Automatic gender detection is a process of determining the gender of a human according to the characteristic properties that represent the masculine and feminine attributes of a subject. Automatic gender detection is used in many areas such as customer behaviour analysis, robust security system construction, resource management, human-computer interaction, video games, mobile applications, neuro-marketing etc., in which manual gender detection may be not feasible. In this study, we have developed a fully automatic system that uses the 3D anthropometric measurements of human subjects for gender detection. A Kinect 3D camera was used to recognize the human posture, and body metrics are used as features for classification. To classify the gender, KNN, SVM classifiers and Neural Network were used with the parameters. A unique dataset gathered from 29 female and 31 male (a total of 60 people) participants was used in the experiment and the Leave One Out method was used as the cross-validation approach. The maximum accuracy achieved is 96.77% for SVM with an MLP kernel function.

The paper deals with frequency estimation methods of sine-wave signals for a few signal cycles and consists of two parts. The first part contains a short overview where analytical error formulae for a signal distorted by noise and harmonics are presented. These formulae are compared with other accurate equations presented previously by the authors which are even more accurate below one cycle in the measurement window. The second part contains a comparison of eight estimation methods (ESPRIT, TLS, Prony LS, a newly developed IpDFT method and four other 3-point IpDFT methods) in respect of calculation time and accuracy for an ideal sine-wave signal, signal distorted by AWGN noise and a signal distorted by harmonics. The number of signal cycles is limited from 0.1 to 3 or 5. The results enable to select the most accurate/ fastest estimation method in various measurement conditions. Parametric methods are more accurate but also much slower than IpDFT methods (up to 3000 times for the number of samples equal to 5000). The presented method is more accurate than other IpDFT methods and much faster than parametric methods, which makes it possible to use it as an alternative, especially in real-time applications.

The paper deals with multiple soft fault diagnosis of analogue circuits. A method for diagnosis of linear circuits is developed, belonging to the class of the fault verification techniques. The method employs a measurement test performed in the frequency domain, leading to the nonlinear least squares problem. To solve this problem the Powell minimization method is applied. The diagnostic method is adapted to real circumstances, taking into account deviations of fault-free parameters and measurement uncertainty. Two examples of electronic circuits encountered in practice demonstrate that the method is efficient for diagnosis of middle-sized circuits. Although the method is dedicated to linear circuits it can be adapted to multiple soft fault diagnosis of nonlinear ones. It is illustrated by an example of a CMOS circuit designed in a sub-micrometre technology.

This work shows a time-domain method for the discrimination and digitization of parameters of voltage pulses coming from optical detectors, taking into account the presence of electronic noise and afterpulsing. Our scheme is based on an FPGA-based time-to-digital converter as well as an adjustable-threshold comparator complemented with commercial elements. Here, the design, implementation and optimization of a multiphase TDC using delay lines shorter than a single clock period is also described. The performance of this signal processing system is discussed through the results from the statistical code density test, statistical distributions of measurements and information gathered from an optical detector. Unlike dual voltage threshold discriminators or constant-fraction discriminators, the proposed method uses amplitude and time information to define an adjustable discrimination window that enables the acquisition of spectra.

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.

The paper presents the properties of a strain sensor, which was made using the micro hole collapse method and operates in the configuration of a Mach-Zehnder modal interferometer with a PM-1550-01 polarization maintaining photonic crystal fibre. The sensor’s transfer curve was determined analytically. Its strain sensitivity, determined from measurements, decreases slightly with increasing wavelength and is in a range from −2.01 to −2.23 pm/µε in the wavelength range 1520–1580 nm. Based on the Fourier analysis of the wavelength spectrum of the constructed sensor, the difference of the group refractive indices of the core and the cladding of the photonic crystal fibre was determined, which are in a range from 7.45·10−3 to 1.01·10−2. The temperature sensitivity of the sensor, determined on the basis of measurements performed in a range from 23 to 60◦C, is positive and equals 5.9 pm/K.

Development of facial recognition or expression recognition algorithms requires input data to thoroughly test the performance of algorithms in various conditions. Researchers are developing various methods to face challenges like illumination, pose and expression changes, as well as facial disguises. In this paper, we propose and establish a dataset of thermal facial images, which contains a set of neutral images in various poses as well as a set of facial images with different posed expressions collected with a thermal infrared camera. Since the properties of face in the thermal domain strongly depend on time, in order to show the impact of aging, collection of the dataset has been repeated and a corresponding set of data is provided. The paper describes the measurement methodology and database structure. We present baseline results of processing using state-of-the-art facial descriptors combined with distance metrics for thermal face reidentification. Three selected local descriptors, a histogram of oriented gradients, local binary patterns and local derivative patterns are used for elementary assessment of the database. The dataset offers a wide range of capabilities – from thermal face recognition to thermal expression recognition.

