Case Study: Worldwide Semiconductor Company achieves smarter drone-based crop monitoring with Auriga

Drone Software for a Smart Agriculture Project

Worldwide Semiconductor Company, a global leader in the semiconductor industry, worked with Auriga to develop drone software for a smart agriculture project. The goal was to monitor the condition of trees in an apple garden using a software utility built with data science, IoT, Linux, machine learning, and Python.

Auriga implemented dataflow programming with TensorFlow, deployed an SSD-Mobilenet neural network, and built tools for object identification, labeling, augmentation, and data formatting, along with a GUI. As a result, video-card data calculations were accelerated up to 4x, the machine learning and deep learning models were improved with a significantly larger object database, and the end-user application became easier to use.