Search

Image recognition has long been one of the research hotspots in computer vision tasks. The development of deep learning is rapid in recent years, and convolutional neural networks usually need to be designed with fixed resources. If sufficient resources are available, the model can be scaled up to achieve higher accuracy, for example, VggNet, ResNet, GoogLeNet, etc. Although the accuracy of large-scale models has been improved, the following problems will occur with the expansion of model scale: (1) There may be over-fitting; (2) increasing model parameters; (3) slow model convergence. This paper proposes a design method for a modular convolutional neural network model which solves the problem of over-fitting and large model parameters by connecting multiple modules in parallel. Moreover, each module contains several submodules (three submodules in this paper) and fuses the features extracted from the submodules. The model convergence can be accelerated by using the fused features (the fused features contain more image information). In this study, we add a gate unit based on the attention mechanism to the model, which aims to optimize the structure of the model (select the optimal number of modules), allowing the model to select an optimum network structure by learning and dynamically reducing FLOPs (floating-point operations per second) of the model. Compared to VggNet, ResNet, and GoogLeNet, the structure of the model proposed in this paper is simple and the parameters are small. The proposed model achieves good results in the Kaggle datasets Cats-vs.-Dogs (99.3%), 10-Monkey Species (99.26%), and Birds-400 (99.13%). [ABSTRACT FROM AUTHOR]

Due to the sharp increase in household waste, its separate collection is essential in order to reduce the huge amount of household waste, since it is difficult to recycle trash without separate collection. However, since it is costly and time-consuming to separate trash manually, it is crucial to develop an automatic system for separate collection using deep learning and computer vision. In this paper, we propose two Anchor-free-based Recyclable Trash Detection Networks (ARTD-Net) which can recognize overlapped multiple wastes of different types efficiently by using edgeless modules: ARTD-Net1 and ARTD-Net2. The former is an anchor-free based one-stage deep learning model which consists of three modules: centralized feature extraction, multiscale feature extraction and prediction. The centralized feature extraction module in backbone architecture focuses on extracting features around the center of the input image to improve detection accuracy. The multiscale feature extraction module provides feature maps of different scales through bottom-up and top-down pathways. The prediction module improves classification accuracy of multiple objects based on edge weights adjustments for each instance. The latter is an anchor-free based multi-stage deep learning model which can efficiently finds each of waste regions by additionally exploiting region proposal network and RoIAlign. It sequentially performs classification and regression to improve accuracy. Therefore, ARTD-Net2 is more accurate than ARTD-Net1, while ARTD-Net1 is faster than ARTD-Net2. We shall show that our proposed ARTD-Net1 and ARTD-Net2 methods achieve competitive performance in mean average precision and F1 score compared to other deep learning models. The existing datasets have several problems that do not deal with the important class of wastes produced commonly in the real world, and they also do not consider the complex arrangement of multiple wastes with different types. Moreover, most of the existing datasets have an insufficient number of images with low resolution. We shall present a new recyclables dataset which is composed of a large number of high-resolution waste images with additional essential classes. We shall show that waste detection performance is improved by providing various images with the complex arrangement of overlapped multiple wastes with different types. [ABSTRACT FROM AUTHOR]

To ensure the safe operation of highway traffic lines, given the imperfect feature extraction of existing road pit defect detection models and the practicability of detection equipment, this paper proposes a lightweight target detection algorithm with enhanced feature extraction based on the YOLO (You Only Look Once) algorithm. The BIFPN (Bidirectional Feature Pyramid Network) network structure is used for multi-scale feature fusion to enhance the feature extraction ability, and Varifocal Loss is used to optimize the sample imbalance problem, which improves the accuracy of road defect target detection. In the evaluation test of the model in the constructed PCD1 (Pavement Check Dataset) dataset, the mAP@.5 (mean Average Precision when IoU = 0.5) of the BV-YOLOv5S (BiFPN Varifocal Loss-YOLOv5S) model increased by 4.1%, 3%, and 0.9%, respectively, compared with the YOLOv3-tiny, YOLOv5S, and B-YOLOv5S (BiFPN-YOLOv5S; BV-YOLOv5S does not use the Improved Focal Loss function) models. Through the analysis and comparison of experimental results, it is proved that the proposed BV-YOLOv5S network model performs better and is more reliable in the detection of pavement defects and can meet the needs of road safety detection projects with high real-time and flexibility requirements. [ABSTRACT FROM AUTHOR]

