Remotely sensed data have been used in crop condition monitoring for decades. Traditionally, crop growth conditions were assessed by comparing Normalized Difference Vegetation Index (NDVI) of the current year and past years at a pixel scale on the same calendar day. The assumption of this comparison is that the different years’ crops were at the same growing stage on the same day. However, this assumption is often violated in reality. This paper proposes to combine remotely sensed data and meteorological data to assess corn growth conditions at the same growth stages at county level. The proposed approach uses the active accumulated temperature (AAT) computed from Daymet, a daily weather data product, to align different years of NDVI time series at the same growth stages estimated from AATs. The study area covers Carroll County, Iowa. The best index slope extraction (BISE) method and Savitzky–Golay filter are used to filter noise and to reconstruct 11 years of corn growing season NDVI time series from 250 m MODIS daily surface reflectance data product (MOD09GQ). The corn growth stages are identified every year with precise Julian dates from AAT time series. The corn growth conditions are assessed based on the aligned growth stages. The validation of the assessed crop conditions is performed based on National Agricultural Statistics Service (NASS) reports. The study indicates that the crop condition assessment results based on aligned growth stages are consistent with the NASS reported results and they are more reliable than the results based on the same calendar days. The proposed method provides not only crop growth condition information but also crop phenology information. Potentially, it can help improve crop yield prediction since it can effectively measure crop growth changes with NDVI and AAT data.
Remote Sensing technology has been used in agricultural statistics since early 1970s in developed countries and since late 1970s in China. It has greatly improved the efficiency with its accurate, timingly and credible information.But agricultural monitoring using remote sensing has not yet been assessed with credible data in China and its accuracy seems not consistent and reliable to many users. The paper reviews different methods and the corresponding assessments of agricultural monitoring using remote sensing in developed countries and China, then assesses the crop area estimating method using Landsat TM remotely sensed data as sampling area in Northeast China. The ground truth is gathered with global positioning system and 40 sampling areas are used to assess the classification accuracy. The error matrix is constructed from which the accuracy is calculated. The producer accuracy, the user accuracy and total accuracy are 89.53%, 95.37% and 87.02% respectively and the correlation coefficient between the ground truth and classification results is 0.96. A new error index is introduced and the average of rice area estimation to the truth data is 0.084. measures how much the RS classification result is positive or negative apart from the truth data .
In the final feature map obtained using a convolutional neural network for remote sensing image segmentation, there are great differences between the feature values of the pixels near the edge of the block and those inside the block; ensuring consistency between these feature values is the key to improving the accuracy of segmentation results. The proposed model uses an edge feature branch and a semantic feature branch called the edge assistant feature network (EFNet). The EFNET model consists of one semantic branch, one edge branch, one shared decoder, and one classifier. The semantic branch extracts semantic features from remote sensing images, whereas the edge branch extracts edge features from remote sensing images and edge images. In addition, the two branches extract five-level features through five sets of feature extraction units. The shared decoder sets up five levels of shared decoding units, which are used to further integrate edge features and deep semantic features. This strategy can reduce the feature differences between the edge pixels and the inner pixels of the object, obtaining a per-pixel feature vector with high inter-class differentiation and intra-class consistency. Softmax is used as the classifier to generate the final segmentation result. We selected a representative winter wheat region in China (Feicheng City) as the study area and established a dataset for experiments. The comparison experiment included three original models and two models modified by adding edge features: SegNet, UNet, and ERFNet, and edge-UNet and edge-ERFNet, respectively. EFNet's recall (91.01%), intersection over union (81.39%), and F1-Score (91.68%) were superior to those of the other methods. The results clearly show that EFNET improves the accuracy of winter wheat extraction from remote sensing images. This is an important basis not only for crop monitoring, yield estimation, and disaster assessment but also for calculating land carrying capacity and analyzing the comprehensive production capacity of agricultural resources.
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.