A computationally efficient multi-domain active learning method for crop mapping using satellite image time-series

Crop mapping through classification of Satellite Image Time-Series (SITS) data can provide precious information for several agricultural applications, such as crop monitoring, yield forecasting, and crop inventory. However, several issues affect the classification performance of SITS data. As one of the most challenging problems, constituent images of time-series provide different levels of information about crops. These differences are the result of dynamic spectral responses of crops and also the variable atmospheric and sensor conditions. The second issue is the unavailability of adequate high-quality samples for training the classifier. In this study, we proposed a novel computationally efficient Multi-Domain Active Learning (MDAL) method which takes advantage of Multiple Kernel Learning (MKL) and Active Learning (AL) algorithms to address these two issues. The proposed method uses MKL algorithms to address the issues associated with different information level of the data, which generally cannot be modelled using the well-known classification algorithms. AL algorithms were also used for semi-automatic selection of training samples. However, most of the MKL algorithms are very computationally demanding. Consequently, using them in the MDAL method can dramatically increase the computational costs. Thus, in this paper, we presented the similarity-based MKL algorithms. Thanks to their low computational complexities, these algorithms are the most suitable MKL algorithms that can be used in the MDAL method. We evaluated the proposed method using two multispectral SITS datasets, acquired by RapidEye and SPOT sensors. The obtained results of the MDAL method for these datasets respectively showed 8.2% and 5.87% increase in the overall accuracy of classification as compared to the accuracy of the standard AL algorithm. The results also showed that in the case of adopting the SimpleMKL algorithm (a common MKL algorithm in the literature) the computational time of the MDAL method is 577 and 474 seconds for RapidEye and SPOT datasets, respectively. However, in the case of adopting the similarity-based MKL algorithms, these computational times respectively decreases to 4 and 2 seconds.     

Correct program parallelisations

International Journal on Software Tools for Technology Transfer - A commonly used approach to develop deterministic parallel programs is to augment a sequential program with compiler directives...     

Fast implementation of dense stereo vision algorithms on a highly parallel SIMD architecture

In this paper, we present faster than real-time implementation of a class of dense stereo vision algorithms on a low-power massively parallel SIMD architecture, the CSX700. With two cores, each with 96 Processing Elements, this SIMD architecture provides a peak computation power of 96 GFLOPS while consuming only 9 Watts, making it an excellent candidate for embedded computing applications. Exploiting full features of this architecture, we have developed schemes for an efficient parallel implementation with minimum of overhead. For the sum of squared differences (SSD) algorithm and for VGA (640 × 480) images with disparity ranges of 16 and 32, we achieve a performance of 179 and 94 frames per second (fps), respectively. For the HDTV (1,280 × 720) images with disparity ranges of 16 and 32, we achieve a performance of 67 and 35 fps, respectively. We have also implemented more accurate, and hence more computationally expensive variants of the SSD, and for most cases, particularly for VGA images, we have achieved faster than real-time performance. Our results clearly demonstrate that, by developing careful parallelization schemes, the CSX architecture can provide excellent performance and flexibility for various embedded vision applications.     

Effective elastic, piezoelectric and dielectric properties of braided fabric composites

Composites based upon 3D textile preforms have found broad structural application. This paper presents an analytical methodology for functional composites using piezoceramic fibers in a 3D braided preform. The effective elastic, piezoelectric and dielectric properties of 2-step braided composites with a polymeric matrix have been investigated. In the analytical approach, the effective properties of the braider and axial yarns of the unit cells are determined first using a 3D connectivity model. Then, the effective properties of the 2-step braided composite are predicted using an averaging technique. Results of a numerical example illustrating the variation of elastic, piezoelectric and dielectric constants with the braider yarn angle are provided. Textile preforming technique in general offers the potential of near net shape forming and 3D fiber placement. The present work provides the analytical basis for 3D piezoceramic textile composites.     

Use of sourdough to reduce phytic acid and improve zinc bioavailability of a traditional flat bread (sangak) from Iran

In this study, effects of 8 different sourdough starters prepared with Saccharomyces cerevisiae, Lactobacillus plantarum, L. acidophilus, and Leuconostoc mesenteroides were investigated on the phytic acid level and mole ratio of phytic acid to zinc in a traditional Iranian bread (sangak). Different sourdough preparations were made and incubated at 30°C for 16 h and added to the dough at 10, 20, and 30% replacement levels. Use of sourdough resulted in a decrease in phytic acid level (also in a decrease in the phytic acid to zinc mole ratio) and corresponding increase in zinc bioavailability index when compared to the commercial sangak bread. The lowest phytic acid concentration and highest zinc bioavailability index were achieved when S. cerevisiae, L. plantarum, and Leu. mesenteroides were used at 30.0% dough replacement with sourdough. This study provides awareness about the negative impacts of higher phytic acid level in the breads, which is particularly the case for sangak bread, and also provides a solution for such issue.     

Three-dimensional numerical modelling of a fast fluidized bed biomass gasifier to generate energy from wastes

The modelling of a biomass fluidized bed gasification system, one of the most effective ways to produce energy from biomass resources and wastes, has been performed in this study. The effect of the turbulence phenomena, including calculations relating to flow turbulence, chemical fuel reactions, and energy and momentum exchange between multiple solid and gas phases, has been taken into account in the current research as a novel approach. A computational fluid dynamics case study model that combines equations with comprehensive geometry has been considered. Results have been compared with published operational records of an existing power plant to validate the model. The solid particle distribution, the velocity of the mixture and gas phase, the turbulent flow viscosity ratio, and the temperature distribution in the model indicated the accuracy of the simulation performance compared with the experimental studies. The production of the molar fraction of the constituent elements of the synthesis gas has been evaluated in transient conditions. Additionally, 35 s after the process began, the system's performance was estimated, and the results indicated the average molecular weights of hydrogen, carbon monoxide, carbon dioxide, and methane are 26%, 23%, 12.5%, and 3.3%, respectively, which presented high precision with the experimental results.     

Potentials of NOX emission reduction methods in SI hydrogen engines: Simulation study

The ever increasing cost of hydrocarbon fuels and more stringent emission standards may resolve challenges in producing hydrogen and using it as an alternative fuel in industries. Internal combustion engines are well-established technology and hydrogen fuel in such engines is considered as an attractive choice in exploiting clean, efficient and renewable hydrogen energy.     

Finite element modeling of perforated PZT-polymer compositest

Perforated PZT-polymer composites with 3-1 connectivity patterns have been fabricated by drilling square holes perpendicular to the poling direction in prepoled PZT blocks and filling the holes with epoxy polymer. The influence of the separation between the holes on the dielectric and piezoelectric properties of the composite was studied. The finite element method has been used to calculate the stress distributions in this composite. The hydrostatic d_h and g_h coefficients have been calculated by summing the individual element polarization. The calculated values are compared with experimental results.     

Document registration using projective geometry

A technique for extracting filled-in information in form documents is presented. The transformation that is required to convert a filled-in form to match the master (blank form) is derived using results from projective geometry. Experimental studies with several forms indicate the proposed technique to be accurate and robust.