Branch Convolutional Neural Network Research Articles

An intelligent energy efficient cooperative MIMO-AF multi-hop and relay based communications for Unmanned Aerial Vehicular networks

Unmanned Aerial Vehicles (UAVs) are recently used for both civilian and military applications in worldwide. Energy Efficiency (EE) is an exceptional design approach for modern communication based systems. New advance technology is needed in order to support UAV applications with reduced energy usage. In this paper, an Energy Efficiency with Hybrid Fuzzy Firefly Algorithm (EE-HFFA) method is introduced for Multiple-Input–Multiple-Output (MIMO) Amplify-And-Forward (AF) systems in which Partial Channel State Information (PCSI) estimation is existing at the relays because of the high speed mobility. A new EE-HFFA algorithm is presented in this research, by means of merging the benefits of the Firefly Algorithm (FA) as well as Differential Evolution (DE). For increasing information sharing both the techniques are implemented in parallel and as a result improve searching efficiency. The outcomes of these two techniques are based upon the fuzzy membership function. For approximation of PCSI for the source node as well as relay nodes, Branch Convolutional Neural Network (B-CNN) classifier is presented to raise the capability of cooperative MIMO-AF systems. EE-optimal source and relay precoding matrices are cooperatively enhanced by means of EE-HFFA. Simulation out comes illustrate that the presented EE-HFFA as well as B-CNN classifier could enhance the EE of MIMO-AF systems with PSCI while matched up with direct/relay link merely precoding optimization.

Fusion High-Resolution Network for Diagnosing ChestX-ray Images

The application of deep convolutional neural networks (CNN) in the field of medical image processing has attracted extensive attention and demonstrated remarkable progress. An increasing number of deep learning methods have been devoted to classifying ChestX-ray (CXR) images, and most of the existing deep learning methods are based on classic pretrained models, trained by global ChestX-ray images. In this paper, we are interested in diagnosing ChestX-ray images using our proposed Fusion High-Resolution Network (FHRNet). The FHRNet concatenates the global average pooling layers of the global and local feature extractors—it consists of three branch convolutional neural networks and is fine-tuned for thorax disease classification. Compared with the results of other available methods, our experimental results showed that the proposed model yields a better disease classification performance for the ChestX-ray 14 dataset, according to the receiver operating characteristic curve and area-under-the-curve score. An ablation study further confirmed the effectiveness of the global and local branch networks in improving the classification accuracy of thorax diseases.

An Ensemble Convolutional Neural Networks for Bearing Fault Diagnosis Using Multi-Sensor Data.

Multi-sensor data fusion is a feasible technique to achieve accurate and robust results in fault diagnosis of rotating machinery under complex conditions. However, the problem of information losses is always ignored during the fusion process. To solve above problem, an ensemble convolutional neural network model is proposed for bearing fault diagnosis. The framework of the proposed model contains three convolutional neural network branches: one multi-channel fusion convolutional neural network branch and two 1-D convolutional neural network branches. The former branch extracts the coupling features based on multi-sensor data and the latter two branches extract the inherent features based on single-sensor data, which can collect comprehensive fault information and reduce information losses. Furthermore, the support vector machine ensemble strategy is employed to fuse the results of multiple branches, which can improve the generalization and robustness of the proposed model. The experiments show that the proposed can obtain more effective and robust results than other methods.

ViraMiner: Deep learning on raw DNA sequences for identifying viral genomes in human samples.

