K-means Classification Method Research Articles

Metal Artifact Reduction in CT Images through Sinogram Data Inpainting

When metallic implants are present within the human body, they frequently introduce metallic artifacts into X-ray CT images. These artifacts can lead to significant distortions, obscuring critical information and potentially degrading the quality of the CT images, thereby impacting diagnostic accuracy for clinicians. In recent years, there has been extensive research aimed at mitigating the challenges posed by metallic artifacts, resulting in the development of multiple solutions to address this issue. In this study, we present an efficient approach for artifact removal. Our method involves utilizing the image reconstructed from a sinogram affected by artifacts to generate a synthesized sinogram, deviating from the conventional acquisition of sinogram data. The key stages of our approach encompass segmentation, sinogram gap-filling, and subsequent image enhancement. To achieve rapid segmentation, we employed a K-means classification method. For the retrieval of missing data, we utilized an interpolation algorithm based on a penalized least squares method. In the final phase of image reconstruction enhancement, we implemented an advanced contrast equalization technique to restore image intensities to their inherent dynamic range. Through rigorous verification using both simulated and clinical data, our method consistently demonstrates a remarkable improvement in image quality.

A bone structure-inspired design of bionic wing rib

This paper presents a method to generative design bionic bone structures wing ribs by simulating the formation process of soap bubble boundary. In this method, the position of the central control point is optimized by the genetic algorithm to simulate the moving process of the bubble, and the control point is simplified by the K-means classification method to improve the computational efficiency. The boundary formed by the extrusion of bubbles is filled with a bone-like grid structure to realize the optimal design of a bone-like rib structure. Compared with the traditional honeycomb sandwich structure, the bone-like rib structure can distribute the stress evenly throughout the structure, thus reducing the stress concentration and improving the bearing capacity of the structure. The method proposed in this paper provides a new idea for a multilevel structure design.

Maturity Analysis of Stock Exchanges in Africa: Prepandemic Sustainability Perspective

This paper focuses on the economic dimension of sustainability by examining the stock exchange interface of financial markets, the influence of capital market stakeholders, and the instruments that contribute to a supportive fiscal framework. Only mature stock exchanges are present in sustainability indices; hence, comparative assessment of stock exchanges is limited and contributes to the complexity of conducting such a study. Utilizing multivariate analysis, this study investigates the potential for African stock exchanges to support sustainability. An empirical study was conducted on a selected sample of 15 African stock exchanges at the end of 2020 using collected 5-year interval data from Q1 of 2021. A total of 22 variables were selected based on their legitimacy to support sustainability. Using exploratory factor analysis, two key sustainability drivers of differentiation and classified exchanges were identified, i.e., hard and soft. K-means classification method verified the results and found that of the four identified homogeneous groups, one—the Johannesburg Stock Exchange, Nigerian Stock Exchange, and the Egyptian Exchange—emerged on top. Two smaller groups had the potential to be strengthened, and the majority group lagged behind. The research demonstrated the importance of identifying key sustainability drivers and examined the materiality of the drivers within an African context.

Assessing dynamic metabolic heterogeneity in non-small cell lung cancer patients via ultra-high sensitivity total-body [18F]FDG PET/CT imaging: quantitative analysis of [18F]FDG uptake in primary tumors and metastatic lymph nodes.

