on the telephone dataset from the 2008 NIST speaker recognition evaluation (SRE), the result shows that the similar performance as that obtained with i-vector baseline system is achieved. Furthermore, fused with i-vector baseline system the proposed method could further improve the system performance; and the equal error rates (EER) for female and male telephone speech datasets are reduced to 6.83% and 4.73%, respectively. © 2016, Science Press. All right reserved.

Loading Chongqing Public Security Bureau collaborators
Loading Chongqing Public Security Bureau collaborators