Vegetation indices and textures in object-based weed detection from UAV imagery

David, Lawrence Charlemagne G.; Ballado, Alejandro H.

doi:10.1109/iccsce.2016.7893584

Cited by 24 publications

(26 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The OBIA classifier produced results equal to the pixel-based classifiers in 2016, but achieved consistently lower accuracies in 2015, probably due to the lack of regular objects (patches) that favour this classifier [3,27].…”

Section: Discussionmentioning

confidence: 97%

“…This study tested the use of vegetation canopy height together with multispectral UAV images in mapping mature S. marianum weeds. Relevant work has used plant height indirectly, with the incorporation of image texture for weed mapping with several classifiers [9,26] and object-based image analysis [27], as well as with the use of estimated plant height for segmentation of crops and weeds into objects but not directly in the classification algorithm [19]. However, none of the above-mentioned studies has evaluated the improvement of the multispectral image classification that was achieved by using the vegetation elevation information.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Incorporating Surface Elevation Information in UAV Multispectral Images for Mapping Weed Patches

et al. 2018

View full text Add to dashboard Cite

Accurate mapping of weed distribution within a field is a first step towards effective weed management. The aim of this work was to improve the mapping of milk thistle (Silybum marianum) weed patches through unmanned aerial vehicle (UAV) images using auxiliary layers of information, such as spatial texture and estimated vegetation height from the UAV digital surface model. UAV multispectral images acquired in the visible and near-infrared parts of the spectrum were used as the main source of data, together with texture that was estimated for the image bands using a local variance filter. The digital surface model was created from structure from motion algorithms using the UAV image stereopairs. From this layer, the terrain elevation was estimated using a focal minimum filter followed by a low-pass filter. The plant height was computed by subtracting the terrain elevation from the digital surface model. Three classification algorithms (maximum likelihood, minimum distance and an object-based image classifier) were used to identify S. marianum from other vegetation using various combinations of inputs: image bands, texture and plant height. The resulting weed distribution maps were evaluated for their accuracy using field-surveyed data. Both texture and plant height have helped improve the accuracy of classification of S. marianum weed, increasing the overall accuracy of classification from 70% to 87% in 2015, and from 82% to 95% in 2016. Thus, as texture is easier to compute than plant height from a digital surface model, it may be preferable to be used in future weed mapping applications.hard to control using herbicides in fields, and its thorny leaves and high posture are a nuisance to grazing animals in pastures. It grows in dense strands and is more competitive than grass weeds, growing taller to overshadow adjacent species that compete for light [1].In recent years, unmanned aerial vehicles (UAVs) have proven to be useful in agricultural management. Their applications vary, but usually involve remote sensing, mapping, land modelling and precision agriculture. The most common use of UAVs in weed management is to acquire multispectral images in the visible and infrared spectrum for weed identification and mapping, however, a number of other potential uses have emerged, including the application of pesticides and other agrochemicals by the UAV itself [2]. Mapping weeds using UAVs provides high spatial resolutions, usually on the order of a few centimetres. This is vital in order to detect small weeds, weeds at a young stage, or within low crop cover with high background soil signal. In addition to the advantage of spatial resolution, UAVs offer temporal flexibility and can collect data at any time that fits the user's requirements, as compared to the satellite images. Moreover, UAVs can acquire images during overcast conditions, thus bypassing a frequent problem with satellite remote sensing, although cloud shadows and varying illumination conditions during image acquisition may affect data quality [3,4].A chall...

show abstract

Section: Discussionmentioning

confidence: 97%

Section: Discussionmentioning

confidence: 99%

Incorporating Surface Elevation Information in UAV Multispectral Images for Mapping Weed Patches

et al. 2018

View full text Add to dashboard Cite

show abstract

“…O índice NGRDI fornece indicações da variabilidade espectral dos dosséis da vegetação no plano horizontal (LI et al, 2016). Este índice de vegetação foi originalmente proposto devido ao seu potencial na discriminação espectral de espécies invasoras em cultivos (BARRERO; PERDOMO, 2018;DAVID;BALLADO, 2016). O cálculo do índice NGRDI assim como a plotagem foi realizado no software R 3.5.1 (R CORE TEAM, 2018) com os pacotes raster (HIJMANS, 2019), grDevices (R CORE TEAM, 2018) e RColorBrewer (NEUWIRTH, 2014).…”

Section: íNdices De Vegetação (Iv)unclassified

Aeronave Remotamente Pilotada De Baixo Custo No Estudo De Plantas Invasoras Em Áreas De Cerrado

