GIS 4035 Module 5: Unsupervised and Supervised Image Classification
This week we spent time working on Image classifications through supervised and unsupervised methods. The first activity that I worked on was the unsupervised classifications which didn't seem too hard. Most of the information seemed relatively easy to remember (for the incoming final project due). Recoding the classes was an interesting process just because you get to see all of the signatures you created be condensed down into smaller categories. You also use recoding and signatures in supervised classifications so it was cool to see the transition from one to the other was really simple. The issues with this lab came with the color coding of my newly recoded signatures. I felt as though the colors were too close together and caused too much spectral confusion overall. What I ended up doing was changing a few of the features colors so that way they stood out a little better and wasn't so congested on the map. I know the risk in doing so could cost me, but in order to maintain a visually pleasing and easy to understand map I took a leap of faith. Other than that (and turning in the project late), the exercise seemed ok. I am eager to start the final project and apply the knowledge that I have learned over these past few months to a new project!
Comments
Post a Comment