Advanced search
Start date

Pattern recognition in crops from the combination of classifiers and clusterers


Unmanned Aerial Vehicles (UAVs) are typically used in the mapping of agricultural areas due to their low cost and flexibility in data collection. Mapped areas help to detect diseases affecting planted areas, such as those for the sugarcane culture. From this perspective, the identification of several patterns in data collected by UAVs typically depends on the use of methods and algorithms based on Supervised Machine Learning. However, such methods and algorithms require significant amount of labeled data to induce accurate classification models with good generalization capability, which is rarely achieved in agricultural applications. Such a limitation makes room for the study and development of classifiers that, even induced from a small set of labeled data, are able to provide relevant results in pattern recognition. Our starting point will be the investigation of approaches for combining classifiers and clusterers, whose the underlying assumption is compensating the lack of labeled data with the information of clusters found on data. From these approaches, we will develop a method/algorithm to refine classification models fed by visual data captured by UAV in order to detect the Migdolus fryanus beetle larvae in sugarcane culture. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items

Scientific publications
(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)
COLETTA, LUIZ F. S.; PONTI, MOACIR; HRUSCHKA, EDUARDO R.; ACHARYA, AYAN; GHOSH, JOYDEEP. Combining clustering and active learning for the detection and learning of new image classes. Neurocomputing, v. 358, p. 150-165, . (17/00357-7)

Please report errors in scientific publications list by writing to: