Prediction of the Kiwifruit Decline Syndrome in Diseased Orchards by Remote Sensing

Eight years after the first record in Italy, Kiwifruit Decline (KD), a destructive disease causing root rot, has already affected more than 25% of the area under kiwifruit cultivation in Italy. Diseased plants are characterised by severe decay of the fine roots and sudden wilting of the canopy, which is only visible after the season’s first period of heat (July–August). The swiftness of symptom appearance prevents correct timing and positioning for sampling of the disease, and is therefore a barrier to aetiological studies. The aim of this study is to test the feasibility of thermal and multispectral imaging for the detection of KD using an unsupervised classifier. Thus, RGB, multispectral and thermal data from a kiwifruit orchard, with healthy and diseased plants, were acquired simultaneously during two consecutive growing seasons (2017–2018) using an Unmanned Aerial Vehicle (UAV) platform. Data reduction was applied to the clipped areas of the multispectral and thermal data from the 2017 survey. Reduced data were then classified with two unsupervised algorithms, a K-means and a hierarchical method. The plant vigour (canopy size and presence/absence of wilted leaves) and the health shifts exhibited by asymptomatic plants between 2017 and 2018 were evaluated from RGB data via expert assessment and used as the ground truth for cluster interpretation. Multispectral data showed a high correlation with plant vigour, while temperature data demonstrated a good potential use in predicting health shifts, especially in highly vigorous plants that were asymptomatic in 2017 and became symptomatic in 2018. The accuracy of plant vigour assessment was above 73% when using multispectral data, while clustering of the temperature data allowed the prediction of disease outbreak one year in advance, with an accuracy of 71%. Based on our results, the unsupervised clustering of remote sensing data could be a reliable tool for the identification of sampling areas, and can greatly improve aetiological studies of this new disease in kiwifruit.

Download Full-text

Groundwater Potential Zones in Relation to Catchment Condition in Orenburg, Russia

Information Technology and Nanotechnology ◽

10.18287/1613-0073-2019-2391-60-65 ◽

2019 ◽

pp. 60-65

Author(s):

K Choudhary ◽

M S Boori ◽

A Kupriyanov

Keyword(s):

Remote Sensing ◽

Land Use ◽

Remote Sensing Data ◽

Ground Truth ◽

Groundwater Potential ◽

Groundwater Potential Zone ◽

Potential Zone ◽

Ground Truth Data ◽

Groundwater Availability ◽

Sensing Data

The main objective of this study was to detect groundwater availability for agriculture in the Orenburg, Russia. Remote sensing data (RS) and geographic information system (GIS) were used to locate potential zones for groundwater in Orenburg. Diverse maps such as a base map, geomorphological, geological structural, lithology, drainage, slope, land use/cover and groundwater potential zone were prepared using the satellite remote sensing data, ground truth data, and secondary data. ArcGIS software was utilized to manipulate these data sets. The groundwater availability of the study was classified into different classes such as very high, high, moderate, low and very low based on its hydro-geomorphological conditions. The land use/cover map was prepared using a digital classification technique with the limited ground truth for mapping irrigated areas in the Orenburg, Russia.

Download Full-text

A Review of the Application of Remote Sensing Data for Abandoned Agricultural Land Identification with Focus on Central and Eastern Europe

Remote Sensing ◽

10.3390/rs11232759 ◽

2019 ◽

Vol 11 (23) ◽

pp. 2759 ◽

Cited By ~ 6

Author(s):

Tomáš Goga ◽

Ján Feranec ◽

Tomáš Bucha ◽

Miloš Rusnák ◽

Ivan Sačkov ◽

...

