scholarly journals Rice Mapping in Training Sample Shortage Regions Using a Deep Semantic Segmentation Model Trained on Pseudo-Labels

2022 ◽  
Vol 14 (2) ◽  
pp. 328
Author(s):  
Pengliang Wei ◽  
Ran Huang ◽  
Tao Lin ◽  
Jingfeng Huang

A deep semantic segmentation model-based method can achieve state-of-the-art accuracy and high computational efficiency in large-scale crop mapping. However, the model cannot be widely used in actual large-scale crop mapping applications, mainly because the annotation of ground truth data for deep semantic segmentation model training is time-consuming. At the operational level, it is extremely difficult to obtain a large amount of ground reference data by photointerpretation for the model training. Consequently, in order to solve this problem, this study introduces a workflow that aims to extract rice distribution information in training sample shortage regions, using a deep semantic segmentation model (i.e., U-Net) trained on pseudo-labels. Based on the time series Sentinel-1 images, Cropland Data Layer (CDL) and U-Net model, the optimal multi-temporal datasets for rice mapping were summarized, using the global search method. Then, based on the optimal multi-temporal datasets, the proposed workflow (a combination of K-Means and random forest) was directly used to extract the rice-distribution information of Jiangsu (i.e., the K–RF pseudo-labels). For comparison, the optimal well-trained U-Net model acquired from Arkansas (i.e., the transfer model) was also transferred to Jiangsu to extract local rice-distribution information (i.e., the TF pseudo-labels). Finally, the pseudo-labels with high confidences generated from the two methods were further used to retrain the U-Net models, which were suitable for rice mapping in Jiangsu. For different rice planting pattern regions of Jiangsu, the final results showed that, compared with the U-Net model trained on the TF pseudo-labels, the rice area extraction errors of pseudo-labels could be further reduced by using the U-Net model trained on the K–RF pseudo-labels. In addition, compared with the existing rule-based rice mapping methods, he U-Net model trained on the K–RF pseudo-labels could robustly extract the spatial distribution information of rice. Generally, this study could provide new options for applying a deep semantic segmentation model to training sample shortage regions.

2021 ◽  
Author(s):  
Michael Tarasiou

This paper presents DeepSatData a pipeline for automatically generating satellite imagery datasets for training machine learning models. We also discuss design considerations with emphasis on dense classification tasks, e.g. semantic segmentation. The implementation presented makes use of freely available Sentinel-2 data which allows the generation of large scale datasets required for training deep neural networks (DNN). We discuss issues faced from the point of view of DNN training and evaluation such as checking the quality of ground truth data and comment on the scalability of the approach.


2021 ◽  
Author(s):  
Michael Tarasiou

This paper presents DeepSatData a pipeline for automatically generating satellite imagery datasets for training machine learning models. We also discuss design considerations with emphasis on dense classification tasks, e.g. semantic segmentation. The implementation presented makes use of freely available Sentinel-2 data which allows the generation of large scale datasets required for training deep neural networks (DNN). We discuss issues faced from the point of view of DNN training and evaluation such as checking the quality of ground truth data and comment on the scalability of the approach.


2021 ◽  
Vol 13 (9) ◽  
pp. 5274
Author(s):  
Xinyang Yu ◽  
Younggu Her ◽  
Xicun Zhu ◽  
Changhe Lu ◽  
Xuefei Li

Development of a high-accuracy method to extract arable land using effective data sources is crucial to detect and monitor arable land dynamics, servicing land protection and sustainable development. In this study, a new arable land extraction index (ALEI) based on spectral analysis was proposed, examined by ground truth data, and then applied to the Hexi Corridor in northwest China. The arable land and its change patterns during 1990–2020 were extracted and identified using 40 Landsat TM/OLI images acquired in 1990, 2000, 2010, and 2020. The results demonstrated that the proposed method can distinguish arable land areas accurately, with the User’s (Producer’s) accuracy and overall accuracy (kappa coefficient) exceeding 0.90 (0.88) and 0.89 (0.87), respectively. The mean relative error calculated using field survey data obtained in 2012 and 2020 was 0.169 and 0.191, respectively, indicating the feasibility of the ALEI method in arable land extracting. The study found that arable land area in the Hexi Corridor was 13217.58 km2 in 2020, significantly increased by 25.33% compared to that in 1990. At 10-year intervals, the arable land experienced different change patterns. The study results indicate that ALEI index is a promising tool used to effectively extract arable land in the arid area.


