Search

Collaborative fault diagnosis has become a hot research topic in fault detection and identification, greatly benefiting from emerging multisensory fusion techniques and newly developed convolutional neural network (CNN) models. Existing CNN models take advantage of various fusion techniques to identify machine health status by utilizing multiple sensory signals. Nevertheless, a few of them are able to simultaneously explore modality-specific features and intrinsic shared features among multi-source signals, limiting the capability of the exploration of multisource data. To address this issue, this paper proposes a novel convolutional network called a graph-guided collaborative convolutional neural network (GGCN) for highly-effective fault diagnosis of electromechanical systems. The main contributions of this study include: (1) developing a novel graph-guided CNN algorithm for collaborative fault detection; (2) establishing a graph reasoning fusion module (GRFM) to explore the inherent correlations between multisource signals; and (3) advancing the current approaches by taking into account both the distribution gap and the intrinsic correlation between different signals simultaneously. The developed GGCN is expected to shed new light on collaborative fault diagnosis using the graph-convolution-based intermediate fusion scheme. Two experimental datasets namely, the cylindrical rolling bearing dataset and the planetary gearbox dataset, are applied in this paper to verify the efficacy of the GGCN. Experimental results demonstrate that GGCN outperforms seven other state-of-the-art approaches, particularly under noisy conditions. [ABSTRACT FROM AUTHOR]

• We present a new EEG data set and evaluation tasks with well-formed annotations. • We propose a novel, fully end-to-end deep model (CEEDNet) for screening EEG signals. • CEEDNet aims to bring all functions for EEG analysis in a seamless learnable fashion. • The proposed CEEDNet significantly improves accuracy compared to existing methods. • Extensive experiments and analyses provide the in-depth property of our CEEDNet. Based on the Chung-Ang University Hospital EEG (CAUEEG) dataset, this paper presents a new fully end-to-end deep learning approach for screening EEG signals, called the C AU E EG E nd-to-end D eep neural Net work (CEEDNet). The core idea of CEEDNet is to combine all the functional elements used to analyze EEG signals in a seamless learnable fashion. CEEDNet pursues to utilize the domain characteristics of EEG signals while minimizing unnecessary human intervention. On the CAUEEG-Dementia and CAUEEG-Abnormal evaluation tasks, CEEDNet produced a significant im provement in accuracy and other metrics compared with existing methods. [Display omitted] For automatic EEG diagnosis, this paper presents a new EEG data set with well-organized clinical annotations called Chung-Ang University Hospital EEG (CAUEEG), which has event history, patient's age, and corresponding diagnosis labels. We also designed two reliable evaluation tasks for the low-cost, non-invasive diagnosis to detect brain disorders: i) CAUEEG-Dementia with normal, mci , and dementia diagnostic labels and ii) CAUEEG-Abnormal with normal and abnormal. Based on the CAUEEG dataset, this paper proposes a new fully end-to-end deep learning model, called the C AU E EG E nd-to-end D eep neural Net work (CEEDNet). CEEDNet pursues to bring all the functional elements for the EEG analysis in a seamless learnable fashion while restraining non-essential human intervention. Extensive experiments showed that our CEEDNet significantly improves the accuracy compared with existing methods, such as machine learning methods and Ieracitano-CNN (Ieracitano et al., 2019), due to taking full advantage of end-to-end learning. The high ROC-AUC scores of 0.9 on CAUEEG-Dementia and 0.86 on CAUEEG-Abnormal recorded by our CEEDNet models demonstrate that our method can lead potential patients to early diagnosis through automatic screening. [ABSTRACT FROM AUTHOR]

