Interpolation pour l'augmentation de donnees : Application \`a la gestion des adventices de la canne a sucre a la Reunion

TL;DR

This paper investigates interpolation techniques like Gaussian processes and kriging for augmenting geo-referenced data to improve weed prediction in sugarcane fields, emphasizing spatial consistency and predictive performance.

Contribution

It compares the effectiveness of GP and kriging interpolation methods for data augmentation in spatial weed prediction tasks, highlighting the superior performance of GP-based approaches.

Findings

GP-COMB significantly improves regression performance with less data

Kriging offers more homogeneous spatial coverage

Interpolation enhances model robustness with limited data

Abstract

Data augmentation is a crucial step in the development of robust supervised learning models, especially when dealing with limited datasets. This study explores interpolation techniques for the augmentation of geo-referenced data, with the aim of predicting the presence of Commelina benghalensis L. in sugarcane plots in La R\'eunion. Given the spatial nature of the data and the high cost of data collection, we evaluated two interpolation approaches: Gaussian processes (GPs) with different kernels and kriging with various variograms. The objectives of this work are threefold: (i) to identify which interpolation methods offer the best predictive performance for various regression algorithms, (ii) to analyze the evolution of performance as a function of the number of observations added, and (iii) to assess the spatial consistency of augmented datasets. The results show that GP-based methods, in particular with combined kernels (GP-COMB), significantly improve the performance of regression algorithms while requiring less additional data. Although kriging shows slightly lower performance, it is distinguished by a more homogeneous spatial coverage, a potential advantage in certain contexts.