2020 ◽  
Vol 36 (10) ◽  
pp. 3011-3017 ◽  
Author(s):  
Olga Mineeva ◽  
Mateo Rojas-Carulla ◽  
Ruth E Ley ◽  
Bernhard Schölkopf ◽  
Nicholas D Youngblut

Abstract Motivation Methodological advances in metagenome assembly are rapidly increasing in the number of published metagenome assemblies. However, identifying misassemblies is challenging due to a lack of closely related reference genomes that can act as pseudo ground truth. Existing reference-free methods are no longer maintained, can make strong assumptions that may not hold across a diversity of research projects, and have not been validated on large-scale metagenome assemblies. Results We present DeepMAsED, a deep learning approach for identifying misassembled contigs without the need for reference genomes. Moreover, we provide an in silico pipeline for generating large-scale, realistic metagenome assemblies for comprehensive model training and testing. DeepMAsED accuracy substantially exceeds the state-of-the-art when applied to large and complex metagenome assemblies. Our model estimates a 1% contig misassembly rate in two recent large-scale metagenome assembly publications. Conclusions DeepMAsED accurately identifies misassemblies in metagenome-assembled contigs from a broad diversity of bacteria and archaea without the need for reference genomes or strong modeling assumptions. Running DeepMAsED is straight-forward, as well as is model re-training with our dataset generation pipeline. Therefore, DeepMAsED is a flexible misassembly classifier that can be applied to a wide range of metagenome assembly projects. Availability and implementation DeepMAsED is available from GitHub at https://github.com/leylabmpi/DeepMAsED. Supplementary information Supplementary data are available at Bioinformatics online.


2018 ◽  
Vol 10 (12) ◽  
pp. 1907 ◽  
Author(s):  
Luís Pádua ◽  
Pedro Marques ◽  
Jonáš Hruška ◽  
Telmo Adão ◽  
Emanuel Peres ◽  
...  

This study aimed to characterize vineyard vegetation thorough multi-temporal monitoring using a commercial low-cost rotary-wing unmanned aerial vehicle (UAV) equipped with a consumer-grade red/green/blue (RGB) sensor. Ground-truth data and UAV-based imagery were acquired on nine distinct dates, covering the most significant vegetative growing cycle until harvesting season, over two selected vineyard plots. The acquired UAV-based imagery underwent photogrammetric processing resulting, per flight, in an orthophoto mosaic, used for vegetation estimation. Digital elevation models were used to compute crop surface models. By filtering vegetation within a given height-range, it was possible to separate grapevine vegetation from other vegetation present in a specific vineyard plot, enabling the estimation of grapevine area and volume. The results showed high accuracy in grapevine detection (94.40%) and low error in grapevine volume estimation (root mean square error of 0.13 m and correlation coefficient of 0.78 for height estimation). The accuracy assessment showed that the proposed method based on UAV-based RGB imagery is effective and has potential to become an operational technique. The proposed method also allows the estimation of grapevine areas that can potentially benefit from canopy management operations.


2016 ◽  
Author(s):  
Anwar Abdelrahman Aly ◽  
Abdulrasoul Mosa Al-Omran ◽  
Abdulazeam Shahwan Sallam ◽  
Mohammad Ibrahim Al-Wabel ◽  
Mohammad Shayaa Al-Shayaa

Abstract. Vegetation cover (VC) changes detection is essential for a better understanding of the interactions and interrelationships between humans and their ecosystem. Remote sensing (RS) technology is one of the most beneficial tools to study spatial and temporal changes of VC. A case study has been conducted in the agro-ecosystem (AE) of Al-Kharj, in the centre of Saudi Arabia. Characteristics and dynamics of VC changes during a period of 26 years (1987–2013) were investigated. A multi-temporal set of images was processed using Landsat images; Landsat4 TM 1987, Landsat7 ETM+ 2000, and Landsat8 2013. The VC pattern and changes were linked to both natural and social processes to investigate the drivers responsible for the change. The analyses of the three satellite images concluded that the surface area of the VC increased by 107.4 % between 1987 and 2000, it was decreased by 27.5 % between years 2000 and 2013. The field study, review of secondary data and community problem diagnosis using the participatory rural appraisal (PRA) method suggested that the drivers for this change are the deterioration and salinization of both soil and water resources. Ground truth data indicated that the deteriorated soils in the eastern part of the Al-Kharj AE are frequently subjected to sand dune encroachment; while the south-western part is frequently subjected to soil and groundwater salinization. The groundwater in the western part of the ecosystem is highly saline, with a salinity ≥ 6 dS m−1. The ecosystem management approach applied in this study can be used to alike AE worldwide.


2020 ◽  
Author(s):  
Moussa Issaka ◽  
Walter Christian ◽  
Michot Didier ◽  
Pichelin Pascal ◽  
Nicolas Hervé ◽  
...  