Despite its clinical importance, detection of highly divergent or yet unknown viruses is a major challenge. When human samples are sequenced, conventional alignments classify many assembled contigs as "unknown" since many of the sequences are not similar to known genomes. In this work, we developed ViraMiner, a deep learning-based method to identify viruses in various human biospecimens. ViraMiner contains two branches of Convolutional Neural Networks designed to detect both patterns and pattern-frequencies on raw metagenomics contigs. The training dataset included sequences obtained from 19 metagenomic experiments which were analyzed and labeled by BLAST. The model achieves significantly improved accuracy compared to other machine learning methods for viral genome classification. Using 300 bp contigs ViraMiner achieves 0.923 area under the ROC curve. To our knowledge, this is the first machine learning methodology that can detect the presence of viral sequences among raw metagenomic contigs from diverse human samples. We suggest that the proposed model captures different types of information of genome composition, and can be used as a recommendation system to further investigate sequences labeled as "unknown" by conventional alignment methods. Exploring these highly-divergent viruses, in turn, can enhance our knowledge of infectious causes of diseases.

Deep Image Similarity Measurement Based on the Improved Triplet Network with Spatial Pyramid Pooling

Image similarity measurement is a fundamental problem in the field of computer vision. It is widely used in image classification, object detection, image retrieval, and other fields, mostly through Siamese or triplet networks. These networks consist of two or three identical branches of convolutional neural network (CNN) and share their weights to obtain the high-level image feature representations so that similar images are mapped close to each other in the feature space, and dissimilar image pairs are mapped far from each other. Especially, the triplet network is known as the state-of-the-art method on image similarity measurement. However, the basic CNN can only handle fixed-size images. If we obtain a fixed size image via cutting or scaling, the information of the image will be lost and the recognition accuracy will be reduced. To solve the problem, this paper has proposed the triplet spatial pyramid pooling network (TSPP-Net) through combing the triplet convolution neural network with the spatial pyramid pooling. Additionally, we propose an improved triplet loss function, so that the network model can realize twice distance learning by only inputting three samples at one time. Through the theoretical analysis and experiments, it is proved that the TSPP-Net model and the improved triple loss function can improve the generalization ability and the accuracy of image similarity measurement algorithm.

Improved Lane Detection With Multilevel Features in Branch Convolutional Neural Networks

Existing smart vehicles heavily depend on success of precise positioning, optical radar, visual detection and recognition to determine their road conditions and perfect routes. The visual-based approach is the simplest and most effective enabling technology to reach the goal. In performing such an overtaking maneuver, this vision-based driving assistant system must be able to precisely recognize the lane marks first. The traditional approach needs a classifier with hand-crafted features and an adjusted threshold to achieve a robust lane detection in various environment conditions. In this paper, we adopt the deep learning approach to achieve the robust lane detection. The proposed lane detection network inspired by LaneNet model, which uses semantic segmentation concepts utilizes multiple level features of the encoder and designs enhanced binary segmentation and reduced pixel embedding branches. By reduction of computation in decoders, the proposed network effectively utilizes multilevel features to precisely predict the high quality lane maps. The experiments on Tusimple and CuLane datasets verify that the proposed lane detection network achieves better accuracy performance than LaneNet and faster computation than the existed methods for real-time applications.

PCCN: Parallel Cross Convolutional Neural Network for Abnormal Network Traffic Flows Detection in Multi-Class Imbalanced Network Traffic Flows

Network attack behavior detection using deep learning is an important research topic in the field of network security. Currently, there are still many challenges in detecting multi-class imbalanced abnormal traffic data. This paper proposed a new intrusion detection network based on deep learning, named parallel cross convolutional neural network (PCCN), to improve the detection performance of imbalanced abnormal flows. By fusing the flow features learned from the two branch convolutional neural networks (CNN), PCCN can better learn the flow features with fewer samples, to improve the detection results of the imbalanced abnormal flows. We proposed an improved feature extraction method of the original flow to extract multi-class flow features at the same time. The proposed algorithm not only reduces the number of useless elements for network learning, but also accelerates network convergence. In addition, we proposed four improved versions of the PCCN network structure to meet the real-time requirements of network intrusion detection in the current big data computing. These networks can achieve almost the same detection results as the PCCN, but greatly reduce the detection time of data. Through the analysis of high-order evaluation metrics, the proposed PCCN algorithm is significantly better than the traditional machine learning algorithms. Compared with the current hierarchical network model, PCCN can also achieve better performance in term of overall accuracy.