In recent years, the rapid development of convolutional neural networks (CNNs) has significantly advanced the progress of intelligent fault diagnosis. Most currently-available CNN-based diagnostic models are built on the premise that the monitored machine operates under stable conditions. However, in real-world scenarios, rotary machinery usually operates at varying speeds, making the fault-related pulse features susceptible to noise oversaturation. To extract discriminative features from mechanical signals under non-stationary conditions, a global contextual feature aggregation network (GCFAN) is developed in this paper. To begin with, a global contextual module (GCM) is embedded in the CNN architecture to explore multimodal features. Then, a multiscale attention module (MSAM) is introduced to guide the model to focus on global and local discriminative information. Further, a multiscale feature enhancement module (MFEM) is established to enlarge the receptive field and eliminate useless features. Finally, the GCFAN architecture is constructed based on these improvements. To achieve favourable diagnostic results under fluctuating variable speed conditions, we apply the label smoothing algorithm and the AMSGrad algorithm to assist the training of the model. Two case studies using the benchmark variable speed bearing dataset and the HF-MS variable speed gearbox dataset were carried out to test the practicality of the developed approach. Experimental results demonstrated that the developed GCFAN performs better than seven state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Convolutional neural networks have been proposed as an approach for classifying data corresponding to a variety of datasets. Indeed, developments in data diversity and information technology have increased the complexity of deep learning algorithms. Numerous trained models have been proposed for supporting complex algorithms and data detachment with high accuracy. Convolutional operations increase when the convolution depth of neural networks increases. Thus, employing deep convolutional networks is challenging regarding energy consumption, bandwidth, memory requirements, and memory access. Different types of on-chip communication platforms and traffic distribution methods are effective in the improvement of memory access and energy consumption induced by data transfer. Also, dataflow mapping methods have an impressive effect on reducing or increasing delay and energy consumption caused by exchanging data between cores of a communication network. Different methods have been proposed to dataflow mapping on various networks for reducing total hop counts that led to improve performance and cost. Dataflow mapping approach can affect performance improvement of the inference phase in neural networks. This paper proposes various traffic patterns by considering different memory access mechanisms for traffic distribution of a trained AlexNet model on mesh topology. We propose a flow mapping method (FMM) on the mesh to determine the data flow efficiency of different traffic patterns on energy consumption. The FMM reduced energy and total flow by approximately 17.86% and 34.16%, respectively, using different traffic patterns. Thus, the FMM improved the performance of AlexNet traffic distribution while the impact on data flow reduced energy consumption. [ABSTRACT FROM AUTHOR]

Scientific evidence shows that acoustic analysis could be an indicator for diagnosing COVID-19. From analyzing recorded breath sounds on smartphones, it is discovered that patients with COVID-19 have different patterns in both the time domain and frequency domain. These patterns are used in this paper to diagnose the infection of COVID-19. Statistics of the sound signals, analysis in the frequency domain, and Mel-Frequency Cepstral Coefficients (MFCCs) are then calculated and applied in two classifiers, k-Nearest Neighbors (kNN) and Convolutional Neural Network (CNN), to diagnose whether a user is contracted with COVID-19 or not. Test results show that, amazingly, an accuracy of over 97% could be achieved with a CNN classifier and more than 85% on kNN with optimized features. Optimization methods for selecting the best features and using various metrics to evaluate the performance are also demonstrated in this paper. Owing to the high accuracy of the CNN model, the CNN model was implemented in an Android app to diagnose COVID-19 with a probability to indicate the confidence level. The initial medical test shows a similar test result between the method proposed in this paper and the lateral flow method, which indicates that the proposed method is feasible and effective. Because of the use of breath sound and tested on the smartphone, this method could be used by everybody regardless of the availability of other medical resources, which could be a powerful tool for society to diagnose COVID-19. [ABSTRACT FROM AUTHOR]

Image-based facial age estimation is considered an intractable problem because aging characteristics are hard to obtain. Most previous works have focused on extracting age-related features, but rarely explored which local region plays an important role. Several works combine local face regions with global face to estimate age in a heuristic way, where the local regions are uniformly cropped for each individual. In this paper, we design an individual adaptive segmentation of local regions of interest to perform personalized local features extraction and build hierarchical age features by erasing the local regions of interest iteratively for each individual. A joint multi-input and multi-output (MIMO) network for multi-task learning of age classification and regression tasks is designed by combining global features and personalized local features as inputs. In addition, we conduct extensive experiments to validate the effectiveness of the proposed method for age estimation, which beats most state-of-the-art methods in three public datasets and also works well for gender and race estimation. • An individual adaptive local region segmentation technique is proposed. • Personalized local features instead of uniformly cropped ones are extracted. • A joint multi-input and multi-output network is designed. • Our method can be well applied to age, gender and race estimation. [ABSTRACT FROM AUTHOR]