Keyword(s):

Remote Sensing ◽

World Literature ◽

Agricultural Land ◽

Remote Sensing Data ◽

Field Research ◽

Radar Data ◽

Expert Assessment ◽

Biomass Estimation ◽

Agricultural Abandonment ◽

Expert Analysis

This study aims to analyze and assess studies published from 1992 to 2019 and listed in the Web of Science (WOS) and Current Contents (CC) databases, and to identify agricultural abandonment by application of remote sensing (RS) optical and microwave data. We selected 73 studies by applying structured queries in a field tag form and Boolean operators in the WOS portal and by expert analysis. An expert assessment yielded the topical picture concerning the definitions and criteria for the identification of abandoned agricultural land (AAL). The analysis also showed the absence of similar field research, which serves not only for validation, but also for understanding the process of agricultural abandonment. The benefit of the fusion of optical and radar data, which supports the application of Sentinel-1 and Sentinel-2 data, is also evident. Knowledge attained from the literary sources indicated that there exists, in the world literature, a well-covered problem of abandonment identification or biomass estimation, as well as missing works dealing with the assessment of the natural accretion of biomass in AAL.

Download Full-text

Measurement of nitrogen content in rice by inversion of hyperspectral reflectance data from an unmanned aerial vehicle

Ciência Rural ◽

10.1590/0103-8478cr20180008 ◽

2018 ◽

Vol 48 (6) ◽

Cited By ~ 1

Author(s):

Du Wen ◽

Xu Tongyu ◽

Yu Fenghua ◽

Chen Chunling

Keyword(s):

Remote Sensing ◽

Nitrogen Content ◽

Unmanned Aerial Vehicle ◽

Gaussian Process Regression ◽

Remote Sensing Data ◽

Principal Component ◽

Ground Truth ◽

Hyperspectral Remote Sensing ◽

Inversion Algorithm ◽

Aerial Vehicle

ABSTRACT: The Nitrogen content of rice leaves has a significant effect on growth quality and crop yield. We proposed and demonstrated a non-invasive method for the quantitative inversion of rice nitrogen content based on hyperspectral remote sensing data collected by an unmanned aerial vehicle (UAV). Rice canopy albedo images were acquired by a hyperspectral imager onboard an M600-UAV platform. The radiation calibration method was then used to process these data and the reflectance of canopy leaves was acquired. Experimental validation was conducted using the rice field of Shenyang Agricultural University, which was classified into 4 fertilizer levels: zero nitrogen, low nitrogen, normal nitrogen, and high nitrogen. Gaussian process regression (GPR) was then used to train the inversion algorithm to identify specific spectral bands with the highest contribution. This led to a reduction in noise and a higher inversion accuracy. Principal component analysis (PCA) was also used for dimensionality reduction, thereby reducing redundant information and significantly increasing efficiency. A comparison with ground truth measurements demonstrated that the proposed technique was successful in establishing a nitrogen inversion model, the accuracy of which was quantified using a linear fit (R2=0.8525) and the root mean square error (RMSE=0.9507). These results support the use of GPR and provide a theoretical basis for the inversion of rice nitrogen by UAV hyperspectral remote sensing.

Download Full-text

BENCHMARKING DEEP LEARNING FRAMEWORKS FOR THE CLASSIFICATION OF VERY HIGH RESOLUTION SATELLITE MULTISPECTRAL DATA

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprsannals-iii-7-83-2016 ◽

2016 ◽

Vol III-7 ◽

pp. 83-88 ◽

Cited By ~ 24

Author(s):

M. Papadomanolaki ◽

M. Vakalopoulou ◽

S. Zagoruyko ◽

K. Karantzalos

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

High Resolution ◽

Remote Sensing Data ◽

Multispectral Data ◽

Sensing Data ◽

Multispectral Remote Sensing ◽

High Level ◽

Learning Frameworks

In this paper we evaluated deep-learning frameworks based on Convolutional Neural Networks for the accurate classification of multispectral remote sensing data. Certain state-of-the-art models have been tested on the publicly available SAT-4 and SAT-6 high resolution satellite multispectral datasets. In particular, the performed benchmark included the AlexNet, AlexNet-small and VGG models which had been trained and applied to both datasets exploiting all the available spectral information. Deep Belief Networks, Autoencoders and other semi-supervised frameworks have been, also, compared. The high level features that were calculated from the tested models managed to classify the different land cover classes with significantly high accuracy rates i.e., above 99.9%. The experimental results demonstrate the great potentials of advanced deep-learning frameworks for the supervised classification of high resolution multispectral remote sensing data.

