Data acquisition is the first step in the spatial data decision making process. Depending on the decision that has to be made a choice has to be made between attribute, vector or image data. Typically projects will require a combination of all these forms of data.
Mass Data Acquisition
Once data has been acquired for a scene, the next step is to construct models that mathematically describe the objects or phenomena captured in the scene. These models allow for (i) objects and phenomena to be analysed, (ii) the making of predictions, and (iii) the visualisation of trends in the spatial data.
For the purpose of modeling, the spatial data will have to be (i) pre-processed, (ii) processed and finally (iii) post-processed.
Pre-processing includes the conversion of the data, the structuring of the data, the coordinate transformation of the data. In the processing stage a variety of models are applied to the data and those models that best fit the data are selected. This modeling can be mathematical, statistical. Furthermore the models can be parameterised or non-parameterised (for example neural networks). In the post-processing stage the models are packaged for use by decision makers.
|Transformations||Coordinate Transformations, ...|
|Mathematical modeling||Functional modeling, Statistical modeling, Machine Learning, Optimization, ...|
Data is a discrete representation of objects and phenomena in the real world. Modeling uses the data to propose functions/models that represent the real world objects and phenomena.
With the aid of models we are now able to analyse the behaviour of phenomena in the real world. This analysis includes (i) searching for trends in the data, (ii) investigating the iteraction between phenomena, (iii) predicting future trends, and (iv) mining the data for hidden structures.
|Mathematical Modeling||Functional analysis, Machine Learning, ...|
The final stage in preparation for decision making is visualisation. Raw information can be overwhelming for the human mind. For this reason it is necessary to present the data in a way that conveys to the user salient information. For this reason visualisation is a necesary final step in the analysis process.
There are a range of visual techniques that can be used to display information. The choice of technique depends on the quantities being visualised, the dimensionality of the data, and the volume of data to be visualised.
Another consideration in the visualisation of data is the desired level of interactivity and the platform on which the data will be displayed. While most visualisations are static, understanding the underlying structures in high dimensional data may require interactive visualisations. These visualisations can be generated for desktops or the web.
|Visualisation of data||Scalar fields, Vector fields, ...|
|Renderings||Static, Fly-throughs, Animations, ...|
|Platforms||Desktop, Web, ...|