Deep Residual Convolutional Network Research Articles

Complex Human Activity Recognition Based on Spatial LSTM and Deep Residual Convolutional Network Using Wearable Motion Sensors

Complex Human Activity Recognition Based on Spatial LSTM and Deep Residual Convolutional Network Using Wearable Motion Sensors

A maize seed variety identification method based on improving deep residual convolutional network.

Seed quality and safety are related to national food security, and seed variety purity is an essential indicator in seed quality detection. This study established a maize seed dataset comprising 5877 images of six different types and proposed a maize seed recognition model based on an improved ResNet50 framework. Firstly, we introduced the ResStage structure in the early stage of the original model, which facilitated the network's learning process and enabled more efficient information propagation across the network layers. Meanwhile, in the later residual blocks of the model, we introduced both the efficient channel attention (ECA) mechanism and depthwise separable (DS) convolution, which reduced the model's parameter cost and enabled the capturing of more precise and detailed features. Finally, a Swish-PReLU mixed activation function was introduced globally to improve the overall predictive power of the model. The results showed that our model achieved an impressive accuracy of 91.23% in corn seed classification, surpassing other related models. Compared with the original model, our model improved the accuracy by 7.07%, reduced the loss value by 0.19, and decreased the number of parameters by 40%. The research suggested that this method can efficiently classify corn seeds, holding significant value in seed variety identification.

Tool wear condition monitoring based on multi-sensor integration and deep residual convolution network

In order to guarantee machining quality and production efficiency and as well as to reduce downtime and cost, it is essential to receive information about the cutting tool condition and change it in time if necessary. To this end, tool condition monitoring systems are of great significance. Based on the data fusion technology of multi-sensor integration and the learning ability of deep residual network, a tool wear monitoring system is proposed in this paper. For data acquisition, multiple sensors are used to collect vibration, noise and acoustic emission signals in the machining process. For signal processing, the data fusion method of angular summation of matrices is used, so that the multi-source data are fused into two-dimensional pictures. Then, the feature extraction module of the proposed system uses the pictures as input and takes advantage of the deep residual convolutional network in extracting the deep features from the pictures, and finally the identification module completes the recognition of the tool wear type. To verify its feasibility, a testing bed is built on a CNC milling machine, and multi-sensor data are collected during the machining process of a simple workpiece with one of the two selected cutting tools each time. The proposed system is trained using the collected data and then is used to monitor the wear condition of a new cutting tool by collecting real-time data under similar condition. The results show that the accuracy of wear state recognition is as high as 90%, which verifies the feasibility of the proposed tool wear monitoring system.

DEMO2: Assemble multi-domain protein structures by coupling analogous template alignments with deep-learning inter-domain restraint prediction.

Most proteins in nature contain multiple folding units (or domains). The revolutionary success of AlphaFold2 in single-domain structure prediction showed potential to extend deep-learning techniques for multi-domain structure modeling. This work presents a significantly improved method, DEMO2, which integrates analogous template structural alignments with deep-learning techniques for high-accuracy domain structure assembly. Starting from individual domain models, inter-domain spatial restraints are first predicted with deep residual convolutional networks, where full-length structure models are assembled using L-BFGS simulations under the guidance of a hybrid energy function combining deep-learning restraints and analogous multi-domain template alignments searched from the PDB. The output of DEMO2 contains deep-learning inter-domain restraints, top-ranked multi-domain structure templates, and up to five full-length structure models. DEMO2 was tested on a large-scale benchmark and the blind CASP14 experiment, where DEMO2 was shown to significantly outperform its predecessor and the state-of-the-art protein structure prediction methods. By integrating with new deep-learning techniques, DEMO2 should help fill the rapidly increasing gap between the improved ability of tertiary structure determination and the high demand for the high-quality multi-domain protein structures. The DEMO2 server is available at https://zhanggroup.org/DEMO/.

Electroencephalogram-Based Motor Imagery Classification Using Deep Residual Convolutional Networks.

