As part of an academic activity in the "GEOG 190: Map, Aerial Photo, and Satellite Image Interpretation" unit, I performed digital image classification on a Landsat 8 image through MultiSpec, QGIS Semi-automatic Classification Plugin (SCP), and Google Earth Engine (GEE). The resulting supervised/trained outputs are classified according to particular land cover types.

1. Working with MultiSpec (image #1) included preparing the Set Project Window, collecting samples for the First Training Fields (Subspectral Classes), defining classes/field descriptions/information classes (Land Cover Category), running the classification, and assessing the class performance (accuracy) of the resulting output.

2. Working with the QGIS SCP (image #2) included practicing data preprocessing workflows, preparing WMS, collecting the initial samples for training fields, learning the image classification workflow, and utilizing maximum likelihood algorithm for classification.

3. Working with Google Earth Engine (image #3) included successful execution of prepared scripts and troubleshooting/debugging said scripts. The processes included for this output are Landsat image collection, establishing a cloud masking layer, and instantiating clusters for training.