Maize and weed classification using color indices with support vector data description in outdoor fields

Zheng, Yang; Zhu, Qibing; Huang, Min; Guo, Ya; Qin, Jianwei

doi:10.1016/j.compag.2017.07.028

Cited by 105 publications

(60 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This method is implemented at low resolution and operating speeds (average ground speed = 0.09 m/s). Plants were identified in maize fields using a selection of nine colour indices combined with support vector data description in [41] achieving up to 90.79% classification, but with significant variation in results due to weather and time of day. By comparison, our system has a lower classification rate, but provides real-time identification at high resolution for individual plant treatment, whilst correcting target position data through the height estimation algorithm.…”

Section: Resultsmentioning

confidence: 99%

High Speed Crop and Weed Identification in Lettuce Fields for Precision Weeding

Elstone

How

Brodie

et al. 2020

Sensors

View full text Add to dashboard Cite

Precision weeding can significantly reduce or even eliminate the use of herbicides in farming. To achieve high-precision, individual targeting of weeds, high-speed, low-cost plant identification is essential. Our system using the red, green, and near-infrared reflectance, combined with a size differentiation method, is used to identify crops and weeds in lettuce fields. Illumination is provided by LED arrays at 525, 650, and 850 nm, and images are captured in a single-shot using a modified RGB camera. A kinematic stereo method is utilised to compensate for parallax error in images and provide accurate location data of plants. The system was verified in field trials across three lettuce fields at varying growth stages from 0.5 to 10 km/h. In-field results showed weed and crop identification rates of 56% and 69%, respectively. Post-trial processing resulted in average weed and crop identifications of 81% and 88%, respectively.

show abstract

Section: Resultsmentioning

confidence: 99%

High Speed Crop and Weed Identification in Lettuce Fields for Precision Weeding

Elstone

How

Brodie

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…In the first interval, an increase in the values of overall accuracy and kappa coefficient were observed as DAS increased, which is due to an increase in leaf area and vegetation cover index. In the second interval, there is a decrease in vegetation index accuracy, which, according to Zheng et al (2017), is related to a decrease of plant size, leaf area, cover index, and leaf dryness. However, the index WI showed a great variation over crop cycle, always presenting a less accuracy when compared to the others.…”

Section: Classification Assessment Of Vegetation Indices In the Visibmentioning

confidence: 99%

“…According to Motohka et al (2010), Hunt et al (2005), and Tucker (1979), the applicability of vegetation indices to different crops may be limited to a certain stage of plant development. Zheng et al (2017) showed that the vegetation indices EXG, CIVE, and EXGR allowed a vegetation segmentation in corn, in which the highest accuracy values were obtained at the first crop development stages. (2017) showed that the vegetation indices EXG, CIVE, and EXGR allowed a vegetation segmentation in corn, in which the highest accuracy values were obtained at the first crop development stages.…”

Section: Classification Assessment Of Vegetation Indices In the Visibmentioning

confidence: 99%

“…The permanent monitoring of vegetation cover is important to guarantee a sustainable management of agricultural activities, with a relevant role in the reduction of water erosion (CARDOSO et al, 2012;FAUSTOLO et al, 2017;GUIMARAES et al, 2017;SPERANDIO et al, 2012). This monitoring can be carried out by different indicators (CARDOSO et al, 2012;LIMA et al, 2014;MARRERO et al, 2009;PASSOS et al, 2015) such as vegetation cover indices, which represent the percentage of area covered by vegetation in relation to the total surface area of study (ZHONGMING et al, 2010).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Determination of vegetation cover index under different soil management systems of cover plants by using an unmanned aerial vehicle with an onboard digital photographic camera

