To answer your question, let us take the following example. Let us assume that we need to monitor a certain agricultural crop (e.g., Sugar Cane) over a period of 12 months and each month is covered by one satellite image (e. 'g., Landsat8). Also assume that we are looking for a particular area that has a known boudary (e.g., GIS shape file or Google KMZ). If we extract this boundary from the first image, then we will end up with spatial data from the first image and if we extract the corresponding boundary from the twelev images, then we will end up with the extraction of temporal and spatial data. In other words, the word, spatial refers to the collected image data over time or more preciely,t ime sequence . In general, your question is typically phrased under thge ame of "Spatio-temporal" data extraction in remote sensing. From an interactive point of view, the best way to extract Spatio-temporal image data is to use a GIS software and delineate your polgyon of interest by digitization in one of the images of your particular project. Then extract the relevant pixels or intensities from all images that define your sequence. Otherwise you can build a computer program to do this Spatio-temporal extraction.
Thank you for your beautiful explanation. I am not familiar with GIS thats why asking you this dumb question. How do I extract particular agricultural crop (e.g. sugar cane) from the satellite image? Is there any color code or extract information? Thanks