Numerical Computation-Based NDVI Calculation for Multispectral Image

Abstract: Diminishing of green areas (vegetation, park, forest, etc.) influences the climate change and the quality of life in an area. Indonesia has large are, hence, it is difficult to find the number of vegetation and without the computer-based calculation accurately. Also, nowadays, the remote sensing technology has been widely used to capture the large region using satellite sensor, e.g., Landsat, Sentinel, Ikonos, etc. from websites that give free access like United States Geological Survey (USGS). Many vertical applications can be used to calculate percentage of vegetation, e.g., IDRISI, Dyna-Clue, eCognition, etc., but in this study, a Matlab-based calculation was used. This programming language will convert a data into a mat-file that can be integrated in a single datastore, hence the program will execute the data efficiently. In a datastore, many regions can be calculated simultaneously using the big data facility in Matlab. The study shows the calculation using a special data format (mat-file) has good accuracy and fast although in this study multispectral data were used before conversion into normalized difference vegetation index (NDVI).

Keywords: Satellite Imagery, Multispectral, Matlab, NDVI, Near Infrared, sensors