The classification of electroencephalogram (EEG) signals is of significant importance in brain-computer interface (BCI) systems. Aiming to achieve intelligent classification of motor imagery EEG types with high accuracy, a classification methodology using the wavelet packet decomposition (WPD) and the proposed deep residual convolutional networks (DRes-CNN) is proposed. Firstly, EEG waveforms are segmented into sub-signals. Then the EEG signal features are obtained through the WPD algorithm, and some selected wavelet coefficients are retained and reconstructed into EEG signals in their respective frequency bands. Subsequently, the reconstructed EEG signals were utilized as input of the proposed deep residual convolutional networks to classify EEG signals. Finally, EEG types of motor imagination are classified by the DRes-CNN classifier intelligently. The datasets from BCI Competition were used to test the performance of the proposed deep learning classifier. Classification experiments show that the average recognition accuracy of this method reaches 98.76%. The proposed method can be further applied to the BCI system of motor imagination control.

Deep learning based hybrid sequence modeling for optical response retrieval in metasurfaces for STPV applications

A standardized hybrid deep-learning model based on a combination of a deep convolutional network and a recurrent neural network is proposed to predict the optical response of metasurfaces considering their shape and all the important dimensional parameters (such as periodicity, height, width, and aspect ratio) simultaneously. It is further used to aid the design procedure of the key components of solar thermophotovoltaics (STPVs), i.e., metasurface based perfect solar absorbers and selective emitters. Although these planar meta-absorbers and meta-emitters offer an ideal platform to realize compact and efficient STPV systems, a conventional procedure to design these is time taking, laborious, and computationally exhaustive. The optimization of such planar devices needs hundreds of EM simulations, where each simulation requires multiple iterations to solve Maxwell's equations on a case-by-case basis. To overcome these challenges, we propose a unique deep learning-based model that generates the most likely optical response by taking images of the unit cells as input. The proposed model uses a deep residual convolutional network to extract the features from the images followed by a gated recurrent unit to infer the desired optical response. Two datasets having considerable variance are collected to train the proposed network by simulating randomly shaped nanostructures in CST microwave studio with periodic boundary conditions over the desired wavelength ranges. These simulations yield the optical absorption/emission response as the target labels. The proposed hybrid configuration and transfer learning provide a generalized model to infer the absorption/emission spectrum of solar absorbers/emitters within a fraction of seconds with high accuracy, regardless of its shape and dimensions. This accuracy is defined by the regression metric mean square error (MSE), where the minimum MSE achieved for absorbers and emitters test datasets are 7.3 × 10−04 and 6.2 × 10−04 respectively. The trained model can also be fine-tuned to predict the absorption response of different thin film refractory materials. To enhance the diversity of the model. Thus it aids metasurface design procedure by replacing the conventional time-consuming and computationally exhaustive numerical simulations and electromagnetic (EM) software. The comparison of the average simulation time (for 10 samples) and the average DL model prediction time shows that the DL model works about 98% faster than the conventional simulations. We believe that the proposed methodology will open new research directions towards more challenging optimization problems in the field of electromagnetic metasurfaces.

Deducing high-accuracy protein contact-maps from a triplet of coevolutionary matrices through deep residual convolutional networks.

The topology of protein folds can be specified by the inter-residue contact-maps and accurate contact-map prediction can help ab initio structure folding. We developed TripletRes to deduce protein contact-maps from discretized distance profiles by end-to-end training of deep residual neural-networks. Compared to previous approaches, the major advantage of TripletRes is in its ability to learn and directly fuse a triplet of coevolutionary matrices extracted from the whole-genome and metagenome databases and therefore minimize the information loss during the course of contact model training. TripletRes was tested on a large set of 245 non-homologous proteins from CASP 11&12 and CAMEO experiments and outperformed other top methods from CASP12 by at least 58.4% for the CASP 11&12 targets and 44.4% for the CAMEO targets in the top-L long-range contact precision. On the 31 FM targets from the latest CASP13 challenge, TripletRes achieved the highest precision (71.6%) for the top-L/5 long-range contact predictions. It was also shown that a simple re-training of the TripletRes model with more proteins can lead to further improvement with precisions comparable to state-of-the-art methods developed after CASP13. These results demonstrate a novel efficient approach to extend the power of deep convolutional networks for high-accuracy medium- and long-range protein contact-map predictions starting from primary sequences, which are critical for constructing 3D structure of proteins that lack homologous templates in the PDB library.

