Search results

This paper aims to propose a solution for detecting and grading diabetic retinopathy (DR) in retinal images using a convolutional neural network (CNN)-based approach. It could classify input retinal images into a normal class or an abnormal class, which would be further split into four stages of abnormalities automatically.

The proposed solution is developed based on a newly proposed CNN architecture, namely, DeepRoot. It consists of one main branch, which is connected by two side branches. The main branch is responsible for the primary feature extractor of both high-level and low-level features of retinal images. Then, the side branches further extract more complex and detailed features from the features outputted from the main branch. They are designed to capture details of small traces of DR in retinal images, using modified zoom-in/zoom-out and attention layers.

The proposed method is trained, validated and tested on the Kaggle dataset. The regularization of the trained model is evaluated using unseen data samples, which were self-collected from a real scenario from a hospital. It achieves a promising performance with a sensitivity of 98.18% under the two classes scenario.

The new CNN-based architecture (i.e. DeepRoot) is introduced with the concept of a multi-branch network. It could assist in solving a problem of an unbalanced dataset, especially when there are common characteristics across different classes (i.e. four stages of DR). Different classes could be outputted at different depths of the network.

In a vehicle power plant comprising a pair of coaxial controllable‐pitch propellers and means for rotating said propellers independently in opposite directions, means for holding one of said propellers against rotation, and means for turning the blades of the propeller so held to a pitch position at which aerodynamic forces acting on said propeller during flight tend to turn said propeller in the same direction as the direction of rotation of the other propeller.

The purpose of study is to examine the dual nature of the branches for the problem of Darcy–Forchheimer porous medium flow of rotating nanofluid on a linearly stretching/shrinking surface under the field of magnetic influence. The dual nature of the branches confronts the uniqueness and existence theorem, moreover, mathematically it is a great achievement. For engineering purposes, this study applied a linear stability test on the multiple branches to determine which solution is physically reliable (stable).

Nanofluid model has been developed with the help of Buongiorno model. The partial differential equations in space coordinates for the law of conservation of mass, momentum and energy have been transformed into ordinary differential equations by introducing the similarity variables. Two numerical techniques, namely, the shooting method in Maple software and the three-stage Lobatto IIIA method in Matlab software, have been used to find multiple branches and to accomplish stability analysis, respectively.

The parametric investigation has been executed to find the multiple branches and explore the effects on skin friction, Sherwood number, Nusselt number, concentration and temperature profiles. The findings exhibited the presence of dual branches only in the case of a shrinking sheet.

The originality of work is a determination of multiple branches and the performance of the stability analysis of the branches. It has also been confirmed that such a study has not yet been considered in the previous literature.

The purpose of this study is to find the multiple branches of the three-dimensional flow of Cu-Al₂ O₃/water rotating hybrid nanofluid perfusing a porous medium over the stretching/shrinking surface. The extended model of Darcy due to Forchheimer and Brinkman has been considered to make the hybrid nanofluid model over the pores by considering the porosity and permeability effects.

The Tiwari and Das model with the thermophysical properties of spherical particles for efficient dynamic viscosity of the nanoparticle is used. The linear similarity transformations are applied to convert the partial differential equations into ordinary differential equations (ODEs). The system of governing ODEs is solved by using the three-stage Lobatto IIIa scheme in MATLAB for evolving parameters.

The system of governing ODEs produces dual branches. A unique stable branch is identified with help of stability analysis. The reduced heat transfer rate has been shown to increase with the reduced ϕ2 in both branches. Further, results revealed that the presence of multiple branches depends on the ranges of porosity, suction and stretching/shrinking parameters for the particular value of the rotating parameter.

Dual branches of the three-dimensional flow of Cu-Al₂ O₃/water rotating hybrid nanofluid have been found. Therefore, stability analysis of the branches is also conducted to know which branch is appropriate for the practical applications. To the best of the authors’ knowledge, this research is novel and there is no previously published work relevant to the present study.

The purpose of this paper is to focus on the design of a dual-branch balance saliency model based on fully convolutional network (FCN) for automatic fabric defect detection, and improve quality control in textile manufacturing.

This paper proposed a dual-branch balance saliency model based on discriminative feature for fabric defect detection. A saliency branch is firstly designed to address the problems of scale variation and contextual information integration, which is realized through the cooperation of a multi-scale discriminative feature extraction module (MDFEM) and a bidirectional stage-wise integration module (BSIM). These modules are respectively adopted to extract multi-scale discriminative context information and enrich the contextual information of features at each stage. In addition, another branch is proposed to balance the network, in which a bootstrap refinement module (BRM) is trained to guide the restoration of feature details.

To evaluate the performance of the proposed network, we conduct extensive experiments, and the experimental results demonstrate that the proposed method outperforms state-of-the-art (SOTA) approaches on seven evaluation metrics. We also conduct adequate ablation analyses that provide a full understanding of the design principles of the proposed method.

The dual-branch balance saliency model was proposed and applied into the fabric defect detection. The qualitative and quantitative experimental results show the effectiveness of the detection method. Therefore, the proposed method can be used for accurate fabric defect detection and even surface defect detection of other industrial products.

This study aims to study the mixed convection flow and heat transfer of Al₂O₃-Cu/water hybrid nanofluid over a vertical plate. Governing equations for conservation of mass, momentum and energy for the hybrid nanofluid over a vertical flat plate are introduced.

