Change & Time Series Analysis

• Compare two quantitative images of the same variable for different dates. Output simple difference, percent change, standardized difference (z values), or standardized classes.
• Compare two quantitative images of the same variable for different dates by computing their ratio or natural log ratio.
• Compare two-band sets of images for two dates to create both magnitude and direction (character) of change images (change vector analysis).
• Address problems of non-comparability due to sensor changes between two images from different dates.
• Calculate the overall and per-category Kappa Index of Agreement to facilitate change analysis with categorical data.
• Examine a graphic cross-section profile along a linear feature.
• Calculate Standardized Principal Components on a group of images with the production of an equal number of resulting components. Loadings graphs are output.
• Calculate the Pearson Product Moment Coefficient of Correlation through a time series of images for each pixel of an image.
• Create video clips of the images in a time series for visual change detection.
• Analyze two qualitative land cover images from different dates and produce a Markovian transition matrix, a transition areas matrix, and a set of conditional probability images.
• Create a stochastic land cover map by evaluating the conditional probabilities that each land cover can exist at each pixel location against a rectilinear random distribution of probabilities.
• Redistribute the conditional probabilities of a particular land cover type according to a designated pattern.
• Use cellular automata, typically for dynamic modeling where the future state of a pixel depends upon its current state and the states of its neighbors.
• Combine cellular automata and the Markov change land cover prediction procedure to add an element of spatial contiguity as well as knowledge of the likely spatial distribution of transitions to Markov change analysis.
• Geomod - A spatial modeler of land use change.
• Compute statistics for measuring the similarity between two qualitative images, including specialized Kappa measures that discriminate between errors of quantity and errors of location.
• Measure the correspondence between a quantitative modeled image showing the likelihood that a particular class exists and a Boolean image of that class as it actually occurs.