Download Full-text

A Method of Using WRF-Simulated Surface Temperature to Estimate Daily Evapotranspiration

Journal of Applied Meteorology and Climatology ◽

10.1175/jamc-d-19-0287.1 ◽

2020 ◽

Vol 59 (5) ◽

pp. 901-914 ◽

Cited By ~ 1

Author(s):

Dakang Wang ◽

Yan Liu ◽

Tao Yu ◽

Yin Zhang ◽

Qixin Liu ◽

...

Keyword(s):

Remote Sensing ◽

Surface Temperature ◽

Continuous Time ◽

Land Surface ◽

Remote Sensing Data ◽

Temperature Data ◽

Model Data ◽

Simulated Surface ◽

Surface Temperature Data ◽

Better Than

AbstractSurface temperature is one of the key parameters for estimating regional evapotranspiration (ET) based on the Surface Energy Balance System (SEBS) model using remote sensing data. However, continuous daily remote sensing surface temperature data are often not available due to the weather and environmental conditions. This paper proposed a scheme to obtain reliable ET that estimating ET using WRF-simulated surface skin temperature (TSK) and then modifying the deviation using the normalized difference vegetation index (NDVI). This study aims to explore whether the model data can be a viable option when the continuous-time-series remote sensing surface temperature is missing for estimating ET. Comparison results show that the correlation between WRF TSK and the measured temperature of the 2-cm soil (Ts) is better than MODIS land surface temperature (LST) in the study area, while the correlation between MODIS LST and the measured surface radiation temperature (IRT) is better than WRF TSK. The MODIS LST is significantly higher than Ts, and the WRF TSK is closer to Ts than MODIS LST. However, the ET calculated using WRF TSK was not good, exhibiting relatively high ET in the whole area and a poor correlation with the measurements, whereby R2, RMSE, and the percent bias (PBIAS) were equal to 0.1256, 5.2783 mm, and −202.17%, respectively. According to the principle of land surface process simulation in WRF, this paper proposes using NDVI to modify ET calculated using TSK. The comparison between the modified ET and the measurements exhibited a relatively good correlation, with R2 = 0.7532, RMSE = 1.0993 mm, and PBIAS = −17.9%. Therefore, the model surface temperature data can be used to estimate continuous-time-series regional ET when NDVI is used to modify the deviation, which indicates the surface temperature data simulated by the WRF Model can become the optional data for estimating ET and compensate for the shortcoming of poor time continuity of remote sensing data, further expanding the application prospects of meteorological model data in the remote sensing field.

Download Full-text

Taking the pulse of the Amazon rainforest by fusing multitemporal Sentinel 1 and 2 data for advanced deep-learning

10.5194/egusphere-egu21-3749 ◽

2021 ◽

Author(s):