DeepDist: real-value inter-residue distance prediction with deep residual convolutional network

BackgroundDriven by deep learning, inter-residue contact/distance prediction has been significantly improved and substantially enhanced ab initio protein structure prediction. Currently, most of the distance prediction methods classify inter-residue distances into multiple distance intervals instead of directly predicting real-value distances. The output of the former has to be converted into real-value distances to be used in tertiary structure prediction.ResultsTo explore the potentials of predicting real-value inter-residue distances, we develop a multi-task deep learning distance predictor (DeepDist) based on new residual convolutional network architectures to simultaneously predict real-value inter-residue distances and classify them into multiple distance intervals. Tested on 43 CASP13 hard domains, DeepDist achieves comparable performance in real-value distance prediction and multi-class distance prediction. The average mean square error (MSE) of DeepDist’s real-value distance prediction is 0.896 Å2 when filtering out the predicted distance ≥ 16 Å, which is lower than 1.003 Å2 of DeepDist’s multi-class distance prediction. When distance predictions are converted into contact predictions at 8 Å threshold (the standard threshold in the field), the precision of top L/5 and L/2 contact predictions of DeepDist’s multi-class distance prediction is 79.3% and 66.1%, respectively, higher than 78.6% and 64.5% of its real-value distance prediction and the best results in the CASP13 experiment.ConclusionsDeepDist can predict inter-residue distances well and improve binary contact prediction over the existing state-of-the-art methods. Moreover, the predicted real-value distances can be directly used to reconstruct protein tertiary structures better than multi-class distance predictions due to the lower MSE. Finally, we demonstrate that predicting the real-value distance map and multi-class distance map at the same time performs better than predicting real-value distances alone.

Unboxing AI - Radiological Insights Into a Deep Neural Network for Lung Nodule Characterization

To explain predictions of a deep residual convolutional network for characterization of lung nodule by analyzing heat maps. A 20-layer deep residual CNN was trained on 1245 Chest CTs from National Lung Screening Trial (NLST) trial to predict the malignancy risk of a nodule. We used occlusion to systematically block regions of a nodule and map drops in malignancy risk score to generate clinical attribution heatmaps on 103 nodules from Lung Image Database Consortium image collection and Image Database Resource Initiative (LIDC-IDRI) dataset, which were analyzed by a thoracic radiologist. The features were described as heat inside nodule -bright areas inside nodule, peripheral heat continuous/interrupted bright areas along nodule contours, heat in adjacent plane -brightness in scan planes juxtaposed with the nodule, satellite heat - a smaller bright spot in proximity to nodule in the same scan plane, heat map larger than nodule bright areas corresponding to the shape of the nodule seen outside the nodule margins and heat in calcification. These six features were assigned binary values. This feature vector was fedinto a standard J48 decision tree with 10-fold cross-validation, which gave an 85 % weighted classification accuracy with a 77.8% True Positive (TP) rate, 8% False Positive (FP) rate for benign cases and 91.8% TP and 22.2% FP rates for malignant cases. Heat Inside nodule was more frequently observed in nodules classified as malignant whereas peripheral heat, heat in adjacent plane, and satellite heat were more commonly seen in nodules classified as benign. We discuss the potential ability of a radiologist to visually parse the deep learning algorithm generated "heat map" to identify features aiding classification.

