Classification of water masses in the area investigated during the 1981 FIBEX Expedition and two winter expeditions at the "H. Arctowski" Station using the method of Empirical Orthogonal Functions (EOF) is presented. Four basic water masses (warm and cold Bellinghausen Sea surface waters, surface Weddell Sea waters, Circumpolar Warm Deep Water (CWDW) and the transitional zone) were observed in the area and a significant dependence of water masses distribution ón depth was found. A strong winter increase in the Weddell Sea waters influence was recorded.
This paper contains the results of phytosociological studies carried out on the model fragment of Spitsbergen tundra at Bellsund. In the area of 4800 m2 19 plant communities have been distinguished through association analysis and these communities, in turn, have been compared according to cluster analysis. Also, ecological groups of species have been distinguished.
To avoid of manipulating search engines results by web spam, anti spam system use machine learning techniques to detect spam. However, if the learning set for the system is out of date the quality of classification falls rapidly. We present the web spam recognition system that periodically refreshes the learning set to create an adequate classifier. A new classifier is trained exclusively on data collected during the last period. We have proved that such strategy is better than an incrementation of the learning set. The system solves the starting–up issues of lacks in learning set by minimisation of learning examples and utilization of external data sets. The system was tested on real data from the spam traps and common known web services: Quora, Reddit, and Stack Overflow. The test performed among ten months shows stability of the system and improvement of the results up to 60 percent at the end of the examined period.
Definition of a composite  describes an ideal composite material with perfect structure. In real composite materials, structure is usually imperfect – composites contain various types of defects [2, 3–5], especially as the casted composites are of concern. The reason for this is a specific structure of castings, related to course of the manufacturing process. In case of metal matrix composite castings, especially regarding these manufactured by saturation, there is no classification of these defects [2, 4]. Classification of defects in castings of classic materials (cast iron, cast steel, non-ferrous alloys) is insufficient and requires completion of specific defects of mentioned materials. This problem (noted during manufacturing metal matrix composite castings with saturated reinforcement in Institute of Basic Technical Sciences of Maritime University Szczecin) has become a reason of starting work aimed at creating such classification. As a result, this paper was prepared. It can contribute to improvement of quality of studied materials and, as a consequence, improve the environment protection level.
This investigation focuses on the modern Ukrainian surnames derived from a person’s appearance. The author analyzed different approaches to the systematization of such surnames that were applied in onomastic research, finding some differences in these classifications. For example, there is some controversy as to the scope (content) of this group. For example, some investigators do not include surnames derived from proper names of persons on the basis of their strength, health, clothing and so on as a part of this group surnames. On the other hand, some other researchers believe that it is necessary. Doubts have also been expressed as to the appropriacy of including some surnames which are derived from the names that described the gait of the person or their gender. Developing our approach to the classification of surnames derived from the proper names of persons on the basis of appearance, the author used the achievements of gabitology that uses characteristics of human appearance for the identification of the person. Fixed surnames were distributed according to the semantics of lexemes in the following subgroups: general physical (they can characterize the appearance of the person in general), anatomical (they can characterize body type, organs of the human body) and functional (characteristics of human appearance that are seen in motion). Surnames were also differentiated within each subgroup. For example, the author identified the surnames derived from the proper names whose meaning are connected with the features of human growth, body type (skinny build, slim build, chunky, fat), unusual shape, size, and color of organs of the human body. The important thing to note is that the namegivers focused on those characteristics of human appearance that were original, relatively constant, and helped to identify a person.
The compactness of dimension stone blocks was previously controlled through various methods that were partially based on personal experiences, acoustic and visual observance of materials. With the development of technology, the ultrasonic pulse method is frequently used for the examination of stone test pieces and with an analysis of acquired data through the tomography method, the compactness is determined. The monolith stone blocks that are found at a site contain hidden discontinuities. The technique of data acquisition and the use of various instruments enable a good overview of the block interior. With an increased number of measurements, a suitable classification is prepared that helps reduce modification costs and increases the quality of stone blocks. The control methodology of compactness is based on the passage of longitudinal waves through the stone block without damaging the block during control. High differences in speed show irregularities in the material. With the observation system, we can prepare a tomography of the measured profiles that show us the locations of irregularities that should be observed more closely. During in situ measurements, the data for comparison with measured results are acquired. Determination of critical locations is of extreme importance before the processing of the block into smaller stone products or during the reconstruction of older stone elements or sculptures. The purpose of “in situ” measurements is to prepare a simple and fast method for the evaluation of materials compactness and for production work.