Melanie Brandmeier ◽

Eya Cherif

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

Land Cover ◽

Supervised Learning ◽

Large Scale ◽

Remote Sensing Data ◽

Ground Truth ◽

Brazilian Amazon ◽

Sensing Data ◽

Weakly Supervised

Degradation of large forest areas such as the Brazilian Amazon due to logging and fires can increase the human footprint way beyond deforestation. Monitoring and quantifying such changes on a large scale has been addressed by several research groups (e.g. Souza et al. 2013) by making use of freely available remote sensing data such as the Landsat archive. However, fully automatic large-scale land cover/land use mapping is still one of the great challenges in remote sensing. One problem is the availability of reliable &#8220;ground truth&#8221; labels for training supervised learning algorithms. For the Amazon area, several landcover maps with 22 classes are available from the MapBiomas project that were derived by semi-automatic classification and verified by extensive fieldwork (Project MapBiomas). These labels cannot be considered real ground-truth as they were derived from Landsat data themselves but can still be used for weakly supervised training of deep-learning models that have a potential to improve predictions on higher resolution data nowadays available. The term weakly supervised learning was originally coined by (Zhou 2017) and refers to the attempt of constructing predictive models from incomplete, inexact and/or inaccurate labels as is often the case in remote sensing. To this end, we investigate advanced deep-learning strategies on Sentinel-1 timeseries and Sentinel-2 optical data to improve large-scale automatic mapping and monitoring of landcover changes in the Amazon area. Sentinel-1 data has the advantage to be resistant to cloud cover that often hinders optical remote sensing in the tropics.We propose new architectures that are adapted to the particularities of remote sensing data (S1 timeseries and multispectral S2 data) and compare the performance to state-of-the-art models.&#160; Results using only spectral data were very promising with overall test accuracies of 77.9% for Unet and 74.7% for a DeepLab implementation with ResNet50 backbone and F1 measures of 43.2% and 44.2% respectively.&#160; On the other hand, preliminary results for new architectures leveraging the multi-temporal aspect of &#160;SAR data have improved the quality of mapping, particularly for agricultural classes. For instance, our new designed network AtrousDeepForestM2 has a similar quantitative performances as DeepLab &#160;(F1 of 58.1% vs 62.1%), however it produces better qualitative land cover maps.To make our approach scalable and feasible for others, we integrate the trained models in a geoprocessing tool in ArcGIS that can also be deployed in a cloud environment and offers a variety of post-processing options to the user.Souza, J., Carlos M., et al. (2013). "Ten-Year Landsat Classification of Deforestation and Forest Degradation in the Brazilian Amazon." Remote Sensing 5(11): 5493-5513.&#160;&#160;&#160;Zhou, Z.-H. (2017). "A brief introduction to weakly supervised learning." National Science Review 5(1): 44-53."Project MapBiomas - Collection&#160; 4.1 of Brazilian Land Cover & Use Map Series, accessed on January 2020 through the link: https://mapbiomas.org/colecoes-mapbiomas?cama_set_language=en"

Download Full-text

Enhancement of Component Images of Multispectral Data by Denoising with Reference

Remote Sensing ◽

10.3390/rs11060611 ◽

2019 ◽

Vol 11 (6) ◽

pp. 611 ◽

Cited By ~ 5

Author(s):

Sergey Abramov ◽

Mikhail Uss ◽

Vladimir Lukin ◽

Benoit Vozel ◽

Kacem Chehdi ◽

...

Keyword(s):

Remote Sensing ◽

High Efficiency ◽

Real Life ◽

Remote Sensing Data ◽

Nonlinear Transformations ◽

High Quality ◽

Multispectral Data ◽

Sensing Data ◽

Reference Images ◽

Multispectral Remote Sensing

Multispectral remote sensing data may contain component images that are heavily corrupted by noise and the pre-filtering (denoising) procedure is often applied to enhance these component images. To do this, one can use reference images—component images having relatively high quality and that are similar to the image subject to pre-filtering. Here, we study the following problems: how to select component images that can be used as references (e.g., for the Sentinel multispectral remote sensing data) and how to perform the actual denoising. We demonstrate that component images of the same resolution as well as component images of a better resolution can be used as references. To provide high efficiency of denoising, reference images have to be transformed using linear or nonlinear transformations. This paper proposes a practical approach to doing this. Examples of denoising tests and real-life images demonstrate high efficiency of the proposed approach.

Download Full-text

Do Game Data Generalize Well for Remote Sensing Image Segmentation?

Remote Sensing ◽

10.3390/rs12020275 ◽

2020 ◽

Vol 12 (2) ◽

pp. 275 ◽

Cited By ~ 1

Author(s):