A deep residual convolutional neural network for automatic lithological facies identification in Brazilian pre-salt oilfield wellbore image logs

Field characterization in oil industry is a challenging task which aims to determine or estimate some of the petrophysical properties of a reservoir. These properties are expected to be later used in a non-deterministic workflow (which might depend on subjective interpretations carried out by each petrophysicist) whose aim is to answer questions such as whether exploiting the reservoir is feasible or not, how much hydrocarbon can be extracted from it and whether the geology of the reservoir will stand the stresses of exploitation without collapsing. Three properties stand-out amongst all the rest, as they are closely correlated to the amount of hydrocarbon present in the reservoir and also to the feasibility of its exploitation: permeability, porosity, and lithology. Obtaining reliable and robust measurements of these properties requires extracting core samples from the reservoir, which is, however, a very resource-consuming task. Hence, it is common to use other tools (such as wirelogging tools or borehole imagers) to extract several other properties from the reservoir, in an attempt to obtain information that might help experts to estimate these properties on those intervals at which no core sample could be extracted. In this context, estimating lithofacies helps petrophysicists to automatize the process of identifying the lithology of the reservoir. Previous work (Basu et al., 2002; Chai et al., 2009; Linek et al., 2007; Newberry et al., 2004; Hall et al., 1996) used a considerable amount of wirelog data and borehole image logs to induce the lithology of the reservoir. In the work presented in this paper, we introduce a new methodology for automatic reservoir lithofacies identification. Our model relies only on ultrasonic and microresistivity borehole image logs as inputs, which are characterized by a deep residual convolutional network which then infers the probability of each sample to be classified as each type of previously defined lithofacies classes. The method presented in this work was tested on a carbonate well from the Brazilian pre-salt oilfields, and it allowed us to obtain an average classification accuracy of 81.45% and an average area under the ROC curve of 92.70% for all classes, for the blind test sample.

Lightweight Deep Residual CNN for Fault Diagnosis of Rotating Machinery Based on Depthwise Separable Convolutions

This paper proposes an efficient and noise-insensitive end-to-end lightweight deep learning method. The method synthesizes the characteristics of a frequency domain transform and a deep convolutional neural network. The former can extract multiscale information in vibration signal processing and the latter has a good classification performance, data-driven, and high transfer-learning ability. A vibration signal is decomposed into a pyramidal wavelet packet, and each sub-band coefficient is used as an input of a channel in the deep network. A deep residual convolutional network based on a separable convolution and concatenated rectified linear unit (CReLU) lightweight convolution technology is used for fault diagnosis. The proposed algorithm is compared with related deep learning algorithms using two bearing datasets produced by Case Western Reserve University (CWRU) and the Center for Intelligent Maintenance Systems (IMS), University of Cincinnati. Compared with the existing algorithms, the experimental results show that the comprehensive performance of the algorithm proposed in this paper is “small, light, and fast,” and satisfactory diagnostic results are obtained in the fault diagnosis of rotating machinery.

Estimating Full Regional Skeletal Muscle Fibre Orientation from B-Mode Ultrasound Images Using Convolutional, Residual, and Deconvolutional Neural Networks

This paper presents an investigation into the feasibility of using deep learning methods for developing arbitrary full spatial resolution regression analysis of B-mode ultrasound images of human skeletal muscle. In this study, we focus on full spatial analysis of muscle fibre orientation, since there is an existing body of work with which to compare results. Previous attempts to automatically estimate fibre orientation from ultrasound are not adequate, often requiring manual region selection, feature engineering, providing low-resolution estimations (one angle per muscle) and deep muscles are often not attempted. We build upon our previous work in which automatic segmentation was used with plain convolutional neural network (CNN) and deep residual convolutional network (ResNet) architectures, to predict a low-resolution map of fibre orientation in extracted muscle regions. Here, we use deconvolutions and max-unpooling (DCNN) to regularise and improve predicted fibre orientation maps for the entire image, including deep muscles, removing the need for automatic segmentation and we compare our results with the CNN and ResNet, as well as a previously established feature engineering method, on the same task. Dynamic ultrasound images sequences of the calf muscles were acquired (25 Hz) from 8 healthy volunteers (4 male, ages: 25–36, median 30). A combination of expert annotation and interpolation/extrapolation provided labels of regional fibre orientation for each image. Neural networks (CNN, ResNet, DCNN) were then trained both with and without dropout using leave one out cross-validation. Our results demonstrated robust estimation of full spatial fibre orientation within approximately 6° error, which was an improvement on previous methods.