VCS-1 tree detection and NDVI computation

The Intel Realsense D435i Camera streams an image of the surrounding landscape to the VCS-1 allowing the Neural Network to check for vegetation. Plotting the coordinates of the trees, the VCS-1 can calculate the region of interest so that the “Normalized Difference Vegetation Index” (NDVI) can be calculated.

The Neural Network is able to effectively predict objects on the fly. By inspecting images frame by frame, objects are able to be identified even if they were only inside of the stream for a split second. This means that smaller objects such as tree stumps can be identified alongside larger objects such as Trees.

The processing power of the VCS-1 makes real-time object classification possible enabling computation of the NDVI. Without the use of VCS-1 there would have been an increased chance of false positives leading to incorrect statistics being achieved from surveys. The portability of the hardware and low power usage makes it a far more versatile solution compared to other options like bulky laptops.