Jurandir Zullo

A New Time Series Mining Approach Applied to Multitemporal Remote Sensing Imagery

In this paper, we present a novel unsupervised algorithm, called CLimate and rEmote sensing Association patteRns Miner, for mining association patterns on heterogeneous time series from climate and remote sensing data integrated in a remote sensing information system developed to improve the monitoring of sugar cane fields. The system, called RemoteAgri, consists of a large database of climate data and low-resolution remote sensing images, an image preprocessing module, a time series extraction module, and time series mining methods. The preprocessing module was projected to perform accurate geometric correction, what is a requirement particularly for land and agriculture applications of satellite images. The time series extraction is accomplished through a graphical interface that allows easy interaction and high flexibility to users. The time series mining method transforms series to symbolic representation in order to identify patterns in a multitemporal satellite images and associate them with patterns in other series within a temporal sliding window. The validation process was achieved with agroclimatic data and NOAA-AVHRR images of sugar cane fields. Results show a correlation between agroclimatic time series and vegetation index images. Rules generated by our new algorithm show the association patterns in different periods of time in each time series, pointing to a time delay between the occurrences of patterns in the series analyzed, corroborating what specialists usually forecast without having the burden of dealing with many data charts.

In-flight absolute calibration of the Landsat-5 TM on the test site Salar de Uyuni

In Brazil, the increase of the application of quantitative approaches in the natural resources studies using remote sensing technology has required knowledge about the radiometric conditions of remote sensors as the Thematic Mapper (TM) and the Enhanced TM Plus, for instance. The establishment of a correlation between radiometric data and biophysical and geophysical ones has become a frequent need in the Brazilian remote sensing community, and it has increased the demand of calibration coefficients in order to transform digital numbers to physical values like radiance and reflectance. Since the China-Brazil Environmental Remote Sensing Satellite became a reality, the necessity to perform calibration campaigns increased significantly. Following Price and other researchers suggestions, an in-flight absolute calibration of the Landsat-5 data was carried out in the Salar de Uyuni, Bolivia. It was only possible to determine calibration coefficients for bands TM2, TM3, and TM4 due to the saturation of band TM1 and surface moisture conditions that impacted the TM5 and TM7. The methodology applied here seemed to be sufficient to determine valid calibration coefficients for orbital sensors.

Data Clustering using Self-Organizing Maps segmented by Mathematic Morphology and Simplified Cluster Validity Indexes: an application in remotely sensed images

This paper presents a cluster analysis method which automatically finds the number of clusters as well as the partitioning of a data set without any type of interaction with the user. The data clustering is made using the self-organizing (or Kohonen) map (SOM). Different partitions of the trained SOM are obtained from different segmentations of the U-matrix (a neuron-distance image) that are generated by means of mathematical morphology techniques. The different partitions of the trained SOM produce different partitions for the data set which are evaluated by cluster validity indexes. To reduce the computational cost of the cluster analysis process this work also proposes the simplification of cluster validity indexes using the statistical properties of the SOM. The proposed methodology is applied in the cluster analysis of remotely sensed images.

Characterization of the Salar de Uyuni for in-orbit satellite calibration

Field work was carried out on June 8 and 9, 1999 to evaluate the use of the Salar de Uyuni as a test site for in-orbit satellite calibration. A dataset of ten Thematic Mapper (TM) images, from 1988-1997, was used to select three test points based on the analysis of the temporal stability of the reflectance of Salars surface. Bidirectional reflectance factor (BRF) values of Salars surface within the precision suitable for vicarious calibration procedures were obtained using a CE313-2/CIMEL radiometer. In spite of seeming visually homogeneous, the BRF values of one test point have presented significative statistical differences with the two others. Atmospheric characterization was possible with a sunphotometer CE317/CIMEL showing the low importance of the atmospheric effects in the image acquisition. The results confirm that the Salar de Uyuni has the characteristics pointed out by many authors as suitable for a vicarious calibration site, specially from April to November because of the reduced rainfall occurrence. The main disadvantages are the difficult access and the critical period for data collecting in the rainy season from November to March. An angular reflectance variation study is recommended in order to evaluate its Lambertian properties.

Analysis of NDVI time series using cross-correlation and forecasting methods for monitoring sugarcane fields in Brazil

Brazil is the largest sugarcane producer in the world and has a privileged position to attend to national and international marketplaces. To maintain the high production of sugarcane, it is fundamental to improve the forecasting models of crop seasons through the use of alternative technologies, such as remote sensing. Thus, the main purpose of this article is to assess the results of two different statistical forecasting methods applied to an agroclimatic index (the water requirement satisfaction index; WRSI) and the sugarcane spectral response (normalized difference vegetation index; NDVI) registered on National Oceanic and Atmospheric Administration Advanced Very High Resolution Radiometer (NOAA-AVHRR) satellite images. We also evaluated the cross-correlation between these two indexes. According to the results obtained, there are meaningful correlations between NDVI and WRSI with time lags. Additionally, the adjusted model for NDVI presented more accurate results than the forecasting models for WRSI. Finally, the analyses indicate that NDVI is more predictable due to its seasonality and the WRSI values are more variable making it difficult to forecast.