The work proposes a new method for vehicle classification, which allows treating vehicles uniformly at the stage of defining the vehicle classes, as well as during the classification itself and the assessment of its correctness. The sole source of information about a vehicle is its magnetic signature normalised with respect to the amplitude and duration. The proposed method allows defining a large number (even several thousand) of classes comprising vehicles whose magnetic signatures are similar according to the assumed criterion with precisely determined degree of similarity. The decision about the degree of similarity and, consequently, about the number of classes, is taken by a user depending on the classification purpose. An additional advantage of the proposed solution is the automated defining of vehicle classes for the given degree of similarity between signatures determined by a user. Thus the human factor, which plays a significant role in currently used methods, has been removed from the classification process at the stage of defining vehicle classes. The efficiency of the proposed approach to the vehicle classification problem was demonstrated on the basis of a large set of experimental data.
Traffic classification is an important tool for network management. It reveals the source of observed network traffic and has many potential applications e.g. in Quality of Service, network security and traffic visualization. In the last decade, traffic classification evolved quickly due to the raise of peer-to-peer traffic. Nowadays, researchers still find new methods in order to withstand the rapid changes of the Internet. In this paper, we review 13 publications on traffic classification and related topics that were published during 2009-2012. We show diversity in recent algorithms and we highlight possible directions for the future research on traffic classification: relevance of multi-level classification, importance of experimental validation, and the need for common traffic datasets.
The paper presents the results of a study of cyanobacteria and green algae assemblages occurring in various tundra types determined on the basis of mosses and vascular plants and habitat conditions. The research was carried out during summer in the years 2009–2013 on the north sea−coast of Hornsund fjord (West Spitsbergen, Svalbard Archipelago). 58 sites were studied in various tundra types differing in composition of vascular plants, mosses and in trophy and humidity. 141 cyanobacteria and green algae were noted in the research area in total. Cyanobacteria and green algae flora is a significant element of many tundra types and sometimes even dominate there. Despite its importance, it has not been hitherto taken into account in the description and classification of tundra. The aim of the present study was to demonstrate the legitimacy of using phycoflora in supplementing the descriptions of hitherto described tundra and distinguishing new tundra types. Numeric hierarchical−accumulative classification (MVSP 3.1 software) methods were used to analyze the cyanobacterial and algal assemblages and their co−relations with particular tundra types. The analysis determined dominant and distinctive species in the communities in concordance with ecologically diverse types of tundra. The results show the importance of these organisms in the composition of the vegetation of tundra types and their role in the ecosystems of this part of the Arctic.
This study addresses the problem of magnetic field emission produced by the laptop computers. Although, the magnetic field is spread over the entire frequency spectrum, the most dangerous part of it to the laptop users is the frequency range from 50 to 500 Hz, commonly called the extremely low frequency magnetic field. In this frequency region the magnetic field is characterized by high peak values. To examine the influence of laptop’s magnetic field emission in the office, a specific experiment is proposed. It includes the measurement of the magnetic field at six laptop’s positions, which are in close contact to its user. The results obtained from ten different laptop computers show the extremely high emission at some positions, which are dependent on the power dissipation or bad ergonomics. Eventually, the experiment extracts these dangerous positions of magnetic field emission and suggests possible solutions.
