An iterative outlier elimination procedure based on hypothesis testing, commonly known as Iterative Data Snooping (IDS) among geodesists, is often used for the quality control of modern measurement systems in geodesy and surveying. The test statistic associated with IDS is the extreme normalised least-squares residual. It is well-known in the literature that critical values (quantile values) of such a test statistic cannot be derived from well-known test distributions but must be computed numerically by means of Monte Carlo. This paper provides the first results on the Monte Carlo-based critical value inserted into different scenarios of correlation between outlier statistics. From the Monte Carlo evaluation, we compute the probabilities of correct identification, missed detection, wrong exclusion, over-identifications and statistical overlap associated with IDS in the presence of a single outlier. On the basis of such probability levels, we obtain the Minimal Detectable Bias (MDB) and Minimal Identifiable Bias (MIB) for cases in which IDS is in play. The MDB and MIB are sensitivity indicators for outlier detection and identification, respectively. The results show that there are circumstances in which the larger the Type I decision error (smaller critical value), the higher the rates of outlier detection but the lower the rates of outlier identification. In such a case, the larger the Type I Error, the larger the ratio between the MIB and MDB. We also highlight that an outlier becomes identifiable when the contributions of the measures to the wrong exclusion rate decline simultaneously. In this case, we verify that the effect of the correlation between outlier statistics on the wrong exclusion rate becomes insignificant for a certain outlier magnitude, which increases the probability of identification.

No abstract

Attaining reliable and timely agricultural estimates is very important everywhere, and in Brazil, due to its characteristics, this is especially true. In this study, estimations of crop production were made based on the temporal profiles of the Enhanced Vegetation Index (EVI) obtained from Moderate Resolution Imaging Spectroradiometer (MODIS) images. The objective was to evaluate the coupled model (CM) performance of crop area and crop yield estimates based solely on MODIS/EVI as input data in Rio Grande do Sul State, which is characterized by high variability in seasonal soybean yields, due to different crop development conditions. The resulting production estimates from CM were compared to official agricultural statistics of Brazilian Institute of Geography and Statistics (IBGE) and the National Company of Food Supply (CONAB) at different levels from 2000/2001 to 2010/2011 crop years. Results obtained with CM indicate that its application is able to generate timely production estimates for soybean both at municipality and local levels. Validation estimates with CM at State level obtained R 2 = 0.95. Combining all cropping years at municipality level, estimates were highly correlated to official statistics from IBGE, with R 2 = 0.91 and RMSD = 10,840 tons. Spatially interpolated comparisons of yield maps obtained from the CM estimates and IBGE data also showed visual similarity in their spatial distribution. Local level comparisons were performed and presented R 2 = 0.95. Implications of this work point out that time-series analysis of production estimates are able to provide anticipated spatial information prior to the soybean harvest.

Estimations of crop area were made based on the temporal profiles of the Enhanced Vegetation Index (EVI) obtained from moderate resolution imaging spectroradiometer (MODIS) images. Evaluation of the ability of the MODIS crop detection algorithm (MCDA) to estimate soybean crop areas was performed for fields in the Mato Grosso state, Brazil. Using the MCDA approach, soybean crop area estimations can be provided for December (first forecast) using images from the sowing period and for February (second forecast) using images from the sowing period and the maximum crop development period. The area estimates were compared to official agricultural statistics from the Brazilian Institute of Geography and Statistics (IBGE) and from the National Company of Food Supply (CONAB) at different crop levels from 2000/2001 to 2010/2011. At the municipality level, the estimates were highly correlated, with R 2 = 0.97 and RMSD = 13,142 ha. The MCDA was validated using field campaign data from the 2006/2007 crop year. The overall map accuracy was 88.25%, and the Kappa Index of Agreement was 0.765. By using pre-defined parameters, MCDA is able to provide the evolution of annual soybean maps, forecast of soybean cropping areas, and the crop area expansion in the Mato Grosso state.

Audio-visual Speech Recognition has been an active area of research lately. A bit, and yet unsolved, part of this problem is the visual only recognition, or lip reading. Considering an image sequence of a person pronouncing a word, a full image analysis solution would have to segment the mouth area, extract relevant features, and use them to be able to classify the word from those visual features. In this paper we approach this problem by proposing a segmentation technique for the lips contours together with a set of features based on the extracted contours which is able to perform lip reading with promising results. We have collected visual speech sequences in our lab and show the results here for a set of ten words in Brazilian Portuguese, spoken by different speakers in more than 150 samples. The approach can be extended and applied to other spoken languages as well.