Brazil's 2001 energy crisis monitored from space

Data sensed by the US Air Force Defence Meteorological Satellite Program (DMSP) Operational Linescan System (OLS) during the years 2000 and 2001 in Brazil were tested as a tool to monitor reduction of nocturnal lighting. This particular timing was examined as the Brazilian population and industry were forced to reduce electric power consumption by 20% during 2001, in relation to 2000, for a period of several months, starting officially on 1 June 2001. Large urban agglomerates were compelled to switch off city lights by at least the same amount. The Distrito Federal (DF), including the Brazilian capital, Brasilia, was one of the primary areas where the government actively sought electric power consumption reductions. Using the DF as a study case, we demonstrate that the mean grey levels derived from averaging DMSP-OLS data acquired over urban centres appear to be a useful index to monitor relative oscillations in energy consumption.

Surface ozone study in Campinas, Sao Paulo, Brazil

Abstract The first results of the surface ozone study in Campinas, Sao Paulo state, Brazil, are presented. During the local winter (dry season) the photochemical ozone production is found to be a regular process with the afternoon ozone mixing ratio maximum ranging from 10 to 15 ppbv during totally cloudy days to 40–60 ppbv during clear sky weather. Several high-value ozone episodes with ozone mixing ratios from 80 to 140 ppbv have been registered. At the beginning of the wet season the ozone concentration in Campinas did not decrease significantly as in the Amazonia forest region, but diurnal variations became more complicated with sharp dropouts during the days with rainfall and other fast changes of meteorological conditions. The fast irregular pulsations of ozone concentration with periods from the first tens of seconds to tens of minutes have been registered. Possible explanations of the nature of pulsations are briefly discussed. Photochemical ozone production by urban plumes of Campinas or Sao Paulo is named as a first possible source of the elevated ozone concentration and a local biomass burning is suggested as an alternative or an additional source.

New DTW-based method to similarity search in sugar cane regions represented by climate and remote sensing time series

Brazil is an important sugar cane producer, which is the main resource for ethanol production, a renewable source of energy. This agricultural commodity is important to the country economy, becoming fundamental to improve models that assist the crops monitoring process. Vegetation indexes originated from remote sensing images and agrometeorological indexes can be combined to represent sugar cane fields in a regional scale. However, finding different regions with similar patterns to classify or analyze their characteristics is a non-trivial task. Accordingly, this paper presents a method to find similar sugar cane fields represented by series of vegetation and agrometeorological indexes. The proposed method combines a weighted distance function with an algorithm to find similar objects. Results were coincident in the most cases with the classification done by experts, finding regions with similar characteristics of climate and productivity. Consequently, this approach can help in decision making processes by agricultural entrepreneurs.

CLEARMiner: a new algorithm for mining association patterns on heterogeneous time series from climate data

Recently, improvements in sensor technology contributed to increasing in spatial data acquisition. The use of remote sensing in many countries and states, where agricultural business is a large part of their gross income, can provide a valuable source to improve their economy. The combination of climate and remote sensing data can reveal useful information, which can help researchers to monitor and estimate the production of agricultural crops. Data mining techniques are the main tools to analyze and extract relationships and patterns. In this context, this paper presents a new algorithm for mining association patterns in Geo-referenced databases of climate and satellite images. The CLEARMiner (CLimatE Association patteRns Miner) algorithm identifies patterns in a time series and associates them with patterns in other series within a temporal sliding window. Experiments were performed with synthetic and real data of climate and NOAA-AVHRR sensor for sugar cane fields. Results show a correlation between agroclimate time series and vegetation index images. Rules generated by our new algorithm show the association patterns in different periods of time in each time series, pointing to a time delay between the occurrences of patterns in the series analyzed, corroborating what specialists usually forecast having the burden of dealing with many data charts.

A neural architecture for the classification of remote sensing imagery with advanced learning algorithms

This work presents an artificial neural networks based architecture for the classification of remote sensing (RS) multispectral imagery. The architecture consists of two processing modules: an image feature extraction module using Kohonen self-organizing map and a classification module using multilayer perceptron network. The architecture was developed aiming at two specific goals: to exploit the advantages of unsupervised learning for feature extraction, and the testing of techniques to increase the learning algorithms performance concerning training time. To test the applicability of this work, the architecture was applied to the classification of a LANDSAT/TM image segment from a pre-selected testing area and its performance was compared with that of a maximum likelihood classifier, conventionally used for RS multispectral images classification.