In this paper, a new Multi-Layer Perceptron Neural Network (MLP NN) classifier is proposed for classifying sonar targets and non-targets from the acoustic backscattered signals. Besides the capabilities of MLP NNs, it uses Back Propagation (BP) and Gradient Descent (GD) for training; therefore, MLP NNs face with not only impertinent classification accuracy but also getting stuck in local minima as well as lowconvergence speed. To lift defections, this study uses Adaptive Best Mass Gravitational Search Algorithm (ABGSA) to train MLP NN. This algorithm develops marginal disadvantage of the GSA using the bestcollected masses within iterations and expediting exploitation phase. To test the proposed classifier, this algorithm along with the GSA, GD, GA, PSO and compound method (PSOGSA) via three datasets in various dimensions will be assessed. Assessed metrics include convergence speed, fail probability in local minimum and classification accuracy. Finally, as a practical application assumed network classifies sonar dataset. This dataset consists of the backscattered echoes from six different objects: four targets and two non-targets. Results indicate that the new classifier proposes better output in terms of aforementioned criteria than whole proposed benchmarks.
The aim of this study was to investigate the possible use of geoinformatics tools and generally available geodata for mapping land cover/use on the reclaimed areas. The choice of subject was dictated by the growing number of such areas and the related problem of their restoration. Modern technology, including GIS, photogrammetry and remote sensing are relevant in assessing the reclamation effects and monitoring of changes taking place on such sites. The LULC classes mapping, supported with thorough knowledge of the operator, is useful tool for the proper reclamation process evaluation. The study was performed for two post-mine sites: reclaimed external spoil heap of the sulfur mine Machów and areas after exploitation of sulfur mine Jeziórko, which are located in the Tarnobrzeski district. The research materials consisted of aerial orthophotos, which were the basis of on-screen vectorization; LANDSAT satellite images, which were used in the pixel and object based classification; and the CORINE Land Cover database as a general reference to the global maps of land cover and land use.
In general, currently employed vehicle classification algorithms based on the magnetic signature can distinguish among only a few vehicle classes. The work presents a new approach to this problem. A set of characteristic parameters measurable from the magnetic signature and limits of their uncertainty intervals are determined independently for each predefined class. The source of information on the vehicle parameters is its magnetic signature measured in a system that enables independent measurement of two signals, i.e. changes in the active and reactive component of the inductive loop impedance caused by a passing vehicle. These innovations result in high selective classification system, which utilizes over a dozen vehicle classes. The evaluation of the proposed approach was carried out for good vehicles consisting of 2-axle tractor and a 3-axle semi-trailer.
A purpose of the present study is an evaluation of various models of classification of the South branch of the Cushitic languages. The South Cushitic languages are studied in their narrow sense here, i.e. without Dahalo and Ma’a, although their probable cognates are registered.
The systematic position of Sorbus population occurring in the Pieniny Mts. is controversial. To verify its taxonomic status we studied the ITS sequence of closely related species of the S. aria group: Sorbus sp. from the Pieniny Mts., S. aria from the Tatra Mts., S. graeca from the Balkans, and other well-distinguished native Polish Sorbus species (S. aria, S. aucuparia, S. intermedia and S. torminalis). As a reference we examined Sorbus populations closest to the Pieniny Mts. where S. graeca was reported to occur, in Slovakia. The results indicate that the Sorbus plants found in the Pieniny Mts. differ genetically from those in the Tatra Mts. but are identical to those collected from the Vihorlat Mts. in Slovakia and are closely related to S. graeca from the Balkans
Accurate network fault diagnosis in smart substations is key to strengthening grid security. To solve fault classification problems and enhance classification accuracy, we propose a hybrid optimization algorithm consisting of three parts: anti-noise processing (ANP), an improved separation interval method (ISIM), and a genetic algorithm-particle swarm optimization (GA-PSO) method. ANP cleans out the outliers and noise in the dataset. ISIM uses a support vector machine (SVM) architecture to optimize SVM kernel parameters. Finally, we propose the GA-PSO algorithm, which combines the advantages of both genetic and particle swarm optimization algorithms to optimize the penalty parameter. The experimental results show that our proposed hybrid optimization algorithm enhances the classification accuracy of smart substation network faults and shows stronger performance compared with existing methods.
Five models and methodology are discussed in this paper for constructing classifiers capable of recognizing in real time the type of fuel injected into a diesel engine cylinder to accuracy acceptable in practical technical applications. Experimental research was carried out on the dynamic engine test facility. The signal of in-cylinder and in-injection line pressure in an internal combustion engine powered by mineral fuel, biodiesel or blends of these two fuel types was evaluated using the vibro-acoustic method. Computational intelligence methods such as classification trees, particle swarm optimization and random forest were applied.
