Weed mapping using a machine vision system

Abstract

-Weed mapping is a useful tool for site-specific herbicide applications. The objectives of this study were (1) to determine the percentage of land area covered by weeds in no-till and conventionally tilled fields of common bean using digital image processing and geostatistics, and (2) to compare two types of cameras. Two digital cameras (color and infrared) and a differential GPS were affixed to a center pivot structure for image acquisition. Sample field images were acquired in a regular grid pattern, and the images were processed to estimate the percentage of weed cover. After calculating the georeferenced weed percentage values, maps were constructed using geostatistical techniques. Based on the results, color images are recommended for mapping the percentage of weed cover in no-till systems, while infrared images are recommended for weed mapping in conventional tillage systems.Keywords: precision agriculture, geostatistic, common bean. RESUMO -Mapeamento de plantas daninhas é uma ferramenta útil em aplicações localizadas de herbicidas. Os objetivos deste trabalho foram: determinar o percentual da cobertura vegetal de plantas daninhas em uma lavoura de feijão sob os sistemas de plantio direto e convencional, usando processamento digital de imagens e geoestatística; e comparar dois tipos de câmera. O sistema de visão artificial era composto por duas câmeras digitais (colorida e infravermelha), acopladas à estrutura móvel do pivô central, e um DGPS. As imagens adquiridas representavam amostras da área

Phytosociological variability of weeds in soybean field

“…The technique was effective; however, it becomes more time consuming with the increasing complexity of the spatial distribution of weeds in the area. Recently, with the use of optical sensors in Brazilian agriculture, some studies have shown promising results for use in weed mapping (Silva Jr. et al, 2012;Merotto Jr. et al, 2012).…”

“…As for the automatic collection of data, one can use color cameras and infrared coupled to a land (Silva Jr. et al, 2012) or air equipment (Candón et al, 2012b) and sensors and vegetation index (Chang et al, 2004;Merotto Jr. et al, 2012). On the other hand, in the conventional sampling meshes are commonly used sample, dividing the area into smaller squares, where the georeferenced samples of the weed are held (Merotto Jr. & Bredemeier, 2010).…”

Use of artificial vision techniques for diagnostic of nitrogen nutritional status in maize plants

“…With the obtained results, it was found global percentage of right 76% with reliable Kappa index. Silva Júnior et al (2012) determined the percentage of vegetation cover of weeds plants in the crop beans, under the no-tillage and conventional, using digital image processing and geostatistics. Abrahão et al (2013) conducted a study of classifiers based on different combinations of bands and spectral indices of original images to discriminate foliar nitrogen and chlorophyll, also in the crop beans.…”

Precision Agriculture in Brazil: The Trajectory of 25 Years of Scientific Research

“…Using the principles of the proximal sensing of plants, Merotto et al [102] evaluated vegetation indices obtained by commercial sensors as a strategy for determining the location of weeds in the field. Likewise, Silva Jr. et al [103] used a computer vision system to analyze images and to differentiate weed species from dry bean plants. The first study related to pest mapping and control in Brazil was carried out by Farias et al [104], who analyzed the spatial distribution of soil nematode infestation in cotton.…”