A Hybrid Quantum-Classical Machine Learning Approach to Vision Sensor Data Analysis in Aerospace Applications

Aerospace vehicle detection, localization, and subsequent parametric computations such as range and altitude in frame-based data analysis acquired through vision sensors such as cameras can be a challenging problem due to a variety of factors including but not limited to the size and shape of the vehicle, the distance and position of the vehicle, lighting conditions, and frame resolution. A combination of machine learning and computer vision techniques are generally used to address these situations. In this paper, we demonstrate a workflow using quantum-enhanced, transfer and classical machine learning to detect and localize aerospace vehicles and compute parameters such as range and altitude from color histograms. Additionally, we describe a data synthesis approach for training machine learning algorithms against aerospace imagery. The purpose of our study is not to present the superiority of a technique but rather the feasibility and practicality of incorporating quantum enhancement in aerospace data analytics, data synthesis through simulation programs, and parametric computations through color histograms.