Bi-stream CNN Down Syndrome screening model based on genotyping array

BackgroundHuman Down syndrome (DS) is usually caused by genomic micro-duplications and dosage imbalances of human chromosome 21. It is associated with many genomic and phenotype abnormalities. Even though human DS occurs about 1 per 1,000 births worldwide, which is a very high rate, researchers haven’t found any effective method to cure DS. Currently, the most efficient ways of human DS prevention are screening and early detection.MethodsIn this study, we used deep learning techniques and analyzed a set of Illumina genotyping array data. We built a bi-stream convolutional neural networks model to screen/predict the occurrence of DS. Firstly, we built image input data by converting the intensities of each SNP site into chromosome SNP maps. Next, we proposed a bi-stream convolutional neural network (CNN) architecture with nine layers and two branch models. We further merged two CNN branch models into one model in the fourth convolutional layer, and output the prediction in the last layer.ResultsOur bi-stream CNN model achieved 99.3% average accuracies, and very low false-positive and false-negative rates, which was necessary for further applications in disease prediction and medical practice. We further visualized the feature maps and learned filters from intermediate convolutional layers, which showed the genomic patterns and correlated SNPs variations in human DS genomes. We also compared our methods with other CNN and traditional machine learning models. We further analyzed and discussed the characteristics and strengths of our bi-stream CNN model.ConclusionsOur bi-stream model used two branch CNN models to learn the local genome features and regional patterns among adjacent genes and SNP sites from two chromosomes simultaneously. It achieved the best performance in all evaluating metrics when compared with two single-stream CNN models and three traditional machine-learning algorithms. The visualized feature maps also provided opportunities to study the genomic markers and pathway components associated with Human DS, which provided insights for gene therapy and genomic medicine developments.

Hyperspectral and Multispectral Image Fusion via Deep Two-Branches Convolutional Neural Network

Enhancing the spatial resolution of hyperspectral image (HSI) is of significance for applications. Fusing HSI with a high resolution (HR) multispectral image (MSI) is an important technology for HSI enhancement. Inspired by the success of deep learning in image enhancement, in this paper, we propose a HSI-MSI fusion method by designing a deep convolutional neural network (CNN) with two branches which are devoted to features of HSI and MSI. In order to exploit spectral correlation and fuse the MSI, we extract the features from the spectrum of each pixel in low resolution HSI, and its corresponding spatial neighborhood in MSI, with the two CNN branches. The extracted features are then concatenated and fed to fully connected (FC) layers, where the information of HSI and MSI could be fully fused. The output of the FC layers is the spectrum of the expected HR HSI. In the experiment, we evaluate the proposed method on Airborne Visible Infrared Imaging Spectrometer (AVIRIS), and Environmental Mapping and Analysis Program (EnMAP) data. We also apply it to real Hyperion-Sentinel data fusion. The results on the simulated and the real data demonstrate that the proposed method is competitive with other state-of-the-art fusion methods.

Fast Branch Convolutional Neural Network for Traffic Sign Recognition

In this paper, we propose a novel framework for speeding up the test-time of traffic sign recognition, which is named Branch Convolution Neural Network. It is the first time to introduce a branch-output mechanism into a deep Convolution Neural Network. Our model has an accuracy as high as a deep convolution neural network model, while it performs faster at the same condition during test stage. It is a significantly accelerated framework for designing a real-time deep neural network system. We present a detail process to change a regular pre-trained Convolution Neural Network into a Branch Convolution Neural Network: train several simple branch classifiers, bias classifiers and optimize branches. Experiment applied on GTSRB shows that large number of traffic signs are unnecessary to go through all layers in a deep model and they can be separated out in a relative shallow neural network. This framework speeds up the recognition progress, while keeping the accuracy within an extremely minor drop.