Hyperspectral remote sensing images (HRSI) have the characteristics of foreign objects with the same spectrum. As it is difficult to label samples manually, the hyperspectral remote sensing images are understood to be typical "small sample" datasets. Deep neural networks can effectively extract the deep features from the HRSI, but the classification accuracy mainly depends on the training label samples. Therefore, the stacked convolutional autoencoder network and transfer learning strategy are employed in order to design a new stacked convolutional autoencoder network model transfer (SCAE-MT) for the purposes of classifying the HRSI in this paper. In the proposed classification method, the stacked convolutional au-to-encoding network is employed in order to effectively extract the deep features from the HRSI. Then, the transfer learning strategy is applied to design a stacked convolutional autoencoder network model transfer under the small and limited training samples. The SCAE-MT model is used to propose a new HRSI classification method in order to solve the small samples of the HRSI. In this study, in order to prove the effectiveness of the proposed classification method, two HRSI datasets were chosen. In order to verify the effectiveness of the methods, the overall classification accuracy (OA) of the convolutional self-coding network classification method (CAE), the stack convolutional self-coding network classification method (SCAE), and the SCAE-MT method under 5%, 10%, and 15% training sets are calculated. When compared with the CAE and SCAE models in 5%, 10%, and 15% training datasets, the overall accuracy (OA) of the SCAE-MT method was improved by 2.71%, 3.33%, and 3.07% (on average), respectively. The SCAE-MT method is, thus, clearly superior to the other methods and also shows a good classification performance. [ABSTRACT FROM AUTHOR]

Although gait recognition has been greatly improved by efforts from many researchers in recent years, its performance is still unsatisfactory due to the lack of gait information under the real scenariowhere only one or two images may be used for recognition. In this paper, a new gait recognition framework is brought about which can combine the long-short-term attention modules on silhouette images over the whole sequence and the real human physiological information calculated by a monocular image. The contributions of this work include the following: (1) Fusing the global long-term attention (GLTA) and local short-term attention (LSTA) over the whole query sequence to improve the gait recognition accuracy, where both the short-term gait feature (from two or three frames) and long-term feature (from the whole sequence) are extracted; (2) presenting a method to calculate the real personal static and dynamic physiological features through a single monocular image; (3) by efficiently applying the human physiological information, a new physiological feature extraction (PFE) network is proposed to concatenate the physiological information with silhouette for gait recognition. Through the experiments between the CASIA-B and Multi-state Gait datasets, the effectiveness and efficiency of the proposed method are proven. Under three different walking conditions of the CASIA-B dataset, the mean accuracy of rank-1 in our method is up to 89.6%, and in the Multi-state Gait dataset, wearing different clothes, the mean accuracy of rank-1 in our method is 2.4% higher than the other works. [ABSTRACT FROM AUTHOR]

Recent advancements in applications of deep neural network for bearing fault diagnosis under variable operating conditions have shown promising outcomes. However, these approaches are limited in practical applications due to the complexity of neural networks, which require substantial computational resources, thereby hindering the advancement of automated diagnostic tools. To overcome these limitations, this study introduces a new fault diagnosis framework that incorporates a tri-channel preprocessing module for multidimensional feature extraction, coupled with an innovative diagnostic architecture known as the Lightweight Ghost Enhanced Feature Attention Network (GEFA-Net). This system is adept at identifying rolling bearing faults across diverse operational conditions. The FFE module utilizes advanced techniques such as Fast Fourier Transform (FFT), Frequency Weighted Energy Operator (FWEO), and Signal Envelope Analysis to refine signal processing in complex environments. Concurrently, GEFA-Net employs the Ghost Module and the Efficient Pyramid Squared Attention (EPSA) mechanism, which enhances feature representation and generates additional feature maps through linear operations, thereby reducing computational demands. This methodology not only significantly lowers the parameter count of the model, promoting a more streamlined architectural framework, but also improves diagnostic speed. Additionally, the model exhibits enhanced diagnostic accuracy in challenging conditions through the effective synthesis of local and global data contexts. Experimental validation using datasets from the University of Ottawa and our dataset confirms that the framework not only achieves superior diagnostic accuracy but also reduces computational complexity and accelerates detection processes. These findings highlight the robustness of the framework for bearing fault diagnosis under varying operational conditions, showcasing its broad applicational potential in industrial settings. The parameter count was decreased by 63.74% compared to MobileVit, and the recorded diagnostic accuracies were 98.53% and 99.98% for the respective datasets. [ABSTRACT FROM AUTHOR]

Obstructive sleep apnea (OSA), a prevalent sleep disorder, is intimately associated with various other diseases, particularly cardiovascular conditions. The conventional diagnostic method, nocturnal polysomnography (PSG), despite its widespread use, faces challenges due to its high cost and prolonged duration. Recent developments in electrocardiogram-based diagnostic techniques have opened new avenues for addressing these challenges, although they often require a deep understanding of feature engineering. In this study, we introduce an innovative method for OSA classification that combines a composite deep convolutional neural network model with a multimodal strategy for automatic feature extraction. This approach involves transforming the original dataset into scalogram images that reflect heart rate variability attributes and Gramian angular field matrix images that reveal temporal characteristics, aiming to enhance the diversity and richness of data features. The model comprises automatic feature extraction and feature enhancement components and has been trained and validated on the PhysioNet Apnea-ECG database. The experimental results demonstrate the model's exceptional performance in diagnosing OSA, achieving an accuracy of 96.37%, a sensitivity of 94.67%, a specificity of 97.44%, and an AUC of 0.96. These outcomes underscore the potential of our proposed model as an efficient, accurate, and convenient tool for OSA diagnosis. [ABSTRACT FROM AUTHOR]