Abstract Optical coherence tomography (OCT) can achieve the high-resolution 3D tomography imaging of the retina, which is crucial for the diagnosis of retinal diseases. Currently, the classification of retinal OCT images is mainly conducted by ophthalmologists, which is time consuming and subjective. In this paper, we propose an iterative fusion convolutional neural network (IFCNN) method for the automatic classification of retinal OCT image. In the convolutional neural network (CNN), different convolutional layers contain feature information from different scales. Therefore, the proposed network adopts an iterative fusion strategy, which iteratively combines features in current convolutional layer with those of all previous layers in the CNN network, and thus can jointly utilize the features of different convolutional layers to achieve accurate classification of OCT images. Experimental results on a real retinal OCT dataset and a musculoskeletal radiographs dataset demonstrate the superiority of the proposed method over the traditional CNN and several well-known OCT classification methods. [ABSTRACT FROM AUTHOR]

In recent years, with the raise of the neural network and deep learning, significant progress has been achieved in the field of image recognition. Convolutional Neural Network (CNN) has been widely used in multiple image recognition tasks, but the recognition accuracy still has a lot of room for improvement. In this paper, we proposed a hybrid model CNN-GRNN to improve recognition accuracy. The model uses CNN to extract multilayer image representation and it uses General Regression Neural Network (GRNN) to classify image using the extracted feature. The CNN-GRNN model replace Back propagation (BP) neural network inside CNN with GRNN to improve generalization and robustness of CNN. Furthermore, we validate our model on the Oxford-IIIT Pet Dataset database and the Keck Gesture Dataset, the experiment result indicate that our model is superior to Gray Level Co-occurrency (GLCM),HU invariant moments, CNN and CNN_SVM on small sample dataset. Our model has favorable real-time characteristic at the same time. [ABSTRACT FROM AUTHOR]

Graphical abstract Highlights • A feature visualization for CNNs method is exploited to improve co-segmentation. • Visualization is used as an auxiliary information to discriminate salient regions. • The visualization information is used as an extra energy term in the cost function. • Region occlusion sensitivity is introduced for feature visualization. • An adaptive strategy is designed to detect correct foreground/background regions. This paper addresses the co-segmentation problem using feature visualization for CNNs. Visualization is exploited as an auxiliary information to discriminate salient image regions (dubbed as “heat-regions”) from non-salient ones. Region occlusion sensitivity is proposed for feature visualization. The co-segmentation problem is formulated via a convex quadratic optimization which is initialized by the heat-regions. The information obtained through the visualization is considered as an extra energy term in the cost function. The results of the visualization demonstrate that there exist some heat-regions which are not productive in the co-segmentation. To detect helpful regions among them, an adaptive strategy in the form of an iterative algorithm is proposed according to the consistency among all images. Comparison experiments conducted on two benchmark datasets, iCoseg and MSRC, illustrate the superior performance of the proposed approach over state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Semiology observation and characterization play a major role in the presurgical evaluation of epilepsy. However, the interpretation of patient movements has subjective and intrinsic challenges. In this paper, we develop approaches to attempt to automatically extract and classify semiological patterns from facial expressions. We address limitations of existing computer-based analytical approaches of epilepsy monitoring, where facial movements have largely been ignored. This is an area that has seen limited advances in the literature. Inspired by recent advances in deep learning, we propose two deep learning models, landmark-based and region-based, to quantitatively identify changes in facial semiology in patients with mesial temporal lobe epilepsy (MTLE) from spontaneous expressions during phase I monitoring. A dataset has been collected from the Mater Advanced Epilepsy Unit (Brisbane, Australia) and is used to evaluate our proposed approach. Our experiments show that a landmark-based approach achieves promising results in analyzing facial semiology, where movements can be effectively marked and tracked when there is a frontal face on visualization. However, the region-based counterpart with spatiotemporal features achieves more accurate results when confronted with extreme head positions. A multifold cross-validation of the region-based approach exhibited an average test accuracy of 95.19% and an average AUC of 0.98 of the ROC curve. Conversely, a leave-one-subject-out cross-validation scheme for the same approach reveals a reduction in accuracy for the model as it is affected by data limitations and achieves an average test accuracy of 50.85%. Overall, the proposed deep learning models have shown promise in quantifying ictal facial movements in patients with MTLE. In turn, this may serve to enhance the automated presurgical epilepsy evaluation by allowing for standardization, mitigating bias, and assessing key features. The computer-aided diagnosis may help to support clinical decision-making and prevent erroneous localization and surgery. [ABSTRACT FROM AUTHOR]