This study aimed to quantitatively assess [18F]FDG uptake in primary tumor (PT) and metastatic lymph node (mLN) in newly diagnosed non-small cell lung cancer (NSCLC) using the total-body [18F]FDG PET/CT and to characterize the dynamic metabolic heterogeneity of NSCLC. The 60-min dynamic total-body [18F]FDG PET/CT was performed before treatment. The PTs and mLNs were manually delineated. An unsupervised K-means classification method was used to cluster patients based on the imaging features of PTs. The metabolic features, including Patlak-Ki, Patlak-Intercept, SUVmean, metabolic tumor volume (MTV), total lesion glycolysis (TLG), and textural features, were extracted from PTs and mLNs. The targeted next-generation sequencing of tumor-associated genes was performed. The expression of Ki67, CD3, CD8, CD34, CD68, and CD163 in PTs was determined by immunohistochemistry. A total of 30 patients with stage IIIA-IV NSCLC were enrolled. Patients were divided into fast dynamic FDG metabolic group (F-DFM) and slow dynamic FDG metabolic group (S-DFM) by the unsupervised K-means classification of PTs. The F-DFM group showed significantly higher Patlak-Ki (P < 0.001) and SUVmean (P < 0.001) of PTs compared with the S-DFM group, while no significant difference was observed in Patlak-Ki and SUVmean of mLNs between the two groups. The texture analysis indicated that PTs in the S-DFM group were more heterogeneous in FDG uptake than those in the F-DFM group. Higher T cells (CD3+/CD8+) and macrophages (CD68+/CD163+) infiltration in the PTs were observed in the F-DFM group. No significant difference was observed in tumor mutational burden between the two groups. The dynamic total-body [18F]FDG PET/CT stratified NSCLC patients into the F-DFM and S-DFM groups, based on Patlak-Ki and SUVmean of PTs. PTs in the F-DFM group seemed to be more homogenous in terms of [18F]FDG uptake than those in the S-DFM group. The higher infiltrations of T cells and macrophages were observed in the F-DFM group, which suggested a potential benefit from immunotherapy.

Mapping the habitats of a complex circalittoral rocky shelf in the Cantabrian Sea (south Bay of Biscay)

This work focuses on the study of habitats and communities of a high structural complexity area at different levels and scales. This gives us a better understanding of an area from an ecological point of view and at the same time provides us with tools that will facilitate management measures. It was developed in a complex circalittoral rocky platform of the central Cantabrian Sea (Bay of Biscay). The sampling was carried out using a towed photogrammetric vehicle and a rock dredge, which was used for the identification of the species. The first level of the study was the abiotic characterization of the area and the analysis of the communities. These analysis were developed using the unsupervised k-means classification method. For abiotic characterization we used the variables directly associated with the composition and morphology of the ground, such as backscatter, BPI (Bathymetric Position Index), roughness and slope. Depth was also included to discriminate between the circalittoral and bathyal zones. We obtained 5 different classes, which we related to the ground types observed by photogrammetry. In the analysis of the communities, the cluster was based on the sampling units extracted from the images (∼10 m), from which 5 assemblages were obtained, providing information on the most abundant species of each class supplied by the abiotic study. The second level was carried out considering a management approach and was based on the modeling of the area at lower resolution, more suitable for the analysis of the habitat-fisheries interactions. Thus, the main habitat-forming species (HFS) of the entire circalittoral area were used to perform delta models based on GAMs (Generalize Additive Models). Obtaining the predictions of presence/absence and combining it with the predictions of densities, we got the zero inflated values density-based model. As all the identified habitats have vulnerable benthic species of a certain size settled on rocky bottoms, they can all be considered to belong to the designation 1170 reefs of the Habitats Directive.

The role of environmental legislation and land use patterns on riparian deforestation dynamics in an Amazonian agricultural frontier (MT, Brazil)

Agribusiness expansion and intensification in Brazil have prompted an abrupt change in land use and occupation in the Amazonian agricultural frontier since the 1980s. Considering the increasing suppression of native vegetation, riparian areas represent an important tool for protecting ecosystem services and biodiversity. Although the effects of land tenure and land use on large-scale deforestation in the Amazon have been widely assessed, their roles on riparian deforestation remains poorly explored. Here we assessed two municipalities – Querência (QRC) and São José do Xingu (SJX) – located in an agricultural frontier of the Brazilian Amazon to explore the relationship of illegal deforestation in riparian areas and different types of land use and property-sizes, as well as the impact of the Native Vegetation Protection Law (NVPL, Federal Law 12,651/2012) on environmental compliance. Therefore, we developed a robust geodatabase using hydrographic, land use and land tenure data. Riparian areas protected as Permanent Preservation Areas (PPAs) were delimited and their land cover mapped for 2012 and 2018 using high spatial resolution satellite images and unsupervised K-means classification method. We also applied landscape metrics to analyze riparian PPA structure and dynamics. Our results indicated that NVPL was followed by a downward trend in the riparian vegetation deficits in all land use types and property-sizes, but it did not stop new clearings. Although riparian PPAs in minifundios (≤ 80 ha) and agrarian reform settlements tended to concentrate higher relative deforestation amounts, large-sized farms were responsible for most of the absolute extent of riparian deforestation in both years, accounting for 76–78% in QRC and 93% in SJX. They were also the main drivers of new clearings, for which account for 71% in QRC and 86% in SJX. The impact of land use on riparian deforestation was not homogeneous among properties, possibly reflecting different levels of technological investment and management techniques. In the so-called consolidated areas, in which the riparian PPA minimum width was reduced by NVPL, decreases in deforestation between 2012 and 2018 were lower. In these areas, vegetation coverage did not exceed 23% in any of the study areas. In the riparian PPAs that was not under consolidated areas, the vegetation coverage was of at least 85% of the area. Local environmental governance may also have affected the riparian deforestation dynamics, in which stricter monitoring and law enforcement lead to lower deforestation extents. Finally, landscape metrics revealed the importance of managing riparian areas at the landscape level, as local improvements did not necessarily result in connectivity gains.