Land use/land cover (LULC) maps are important datasets in various environmental projects. Our aim was to demonstrate how GEOBIA framework can be used for integrating different data sources and classification methods in context of LULC mapping.We presented multi-stage semi-automated GEOBIA classification workflow created for LULC mapping of Tuszyma Forestry Management area based on multi-source, multi-temporal and multi-resolution input data, such as 4 bands- aerial orthophoto, LiDAR-derived nDSM, Sentinel-2 multispectral satellite images and ancillary vector data. Various classification methods were applied, i.e. rule-based and Random Forest supervised classification. This approach allowed us to focus on classification of each class ‘individually’ by taking advantage from all useful information from various input data, expert knowledge, and advanced machine-learning tools. In the first step, twelve classes were assigned in two-steps rule-based classification approach either vector-based, ortho- and vector-based or orthoand Lidar-based. Then, supervised classification was performed with use of Random Forest algorithm. Three agriculture-related LULC classes with vegetation alternating conditions were assigned based on aerial orthophoto and Sentinel-2 information. For classification of 15 LULC classes we obtained 81.3% overall accuracy and kappa coefficient of 0.78. The visual evaluation and class coverage comparison showed that the generated LULC layer differs from the existing land cover maps especially in relative cover of agriculture-related classes. Generally, the created map can be considered as superior to the existing data in terms of the level of details and correspondence to actual environmental and vegetation conditions that can be observed in RS images.
We evaluated the performance of nine machine learning regression algorithms and their ensembles for sub-pixel estimation of impervious areas coverages from Landsat imagery. The accuracy of imperviousness mapping in individual time points was assessed based on RMSE, MAE and R 2 . These measures were also used for the assessment of imperviousness change intensity estimations. The applicability for detection of relevant changes in impervious areas coverages at sub-pixel level was evaluated using overall accuracy, F-measure and ROC Area Under Curve. The results proved that Cubist algorithm may be advised for Landsat-based mapping of imperviousness for single dates. Stochastic gradient boosting of regression trees (GBM) may be also considered for this purpose. However, Random Forest algorithm is endorsed for both imperviousness change detection and mapping of its intensity. In all applications the heterogeneous model ensembles performed at least as well as the best individual models or better. They may be recommended for improving the quality of sub-pixel imperviousness and imperviousness change mapping. The study revealed also limitations of the investigated methodology for detection of subtle changes of imperviousness inside the pixel. None of the tested approaches was able to reliably classify changed and non-changed pixels if the relevant change threshold was set as one or three percent. Also for five percent change threshold most of algorithms did not ensure that the accuracy of change map is higher than the accuracy of random classifier. For the threshold of relevant change set as ten percent all approaches performed satisfactory.
In this work nine non-linear regression models were compared for sub-pixel impervious surface area mapping from Landsat images. The comparison was done in three study areas both for accuracy of imperviousness coverage evaluation in individual points in time and accuracy of imperviousness change assessment. The performance of individual machine learning algorithms (Cubist, Random Forest, stochastic gradient boosting of regression trees, k-nearest neighbors regression, random k-nearest neighbors regression, Multivariate Adaptive Regression Splines, averaged neural networks, and support vector machines with polynomial and radial kernels) was also compared with the performance of heterogeneous model ensembles constructed from the best models trained using particular techniques. The results proved that in case of sub-pixel evaluation the most accurate prediction of change may not necessarily be based on the most accurate individual assessments. When single methods are considered, based on obtained results Cubist algorithm may be advised for Landsat based mapping of imperviousness for single dates. However, Random Forest may be endorsed when the most reliable evaluation of imperviousness change is the primary goal. It gave lower accuracies for individual assessments, but better prediction of change due to more correlated errors of individual predictions. Heterogeneous model ensembles performed for individual time points assessments at least as well as the best individual models. In case of imperviousness change assessment the ensembles always outperformed single model approaches. It means that it is possible to improve the accuracy of sub-pixel imperviousness change assessment using ensembles of heterogeneous non-linear regression models.