Pessi¹,

José²,

Mioto³

et al. 2020

NAT

View full text Add to dashboard Cite

O objetivo dessa pesquisa é analisar se o método CHIS (Canopy Height Invasive Species) representa uma rotina de classificação assertiva na identificação de espécies invasoras a partir de imagens RGB em área de Cerrado com evidência de perturbação. A metodologia empregada foi a produção dos modelos de elevação MDS (Modelo Digital de Superfície) e MDT (Modelo Digital do Terreno) a partir das imagens coletadas em campo com drone e posteriormente processadas no software PhotoScan. A produção do CHIS foi a partir da subtração dos modelos MDS e MDT. Para aferir a precisão do modelo CHIS foram gerados dois modelos convencionais para comparação: classificação não supervisionada K-means e índice de vegetação NGRDI (Normalized Red-Green Difference Index). A comparação entre os modelos se deu em duas áreas amostrais escolhidas de forma não aleatória. Ao final foi aplicado teste de acurácia, correlação e Cohen’s Kappa. Os resultados demonstram que o modelo CHIS obteve os melhores resultados na identificação de espécies invasoras quando comparado com os modelos K-means e NGRDI. Os testes de acurácia para o modelo CHIS na área amostral 1 e 2 foi de 0,973 e 0,9 respectivamente; K-means 0,209 e 0,6; NGRDI 0,795 e 0,518. O modelo CHIS demonstrou ser promissor na identificação de espécies invasoras em áreas perturbadas quando comparado com modelos convencionalmente usados.Palavras-chave: ARP; gestão ambiental; sensoriamento remoto; CHIS. REMOTELY PILOTED AIRCRAFT (DRONE) OF LOW COST IN THE INVASIVE SPECIES STUDY IN CERRADO AREAS ABSTRACT: The objective of this research is to analyze if the CHIS (Canopy Height Invasive Species) method represents an assertive classification routine in the identification of invasive species from RGB images in Cerrado area with evidence of disturbance. The methodology used was the production of the DSM (Digital Surface Model) and DTM (Digital Terrain Model) elevation models from the images collected in the drone field and later processed in the PhotoScan software. The production of the CHIS was based on the subtraction of the DSM and DTM models. To verify the accuracy of the CHIS model two conventional models were generated for comparison: unsupervised K-means classification and NGRDI (Normalized Red-Green Difference Index) vegetation index. The comparison between the models occurred in two sample areas chosen in a non-random manner. At the end, it was applied test of accuracy, correlation and Cohen's Kappa. The results demonstrate that the CHIS model obtained the best results in the identification of invasive species when compared with the K-means and NGRDI models. The accuracy tests for the CHIS model in sample area 1 and 2 were 0,973 and 0,9 respectively; K-means 0,209 and 0,6; NGRDI 0,795 and 0,518. The CHIS model has been shown to be promising in the identification of invasive species in disturbed areas when compared to conventionally used models.Keywords: RPA; environmental management; remote sensing; CHIS.

show abstract

“…Revised works showed a good variety of classifiers. Support Vector Machines (SVM), as a popular supervised classification algorithm, was used on [10,11,12,13,14,15]. Another recurrent classification method found in our review was decision trees, like the random forest or C4.5 [16,16,10,14].…”

Section: State Of the Artmentioning

confidence: 99%

“…Some approaches rely only on color/spectral information, for example, [15] used 33 different spectral bands as feature space. Gray level cooccurrence matrix (GLCM) was the most used method to extract texture information [13,14,4]. This method consists of a matrix that is defined over an image to be the distribution of co-occurring pixel values at a given offset.…”

Section: State Of the Artmentioning

confidence: 99%

Comparison of Classical Computer Vision vs. Convolutional Neural Networks for Weed Mapping in Aerial Images

Júnior

Monteiro

Ribeiro

et al. 2020

RITA

View full text Add to dashboard Cite

In this paper, we present a comparison between convolutional neural networks and classicalcomputer vision approaches, for the specific precision agriculture problem of weed mapping on sugarcane fields aerial images. A systematic literature review was conducted to find which computer vision methods are being used on this specific problem. The most cited methods were implemented, as well as four models of convolutional neural networks. All implemented approaches were tested using the same dataset, and their results were quantitatively and qualitatively analyzed. The obtained results were compared to a human expert made ground truth, for validation. The results indicate that the convolutional neural networks present better precision and generalize better than the classical models

show abstract

Vegetation indices and textures in object-based weed detection from UAV imagery

Cited by 24 publications

References 15 publications

Incorporating Surface Elevation Information in UAV Multispectral Images for Mapping Weed Patches

Incorporating Surface Elevation Information in UAV Multispectral Images for Mapping Weed Patches

Aeronave Remotamente Pilotada De Baixo Custo No Estudo De Plantas Invasoras Em Áreas De Cerrado

Comparison of Classical Computer Vision vs. Convolutional Neural Networks for Weed Mapping in Aerial Images

Contact Info

Product

Resources

About