Search

Videos are considered as most trustworthy means of communication in the present digital era. The advancement in multimedia technology has made video content sharing and manipulation very easy. Hence, the video authenticity is a challenging task for the research community. Video forensics refer to uncovering the forgery traces. The detection of spatiotemporal object‐removal forgery in surveillance videos is crucial for judicial forensics, as the presence of objects in the video has significant information as legal evidence. The author proposes a passive max–median averaging motion residual algorithm for revealing the forgery traces, successfully giving visible object‐removal traces followed by a deep learning approach, YOLO‐V8, for forged region localization. YOLO‐V8 is the latest deep learning model, which has a wide scope for real‐time application. The proposed method utilizes YOLO‐V8 for object‐removal forgery in surveillance videos. The network is trained on the SYSU‐OBJFORG dataset for object‐removal forged region localization in videos. The fine‐tuned YOLO‐V8 successfully classifies and localizes the object‐removal tampered region with an F1‐score of 0.99 and a precision of 0.99. The observed high confidence score of the bounding box around the forged region makes the model reliable. This fine‐tuned YOLO‐V8 would be a better choice in real‐time applications as it solves the complex object‐based forgery detection in videos. The performance of the proposed system is far better than the existing deep learning approach. [ABSTRACT FROM AUTHOR]

The goal of achieving autonomous navigation for agricultural robots poses significant challenges, mostly arising from the substantial natural variations in crop row images as a result of weather conditions and the growth stages of crops. The processing of the detection algorithm also must be significantly low for real-time applications. In order to address the aforementioned requirements, we propose a crop row detection algorithm that has the following features: Firstly, a projective transformation is applied to transform the camera view and a color-based segmentation is employed to distinguish crop and weed from the background. Secondly, a clustering algorithm is used to differentiate between the crop and weed pixels. Lastly, a robust line-fitting approach is implemented to detect crop rows. The proposed algorithm is evaluated throughout a diverse range of scenarios, and its efficacy is assessed in comparison to four distinct existing solutions. The algorithm achieves an overall intersection over union (IOU) of 0.73 and exhibits robustness in challenging scenarios with high weed growth. The experiments conducted on real-time video featuring challenging scenarios show that our proposed algorithm exhibits a detection accuracy of over 90% and is a viable option for real-time implementation. With the high accuracy and low inference time, the proposed methodology offers a viable solution for autonomous navigation of agricultural robots in a crop field without damaging the crop and thus can serve as a foundation for future research.

In recent years, the Internet of Things (IoT) has gradually developed applications such as collecting sensory data and building intelligent services, which has led to an explosion in mobile data traffic. Meanwhile, with the rapid development of artificial intelligence, semantic communication has attracted great attention as a new communication paradigm. However, for IoT devices, however, processing image information efficiently in real time is an essential task for the rapid transmission of semantic information. With the increase of model parameters in deep learning methods, the model inference time in sensor devices continues to increase. In contrast, the Pulse Coupled Neural Network (PCNN) has fewer parameters, making it more suitable for processing real-time scene tasks such as image segmentation, which lays the foundation for real-time, effective, and accurate image transmission. However, the parameters of PCNN are determined by trial and error, which limits its application. To overcome this limitation, an Improved Pulse Coupled Neural Networks (IPCNN) model is proposed in this work. The IPCNN constructs the connection between the static properties of the input image and the dynamic properties of the neurons, and all its parameters are set adaptively, which avoids the inconvenience of manual setting in traditional methods and improves the adaptability of parameters to different types of images. Experimental segmentation results demonstrate the validity and efficiency of the proposed self-adaptive parameter setting method of IPCNN on the gray images and natural images from the Matlab and Berkeley Segmentation Datasets. The IPCNN method achieves a better segmentation result without training, providing a new solution for the real-time transmission of image semantic information.

Abstract Crop residue-based biochar (CRB) has shown great potential for removing trace elements (TEs) from aquatic matrices. Despite the increasing interest in this area, no review has focused specifically on the efficacy of CRB for TEs removal in aquatic environments. This comprehensive review examines the global TEs water contamination status with an emphasis on their sources, compositional metrics for crop residue feedstock (proximate, ultimate, and lignocellulosic properties), and the potential use of CRB for TEs removal in aquatic media. It also evaluates the factors that affect the ability of CRB to remove TEs, such as feedstock type, production conditions, water pH, background electrolytes, water temperature, CRB/water ratio, and underlying pollutant sorption mechanisms. This review also discusses the practical applications of CRB in real water samples and engineering considerations for designing CRB with improved physicochemical properties, treatment efficiencies, and regeneration abilities. Additionally, the cost–benefit and economic assessment of CRB, challenges, and future research directions related to CRB are highlighted to promote research on this sustainable source of biochar. By elucidating the prospects of CRB as an adsorbent, this review emphasizes the need for continued research on its practical implications for environmentally relevant pollutant concentrations. Graphical Abstract