<p>Salinization and alkalinization are worldwide among the soil degradation threats in irrigated schemes affecting soil productivity. Niger River basin irrigated schemes in the Sahel arid zone are no exception (ONAHA, 2011). The use of remote sensing for identifying and evaluating the level of these phenomena is an interesting tool. The launching of the Sentinel2 satellite constellation (2015) brings new perspectives with high spectral and temporal resolutions images. The aim of this study was to develop a methodology for detection of salt-affected soils in this climatic condition.</p><p>To achieve our goal, we used two types of data: remote sensing and ground truth data.</p><p>Two complementary approaches were used: one by observing salinity on bare soil by the use of salinity index (SI) and the other by observing the indirect effects of salinity on the vegetation during eight (8) rice growth phases  using vegetation index NDVI.</p><p>Remote sensing data were acquired from multi temporal sentinel2 images over 4 years (from 11/12/2015 to 30/11/2019). One hundred and fifty seven (157) images were downloaded (one image each 5 days) and corrected from atmospheric effects and some bands resampled to 5 m using python software. The salinity and vegetation indices were calculated. NDVI index was calculated and NDVI integral between NDVI curve and the threshold of 0.21 NDVI calculated for the eight growing cycles.</p><p>Ground truth data were collected in 2019 during the dry growing season (January – may 2019) from 24 calibration plots and 40 validation plots. One hundred and twenty (120) soil samples collected and analyzed for pH and electrical conductivity and finally forty six (46) biomass samples were collected, air dried and weighed for biomass yield and 46 grains samples collected for grain yield.</p><p>NDVI integral proved to be good indicator for yield variations and could distinguish crops behavior according to the growing period. It also makes it possible to distinguish plots which were not cultivated or with weak growth due to strong constraints of which the main one is salinity. It showed also that the effect of salinity on growth differs according to the growing season and the possibility of managing irrigation. Bare soil analysis distinguishes fields with different salinity indexes despite the low number of dates for which bare soil can be observed.</p><p>Ascending Hierarchical Classification (AHC) enabled to identify four classes of NDVI dynamics over time and bare soil salinity index. High saline soils according to direct soil measurements were related to the class characterized by high frequency of no-cultivation during the dry season and low NDVI integral during the wet season. Multi-temporal Sentinel2 images analysis enabled therefore to detect rice crop fields affected by salinity through its influence on crop behavior. This approach will be tested over the whole paddy schemes of the Niger River valley.</p>


Author(s):  
Suppawong Tuarob ◽  
Conrad S. Tucker

The acquisition and mining of product feature data from online sources such as customer review websites and large scale social media networks is an emerging area of research. In many existing design methodologies that acquire product feature preferences form online sources, the underlying assumption is that product features expressed by customers are explicitly stated and readily observable to be mined using product feature extraction tools. In many scenarios however, product feature preferences expressed by customers are implicit in nature and do not directly map to engineering design targets. For example, a customer may implicitly state “wow I have to squint to read this on the screen”, when the explicit product feature may be a larger screen. The authors of this work propose an inference model that automatically assigns the most probable explicit product feature desired by a customer, given an implicit preference expressed. The algorithm iteratively refines its inference model by presenting a hypothesis and using ground truth data, determining its statistical validity. A case study involving smartphone product features expressed through Twitter networks is presented to demonstrate the effectiveness of the proposed methodology.


2012 ◽  
Vol 18 (1) ◽  
pp. 77-85
Author(s):  
Shinya Tanaka ◽  
Tomoaki Takahashi ◽  
Hideki Saito ◽  
Yoshio Awaya ◽  
Toshiro Iehara ◽  
...  

Author(s):  
Marian Muste ◽  
Ton Hoitink

With a continuous global increase in flood frequency and intensity, there is an immediate need for new science-based solutions for flood mitigation, resilience, and adaptation that can be quickly deployed in any flood-prone area. An integral part of these solutions is the availability of river discharge measurements delivered in real time with high spatiotemporal density and over large-scale areas. Stream stages and the associated discharges are the most perceivable variables of the water cycle and the ones that eventually determine the levels of hazard during floods. Consequently, the availability of discharge records (a.k.a. streamflows) is paramount for flood-risk management because they provide actionable information for organizing the activities before, during, and after floods, and they supply the data for planning and designing floodplain infrastructure. Moreover, the discharge records represent the ground-truth data for developing and continuously improving the accuracy of the hydrologic models used for forecasting streamflows. Acquiring discharge data for streams is critically important not only for flood forecasting and monitoring but also for many other practical uses, such as monitoring water abstractions for supporting decisions in various socioeconomic activities (from agriculture to industry, transportation, and recreation) and for ensuring healthy ecological flows. All these activities require knowledge of past, current, and future flows in rivers and streams. Given its importance, an ability to measure the flow in channels has preoccupied water users for millennia. Starting with the simplest volumetric methods to estimate flows, the measurement of discharge has evolved through continued innovation to sophisticated methods so that today we can continuously acquire and communicate the data in real time. There is no essential difference between the instruments and methods used to acquire streamflow data during normal conditions versus during floods. The measurements during floods are, however, complex, hazardous, and of limited accuracy compared with those acquired during normal flows. The essential differences in the configuration and operation of the instruments and methods for discharge estimation stem from the type of measurements they acquire—that is, discrete and autonomous measurements (i.e., measurements that can be taken any time any place) and those acquired continuously (i.e., estimates based on indirect methods developed for fixed locations). Regardless of the measurement situation and approach, the main concern of the data providers for flooding (as well as for other areas of water resource management) is the timely delivery of accurate discharge data at flood-prone locations across river basins.


Sign in / Sign up

Export Citation Format

Share Document