Deep-learning models play a significant role in modern software solutions, with the capabilities of handling complex tasks, improving accuracy, automating processes, and adapting to diverse domains, eventually contributing to advancements in various industries. This study provides a comparative study on deep-learning techniques that can also be deployed on resource-constrained edge devices. As a novel contribution, we analyze the performance of seven Convolutional Neural Network models in the context of data augmentation, feature extraction, and model compression using acoustic data. The results show that the best performers can achieve an optimal trade-off between model accuracy and size when compressed with weight and filter pruning followed by 8-bit quantization. In adherence to the study workflow utilizing the forest sound dataset, MobileNet-v3-small and ACDNet achieved accuracies of 87.95% and 85.64%, respectively, while maintaining compact sizes of 243 KB and 484 KB, respectively. Henceforth, this study concludes that CNNs can be optimized and compressed to be deployed in resource-constrained edge devices for classifying forest environment sounds. [ABSTRACT FROM AUTHOR]

Sensor-based human activity recognition aims to classify human activities or behaviors according to the data from wearable or embedded sensors, leading to a new direction in the field of Artificial Intelligence. When the activities become high-level and sophisticated, such as in the multiple technical skills of playing badminton, it is usually a challenging task due to the difficulty of feature extraction from the sensor data. As a kind of end-to-end approach, deep neural networks have the capacity of automatic feature learning and extracting. However, most current studies on sensor-based badminton activity recognition adopt CNN-based architectures, which lack the ability of capturing temporal information and global signal comprehension. To overcome these shortcomings, we propose a deep learning framework which combines the convolutional layers, LSTM structure, and self-attention mechanism together. Specifically, this framework can automatically extract the local features of the sensor signals in time domain, take the LSTM structure for processing the badminton activity data, and focus attention on the information that is essential to the badminton activity recognition task. It is demonstrated by the experimental results on an actual badminton single sensor dataset that our proposed framework has obtained a badminton activity recognition (37 classes) accuracy of 97.83 % , which outperforms the existing methods, and also has the advantages of lower training time and faster convergence. [ABSTRACT FROM AUTHOR]

A growing number of smart wearable biosensors are operating in the medical IoT environment and those that capture physiological signals have received special attention. Electrocardiogram (ECG) is one of the physiological signals used in the cardiovascular and medical fields that has encouraged researchers to discover new non-invasive methods to diagnose hyperglycemia as a personal variable. Over the years, researchers have proposed different techniques to detect hyperglycemia using ECG. In this paper, we propose a novel deep learning architecture that can identify hyperglycemia using heartbeats from ECG signals. In addition, we introduce a new fiducial feature extraction technique that improves the performance of the deep learning classifier. We evaluate the proposed method with ECG data from 1119 different subjects to assess the efficiency of hyperglycemia detection of the proposed work. The result indicates that the proposed algorithm is effective in detecting hyperglycemia with a 94.53% area under the curve (AUC), 87.57% sensitivity, and 85.04% specificity. That performance represents an relative improvement of 53% versus the best model found in the literature. The high sensitivity and specificity achieved by the 10-layer deep neural network proposed in this work provide an excellent indication that ECG possesses intrinsic information that can indicate the level of blood glucose concentration. [ABSTRACT FROM AUTHOR]

For the purpose of maintaining and prolonging the service life of civil constructions, structural damage must be closely monitored. Monitoring the incidence, formation, and spread of damage is crucial to ensure a structure's ongoing performance. This research proposes a unique approach for multiclass damage detection using acceleration responses based on synchrosqueezing transform (SST) together with deep learning algorithms. In particular, our pipeline is able to classify correctly the time series representing the responses of accelerometers placed on a bridge, which are classified with respect to different types of damage scenarios applied to the bridge. Using benchmark data from the Z24 bridge for multiclass classification for different damage situations, the suggested method is validated. This dataset includes labeled accelerometer measurements from a real-world bridge that has been gradually damaged by various conditions. The findings demonstrate that the suggested approach is successful in exploiting pre-trained 2D convolutional neural networks, obtaining a high classification accuracy that can be further boosted by the application of simple voting methods. [ABSTRACT FROM AUTHOR]

This study presents the results of acoustic emission (AE) measurements and characterization in the loading of biocomposites at room and low temperatures that can be observed in the aviation industry. The fiber optic sensors (FOS) that can outperform electrical sensors in challenging operational environments were used. Standard features were extracted from AE measurements, and a convolutional autoencoder (CAE) was applied to extract deep features from AE signals. Different machine learning methods including discriminant analysis (DA), neural networks (NN), and extreme learning machines (ELM) were used for the construction of classifiers. The analysis is focused on the classification of extracted AE features to classify the source material, to evaluate the predictive importance of extracted features, and to evaluate the ability of used FOS for the evaluation of material behavior under challenging low-temperature environments. The results show the robustness of different CAE configurations for deep feature extraction. The combination of classic and deep features always significantly improves classification accuracy. The best classification accuracy (80.9%) was achieved with a neural network model and generally, more complex nonlinear models (NN, ELM) outperform simple models (DA). In all the considered models, the selected combined features always contain both classic and deep features. [ABSTRACT FROM AUTHOR]