Probabilistic unsupervised classification for large-scale analysis of spectral imaging data

Land cover classification of remote sensing data is a fundamental tool to study changes in the environment such as deforestation or wildfires. A current challenge is to quantify land cover changes with real-time, large-scale data from modern hyper- or multispectral sensors. A range of methods are available for this task, several of them being based on the k-means classification method which is efficient when classes of land cover are well separated. Here a new algorithm, called probabilistic k-means, is presented to solve some of the limitations of the standard k-means. It is shown that the new algorithm performs better than the standard k-means when the data are noisy. If the number of land cover classes is unknown, an entropy-based criterion can be used to select the best number of classes. The proposed new algorithm is implemented in a combination of R and C computer codes which is particularly efficient with large data sets: a whole image with more than 3 million pixels and covering more than 10,000 km2 can be analysed in a few minutes. Four applications with hyperspectral and multispectral data are presented. For the data sets with ground truth data, the overall accuracy of the probabilistic k-means was substantially improved compared to the standard k-means. One of these data sets includes more than 120 million pixels, demonstrating the scalability of the proposed approach. These developments open new perspectives for the large scale analysis of remote sensing data. All computer code are available in an open-source package called sentinel.

Student Grouping Based on Final Exam Values of the Courses with the K-Means Classification Method Using KNIME

In learning, each student must have a different way of learning and learning patterns which will have an impact on the results of their learning evaluation at the end of each semester. Assessing a student who excels in learning is one of them by looking at the score of the final exam results that maybe the student can easily get good grades because they do have expertise in that field or get good grades because they are diligent in studying. The scores of students' final semester exams in several courses will be summarized here in order to be used as a basis for classifying students into several groups, namely smart, average, and less good students.

Aortic aneurysm evaluation by scanning acoustic microscopy and Raman spectroscopy.

Aortic aneurysm is observed as a result of the extensive alteration in the elasticity of the aortic wall due to the breakdown of elastin and collagen. In this study, we studied the feasibility of scanning acoustic microscopy (SAM) and Raman spectroscopy (RS) in characterizing the dilated segments of the aorta from male and female patients with aortic aneurysm. SAM determined the acoustic property variation in the aorta by calculating the acoustic impedance values of aorta samples of 18 patients. RS determined the disease states by analyzing the chemical variation especially in the peaks related to elastin and collagen using the k-means classification method. Consequently, we assume that combining these two techniques in clinics will help to investigate the dilated segment of the aorta with micrometer resolution, which will reduce the possibility of new aneurysm formation due to a segment not excised during the surgery.

Low-light image enhancement algorithm based on an atmospheric physical model

Under low illumination, the colour constancy of human vision can be used for correctly determining the colour of objects according to the fixed reflection coefficient of external light and objects. However, video image acquisition equipment does not implement the colour constancy characteristic of the human visual system. Under low illumination, only a small amount of light is reflected from the surface of the imaged object; as a result, the captured image is underexposed. After statistical analysis of low-light images, these inverted underexposed images appear foggy. Inversion is a uniform and reversible operation that is performed on the entire image. Hereby, a method is proposed for resolving low-light images using conventional physical models. First, a low-light image is inverted for obtaining a foggy image. Subsequently, a pyramid-type dense residual block network and a dark channel prior K-means classification method are applied to the foggy image, to calculate the transmission and atmospheric light. Finally, the parameters obtained from this solution are incorporated into the low-light imaging model to obtain a clear image. We subjectively and qualitatively analysed the experimental results, and used information entropy and average gradient for objective quantitative analysis. We demonstrate that the algorithm improves the overall brightness and contrast of the imaged scenes, and the obtained enhanced images are clear and natural.

