The number of panicles per unit area is a common indicator rice yield and great significance to estimation, breeding, phenotype analysis. Traditional counting methods have various drawbacks, such as long delay times high subjectivity, they are easily perturbed by noise. To improve the accuracy detection in field, we developed implemented panicle system that based on improved region-based fully convolutional networks, use automate rice-phenotype measurements. field experiments were conducted...

Abstract In this paper a new class of finite difference schemes - the Weighted Compact Schemes are proposed. According to idea WENO schemes, Scheme is constructed by combination approximations derivatives on candidate stencils with properly assigned weights so that non-oscillatory property achieve when discontinuities appear. The primitive function reconstruction method ENO applied obtain conservative form Scheme. This scheme not only preserves characteristic standard compact and achieves...

A numerical study is conducted to explore the unsteady nature of fluidic oscillators, which are responsible for generating sweeping jets effective flow control large-scale applications. Two- and three-dimensional Reynolds-averaged Navier–Stokes improved delayed detached-eddy simulations employed simulate resolve structures in internal passages a oscillator an external flowfield quiescent condition. The predicted generated by enlarged actuator with outlet diameter 25 mm using air as working...

A direct numerical simulation of transitional pipe flow is carried out with the help a spectral element method and used to investigate localized regions ‘turbulent’ that are observed in experiments. Two types such can be distinguished: puff slug. The puff, which generally found at low values Reynolds numbers, simulated for Re = 2200 where number based on mean velocity U B diameter D . slug occurs higher it 5000. computations start laminar added prescribed disturbance given axial position...

Leaf coverage is an indicator of plant growth rate and predicted yield, thus it crucial to plant-breeding research. Robust image segmentation leaf from remote-sensing images acquired by unmanned aerial vehicles (UAVs) in varying environments can be directly used for large-scale estimation, a key component high-throughput field phenotyping. We propose image-segmentation method based on machine learning extract relatively accurate information the orthophoto generated after preprocessing. The...

A new globally convergent numerical method is developed for an inverse problem the elliptic equation with unknown potential. The boundary data simulating measurements in optical tomography are generated by running source. Global convergence analysis presented along experiments.

A new numerical imaging algorithm is presented for reconstruction of optical absorption coefficients from near-infrared light data with a continuous-wave source. As continuation our earlier efforts in developing series methods called "globally convergent methods" [J. Opt. Soc. Am. A23, 2388 (2006)], this solves the inverse problem through solution boundary-value Volterra-type integral partial differential equation. We deal here particular issues solving problems an arbitrary convex shape...

A new numerical method for an inverse problem elliptic equation with unknown potential is proposed. In this the point source running along a straight line and source-dependent Dirichlet boundary condition measured as data problem. rigorous convergence analysis shows that converges globally, provided so-called tail function approximated well. This approximation verified in experiments, so global convergence. Applications to medical imaging, imaging of targets on battlefields electrical...

A numerical study is conducted to explore the effect of a single dielectric barrier discharge plasma actuator for controlling turbulent boundary-layer separation on deflected flap high-lift airfoil at chord-based Reynolds number 240,000. An integrated model consisting electrohydrodynamic body force and computational fluid dynamics package called NavyFOAM employed in this study. Comparison current results against experimental data indicates reasonable agreement between two studies baseline...

In recent years, with the continuous innovation of Internet Things technology, image processing technology in has become more and mature. Automated pig raising will mainstream making research intelligent breeding face a change. It important. The traditional model cannot provide suitable growth development environment for pigs. pigs are disturbed by diseases environmental discomforts during process, which increases mortality process consumers guarantee. order to solve problem unfavorable...

Abstract Interpreting time domain reflectometry (TDR) waveforms obtained in soils with non‐uniform water content is an open question. We design a new TDR waveform interpretation model based on convolutional neural networks (CNNs) that can reveal the spatial variations of soil relative permittivity and along sensor. The proposed model, namely TDR‐CNN, constructed three modules. First, geometrical features are extracted simplified version VGG16 network. Second, reflection positions traced...

A method is presented for reconstruction of the optical absorption coefficient from transmission near-infrared data with a cw source. As it distinct other available schemes such as optimization or Newton's iterative method, this resolves inverse problem by solving boundary value Volterra-type integral-differential equation. It demonstrated in numerical studies that technique has better than average stability respect to discrepancy between initial guess and actual unknown coefficient. The...

The commonly used methods for removing metal-induced beam hardening artifacts often rely on the use of high energy photons with either tube voltage or virtual monoenergetic images in dual-energy computed tomography (CT), radiation dose was usually relatively order to generate adequate signals. This retrospective study is designed evaluate application a metal artifact reduction (MAR) algorithm reducing pedicle screw metal-caused post-surgery pediatric low spine CT images.

Core Ideas The simulations of biomass and leaf area index in ORYZA2000 were sensitive to the aerial architecture latitudinal variations extinction coefficient. KDF values calibrated by trial error greatly improved simulated values. could be used evaluate effects different plant architectures on rice growth development. Light distribution light use efficiency a cereal canopy are affected genotypes as well solar altitude at contrasting latitudes. This study was conducted determine whether...

A computational method has been developed to couple the electrohydrodynamic (EHD) body forces induced by dielectric barrier discharge (DBD) actuation with unsteady Reynolds-Averaged Navier–Stokes (URANS) model or large eddy simulation (LES) for incompressible flows. The EHD force is based on solving electrostatic equations electric potential and net charge density. boundary condition density surface obtained from a space–time lumped-element (STLE) circuit an empirical model. DBD–URANS/LES...

The main challenge facing simulation-based hydrodynamic design of naval ships comes from the complexity salient physics involved around ships, which is further compounded by multidisciplinary nature ship applications. Simulation flow using “first principles” computationally very expensive and time-consuming. Other challenges largely pertain to software engineering, ranging architecture, verification validation (V & V), quality assurance. This article presents a computational fluid dynamics...