In this paper, a CNN architecture for object recognition is proposed, aiming at achieving a good processing-quality at the lowest computation-cost. The work includes the design of SdcBlock, a convolution module, for feature extraction, and that of SdcNet, an end-to-end CNN architecture. The module is designed to extract the maximum amount of high-density feature information from a given set of data channels. To this end, successive depthwise convolutions (Sdc) are applied to each group of data to produce feature elements of different filtering orders. To optimize the functionality of these convolutions, a particular pre-and-post-convolution data control is applied. The pre-convolution control is to organize the input channels of the module so that the depthwise convolutions can be performed with a single channel or a combination of multiple data channels, depending on the nature of the data. The post-convolution control is to combine the critical feature elements of different filtering orders to enhance the quality of the convolved results. The SdcNet is mainly composed of cascaded SdcBlocks. The hyper-parameters in the architecture can be adjusted easily so that each module can be tuned to suit its input signals in order to optimize the processing-quality of the entire network. Three different versions of SdcNet have been proposed and tested using CIFAR dataset, and the results demonstrate that the architecture gives a better processing-quality at a significantly lower computation cost, compared with networks performing similar tasks. Two other versions have also been tested with samples from ImageNet to prove the applicability of SdcNet in object recognition with images of ImageNet format. Also, a SdcNet for brain tumor detection has been designed and tested successfully to illustrate that SdcNet can effectively perform the detection with a high computation efficiency. • New convolution modules involving 2 Successive Depthwise Convolutions (SdcBlock). • SdcBlock allowing extracting features of 2 different filtering orders in the module. • New CNN architectures SdcNets based on SdcBlocks for efficient feature extraction. • Simple structures with simple convolution modules for great flexibility. • Test results confirming effectiveness of the SdcNets built for object recognition. [ABSTRACT FROM AUTHOR]

Single image dehazing is a critical image pre-processing step for many practical vision systems. Most existing dehazing methods solve this problem utilizing various of hand-crafted priors or by supervised training on the synthetic hazy image information (such as haze-free image, transmission map and atmospheric light). However, the assumptions on the hand-crafted priors are easily violated and collecting realistic transmission map and atmospheric light are unpractical. In this paper, we propose a novel weakly supervised network based on the multi-level multi-scale block. The proposed network reduces the constraint on the training data and automatically estimates the transmission map and the atmospheric light as well as the intermediate haze-free image without using any realistic transmission map and atmospheric light as supervision. Moreover, the estimated intermediate haze-free image helps to generate accurate transmission map and atmospheric light by embedding the physical-model, which presents reliable restoration of the final haze-free image. In particular, our network also can be trained on the real-world dataset to fine-tune the model and the fine-tuning operation improves the dehazing performance on the real-world dataset. Quantitative and qualitative experimental results demonstrate the proposed method performs on par with the supervised methods. • We propose a weakly supervised single image dehazing method. • The network can simultaneously estimate the transmission map and atmospheric light. • The network can be fine-tuned to improve the real-world dehazing result. [ABSTRACT FROM AUTHOR]

• An iterative local feature learning scheme using both the point correspondences and clustering labels is proposed. • A dual margin loss together with a self-spatial transform learning scheme is used to alleviate the inconsistency problem in matching and clustering results and to learn more robust feature. • Experiments on image-level retrieval and object matching are conducted. This paper focuses on learning the local image region representation via deep neural networks. Existing works mainly learn from matched corresponding image patches, with which the learned feature is too sensitive to the individual local patch matching result and cannot handle aggregation based tasks such as image level retrieval. Thus, we propose to use both the matched corresponding image patches and the clustering result as labels for the network training. To resolve the inconsistency between the matched correspondences and clustering results, we propose a semi-supervised iterative training scheme together with a dual margins loss. Moreover, a jointly learned spatial transform prediction network is utilized to obtain better spatial transform invariance of the learned local features. Using SIFT as the label initializer, experimental results show the comparable or even better performance than the hand-crafted feature, which sheds lights on learning local feature representation in an unsupervised or weakly supervised manner. [ABSTRACT FROM AUTHOR]