Improvement of Methods and Techniques that Subsidize the Analysis of Land Use Changes Over Time

Both natural evolution and human activities result in changes in natural resources and in the terrestrial environment. Observation of the temporal and spatial level of the Earth allows us to understand the interrelationship of the phenomena that cause these changes. The studies carried out in this sense vary according to the need of the researchers and the object to be studied. There is a need to work with a large amount of time in a temporal series of multispectral images in order to obtain more detailed data of the space changes over time. This work aims to contribute to the area of geoprocessing regarding the improvement of methods and techniques that subsidize the analysis of land use changes over time. For this purpose, an approach was proposed in order to allow the analysis of an image with -bands, in -times simultaneously, so the time dimension was added to the model of a common three-dimensional digital image (lines, columns and bands), making it four-dimensional (lines, columns, bands and times). New multispectral and multitemporal data storage formats (TBSQ, TBIL, TBIP) were produced, to test them was developed a two-stage prototype. The first version of the prototype consumed a considerable time to execute the tests, leading to the need for improving the prototype. In the second version of the prototype, a multi-pointer technique was implemented, which brought significant performance gains to the initial version.