| Full text | |
| Author(s): |
Chiuyari Veramendi, Wilbur N.
;
Cruvinel, Paulo E.
;
IEEE
Total Authors: 3
|
| Document type: | Journal article |
| Source: | 2021 IEEE 15TH INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC 2021); v. N/A, p. 5-pg., 2021-01-01. |
| Abstract | |
With the need to increase agricultural production and to avoid loss, this paper presents the development of a new method for counting plants of maize in an agricultural field using spectral images obtained by an UAV, as well as digital processing and semantic modeling techniques. The method is based on the use of the Circular Hough Transform (CHT) in conjunction with the techniques of Backmapping, neighborhood analysis, and a classification of patterns. Both the supper vector machines (SVM) and the neural networks (NN) methods have been evaluated for the classification procedure. Besides, using a computational environment for simulation, previous results have been obtained, i.e., showing not only the usefulness of the direct measures but also an automatic way for the plants identification, counting and height determination of the planted maize. Also, the establishment of a friendly interface has been carried out, which allows the monitoring of the phenological phases involved in the stages of the maize cultivation. (AU) | |
| FAPESP's process: | 17/19350-2 - Advanced digital tool for the agricultural risk management |
| Grantee: | Paulo Estevão Cruvinel |
| Support Opportunities: | Research Grants - Research Partnership for Technological Innovation - PITE |