Dual Branch Research Articles

Dual Branch Autoencoder Network for Spectral-Spatial Hyperspectral Unmixing

Spatial information can play a supporting role in spectral unmixing. In this letter, we propose a dual branch autoencoder network to incorporate spatial-contextual information for spectral-spatial unmixing. The two branches leverage different architectures to efficiently extract spatial information and spectral information. In the first branch, we use fully connected layers to extract spectral information, where the neuron in each layer can capture all spectral features. In the second branch, 2-D convolution is adopted to exploit spatial features, which does not require hand-crafted assumptions compared with conventional methods. Then the extracted features are concatenated and propagated to generate the abundance and reconstruct the pixel. Moreover, to solve the drawbacks of the existing reconstruction functions, we propose a new function termed squared sine distance to improve the convergence quality of the proposed network. Experimental results reveal the effectiveness of our proposed method on both synthetic data and real-world data.

Dual branch network for human pose estimation in dressing scene

Dual branch network for human pose estimation in dressing scene

A Global Spectral–Spatial Feature Learning Network for Semisupervised Hyperspectral Unmixing

Neural networks have been widely applied in hyperspectral unmixing in the past few years. However, most networks only focus on extracting the spectral information or local spectral–spatial correlation of a single pixel. In order to further explore the spectral–spatial information of hyperspectral images (HSIs) and improve the effective use of spatial information in unmixing tasks, an end-to-end semisupervised unmixing network is proposed in this article. At the encoder, the dual branch network performs convolution operations along the spatial dimension and the spectral dimension, respectively, to realize the separated extraction of global spatial and spectral information; the spectral–spatial attention residual module then fuses the separated spectral–spatial features into the joint spectral–spatial information. The encoder is to realize mapping from high-dimensional spectral data to latent abundance representations. The decoder is a simple linear fully connected layer whose weight is fixed to a known spectral library to complete the reconstruction of the HSI. Experiments on two synthetic datasets and two real datasets demonstrate the superior performance of the proposed method, and prove the effectiveness of considering global spatial information in the unmixing task.

HCNNet: A Hybrid Convolutional Neural Network for Spatiotemporal Image Fusion

In recent years, leaps and bounds have developed spatiotemporal fusion (STF) methods for remote sensing (RS) images based on deep learning. However, most existing methods utilize 2D convolution (Conv) to explore features. 3D Conv can explore time-dimensional features, but it requires more memory footprint and is rarely used. In addition, the current STF methods based on convolutional neural networks (CNNs) are mainly the following two: 1) Use 2D Conv to extract features from multiple bands of the input image together and fuse the features to predict the multiband image directly; 2) Use 2D Conv to extract features from individual bands of the image, predict the reflectance data of individual bands, and finally stack the predicted individual bands directly to synthesize the multiband image. The former method does not sufficiently consider the spectral and reflectance differences between different bands, and the latter does not consider the similarity of spatial structures between adjacent bands and the spectral correlation. To solve these problems, we propose a 2D/3D hybrid convolution neural network (CNN) called HCNNet, in which the 2D-CNN branch extracts the spatial information features of single-band image, and the 3D-CNN branch extracts spatiotemporal features of single-band images. After fusing the features of the dual branches, we introduce neighboring band features to share spatial information so that the information is complementary to obtain single-band features and images, and finally stack each single-band image to generate multiband images. Visual assessment and metric evaluation of the three publicly available datasets showed that our method predicted better images compared to the five methods.

Mathematical Analysis of Magnetized Rotating Nanofluid Flow Over nonlinear shrinking surface: Duality and Stability

In this study, the MHD effect on boundary layer rotating flow of a nanofluid is investigated for the multiple branches case. The main focus of current research is to examine flow characteristics on a nonlinear permeable shrinking sheet. Moreover, the governing partial differential equations (PDEs) of the problem considered are reduced into coupled nonlinear ordinary differential equations (ODEs) with the appropriate similarity transformation. Numerical results based on the plotted graphs are gotten by solving ODEs with help of the three-stage Labatto IIIA method in bvp4c solver in MATLAB. To confirm numerical outcomes, current results are compared with previously available outcomes and found in good agreement. Skin friction coefficients, Nusselt and Sherwood numbers, velocity profiles, temperature profiles, and concentration profiles are examined. The results show that dual (no) branches exist in certain ranges of the suction parameter i.e., SSc (SSc). Further, profiles of velocity decrease for rising values of Hartmann number in the upper branch, while reverse trend has been noticed in a lower branch. Profiles of temperature and concentration enhance for the increasing values of thermophoresis in both branches. stability analysis of the branches is also done and noticed that upper branch is a stable branch from both branches. Finally, it is noted that the stable branch has symmetrical behavior with regard to the parameter of rotation.

