Typical geography exercises

Three exercises in geomatics information processing

Three exercises in geomatics information processing

1. Color Image Segmentation

Description of the Satellite Image Segmentation Application

This application is used for satellite image segmentation and facilitates their analysis by allowing users to create masks representing specific elements within the images.

Key features:

The application enables the segmentation of a satellite image based on its color characteristics. To achieve this, users can choose from four distinct color models:

• RGB (Red, Green, Blue): a classic color model suitable for standard images and visually understandable for humans.
• HSV (Hue, Saturation, Value): ideal for separating colors based on their hue and intensity.
• Ycbcr: commonly used in many video formats, excellent for analyzing separate chromatic and brightness components.
• Lab: represents colors more closely to human perception, based on luminance and two opposing color channels.

Once the model is selected, users have access to a series of sliders to manipulate the range of each parameter.

For example, in the RGB model, users can adjust the red, green, and blue levels independently, while in the HSV model, they can set thresholds for hue, saturation, and value. These adjustments allow users to filter the image and highlight only the areas that match the defined ranges.

Potential uses:

This application is ideal for various remote sensing and satellite image analysis applications, such as soil and vegetation mapping, water body detection, land use change monitoring, urban analysis, and more.

Note: This application was originally designed to select and highlight vegetation areas from a satellite image, but it can be used for other purposes. Simply replace "vegetation percentage" with the element for which you wish to create the mask.

2. Calculating a chi²

This livescript provides a MATLAB-based approach to calculating a chi2 in statistics. It allows you to perform the calculation much more quickly and efficiently than by hand.

3. Sampling a point cloud

This livescript lets you sample a point cloud to make it less cumbersome for MATLAB's Lidar Viewer.