RESUMO

This paper explores a data augmentation approach for images of rigid bodies, particularly focusing on electrical equipment and analogous industrial objects. By leveraging manufacturer-provided datasheets containing precise equipment dimensions, we employed straightforward algorithms to generate synthetic images, permitting the expansion of the training dataset from a potentially unlimited viewpoint. In scenarios lacking genuine target images, we conducted a case study using two well-known detectors, representing two machine-learning paradigms: the Viola-Jones (VJ) and You Only Look Once (YOLO) detectors, trained exclusively on datasets featuring synthetic images as the positive examples of the target equipment, namely lightning rods and potential transformers. Performances of both detectors were assessed using real images in both visible and infrared spectra. YOLO consistently demonstrates F1 scores below 26% in both spectra, while VJ's scores lie in the interval from 38% to 61%. This performance discrepancy is discussed in view of paradigms' strengths and weaknesses, whereas the relatively high scores of at least one detector are taken as empirical evidence in favor of the proposed data augmentation approach.

RESUMO

The starting point to understanding cluster properties is the putative global minimum and all the nearby local energy minima; however, locating them is computationally expensive and difficult. The relative populations and spectroscopic properties that are a function of temperature can be approximately computed by employing statistical thermodynamics. Here, we investigate entropy-driven isomers distribution on Be6B11- clusters and the effect of temperature on their infrared spectroscopy and relative populations. We identify the vibration modes possessed by the cluster that significantly contribute to the zero-point energy. A couple of steps are considered for computing the temperature-dependent relative population: First, using a genetic algorithm coupled to density functional theory, we performed an extensive and systematic exploration of the potential/free energy surface of Be6B11- clusters to locate the putative global minimum and elucidate the low-energy structures. Second, the relative populations' temperature effects are determined by considering the thermodynamic properties and Boltzmann factors. The temperature-dependent relative populations show that the entropies and temperature are essential for determining the global minimum. We compute the temperature-dependent total infrared spectra employing the Boltzmann factor weighted sums of each isomer's infrared spectrum and find that at finite temperature, the total infrared spectrum is composed of an admixture of infrared spectra that corresponds to the spectra of the lowest-energy structure and its isomers located at higher energies. The methodology and results describe the thermal effects in the relative population and the infrared spectra.

RESUMO

The aim of this article is to present a new face recognition system based on the fusion of visible and thermal features obtained from the most current local matching descriptors by maximizing face recognition rates through the use of genetic algorithms. The article considers a comparison of the performance of the proposed fusion methodology against five current face recognition methods and classic fusion techniques used commonly in the literature. These were selected by considering their performance in face recognition. The five local matching methods and the proposed fusion methodology are evaluated using the standard visible/thermal database, the Equinox database, along with a new database, the PUCV-VTF, designed for visible-thermal studies in face recognition and described for the first time in this work. The latter is created considering visible and thermal image sensors with different real-world conditions, such as variations in illumination, facial expression, pose, occlusion, etc. The main conclusions of this article are that two variants of the proposed fusion methodology surpass current face recognition methods and the classic fusion techniques reported in the literature, attaining recognition rates of over 97% and 99% for the Equinox and PUCV-VTF databases, respectively. The fusion methodology is very robust to illumination and expression changes, as it combines thermal and visible information efficiently by using genetic algorithms, thus allowing it to choose optimal face areas where one spectrum is more representative than the other.

Assuntos

RESUMO

Velvet worm taxonomy from Santander, Colombia and thermogravimetry, differential scanning calorimetry and infrared spectroscopy of the adhesive secretion (Onychophora: Peripatidae). Onychophoran worms are terrestrial organisms that have changed relatively little since the mid-Cambrian. We collected Macroperipatus geagy in coffee plantations at "Hacienda El Roble", Santander, Colombia, and here redescribe the species based on 15 individuals. A digital three-dimensional reconstruction of the ventral side of the body indicates that the primary and accessory papillae lack a defined distribution pattern. Diagnostic characters: one main tooth and one accessory tooth in the outer jaw, and one main tooth, one accessory tooth and seven denticles in the internal jaw. Measurements: length (X=45.66 mm; SD=26.10), weight (X= 0.95 g; SD= 2.21) and number of lobopods (X= 28.13 pairs; SD= 1.30). We present a taxonomic key for six species of Macroperipatus. The liquid adhesive secretion lost 60% of the initial weight at 70 oC (solid: 13% close to 90 oC). The stability phase of the liquid secretion was within 90-280 oC (100 oC -205 oC in the solid secretion). The starting degradation temperature of the sample was 355 oC. Heat flow changes in the solid and liquid secretions were confirmed by the sample behavior during thermogravimetric analysis. The percentage of β sheets calculated by infrared spectrum was 59%. In comparison with Nephila spiders, the onychophoran secretion lost more weight and entered the phase of degradation at lower temperatures. This secondary structure of proteins gives the onychophoran adhesive secretion a tensile strength and extensibility similar to those of the silk produced by spiders for prey-capture. Rev. Biol. Trop. 57 (3): 567-588. Epub 2009 September 30.

Redescribimos taxonómicamente el gusano Macroperipatus geagy Bouvier 1899, a partir de 15 especímenes recolectados en los cafetales de hacienda El Roble, Santander, Colombia. De acuerdo con la reconstrucción en tres dimensiones de la superficie dorsal del cuerpo, las papilas primarias y accesorias de M. geagy se distribuyen sin formar un patrón determinado. Características diagnósticas: un diente accesorio y un diente principal en la mandíbula externa; y un diente principal, un diente accesorio y siete dentículos en la sierra de la mandíbula interna. Cuerpo: longitud (X=45.66; S=26.10 mm), peso (X= 0.95; S= 2.21) y pares de lobopodos (X= 28.13; S= 1.30). Se elaboró una clave para seis especies de Macroperipatus. La secreción adhesiva en estado líquido perdió 60% del peso inicial al alcanzar los 70 ºC, la muestra sólida perdió el 13% del peso cerca de los 90 ºC, la fase de estabilidad térmica de la secreción estuvo entre los 90 y 280 ºC en la secreción líquida y entre los 100 y 205 ºC en la sólida. La temperatura de inicio de la degradación de las muestras se registró a los 355 ºC. En comparación con la araña Nephila sp., el onicóforo perdió mayor porcentaje de peso y alcanzó la fase de degradación a temperaturas más bajas. Los cambios en el flujo de calor en la secreción adhesiva sólida y líquida de M. geagy fueron confirmados con el comportamiento de las muestras en el análisis de termogravimetría. La proporción de láminas beta calculado en el espectro infrarrojo de la secreción adhesiva fue del 59%. La presencia de este tipo de conformación de la estructura secundaria de las proteínas en la secreción adhesiva de M. geagy conferiría extensibilidad y resistencia a la tracción, en similitud con las fibras de seda tejidas por las arañas para depredación.

Assuntos