Existence of dual solution for micro-polar fluid flow with convective boundary layer in the presence of thermal radiation and suction/injection effects

The goal of this research is to see if there exist dual solutions for the two-dimensional magneto-hydrodynamics (MHD) flow of a micro-polar fluid in a porous media with thermal radiation and suction/injection effects. By considering the porosity and permeability effects, the Darcy model is used to create a micro-polar fluid model over the pores. The nonlinear ordinary differential equations are obtained using an appropriate linear similarity transformation. In terms of the incomplete gamma function, the dual branches of two-dimensional fluid flow are derived in the closed form solution. The analytic closed form solution with duality for all fluid flow parameters, as well as the impacts of the micro-rotation parameter on the Nusselt number due to the dual nature closed form solutions, the impacts of physical parameters on non-dimensional velocity, temperature, skin friction coefficient, micro-rotation, and local Nusselt number are investigated and demonstrated graphically.

Numerical Analysis of Cu + Al 2 O 3 / H 2 O Hybrid Nanofluid of Streamwise and Cross Flow with Thermal Radiation Effect: Duality and Stability

The motion of water conveying copper and aluminum nanoparticles on a heated moving sheet when thermal radiation and stretching/shrinking surface is significant and is investigated in this study to announce the increasing effects of volume fractions, thermal radiation, and moving parameters on this transport phenomenon. Furthermore, the flow of a Cu − Al 2 O 3 /water hybrid nanofluid across a heated moving sheet has been studied in both cross and streamwise directions. Thermal radiation effect is also considered, as this effect along with cross flow has not yet been investigated for the hybrid nanofluid in the published literature. Two distinct types of nanoparticles, namely, Al 2 O 3 (alumina) and Cu (copper), have been used to prepare hybrid nanofluid where water is considered as a base fluid. The system of nonlinear partial differential equations (PDEs) has been transferred to ordinary differential equations (ODEs) by compatible transformations before solving them by employing the III-stage Lobatto-IIIa method in bvp4c solver in MATLAB 2017 software. Temporal stability analysis has been carried out in order to verify stable branch between two branches by obtaining the smallest eigenvalue values. The branches obtained are addressed in depth against every applied parameter using figures and tables. The results show that there are three ranges of branches, no solution exists when λ > λ c , dual branches exist when 0.23 ≤ λ ≤ λ c , and a single solution exists when λ > 0.23 . Moreover, thermal layer thickness declines initially and then enhances in the upper and lower solutions for the higher values of the thermal radiation parameter.

Thermal stability and performances of hybrid nanoparticles for convective heat transfer phenomenon with multiple solutions

The thermal reliability of hybrid nanoparticles is quite impressive which presents many dynamic applications in solar systems, extrusion processes, thermal devices, cooling systems, and many engineering processes. This paper explores the flow of three-dimensional (3D), steady, magnetohydrodynamic of alumina and copper/water hybrid nanofluid over the nonlinear shrinking/stretching surface. A mathematical model was developed by adopting the model of Tiwari and Das in the form of partial differential equations (PDEs), which were then transferred into an analogous set of non-linear ordinary differential equations (ODEs) by employing nonlinear similarity variables. The resulting ODEs were solved by employing abvp4csolver numerically in the MATLAB package. For validation, the collected empirical outcomes were numerically correlated with those of the preceding studies and were in strong alignment. It has been demonstrated that dual branches occur within the different ranges of magnetic, suction, and shrinking/stretching parameters. Dual (no) branches have been discovered when b≤bci (b>bci) and M≥Mci (M<Mci) where i=1,2,3. Because of the non-uniqueness of the branches, a stability analysis was performed, and it was discovered that the upper branch was stable. Further, velocity in y-direction and temperature of hybrid nanofluid increase in both branches when the volume fraction of φCu increases.

