Cyperus esculentus (yellow nutsedge) is one of the world’s worst weeds as it can cause great damage to crops and crop production. To eradicate C. esculentus, early detection is key—a challenging task as it is often confused with other Cyperaceae and displays wide genetic variability. In this study, the objective was to classify C. esculentus clones and morphologically similar weeds. Hyperspectral reflectance between 500 and 800 nm was tested as a measure to discriminate between (I) C. esculentus and morphologically similar Cyperaceae weeds, and between (II) different clonal populations of C. esculentus using three classification models: random forest (RF), regularized logistic regression (RLR) and partial least squares–discriminant analysis (PLS–DA). RLR performed better than RF and PLS–DA, and was able to adequately classify the samples. The possibility of creating an affordable multispectral sensing tool, for precise in-field recognition of C. esculentus plants based on fewer spectral bands, was tested. Results of this study were compared against simulated results from a commercially available multispectral camera with four spectral bands. The model created with customized bands performed almost equally well as the original PLS–DA or RLR model, and much better than the model describing multispectral image data from a commercially available camera. These results open up the opportunity to develop a dedicated robust tool for C. esculentus recognition based on four spectral bands and an appropriate classification model.
<p>Both the sensor viewing angle and the solar angle influence the remote sensing signal of terrestrial ecosystems. This influence is characterized by the bidirectional reflectance distribution function (BRDF). Knowledge of this BRDF is needed to correctly interpret the signal, but can also provide information on vegetation characteristics and structure. Obtaining the BRDF is far from straightforward: at leaf scale, laboratory goniometers can measure reflected radiation over a range of sensor-solar angle; for very homogeneous ecosystems, such as grassland or agricultural cropland, unmanned aerial vehicles (UAVs) can be programmed as giant goniometer, scanning the BRDF of an area of up to a few m&#178;. For heterogeneous ecosystems such as forests, this is not feasible. In this case, BRDF could so far only be derived from theoretical radiation transfer models or semi-empirical models; yet these models do not always agree.</p><p>We here propose a new method for measuring BRDF of forest ecosystems with UAVs, by measuring a star-shaped area of the ecosystem, covering in total about 3600m&#178; and capturing 6 different sensor-solar azimuth angle and three different zenith angles. This approach was applied over two sites of tropical rainforests in Queensland, Australia, with measurements with a RGB camera and a spectrometer. By repeating the flights several times during the day, we were able to test the Helmholtz reciprocity principle &#8211; that states the BRDF function of ecosystems remains the same, regardless of the solar angle &#8211; and are able to increase the range of sensor-solar angles. Our results present the first strictly empirical BRDF of tropical rainforests and confirm the importance of accurate BRDF correction of remote sensing products from forest ecosystems.&#160;</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.