Zhengxia Zou ◽

Tianyang Shi ◽

Wenyuan Li ◽

Zhou Zhang ◽

Zhenwei Shi

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

Video Game ◽

Real World ◽

Remote Sensing Data ◽

Ground Truth ◽

Remote Sensing Image ◽

Learning Model ◽

Image Translation ◽

Deep Learning Model

Despite the recent progress in deep learning and remote sensing image interpretation, the adaption of a deep learning model between different sources of remote sensing data still remains a challenge. This paper investigates an interesting question: do synthetic data generalize well for remote sensing image applications? To answer this question, we take the building segmentation as an example by training a deep learning model on the city map of a well-known video game “Grand Theft Auto V” and then adapting the model to real-world remote sensing images. We propose a generative adversarial training based segmentation framework to improve the adaptability of the segmentation model. Our model consists of a CycleGAN model and a ResNet based segmentation network, where the former one is a well-known image-to-image translation framework which learns a mapping of the image from the game domain to the remote sensing domain; and the latter one learns to predict pixel-wise building masks based on the transformed data. All models in our method can be trained in an end-to-end fashion. The segmentation model can be trained without using any additional ground truth reference of the real-world images. Experimental results on a public building segmentation dataset suggest the effectiveness of our adaptation method. Our method shows superiority over other state-of-the-art semantic segmentation methods, for example, Deeplab-v3 and UNet. Another advantage of our method is that by introducing semantic information to the image-to-image translation framework, the image style conversion can be further improved.

Download Full-text

Spectral Characteristics of the Antarctic Vegetation: A Case Study of Barton Peninsula

Remote Sensing ◽

10.3390/rs13132470 ◽

2021 ◽

Vol 13 (13) ◽

pp. 2470

Author(s):

Junhwa Chi ◽

Hyoungseok Lee ◽

Soon Gyu Hong ◽

Hyun-Cheol Kim

Keyword(s):

Remote Sensing ◽

Spectral Characteristics ◽

Remote Sensing Data ◽

Ground Truth ◽

Spectral Unmixing ◽

Polar Regions ◽

Spectral Library ◽

Sensing Data ◽

Barton Peninsula ◽

Shortwave Infrared

Spectral information is a proxy for understanding the characteristics of ground targets without a potentially disruptive contact. A spectral library is a collection of this information and serves as reference data in remote sensing analyses. Although widely used, data of this type for most ground objects in polar regions are notably absent. Remote sensing data are widely used in polar research because they can provide helpful information for difficult-to-access or extensive areas. However, a lack of ground truth hinders remote sensing efforts. Accordingly, a spectral library was developed for 16 common vegetation species and decayed moss in the ice-free areas of Antarctica using a field spectrometer. In particular, the relative importance of shortwave infrared wavelengths in identifying Antarctic vegetation using spectral similarity comparisons was demonstrated. Due to the lack of available remote sensing images of the study area, simulated images were generated using the developed spectral library. Then, these images were used to evaluate the potential performance of the classification and spectral unmixing according to spectral resolution. We believe that the developed library will enhance our understanding of Antarctic vegetation and will assist in the analysis of various remote sensing data.

Download Full-text

Potencial de las imágenes UAV como datos de verdad terreno para la clasificación de la severidad de quema de imágenes Landsat: aproximaciones a un producto útil para la gestión post incendio

Revista de Teledetección ◽

10.4995/raet.2017.7140 ◽

2017 ◽

pp. 91 ◽

Cited By ~ 12

Author(s):

M. Pla ◽

A. Duane ◽

L. Brotons

Keyword(s):

Remote Sensing ◽

Decision Making ◽

Forest Fires ◽

Fire Severity ◽

Remote Sensing Data ◽

Ground Truth ◽

Linear Regression Models ◽

Kappa Index ◽

Landsat Images ◽

Uav Images

Mapping fire severity is determinant to understand landscape evolution after a wildfire and provides useful information for decision making during post fire management. Quantitative fire severity mapping from relative changes in Normalized Burn Ratio index (RdNBR) is not actually being incorporated into decision making processes, being more useful the categorization in severity levels (high, moderate and low). However, the most common mapping severity methodologies based on the definition of RdNBR thresholds from field information are not always possible due to lack of field data or because the published thresholds are unsatisfactory in new regions. The boom in the use of UAVs (Unmanned Aerial Vehicle) has raised these platforms as potential tools for validation of remote sensing data. This paper presents the potential of UAVs as ground truth information in forest fires. From the photointerpretation of high resolution RGB images, the Aerial Severity Proportion Index (ASPI) has been created. Non-linear regression models between RdNBR and ASPI allows to delimitate of thresholds for the classification of Landsat images and to obtain qualitative severity maps. Validation with random points presents a kappa index of 0,5 and a relative accuracy of 70,8%. Therefore, UAV images become a very useful tool for wildfire severity mapping and for fill the gap between remote sensing information and expensive field ground campaigns

Download Full-text