Y-Net: Dual-branch Joint Network for Semantic Segmentation

Most existing segmentation networks are built upon a “ U -shaped” encoder–decoder structure, where the multi-level features extracted by the encoder are gradually aggregated by the decoder. Although this structure has been proven to be effective in improving segmentation performance, there are two main drawbacks. On the one hand, the introduction of low-level features brings a significant increase in calculations without an obvious performance gain. On the other hand, general strategies of feature aggregation such as addition and concatenation fuse features without considering the usefulness of each feature vector, which mixes the useful information with massive noises. In this article, we abandon the traditional “ U -shaped” architecture and propose Y-Net, a dual-branch joint network for accurate semantic segmentation. Specifically, it only aggregates the high-level features with low-resolution and utilizes the global context guidance generated by the first branch to refine the second branch. The dual branches are effectively connected through a Semantic Enhancing Module, which can be regarded as the combination of spatial attention and channel attention. We also design a novel Channel-Selective Decoder (CSD) to adaptively integrate features from different receptive fields by assigning specific channelwise weights, where the weights are input-dependent. Our Y-Net is capable of breaking through the limit of singe-branch network and attaining higher performance with less computational cost than “ U -shaped” structure. The proposed CSD can better integrate useful information and suppress interference noises. Comprehensive experiments are carried out on three public datasets to evaluate the effectiveness of our method. Eventually, our Y-Net achieves state-of-the-art performance on PASCAL VOC 2012, PASCAL Person-Part, and ADE20K dataset without pre-training on extra datasets.

Remote Sensing Scene Image Classification Based on Dense Fusion of Multi-level Features

For remote sensing scene image classification, many convolution neural networks improve the classification accuracy at the cost of the time and space complexity of the models. This leads to a slow running speed for the model and cannot realize a trade-off between the model accuracy and the model running speed. As the network deepens, it is difficult to extract the key features with a sample double branched structure, and it also leads to the loss of shallow features, which is unfavorable to the classification of remote sensing scene images. To solve this problem, we propose a dual branch multi-level feature dense fusion-based lightweight convolutional neural network (BMDF-LCNN). The network structure can fully extract the information of the current layer through 3 × 3 depthwise separable convolution and 1 × 1 standard convolution, identity branches, and fuse with the features extracted from the previous layer 1 × 1 standard convolution, thus avoiding the loss of shallow information due to network deepening. In addition, we propose a downsampling structure that is more suitable for extracting the shallow features of the network by using the pooled branch to downsample and the convolution branch to compensate for the pooled features. Experiments were carried out on four open and challenging remote sensing image scene data sets. The experimental results show that the proposed method has higher classification accuracy and lower model complexity than some state-of-the-art classification methods and realizes the trade-off between model accuracy and model running speed.

Dual-view Snapshot Compressive Imaging via Optical Flow Aided Recurrent Neural Network

Dual-view snapshot compressive imaging (SCI) aims to capture videos from two field-of-views (FoVs) using a 2D sensor (detector) in a single snapshot, achieving joint FoV and temporal compressive sensing, and thus enjoying the advantages of low-bandwidth, low-power and low-cost. However, it is challenging for existing model-based decoding algorithms to reconstruct each individual scene, which usually require exhaustive parameter tuning with extremely long running time for large scale data. In this paper, we propose an optical flow-aided recurrent neural network for dual video SCI systems, which provides high-quality decoding in seconds. Firstly, we develop a diversity amplification method to enlarge the differences between scenes of two FoVs, and design a deep convolutional neural network with dual branches to separate different scenes from the single measurement. Secondly, we integrate the bidirectional optical flow extracted from adjacent frames with the recurrent neural network to jointly reconstruct each video in a sequential manner. Extensive results on both simulation and real data demonstrate the superior performance of our proposed model in short inference time. The code and data are available at https://github.com/RuiyingLu/OFaNet-for-Dual-view-SCI.

Dual Branch Attention Network for Person Re-Identification.