The similarity transformation approach is used to transform the set of partial differential equations into a set of non-dimensional ordinary differential equations. Finite-deference with collocation method is used to integrate the governing equations for the velocity and temperature profiles.

The results show that dual solutions exist for the case of opposing flow over the plate. Linear stability analysis was performed to identify a stable solution. The stability analysis shows that the lower branch of the solution is always unstable, while the upper branch of the solution is always stable. The results of boundary layer analysis are reported for the various volume fractions of composite nanoparticles and mixed convection parameter. The outcomes show that the composition of nanoparticles can notably influence the boundary layer flow and heat transfer profiles. It is also found that the trend of the variation of surface skin friction and heat transfer for each of the dual solution branches can be different. The critical values of the mixed convection parameter, λ, where the dual solution branches joint together, are also under the influence of the composition of hybrid nanoparticles. For instance, assuming a total volume fraction of 5 per cent for the mixture of Al₂O₃ and Cu nanoparticles, the critical value of mixing parameter of λ changes from −3.1940 to −3.2561 by changing the composition of nanofluids from Al₂O₃ (5 per cent) + Cu (0%) to Al₂O₃ (2.5%) + Cu (2.5 per cent).

The mixed convection stability analysis and heat transfer study of hybrid nanofluids for a stagnation-point boundary layer flow are addressed for the first time. The introduced hybrid nanofluid model and similarity solution are new and of interest in both mathematical and physical points of view.

Deregulation and other factors permit and encourage financial institutions to become more integrated, both within their own (financial) industries, such as banking and insurance, and across these industries. Financial regulators have responded with like integration. As financial institutions increasingly compete with firms from other industries and areas, financial regulators similarly compete more across borders. The resulting competition in financial regulation enhances innovation, choice, and efficiency. The advent of home-run regulation, which in general allows financial institutions to adhere only to the financial regulations of their home area and is spreading across the US and Europe, may allow numerous regulatory regimes within a given market.

This paper aims to theoretically and numerically investigate the steady two-dimensional (2D) Hiemenz flow with heat transfer of Reiner-Rivlin fluid over a linearly stretching/shrinking sheet.

The Navier–Stokes equations are transformed into self-similar equations using appropriate similarity transformations and then solved numerically by using shooting technique. A simple but effective mathematical analysis has been used to prove the existence of a solution for stretching case (λ> 0). Moreover, an attempt has been laid to carry the asymptotic solution behavior for large stretching. The obtained asymptotic solutions are compared with direct numerical solutions, and the comparison is quite remarkable.

It is observed that the self-similar equations exhibit dual solutions within the range [λ_c, −1] of shrinking parameter λ, where λ_c is the turning point from where the dual solutions bifurcate. Unique solution is found for all stretching case (λ > 0). It is noticed that the effects of cross-viscous parameter L and shrinking parameter λ on velocity and thermal fields show opposite character in the dual solution branches. Thus, a linear temporal stability analysis is performed to determine the basic feasible solution. The stability analysis is based on the sign of the smallest eigenvalue, where positive or negative sign leading to a stable or unstable solution. The stability analysis reveals that the first solution is stable that describes the main flow. Increase in cross-viscous parameter L resulting in a significant increment in skin friction coefficient, local Nusselt number and dual solutions domain.

This work’s originality is to examine the combined effects of cross-viscous parameter and stretching/shrinking parameter on skin friction coefficient, local Nusselt number, velocity and temperature profiles of Hiemenz flow over a stretching/shrinking sheet. Although many studies on viscous fluid and nanofluid have been investigated in this field, there are still limited discoveries on non-Newtonian fluids. The obtained results can be used as a benchmark for future studies of higher-grade non-Newtonian flows with several physical aspects. All the generated results are claimed to be novel and have not been published elsewhere.

The purpose of this paper is to study the steady mixed convection hybrid nanofluid flow and heat transfer past a vertical thin needle with prescribed surface heat flux.

The governing partial differential equations are transformed into a set of ordinary differential equations by using a similarity transformation. The transformed equations are then solved numerically using the boundary value problem solver (bvp4c) in Matlab software. The features of the skin friction coefficient and the local Nusselt number as well as the velocity and temperature profiles for different values of the governing parameters are analyzed and discussed.

It is found that dual solutions exist for a certain range of the mixed convection parameter where its critical values decrease with the increasing of the copper (Cu) nanoparticle volume fractions and for the smaller needle size. It is also observed that the increasing of the copper (Cu) nanoparticle volume fractions and the decreasing of the needle size tend to enhance the skin friction coefficient and the local Nusselt number on the needle surface. A temporal stability analysis is performed to determine the stability of the dual solutions in the long run, and it is revealed that only one of them is stable, while the other is unstable.

The problem of hybrid nanofluid flow and heat transfer past a vertical thin needle with prescribed surface heat flux is the important originality of the present study where the dual solutions for the opposing flow are obtained.

Outlines the adaptation process in the distribution channel structure of the retail banking sector as a consequence of the introduction of electronic channels, such as telephone banking, PC banking and Internet banking. Based on responses from 42 retail banks in Denmark, their distribution channel strategies are described and their relation to selected marketing mix elements is examined. Most Danish retail banks attach decisive importance to offering a customer‐friendly PC bank service, whereas fewer of them attach the same importance to telephone, Internet and branch banking. A multiple channel strategy combining several channels is the most popular.