English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

One of the major challenges of agriculture is the rational use of agrochemicals in crop to prevention and protection against diseases, pests and weeds, in order to avoid the excessive application of chemical agents in crops, avoiding damages to the environment and human health. Currently the techniques used to evaluate the quality of the spray are based on the analysis of droplet size that are applied through hydrosensitive labels. These, however, only allow analysis after the application of the agricultural defense. There are commercial solutions based on laser sensing that allow evaluation of hydraulic spray nozzles in the laboratory. These solutions are expensive and restricted to laboratory tests. In this work a test bench was developed, which allows the simulation of the operation of spray nozzles, and a data acquisition, conditioning and analysis system for conducting a laser sensing study applied to the identification of droplet spectra. In addition, a classification method based on Fast Fourier Transform and Artificial Neural Networks was proposed and evaluated. The methodology used for the evaluation of the proposed technique consists of the acquisition and analysis of a light signal from a laser beam intercepted by a water spray created by the nozzle, and classification of the pattern based on the frequency spectrum amplitudes obtained from of the Fourier discrete signal transform. For the classification, a ANN Multilayer Perceptron was used to identify two different spray patterns, classifying the drop spectra into two classes, mist or jet. Cross-validation was performed and the proposed technique presented 91% accuracy. Thus, it was possible to conclude that through the proposed method it is possible to identify different spectra of drops.

Desenvolvimento de um protótipo para classificação de perfis de pulverização utilizando processamento digital de sinais e redes neurais artificiais