As a sub-direction of image retrieval, person re-identification (Re-ID) is usually used to solve the security problem of cross camera tracking and monitoring. A growing number of shopping centers have recently attempted to apply Re-ID technology. One of the development trends of related algorithms is using an attention mechanism to capture global and local features. We notice that these algorithms have apparent limitations. They only focus on the most salient features without considering certain detailed features. People’s clothes, bags and even shoes are of great help to distinguish pedestrians. We notice that global features usually cover these important local features. Therefore, we propose a dual branch network based on a multi-scale attention mechanism. This network can capture apparent global features and inconspicuous local features of pedestrian images. Specifically, we design a dual branch attention network (DBA-Net) for better performance. These two branches can optimize the extracted features of different depths at the same time. We also design an effective block (called channel, position and spatial-wise attention (CPSA)), which can capture key fine-grained information, such as bags and shoes. Furthermore, based on ID loss, we use complementary triplet loss and adaptive weighted rank list loss (WRLL) on each branch during the training process. DBA-Net can not only learn semantic context information of the channel, position, and spatial dimensions but can integrate detailed semantic information by learning the dependency relationships between features. Extensive experiments on three widely used open-source datasets proved that DBA-Net clearly yielded overall state-of-the-art performance. Particularly on the CUHK03 dataset, the mean average precision (mAP) of DBA-Net achieved 83.2%.

Deep Feature Fusion Based Dual Branch Network for X-ray Security Inspection Image Classification

Automatic computer security inspection of X-ray scanned images has an irresistible trend in modern life. Aiming to address the inconvenience of recognizing small-sized prohibited item objects, and the potential class imbalance within multi-label object classification of X-ray scanned images, this paper proposes a deep feature fusion model-based dual branch network architecture. Firstly, deep feature fusion is a method to fuse features extracted from several model layers. Specifically, it operates these features by upsampling and dimension reduction to match identical sizes, then fuses them by element-wise sum. In addition, this paper introduces focal loss to handle class imbalance. For balancing importance on samples of minority and majority class, it assigns weights to class predictions. Additionally, for distinguishing difficult samples from easy samples, it introduces modulating factor. Dual branch network adopts the two components above and integrates them in final loss calculation through the weighted sum. Experimental results illustrate that the proposed method outperforms baseline and state-of-art by a large margin on various positive/negative ratios of datasets. These demonstrate the competitivity of the proposed method in classification performance and its potential application under actual circumstances.

Duality and stability of MHD Darcy–Forchheimer porous medium flow of rotating nanofluid on a linear shrinking/stretching sheet: Buongiorno model

PurposeThe 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).Design/methodology/approachNanofluid 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.FindingsThe 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.Originality/valueThe 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.

A single-branch impedance compression network (icn) optimized by particle swarm optimization algorithm for rf energy harvesting system

Most previous works regarding the impedance compression network ICNs reported in the literature consist of dual branches to carry out the compression functionalities. Although this type of circuits are efficient, they are complex and have large sizes. This work proposes a complex impedance compression network with only a single branch for wireless power transfer (WPT) circuits, operating at 2.4GHz under a wide input power range and a variable load. The design size is about 4x4.2mm2 which thereby, produces a simple, and small design. The simulation results demonstrate that the improvement in efficiency is by 10%, as compared to a design without ICN. The improvement can increased by utilizing the particle swarm optimization PSO algorithm where the best dimensions of the ICN components will be selected relying on the suitable fitness function in the PSO algorithm to the given desired goals. The impedance matching is also enhanced in which it stands behind the conversion efficiency increment. The circuit size is very slightly increased and it becomes 4.2x4.3mm2. The final design has a conversion efficiency about 76%, with more than 15% improvement to a design with only a matching circuit or without using the automation techniques. Eventually, results show that the enhancement is obtained not only the designated input power but also along a range of the input power.

Automatic weight measurement of pigs based on 3D images and regression network

Weighing pigs with scales takes a lot of time and makes pigs stressful easily. It can be avoided by measuring weight based on the depth images. We design a series of preprocessing algorithms including instance segmentation, distance independence, noise reduction, rotation correction. These algorithms are used to eliminate the influence of the surrounding environment and other factors. In order to predict the weight precisely, we build a regression network based on BotNet. Dual branch of 3×3 convolution and MHSA replaces single 3×3 convolution of the fourth block in ResNet. At the end of the network, we use multiple fully connection layers in parallel, then concat them to predict weight. After training on 122736 images, MAE of our network is 6.366 on 5326 test images.