Noise spectroscopy as a highly sensitive method for non-destructive diagnostics of semiconductor devices was applied to solar cells based on crystalline silicon with a view to evaluating the quality and reliability of this solar cell type. The experimental approach was used in a reverse-biased condition where the internal structure of solar cells, as well as pn-junction itself, was electrically stressed and overloaded by a strong electric field. This gave rise to a strong generation of a current noise accompanied by local thermal instabilities, especially in the defect sites. It turned out that local temperature changes could be correlated with generation of flicker noise in a wide frequency range. Furthermore, an electrical breakdown in a nonstable form also occurred in some specific local regions what created micro-plasma noise with a two-level current fluctuation in the form of a Lorentzian-like noise spectrum. The noise research was carried out on both of these phenomena in combination with the spectrally-filtered electroluminescence mapping in the visible/near-infrared spectrum range and the dark lock-in infrared thermography in the far-infrared range. Then the physical origin of the light emission from particular defects was searched by a scanning electron microscope and additionally there was performed an experimental elimination of one specific defect by the focused ion beam milling.

The void fraction is one of the most important parameters characterizing a multiphase flow. The prediction of the performance of any system operating with more than single phase relies on our knowledge and ability to measure the void fraction. In this work, a validated simulation study was performed in order to predict the void fraction independent of the flow pattern in gas-liquid two-phase flows using a gamma ray 60Co source and just one scintillation detector with the help of an artificial neural network (ANN) model of radial basis function (RBF). Three used inputs of ANN include a registered count under Compton continuum and counts under full energy peaks of 1173 and 1333 keV. The output is a void fraction percentage. Applying this methodology, the percentage of void fraction independent of the flow pattern of a gas-liquid two-phase flow was estimated with a mean relative error less than 1.17%. Although the error obtained in this study is almost close to those obtained in other similar works, only one detector was used, while in the previous studies at least two detectors were employed. Advantages of using fewer detectors are: cost reduction and system simplification.

Touch-trigger probes for CNC milling machines usually use wireless communication in the radio or IR band. Additionally they enable triggering signal filtering in order to avoid false triggers of the probe. These solutions cause a delay in trigger signal transmission from the probe to the machine tool controller. This delay creates an additional pre-travel component, which is directly proportional to the measurement speed and which is the cause of a previously observed but not explained increase of the pre-travel with the increase of the measurement speed. In the paper, a method of testing the delay time of triggering signal is described, an example of delay time testing results is presented and the previous, unexplained results of other researchers are analysed in terms of signal transmission delay.

Due to the difficulty of detecting traces of organic acid mixture in an aqueous sample and the complexity of resolving UV-Vis spectra effectively, a combinatory method based on a self-made radical electric focusing solid phase extraction (REFSPE) device, UV-Vis detection and partial least squares (PLS) calculation is proposed here. In this study, REFSPE was used to enhance the extraction process of analytes between the aqueous phase and the membrane phase to enrich the trace of mixed organic acid efficiently. Then, the analytes, which were eluted from the adsorption film by ethanol with the assistance of an ultrasonic cleaning machine, were detected with UV-Vis spectrophotometry. After that, the PLS method was introduced to solve the problem of overlapping peaks in UV-Vis spectra of mixed substances and to quantify each compound. The linearly dependent coefficients between the predicted value of the model and the actual concentration of the sample were all higher than 0.99. The limit values of detection for benzoic acid, phthalic acid and p-toluene sulfonic acid were found at 9.9 µg/L, 12.2 µg/L and 13.8 µg/L with the relative recovery values between 84.8% and 117.9%. The RSD (n = 20) values of each component are 1.17%, 1.11% and 0.86%, respectively. Therefore, the proposed combined method can determine traces of complex materials in an aqueous sample efficiently and has wonderful potential applications.

In this paper, two new sinusoidal signal frequency estimators calculated on the basis of four equally spaced signal samples are presented. These estimators are called four-point estimators. Simulation and experimental research consisting in signal frequency estimation using the invented estimators have been carried out. Simulation has also been performed for frequency tracking. The simulation research was carried out applying the MathCAD computer program that determined samples of a sinusoidal signal disturbed by Gaussian noise. In the experimental research, sinusoidal signal samples were obtained by means of a National Instruments PCI-6024E data acquisition card and an Agilent 33220A function generator. On the basis of the collected samples, the values of four-point estimators invented by the authors and, for comparison, the values of three- and four-point estimators proposed by Vizireanu were determined. Next, estimation errors of the signal frequency were determined. It has been shown that the invented estimators can estimate a signal frequency with greater accuracy.