Journal bearings are the most common type of bearings in which a shaft freely rotates in a metallic sleeve. They find a lot of applications in industry, especially where extremely high loads are involved. Proper analysis of the various bearing faults and predicting the modes of failure beforehand are essential to increase the working life of the bearing. In the current study, the vibration data of a journal bearing in the healthy condition and in five different fault conditions are collected. A feature extraction method is employed to classify the different fault conditions. Automatic fault classification is performed using artificial neural networks (ANN). As the probability of a correct prediction goes down for a higher number of faults in ANN, the method is made more robust by incorporating deep neural networks (DNN) with the help of autoencoders. Training was done using the scaled conjugate gradient algorithm and the performance was calculated by the cross entropy method. Due to the increased number of hidden layers in DNN, it is possible to achieve a high efficiency of 100% with the feature extraction method.
A limited ability to discriminate between different materials is the fundamental problem with all conventional eddy-current-based metal detectors. This paper presents the use, evaluation and classification of nontraditional excitation signals for eddy-current metal detectors to improve their detection and discrimination ability. The presented multi-frequency excitation signals are as follows: a step sweep sine wave, a linear frequency sweep and sin(x)/x signals. All signals are evaluated in the frequency domain. Amplitude and phase spectra and polar graphs of the detector output signal are used for classification and discrimination of the tested objects. Four different classifiers are presented. The classification results obtained with the use of poly-harmonic signals are compared with those obtained with a classical single-tone method. Multifrequency signals provide more detailed information, due to the response function – the frequency characteristic of a detected object, than standard single-tone methods. Based on the measurements and analysis, a metal object can be better distinguished than when using a single-tone method.
Systems of road traffic parameters measurement play a key role in the process of road traffic control, its supervision as well as in gathering and processing information for statistical purposes. Expectations of users of such systems mainly concern automation and provision of measurement continuity, possibility of selection of the measured road traffic parameters and high accuracy along with reliability of obtained results. In order to meet the requirements set for such systems, at the Department of Instrumentation and Measurement of the AGH University of Science and Technology in Cracow a new prototype system of road traffic parameters measurement - Traffic-1 - has been constructed. The innovativeness of the solution is manifested in the structure of the system that can be modified by the user adequately to current measurement needs and in the used algorithms of signals processing. The work contains a brief description of the constructed system with particular focus on the used innovations that are the result of many years of research work of the designers.
The accuracy and reliability of Kalman filter are easily affected by the gross errors in observations. Although robust Kalman filter based on equivalent weight function models can reduce the impact of gross errors on filtering results, the conventional equivalent weight function models are more suitable for the observations with the same noise level. For Precise Point Positioning (PPP) with multiple types of observations that have different measuring accuracy and noise levels, the filtering results obtained with conventional robust equivalent weight function models are not the best ones. For this problem, a classification robust equivalent weight function model based on the t-inspection statistics is proposed, which has better performance than the conventional equivalent weight function models in the case of no more than one gross error in a certain type of observations. However, in the case of multiple gross errors in a certain type of observations, the performance of the conventional robust Kalman filter based on the two kinds of equivalent weight function models are barely satisfactory due to the interaction between gross errors. To address this problem, an improved classification robust Kalman filtering method is further proposed in this paper. To verify and evaluate the performance of the proposed method, simulation tests were carried out based on the GPS/BDS data and their results were compared with those obtained with the conventional robust Kalman filtering method. The results show that the improved classification robust Kalman filtering method can effectively reduce the impact of multiple gross errors on the positioning results and significantly improve the positioning accuracy and reliability of PPP.
The wavelet transform has been successfully used in the area of power quality analysis. There are many published papers with methods for power quality disturbance classification or harmonics measurement, which use wavelet transform. However, the properties of the wavelet transform can drastically vary from the choice of the wavelet. In this paper we analyze the influence of the choice of the wavelet to the accuracy of the power quality classification method and to high frequency harmonics measurements. Additionally to the well known wavelet filters we introduce near perfect reconstruction filter banks. The simulation results indicate that these filter banks are a good choice for classification of power quality disturbances, especially in the presence of noise and for high frequency harmonics measurements