Beniaich

Silva

Avalos

et al. 2019

SCA

View full text Add to dashboard Cite

Determination of vegetation cover index under different soil management systems of cover plants by using an unmanned aerial vehicle with an onboard digital photographic cameraDeterminação do índice de cobertura vegetal em sistemas de manejo do solo com plantas de cobertura, utilizando veículo aéreo não tripulado com câmera fotográfica digital embarcada AbstractThe permanent monitoring of vegetation cover is important to guarantee a sustainable management of agricultural activities, with a relevant role in the reduction of water erosion. This monitoring can be carried out through different indicators such as vegetation cover indices. In this study, the vegetation cover index was obtained using uncalibrated RGB images generated from a digital photographic camera on an unmanned aerial vehicle (UAV). In addition, a comparative study with 11 vegetation indices was carried out. The vegetation indices CIVE and EXG presented a better performance and the index WI presented the worst performance in the vegetation classification during the cycles of jack bean and millet, according to the overall accuracy and Kappa coefficient. Vegetation indices were effective tools in obtaining soil cover index when compared to the standard Stocking method, except for the index WI. Architecture and cycle of millet and jack bean influenced the behavior of the studied vegetation indices. Vegetation indices generated from RGB images obtained by UAV were more practical and efficient, allowing a more frequent monitoring and in a wider area during the crop cycle. ResumoO monitoramento permanente da cobertura vegetal é importante para garantir o manejo sustentável das atividades agrícolas, com relevante papel na redução da erosão hídrica. Este monitoramento pode ser realizado por meio de diferentes indicadores, como os índices de cobertura vegetal. Nesse artigo o índice de cobertura de vegetação foi obtido usando imagens RGB não-calibradas, geradas a partir de câmera fotográfica digital embarcada em um veículo aéreo não tripulado (VANT). Além disso, foi feito um estudo comparativo de 11 índices de vegetação. Os índices de vegetação CIVE e EXG apresentaram melhor desempenho e o índice WI apresentou o pior desempenho na classificação da vegetação durante o ciclo das culturas de feijão-de-porco e milheto, conforme a acurácia global e o coeficiente Kappa. Os índices de vegetação se apresentaram como uma ferramenta eficaz na obtenção dos índices de cobertura de solo, quando comparados ao método padrão de Stocking, exceto para o índice WI. A arquitetura e o ciclo das culturas milheto e o feijão-de-porco influenciaram no comportamento dos índices de vegetação estudados. Os índices de vegetação gerados à partir de imagens RGB obtidas por VANT mostraram ser mais práticos e eficientes, permitindo o monitoramento com maior frequência e abrangência de área durante o ciclo das culturas. Palavras-chave: Índice de cobertura vegetal. Imagem RGB. Índice de vegetação. Veículo aéreo não tripulado.

show abstract

“…The results indicated that the proposed method is effective to recognize two kinds of weed species. Zheng et al (2017) presented a weed recognition method based on a post-processing algorithm. In the method, 9 optimal color features are selected by principal component analysis (PCA) to eliminate the effect of illumination and noise, and SVM classifier is used to recognize weed species from maize fields.…”

Section: Introductionmentioning

confidence: 99%

Combing K-means Clustering and Local Weighted Maximum Discriminant Projections for Weed Species Recognition

Zhang

Guo²,

Wang

2019

Front. Comput. Sci.

View full text Add to dashboard Cite

Weed species recognition is the premise to control weeds in smart agriculture. It is a challenging topic to control weeds in field, because the weeds in field are quite various and irregular with complex background. A recognition method of weed species in crop field is proposed based on Grabcut and local discriminant projections (LWMDP) algorithm. First, Grabcut is used to remove the most background and K-means clustering (KMC) is utilized to segment weeds from the whole image. Then, LWMDP is employed to extract the low-dimensional discriminant features. Finally, the support vector machine (SVM) classifier is adopted to identify weed species. The characteristics of the proposed method are that, (1) Grabcut and KMC are combined to remove the most of background and obtain the clean weed image, which can reduce the burden of the subsequent feature extraction; (2) LWMDP can project the high-dimensional original image into the low-dimensional subspace, such that the different-class data points are mapped as far as possible, while the within-class data points are projected as close as possible, and the matrix inverse computation is ignored in the generalized eigenvalue problem, thus the small sample size (SSS) problem is avoided naturally. The experimental results on the dataset of the weed species images show that the proposed method is effective for weed identification species, and can preliminarily meet the requirements of multi-row spraying of crop based on machine vision.

show abstract

Maize and weed classification using color indices with support vector data description in outdoor fields

Cited by 105 publications

References 20 publications

High Speed Crop and Weed Identification in Lettuce Fields for Precision Weeding

High Speed Crop and Weed Identification in Lettuce Fields for Precision Weeding

Determination of vegetation cover index under different soil management systems of cover plants by using an unmanned aerial vehicle with an onboard digital photographic camera

Combing K-means Clustering and Local Weighted Maximum Discriminant Projections for Weed Species Recognition

Contact Info

Product

Resources

About