An Analysis Method of Crowd Abnormal Behavior for Video Service Robot

Abnormal behavior analysis in crowded scenes is a challenging task in the computer vision field. Determining how to quickly and accurately locate abnormal behavior in a crowd, especially in airports, shopping malls, stations, and other practical application scenarios, has become the key to the timely handling of mass incidents. The authors have developed an embedded video analysis robot capable of local processing for crowd anomaly detection in application scenarios. In recent years, many methods of crowd anomaly detection have been proposed by scholars, but the performance of these methods when applied to video service robots is unsatisfactory. In this paper, an anomaly analysis method based on improved k-means is proposed. This method combines mean shift and the k-means classification method to achieve rapid and accurate crowd anomaly detection. A quantitative experimental evaluation was carried out on a video service robot, and the results demonstrate that the method is effective for the detection of abnormal behavior on multiple and publicly available video sequences.

LeafletAnalyzer, an Automated Software for Quantifying, Comparing and Classifying Blade and Serration Features of Compound Leaves during Development, and among Induced Mutants and Natural Variants in the Legume Medicago truncatula.

Diverse leaf forms ranging from simple to compound leaves are found in plants. It is known that the final leaf size and shape vary greatly in response to developmental and environmental changes. However, changes in leaf size and shape have been quantitatively characterized only in a limited number of species. Here, we report development of LeafletAnalyzer, an automated image analysis and classification software to analyze and classify blade and serration characteristics of trifoliate leaves in Medicago truncatula. The software processes high quality leaf images in an automated or manual fashion to generate size and shape parameters for both blades and serrations. In addition, it generates spectral components for each leaflets using elliptic Fourier transformation. Reconstruction studies show that the spectral components can be reliably used to rebuild the original leaflet images, with low, and middle and high frequency spectral components corresponding to the outline and serration of leaflets, respectively. The software uses artificial neutral network or k-means classification method to classify leaflet groups that are developed either on successive nodes of stems within a genotype or among genotypes such as natural variants and developmental mutants. The automated feature of the software allows analysis of thousands of leaf samples within a short period of time, thus facilitating identification, comparison and classification of leaf groups based on leaflet size, shape and tooth features during leaf development, and among induced mutants and natural variants.

Analiza zróżnicowania wiedzy absolwentów gimnazjów w latach 2002-2013

The aim of the article is to apply selected methods of econometrics (dynamic and spatial) for analysis and assessment of knowledge and skills of junior high schools graduates in the years 2002-2013. In particular, the objective of the analyses was to answer the following questions:- How did junior high school students cope with the tasks and issues contained in the exam in the part concerned with mathematics and natural sciences as well as in the humanities part in subsequent years?- Are there any differences in the results obtained in the exams in the sub-periods of the twelve years under discussion?- Are there any differences in the exam results in the individual sub-periods between voivodships and the parts of the exam?In the analyses, data were used obtained from Student Achievement Analysis Team operating at the Institute for Educational Research[1] which made it possible not only to assess the dynamic changes in the graduates’ exam scores but also to conduct their spatial analysis by voivodships. Linear trend was used for the dynamics analysis, whereas for the studies of spatial changes, the k-means classification method was employed.The analysis showed certain regularities present in the scores achieved by the junior high school graduates:- There was a systematic decline in the graduates’ performance in mathematics and natural sciences in all voivodships in the period under study.- The division of the years 2002-2013 into three periods – good, medium and poor results of junior high school graduates – had no clear impact on the diversification amongst voivodships.- Students from the voivodships of south-eastern Poland achieved better exam scores in mathematics and natural sciences, and humanities.The analysis has cognitive and applicative value. It can be used by local governments in their decision-making process relating to the improvement of the quality of educational services at the junior high school level.[1] http://pwe.ibe.edu.pl/ [accessed on 20.09.2015]

Learning-based compressed sensing for infrared image super resolution

This paper presents an infrared image super-resolution method based on compressed sensing (CS). First, the reconstruction model under the CS framework is established and a Toeplitz matrix is selected as the sensing matrix. Compared with traditional learning-based methods, the proposed method uses a set of sub-dictionaries instead of two coupled dictionaries to recover high resolution (HR) images. And Toeplitz sensing matrix allows the proposed method time-efficient. Second, all training samples are divided into several feature spaces by using the proposed adaptive k-means classification method, which is more accurate than the standard k-means method. On the basis of this approach, a complex nonlinear mapping from the HR space to low resolution (LR) space can be converted into several compact linear mappings. Finally, the relationships between HR and LR image patches can be obtained by multi-sub-dictionaries and HR infrared images are reconstructed by the input LR images and multi-sub-dictionaries. The experimental results show that the proposed method is quantitatively and qualitatively more effective than other state-of-the-art methods.

