Development Of Pattern Recognition Algorithms Research Articles

Problem of Recognition of Strong-Earthquake-Prone Areas: a State-of-the-Art Review

Abstract—The paper addresses almost half a century’s development of pattern recognition algorithms’ application for solving the problem of determination of strong earthquake-prone areas. This approach was named Earthquake-Prone Areas (EPA). The pattern recognition algorithms applied for this purpose, the studied regions, and the methods for assessing the reliability of the obtained results including the theory of dynamic and limit recognition problems are considered. A recently developed alternative method for solving the problem by identifying the clusters of earthquake epicenters is also presented. This method is based on the approaches of the Discrete Mathematical Analysis (DMA) and implemented in the form of the algorithmic system named Formalized Clustering and Zoning (FCAZ). The comparison of the results obtained by the EPA approach and FCAZ system shows their good consistency which provides an additional argument in favor of their reliability. The possibilities for further development and joint application of the EPA approach and FCAZ system and creating, on this basis, an integrated method for systems analysis with the inclusion of artificial intelligence are outlined. In case of success, this method is expected to be used in seismic hazard assessment and planning of earthquake-resistant construction.

Sample Based Fast Adversarial Attack Method

Deep neural network (DNN) brings the rapid development of pattern recognition algorithms. However, a large number of experiments show that there are some vulnerabilities in DNNs. Though many adversarial samples generating algorithms has been proposed, most of them based on some known information of attacked model. We proposed a new fast black-box adversarial attack algorithm purely based on data samples. First, we find the key difference between different classes based on principle component analysis and calculate the difference vector. During attacking, we just drive a sample to the target class (for target adversarial) or the nearest other class (for misclassification adversarial). The minimum modification to create an target adversarial sample is obtained by bi-section line search along the difference vector from current class to target class. For misclassification adversarial attack, the minimum modification among all other classes is given. Experimental results show that the proposed algorithm generating comparable adversarial samples much fast then classical attack algorithms.

Automated interpretation of ANCA patterns - a new approach in the serology of ANCA-associated vasculitis

IntroductionIndirect immunofluorescence (IIF) employing ethanol-fixed neutrophils (ethN) is still the method of choice for assessing antineutrophil cytoplasmic antibodies (ANCA) in ANCA-associated vasculitides (AAV). However, conventional fluorescence microscopy is subjective and prone to high variability. The objective of this study was to evaluate novel pattern recognition algorithms for the standardized automated interpretation of ANCA patterns.MethodsSeventy ANCA-positive samples (20 antimyeloperoxidase ANCA, 50 antiproteinase3 ANCA) and 100 controls from healthy individuals analyzed on ethN and formalin-fixed neutrophils (formN) by IIF were used as a 'training set' for the development of pattern recognition algorithms. Sera from 342 patients ('test set') with AAV and other systemic rheumatic and infectious diseases were tested for ANCA patterns using the novel pattern recognition algorithms and conventional fluorescence microscopy.ResultsInterpretation software employing pattern recognition algorithms was developed enabling positive/negative discrimination and classification of cytoplasmic ANCA (C-ANCA) and perinuclear ANCA (P-ANCA). Comparison of visual reading of the 'test set' samples with automated interpretation revealed Cohen's kappa (κ) values of 0.955 on ethN and 0.929 on formN for positive/negative discrimination. Analysis of the 'test set' with regard to the discrimination between C-ANCA and P-ANCA patterns showed a high agreement for ethN (κ = 0.746) and formN (κ = 0.847). There was no significant difference between visual and automated interpretation regarding positive/negative discrimination on ethN and formN, as well as ANCA pattern recognition (P > 0.05, respectively).ConclusionsPattern recognition algorithms can assist in the automated interpretation of ANCA IIF. Automated reading of ethN and formN IIF patterns demonstrated high consistency with visual ANCA assessment.

Scenario-Driven Automatic Pattern Recognition in Nowcasting

The purpose of this paper is to illustrate how the construction of a knowledge-based system (KBS) to support nowcasting, can be used to guide and facilitate the development of objective pattern recognition algorithms for use with meteorological data. We believe that a KBS based on the semantic interpretation of weather data, using the concept of weather scenarios, can assist the development and use of objective algorithms for pattern recognition in two ways: 1) it focuses the development of pattern recognition algorithms on only those phenomena which are most useful to operational forecasters; 2) its top-down logic constrains when, where, and how objective algorithms should be applied. We first describe our understanding of nowcasting expertise and the use of pattern recognition (“manual”) by human forecasters. We then briefly review the current use of automatic pattern recognition in nowcasting, present the elements within a scenario and discuss a KBS architecture for using scenarios. Finally, we close by discussing the practical benefits of merging a qualitative KBS with algorithmic pattern recognition techniques.