Abstract
Abstract. In the early 1980′s, the European chestnut tree (Castanea sativa, Mill.) assumed an important role in the Portuguese economy. Currently, the Trás-os-Montes region (Northeast of Portugal) concentrates the highest chestnuts production in Portugal, representing the major source of income in the region (€50M-€60M). The recognition of the quality of the Portuguese chestnut varieties has increasing the international demand for both industry and consumer-grade segments. As result, chestnut cultivation intensification has been witnessed, in such a way that widely disseminated monoculture practices are currently increasing environmental disaster risks. Depending on the dynamics of the location of interest, monocultures may lead to desertification and soil degradation even if it encompasses multiple causes and a whole range of consequences or impacts. In Trás-os-Montes, despite the strong increase in the cultivation area, phytosanitary problems, such as the chestnut ink disease (Phytophthora cinnamomi) and the chestnut blight (Cryphonectria parasitica), along with other threats, e.g. chestnut gall wasp (Dryocosmus kuriphilus) and nutritional deficiencies, are responsible for a significant decline of chestnut trees, with a real impact on production. The intensification of inappropriate agricultural practices also favours the onset of phytosanitary problems. Moreover, chestnut trees management and monitoring generally rely on in-field time-consuming and laborious observation campaigns. To mitigate the associated risks, it is crucial to establish an effective management and monitoring process to ensure crop cultivation sustainability, preventing at the same time risks of desertification and land degradation. Therefore, this study presents an automatic method that allows to perform chestnut clusters identification, a key-enabling task towards the achievement of important goals such as production estimation and multi-temporal crop evaluation. The proposed methodology consists in the use of Convolutional Neural Networks (CNNs) to classify and segment the chestnut fruits, considering a small dataset acquired based on digital terrestrial camera.
Highlights
The chestnut (Castanea sativa, Mill.) agro-ecosystem is of great social, economic and landscape importance in north-eastern Portugal, namely to Trás-os-Montes, due to the various resources associated with this crop (e.g. fruit and wood production and mushroom harvesting (Baptista et al, 2010)
In Trás-os-Montes, threats related to phytosanitary problems of biotic and abiotic origins, such as the chestnut ink disease (Phytophthora cinnamomi), and the chestnut blight (Cryphonectria parasitica), among others - e.g. chestnut gall wasp (Dryocosmus kuriphilus) and nutritional deficiencies - are responsible for a significant decline of chestnut trees with a direct impact on production, despite cultivation area growth (Martins et al, 2015)
The proposed methodology consists in the use of Convolutional Neural Networks (CNNs) to classify and segment the chestnut fruits, considering a small dataset acquired based on digital terrestrial camera
Summary
The chestnut (Castanea sativa, Mill.) agro-ecosystem is of great social, economic and landscape importance in north-eastern Portugal, namely to Trás-os-Montes, due to the various resources associated with this crop (e.g. fruit and wood production and mushroom harvesting (Baptista et al, 2010). In previous work (Marques et al, 2019), an approach based on digital image processing for macro detection and monitoring of chestnut trees was proposed, through multiple parameters estimation such as tree identification and counting, individual extraction of tree height, tree crown diameter and area features, using remote sensed imagery acquired by unmanned aerial vehicles. DL is a modern and promising technique for image processing and data analysis, whose application domains include (but are not limited to) plant recognition, leaf and crop type classification and plant diseases detection, among other useful operations with benefits for natural environments mainly populated by vegetation (Kamilaris and Prenafeta-Boldú, 2018). The following sections focus the proposal of a preliminary methodology consisting in the use of Convolutional Neural Networks (CNNs) to classify and roughly segment the chestnut clusters, considering a small dataset acquired based on digital terrestrial camera
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