Construction of quality of life change patterns: example in oncology in a phase III therapeutic trial (FFCD 0307).

ObjectiveQuality of life data in cancerology are often difficult to summarize due to missing data and difficulty to analyze the pattern of evolution in different groups of patients. The aim of this work was to apply a new methodology to construct Quality of Life (QoL) change patterns within patients included in a clinical trial comparing to regimen of treatment in locally advanced eosogastric cancer.Materials and methodsIn this trial, QoL was assessed every 2 months by self-reported EORTC QLQ-C30 questionnaire. Physical dimension scores were analyzed. After multiple imputation of missing data, 27 statistical measures aiming to describe the variation of QoL measures among follow-up were computed for each patient. Based on these measures, patient were grouped into homogenous groups in terms of QoL variation pattern using a K-Means classification method. The mean QoL score at each time was graphically represented in each obtained pattern. Finally, clinical characteristic of patients in each pattern of QoL were described and compared.ResultsThe trial included 416 patients and 1023 questionnaire were collected. 74 % of patients were male with a mean ± SD age of 62 ± 11 years. 43 % of scores were missing. Patients were grouped into four classes of homogeneous QoL variation patterns. 1) a Pattern of 24 (6 %) patients showing improvement in QoL with a mean variation of +10.7 points on the 0–100 scale, 2) a Pattern of 171 (41 %) patients showing a stability 3) two Patterns of 78 (19 %) and 143 (34 %) patients respectively showing a deterioration of QoL with a mean variation of −67.2 and −67.6, respectively. There were no difference between patterns in terms of gender or age. Patients within “degradation” pattern had significantly lower performance status (p = 0.015), higher severe after-effects rate (p < 10-3) and death rate (p < 10-3).ConclusionThis work opens up perspectives for longitudinal data analysis with a high probability of missing values while providing a relevant graphical summary. Patterns of QoL evolution with clinical relevance may help to interpret longitudinal QoL data in Cancer studies.Electronic supplementary materialThe online version of this article (doi:10.1186/s12955-015-0342-1) contains supplementary material, which is available to authorized users.

Built-up Areas Extraction in High Resolution SAR Imagery based on the method of Multiple Feature Weighted Fusion

Abstract. Synthetic aperture radar in the application of remote sensing technology is becoming more and more widely because of its all-time and all-weather operation, feature extraction research in high resolution SAR image has become a hot topic of concern. In particular, with the continuous improvement of airborne SAR image resolution, image texture information become more abundant. It’s of great significance to classification and extraction. In this paper, a novel method for built-up areas extraction using both statistical and structural features is proposed according to the built-up texture features. First of all, statistical texture features and structural features are respectively extracted by classical method of gray level co-occurrence matrix and method of variogram function, and the direction information is considered in this process. Next, feature weights are calculated innovatively according to the Bhattacharyya distance. Then, all features are weighted fusion. At last, the fused image is classified with K-means classification method and the built-up areas are extracted after post classification process. The proposed method has been tested by domestic airborne P band polarization SAR images, at the same time, two groups of experiments based on the method of statistical texture and the method of structural texture were carried out respectively. On the basis of qualitative analysis, quantitative analysis based on the built-up area selected artificially is enforced, in the relatively simple experimentation area, detection rate is more than 90%, in the relatively complex experimentation area, detection rate is also higher than the other two methods. In the study-area, the results show that this method can effectively and accurately extract built-up areas in high resolution airborne SAR imagery.

