Work-in-Progress: Prognostics in Arboriculture Using Computer Vision and Statistical Analysis

Abstract

As engineers, we can analyze and mathematically relate the behaviors present in all life sciences. For example, a brain can be modelled as a neural network, or a heart can be modelled as a closed loop control system. Similarly, entities from the world of arboriculture or the plant kingdom as a whole can be modelled and fitted into mathematical models. With the increased use of high speed computers in recent years, process models have a high scope for getting transformed to their digital twins. The performance analysis of such digital twins can be carried out using various techniques and the future behaviour of these systems can be predicted. The non-linear and stochastic behaviour of inputs for these models can be weighed based on the history of their occurrences and the priority level can be assigned to obtain the performance evaluations. This concept has been applied here in this work, to predict the characteristics of trees in terms of their lifespan and strength. The methods of analysis proposed in this paper use widely available open source software platforms, which make the design dedicated, reliable, and accessible to all.