Configurações para métodos de acesso por escaneamento

The use of access technologies for communication, based on scanning methods, enables new communication opportunities for individuals with severe motor dysfunction. One of the most commom examples of this type of technology is the single switch scanning. Single switch scanning keyboards are often used as augmentative and alternative communication devices for inidividuals with severe mobility restrictions and with compromised speech and writing. They consist of a matrix of keys and simulate the operation of a physical keyboard to write messages. One of the limitations of these systems is their low performance. Low communication rates and considerable errors ocurrence are some of the few problems that users of these devices suffers during daily use. The development and evaluation of new strategies in augmentative and alternative communication are essential to improve the communication opportunities of user who make use of such technology. Thus, this work explores different strategies to increase communication rate and reduce user’s mistakes. Computational and practical analysis were performed for the evaluation of proposed strategies.

Métodos de extração de fósforo e micronutrientes em latossolos com uso de energia ultrassônica

Assessing the soil nutrient availability to plants under lab conditions is one of the main challenges to Soil Fertility and Chemistry, due to the complex behavior and the interaction of the soil properties. Many extractant solutions associated with mechanical forms of agitation have been proposed, showing different correlations with plant growth and nutrients absorption. Using ultrasonic energy is a agitation procedure of the soil:extractant solution suspension (based on the cavitation phenomenon). It allows the establishment of relations between the amount of extracted nutrient and the ultrasonic energy level. Thus, this work aims: to evaluate the effect of cavitation intensity on the extraction of P, Zn, Cu, Mn and Fe in soil samples from five Latosols under different uses around Uberlândia and Uberaba, Minas Gerais State; to obtain extracting curves as function of ultrasonic energy levels; and to obtain an index from extracting curves to expresses the nutrient retention by the soil solid phase. A soil-solution suspension (ratio 1:10) was sonicated using a probe ultrasound equipment under different combinations of power and time: i) 30 W for 35, 70, 140 and 280 s; ii) 50 W for 21, 42, 84 and 168 s; and iii) 70 W for 15, 30, 60 and 120 s. The extractant solutions used were Mehlich-1 (for all elements), Olsen and distilled water for P. After each sonication, P concentration was quantified by molybdenum blue colorimetric method and Zn, Cu, Mn and Fe by flame atomic absorption spectrophotometry. The cavitation intensity did not affect the P extraction, only the total energy applied. The P extraction was influenced by extractant solution, decreasing as follows: Mehlich-1>Olsen>water. In cultivated Latosols, the P extraction increased linearly with ultrasonic energy, and the slope of the 1:1 linear regression reflects the P retention in the soil. The Zn and Fe extractions were influenced only by total energy applied. Mn and Cu extractions were influenced by both cavitation intensity and total ultrasonic energy. Soils containing similar amounts of P, Cu, Zn, Mn, and Fe may have a different extraction rate. Likewise, soils containing different amounts of those elements may have the same extraction rate.

Algorithms for super-resolution of images based on Sparse Representation and Manifolds

lmage super-resolution is defined as a class of techniques that enhance the spatial resolution of images. Super-resolution methods can be subdivided in single and multi image methods. This thesis focuses on developing algorithms based on mathematical theories for single image super­ resolution problems. lndeed, in arder to estimate an output image, we adopta mixed approach: i.e., we use both a dictionary of patches with sparsity constraints (typical of learning-based methods) and regularization terms (typical of reconstruction-based methods). Although the existing methods already per- form well, they do not take into account the geometry of the data to: regularize the solution, cluster data samples (samples are often clustered using algorithms with the Euclidean distance as a dissimilarity metric), learn dictionaries (they are often learned using PCA or K-SVD). Thus, state-of-the-art methods still suffer from shortcomings. In this work, we proposed three new methods to overcome these deficiencies. First, we developed SE-ASDS (a structure tensor based regularization term) in arder to improve the sharpness of edges. SE-ASDS achieves much better results than many state-of-the- art algorithms. Then, we proposed AGNN and GOC algorithms for determining a local subset of training samples from which a good local model can be computed for recon- structing a given input test sample, where we take into account the underlying geometry of the data. AGNN and GOC methods outperform spectral clustering, soft clustering, and geodesic distance based subset selection in most settings. Next, we proposed aSOB strategy which takes into account the geometry of the data and the dictionary size. The aSOB strategy outperforms both PCA and PGA methods. Finally, we combine all our methods in a unique algorithm, named G2SR. Our proposed G2SR algorithm shows better visual and quantitative results when compared to the results of state-of-the-art methods.