Unsupervised classification strategy utilizing an endmember extraction technique for airborne hyperspectral remotely sensed imagery

Remote sensing has become an important source of urban land-use/cover classification, and as a result of their high spatial and spectral resolution, airborne hyperspectral images have been widely used to distinguish different urban classes. However, the previous studies into the classification of urban environments have mainly focused on a supervised scenario, which is limited by the selection of training samples. An unsupervised classification strategy utilizing an endmember extraction technique for airborne hyperspectral imagery is proposed by a combination of endmember extraction and k-means classification. The number of endmembers of the hyperspectral image is first estimated with the hyperspectral signal subspace identification with the minimum error method, and then the simplex growing algorithm is used to extract the endmember spectra that represent the different latent materials in the hyperspectral imagery. These latent materials were further integrated into the predefined number of classes and the k-means classification method was utilized to obtain the final classification map. Different distance measures were experimentally used in the procedure of class integration and classification to investigate the impact of initial cluster centers and further clustering criterion. The proposed strategy was compared with three traditional unsupervised classification methods, k-means, fuzzy k-means, and ISODATA, with two airborne hyperspectral images. The experimental results demonstrate that the proposed approach is more robust and outperforms the other three classification methods, and, hence, provides an innovative perspective for implementing the unsupervised classification of airborne hyperspectral imagery.

An efficient piecewise linear model for predicting activity of caspase-3 inhibitors

Background and purpose of the studyMultimodal distribution of descriptors makes it more difficult to fit a single global model to model the entire data set in quantitative structure activity relationship (QSAR) studies.MethodsThe linear (Multiple linear regression; MLR), non-linear (Artificial neural network; ANN), and an approach based on “Extended Classifier System in Function approximation” (XCSF) were applied herein to model the biological activity of 658 caspase-3 inhibitors.ResultsVarious kinds of molecular descriptors were calculated to represent the molecular structures of the compounds. The original data set was partitioned into the training and test sets by the K-means classification method. Prediction error on the test data set indicated that the XCSF as a local model estimates caspase-3 inhibition activity, better than the global models such as MLR and ANN. The atom-centered fragment type CR2X2, electronegativity, polarizability, and atomic radius and also the lipophilicity of the molecule, were the main independent factors contributing to the caspase-3 inhibition activity.ConclusionsThe results of this study may be exploited for further design of novel caspase-3 inhibitors.

A Practical Target Tracking Technique in Sensor Network Using Clustering Algorithm

Sensor network basically has many intrinsic limitations such as energy consumption, sensor coverage and connectivity, and sensor processing capability. Tracking a moving target in clusters of sensor network online with less complexity algorithm and computational burden is our ultimate goal. Particle filtering (PF) technique, augmenting handoff and K-means classification of measurement data, is proposed to tackle the tracking mission in a sensor network. The handoff decision, an alternative to multi-hop transmission, is implemented for switching between clusters of sensor nodes through received signal strength indication (RSSI) measurements. The measurements being used in particle filter processing are RSSI and time of arrival (TOA). While non-line-of-sight (NLOS) is the dominant bias in tracking estimation/accuracy, it can be easily resolved simply by incorporating K-means classification method in PF processing without any priori identification of LOS/NLOS. Simulation using clusters of sensor nodes in a sensor network is conducted. The dependency of tracking performance with computational cost versus number of particles used in PF processing is also investigated.

Delineating water management zones in a paddy rice field using a Floating Soil Sensing System

Abstract Paddy rice fields are kept inundated during most of the growing period. This requirement is challenging to achieve because of the lack of suitable technologies to detect rapidly percolation prone zones within these fields. The objective of this study was to evaluate a methodology to identify water leakage areas to support precision soil–water management at a within-field level. Therefore, a Floating Sensing System (FloSSy) was designed to record the soil apparent electrical conductivity (ECa) of a paddy field both under dry and inundated conditions using the electromagnetic induction sensor EM38. Comparison of ECa data sets obtained under inundated and dry conditions showed that the ECa measurements under inundated condition (ECa-i) were more strongly related to soil properties due to the absence of variability in soil moisture and the increased stability of the floating sensing platform. Therefore, we proceeded with the ECa-i measurements and grouped them into two classes using a fuzzy k-means classification method. These classes showed significant differences in water infiltration: lower ECa values represented a higher infiltration rate and vice versa. This effect was attributed to differences in soil texture, more specifically the sand content, and its effect on water retention. It was concluded that an ECa-i survey with FloSSy allowed the detection of soil heterogeneity linked to downward water fluxes which has a potential to support precision soil–water management in inundated fields.