Advances in Classification of Crops using Remote Sensing Data

Dhumal, Rajesh K.; Vibhute, Amol D.; Nagne, Ajay D.; Rajendra, Yogesh D.; Kale, K. V.; Mehrotra, S. C.

doi:10.23953/cloud.ijarsg.130

Cited by 14 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods that have been used to locate and measure cropped areas include; soliciting information from farmers (FAO 2016), use of Global Positioning Systems (GPS) (Keita, Carfagna, and Mu'Ammar 2010), digitization of high-resolution satellite and aerial images (FAO 2016), pixel-based and objectbased image classification (Dhumal et al 2013). Amongst these, image classification is regarded as the most time-efficient, cost-effective, and objective technique.…”

Section: Introductionmentioning

confidence: 99%

Enhanced mapping of a smallholder crop farming landscape through image fusion and model stacking

Masiza

Chirima

Hamandawana

et al. 2020

International Journal of Remote Sensing

View full text Add to dashboard Cite

Globally, Smallholder farming systems (SFS) are recognized as one of the most important pillars of rural economic development and poverty alleviation because of their contribution to food security. However, support for this agricultural sector is hampered by lack of reliable information on the distributions and acreage of smallholder fields. This information is essential in not only monitoring food security and informing markets but also in guiding the determination of levels of support required from government by individual farmers. There is urgent need for robust techniques that can be used to cost-effectively and time-efficiently map smallholder crop fields especially in Sub-Saharan Africa and Asia. This study attempts to do this by using an approach in which optical and Synthetic Aperture Radar (SAR) data are systematically combined and classified using Extreme Gradient Boosting (Xgboost). We also investigated model stacking as another technique to improve classification accuracy. We combined Xgboost with Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Networks (ANN), and Naïve Bayes (NB). The combined use of multi-temporal Sentinel-2 bands, spectral indices, and Sentinel-1 produced better results than exclusive use of optical data (α = 0.95, p = 0.0005). Furthermore, stacking of classification algorithms based on model comparisons achieved higher accuracy than stacking the algorithms indiscriminately (α = 0.95, p = 0.0100). Through systematic fusion of SAR and optical data and hyper-parameter tuning of Xgboost, we achieved a maximum classification accuracy of 97.71%, while achieving a maximum accuracy of 96.06% through model stacking. This highlights the importance of multi-sensor data fusion and multi-classifier systems when mapping fragmented agricultural landscapes.

show abstract

Section: Introductionmentioning

confidence: 99%

Enhanced mapping of a smallholder crop farming landscape through image fusion and model stacking

Masiza

Chirima

Hamandawana

et al. 2020

International Journal of Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Yahya et al ( 2021) mentioned that digital image classification is the process of sorting all pixels in an image into a finite number of individual classes based on the spectral information and characteristics of these pixels, which have effective significant information on relevant images for classification purposes and interpretation (Dhumal et al 2015). The result of such processes is shown in Figure 3.…”

Section: Land Use and Land Cover (Lulc) Classificationmentioning

confidence: 99%

Assessment of land use and land cover change on Gum Talha (Acacia seyal var. seyal) forest in Bahar Alarab, Sudan

et al. 2022

View full text Add to dashboard Cite

Abstract. Younis AYI, Jonathan BN, Togun AO, Yasin EHE, Omer SO, Hamad GAI, Aissata SD. 2022. Assessment of land use and land cover change on Gum Talha (Acacia seyal var. seyal) forest in Bahar Alarab, Sudan. Biodiversitas 23: 4549-4560. Evaluating land use and land cover change (LULC) is essential for the sustainable management of natural resources, biodiversity conservation, monitoring of food security, and research related to climate change and ecology. A better assessment of land-use changes is highly needed for further investigation due to increasingly rapid changes in LULC in response to human population growth. The emerging climatic change also has a significant effect on LULC. The objective of this study is to assess land use and land cover in Bahar Alarab, East Darfur State, Sudan, using remote sensing data obtained from satellite images. For assessing the LULC changes, Landsat images of the years 1988, 2002, and 2020 were downloaded and analyzed using QGIS 3.22.1 and ERDAS 2014 software, where supervised classification was applied with GPS point verification, change detection, matrix, and accuracy assessment. The analysis on LULC showed considerable changes during the two study periods, where 2020 had a considerable increase in forest cover in which on that year it occupied 26.44% of the area compared to 2002 and 1988 with 21.27% and 21.45%, respectively. Whereas 2002 the area was covered by a vast herbaceous vegetation (33.41%) compared to 1988 and 2002. Moreover, in 1988, shrubland decreased from 31.95% to 29.06% and 23.67% in 2020 and 2002, respectively. Sparse vegetation covered a considerable area in 2020 (23.61%) compared to 2002 and 1988 (21.65% and 17.71%, respectively). The results highlighted that there was statistically significant correlation between climate factors and LULC. Average temperature was highly positively correlated with sparse vegetation at 98.2%, while rainfall was highly negatively correlated with forest (-96.9%) and sparse vegetation (-88%), and highly positively correlated with herbaceous vegetation (83.5%). This study provides a unique understanding of LULC changes and their implications in management and conservation efforts, as well as a road map for decision makers for sustainable development of LULC in the Bahar Alarab.

show abstract

“…The key difference in the two methods lies in the training stage of supervised classification which involves identifying areas of specific spectral attributes for each land-cover or land-use type of interest to the analyst [12] [17] [18] [19]. In comparison unsupervised image classification into spectral classes is based solely on the natural groupings from the image values.…”

Section: IImentioning

confidence: 99%

“…For present research we are focusing on the Hyperspectral data analysis and its application. The Hyperspectral satellite data analysis is new research in the domain of RS and GIS [19] [22] [23].…”

Section: A Hyperspectral Datamentioning

confidence: 99%

Untitled

2018

IJET

View full text Add to dashboard Cite

Advances in Classification of Crops using Remote Sensing Data

Cited by 14 publications

References 22 publications

Enhanced mapping of a smallholder crop farming landscape through image fusion and model stacking

Enhanced mapping of a smallholder crop farming landscape through image fusion and model stacking

Assessment of land use and land cover change on Gum Talha (Acacia seyal var. seyal) forest in Bahar Alarab, Sudan

Untitled

Contact Info

Product

Resources

About