Multi-granularity scale-aware networks for hard pixels segmentation of pulmonary nodules

Accurate automatic segmentation of pulmonary nodules can greatly assist in the early clinical diagnosis and analysis of lung cancer. However, it remains a challenging task due to (1) the variable scales, complex shapes, and textures heterogeneity of nodules and (2) the similar visual characteristics of nodules and the surrounding environment. These factors make it difficult for a model to capture the representative and distinguishing features of nodules. We define these pixels, which are difficult to segment accurately, as “hard pixels.” To comprehensively tackle these challenges, we focus on the complementarity between surrounding background information and salient nodules information. Accordingly, we propose a Multi-Granularity Scale-Aware Networks (MGSA-Net) for accurate pulmonary nodules segmentation with steps as follows. First, to effectively preserve both global contextual and local fine details information, we unify the representation of the feature about global and patch-level images in a single framework. Second, we introduce a deep scale-aware module (DSAM) in the global stream that could generate multi-scale feature maps with a uniform representational power, which can process more contextual information. Finally, we propose a multi-granularity feature map sharing learning to fuse feature maps from the dual branch at various scales. Benefiting from the rich background information and salient nodules information, the fused features can help simultaneously capture the similarity of nodules and the diversity of background to mine hard pixels. Especially, the two streams are jointly optimized, ensuring they are mutually reinforced and refining current pixels’ predictions by their similar structure boundaries. Extensive experiments on the LIDC-IDRI dataset have demonstrated that the proposed MGSA-Net could surpass most segmentation models and advance the state-of-the-art performance with the dice similarity coefficient (DSC) of 87.32% on the dataset. Code can be available at: https://github.com/ISSE-AILab/MGSA-Net.

Real-time Semantic Segmentation with Context Aggregation Network

With the increasing demand of autonomous systems, pixelwise semantic segmentation for visual scene understanding needs to be not only accurate but also efficient for potential real-time applications. In this paper, we propose Context Aggregation Network, a dual branch convolutional neural network, with significantly lower computational costs as compared to the state-of-the-art, while maintaining a competitive prediction accuracy. Building upon the existing dual branch architectures for high-speed semantic segmentation, we design a high resolution branch for effective spatial detailing and a context branch with light-weight versions of global aggregation and local distribution blocks, potent to capture both long-range and local contextual dependencies required for accurate semantic segmentation, with low computational overheads. We evaluate our method on two semantic segmentation datasets, namely Cityscapes dataset and UAVid dataset. For Cityscapes test set, our model achieves state-of-the-art results with mIOU of 75.9%, at 76 FPS on an NVIDIA RTX 2080Ti and 8 FPS on a Jetson Xavier NX. With regards to UAVid dataset, our proposed network achieves mIOU score of 63.5% with high execution speed (15 FPS).

DBCFace: Towards Pure Convolutional Neural Network Face Detection

Face detection generally requires prior boxes and an extra non-maximum suppression(NMS) post-processing in modern deep learning methods. However, anchor design and anchor matching strategy significantly affect the performance of face detectors, so we have to spend a lot of time on anchor designing for different business scenarios. The other issue is that NMS cannot be easily parallelized and it may become a bottleneck of detection speed. In this paper, we propose a simple yet efficient pure convolutional neural network face detection method, named dual-branch center face detector(DBCFace for short), which solve face detection via a dual branch fully convolutional framework without extra anchor design and NMS. Extensive experiments are conducted on four popular face detection benchmarks, including AFW, PASCAL face, FDDB, and WIDER FACE, demonstrating that our method is comparable with state-of-the-art methods while the speed is faster.

Dual branch solutions (multi-solutions) for nonlinear radiative Falkner–Skan flow of Maxwell nanomaterials with heat and mass transfer over a static/moving wedge

In this study, the numerous solutions to Falkner–Skan flow of a Maxwell fluid with nanoparticles are investigated, considering the nonlinear radiation and magnetic domain. The flow described above can be expressed in accordance with PDEs that are transformed into ODEs by choosing suitable variables of similarity. The fourth- and fifth-order Runge–Kutta–Fehlberg method can be utilized to solve these reduced ODEs by applying the shooting approach. The graphs were drawn to explain the effects of different parameters on different fluid profiles for both the lower- and upper-branch solutions. This study shows that the velocity outlines improve both solutions by increasing local Deborah numbers slightly. Besides, an increase in radiation reduces the thermal gradient for both solutions, thereby reducing the concentration gradient for both solutions contributing to raised Brownian motion and Lewis numbers.