Greetings, does anyone recommend the initial steps to extract the features of an image to enhance a non-image-based dataset?

For example, I have a pollution data set with environmental information and would like to add one or more columns in this dataset from data extracted from an image (satellite or a sewer disposal) to make the data set more robust. Rather than using a set of image data, the idea is to extract information from that image and add it as new columns in the existing dataset.

What kind of information can be extracted?

Any reading recommendations?

Similar questions and discussions