Crop residue-based biochar (CRB) has shown great potential for removing trace elements (TEs) from aquatic matrices. Despite the increasing interest in this area, no review has focused specifically on the efficacy of CRB for TEs removal in aquatic environments. This comprehensive review examines the global TEs water contamination status with an emphasis on their sources, compositional metrics for crop residue feedstock (proximate, ultimate, and lignocellulosic properties), and the potential use of CRB for TEs removal in aquatic media. It also evaluates the factors that affect the ability of CRB to remove TEs, such as feedstock type, production conditions, water pH, background electrolytes, water temperature, CRB/water ratio, and underlying pollutant sorption mechanisms. This review also discusses the practical applications of CRB in real water samples and engineering considerations for designing CRB with improved physicochemical properties, treatment efficiencies, and regeneration abilities. Additionally, the cost–benefit and economic assessment of CRB, challenges, and future research directions related to CRB are highlighted to promote research on this sustainable source of biochar. By elucidating the prospects of CRB as an adsorbent, this review emphasizes the need for continued research on its practical implications for environmentally relevant pollutant concentrations. Highlights: Crop residue-based biochar (CRB) emerged as a top-notch adsorbent for TEs in aquatic systems. The CRB adsorption potential depended on the experimental conditions. The application of CRB to real water samples highlights its potential on a larger scale. Engineer-designed CRB exhibits high treatment efficiencies and regeneration abilities. [ABSTRACT FROM AUTHOR]

In this paper, we address the challenge of ensuring stability in bipedal walking robots and exoskeletons. We explore the feasibility of real-time implementation for the Predicted Step Viability algorithm (PSV), a complex multi-step optimization criterion for planning future steps in bipedal gait. To overcome the high computational cost of the PSV algorithm, we performed an analysis using 11 classification algorithms and a stacking strategy to predict if a step will be stable or not. We generated three datasets of increasing complexity through PSV simulations to evaluate the classification performance. Among the classifiers, k Nearest Neighbors, Support Vector Machine with Radial Basis Function Kernel, Decision Tree, and Random Forest exhibited superior performance. Multi-Layer Perceptron also consistently performed well, while linear-based algorithms showed lower performance. Importantly, the use of stacking did not significantly improve performance. Our results suggest that the feature vector applied with this approach is applicable across various robotic models and datasets, provided that training data is balanced and sufficient points are used. Notably, by leveraging classifiers, we achieved rapid computation of results in less than 1 ms, with minimal computational cost. [ABSTRACT FROM AUTHOR]

Object detection has become a popular tool of deep learning in the era of digital manufacturing. In this study, the most powerful and efficient object detection algorithm, i.e., You Only Look Once (YOLO) algorithm, was used to detect anomalies in deposited beads of wire-arc additive manufacturing (WAAM) using melt pool images. This study used the latest version of YOLO algorithm to train and validate the custom image dataset of the melt pool obtained by conducting experiments using a robotic-controlled WAAM. The mean average precision (mAP) for the "Regular bead" class and the "Irregular bead" class reached 99% at an Intersection over Union (IoU) threshold of 0.5, for both training and validation. When the model was tested for new or unseen datasets by conducting four new experimental trials, the mAP value for the "Regular bead" class reached 98.47% and for the "Irregular bead" class reached 96.68% at an average processing time of 0.014 s/frame. The object detection algorithm YOLO has shown an excellent processing time of 15 ms per frame, which shows its potential for real-time application in the manufacturing industry. [ABSTRACT FROM AUTHOR]

The given text discusses a road segmentation method called MSAN that uses multi-modal scale-aware attentions to segment roads in real-time for autonomous driving. The method combines RGB and depth information to improve prediction results and outperforms single modality models. The MSAN architecture consists of two attention modules, one for feature fusion across modalities and scales and another for scale-aware fusion. The experiments show that MSAN achieves high performance and efficiency on multiple datasets. However, the method relies heavily on labeled data, which limits its adaptability to different scenarios. Future research aims to reduce data dependency through few-shot learning strategies. [Extracted from the article]

Violent attacks have been one of the hot issues in recent years. In the presence of closed-circuit televisions (CCTVs) in smart cities, there is an emerging challenge in apprehending criminals, leading to a need for innovative solutions. In this paper, the propose a model aimed at enhancing real-time emergency response capabilities and swiftly identifying criminals. This initiative aims to foster a safer environment and better manage criminal activity within smart cities. The proposed architecture combines an image-to-image stable diffusion model with violence detection and pose estimation approaches. The diffusion model generates synthetic data while the object detection approach uses YOLO v7 to identify violent objects like baseball bats, knives, and pistols, complemented by MediaPipe for action detection. Further, a long short-term memory (LSTM) network classifies the action attacks involving violent objects. Subsequently, an ensemble consisting of an edge device and the entire proposed model is deployed onto the edge device for real-time data testing using a dash camera. Thus, this study can handle violent attacks and send alerts in emergencies. As a result, our proposed YOLO model achieves a mean average precision (MAP) of 89.5% for violent attack detection, and the LSTM classifier model achieves an accuracy of 88.33% for violent action classification. The results highlight the model's enhanced capability to accurately detect violent objects, particularly in effectively identifying violence through the implemented artificial intelligence system. [ABSTRACT FROM AUTHOR]

Introduction: Date palm species classification is important for various agricultural and economic purposes, but it is challenging to perform based on images of date palms alone. Existing methods rely on fruit characteristics, which may not be always visible or present. In this study, we introduce a new dataset and a new model for image-based date palm species classification. Methods: Our dataset consists of 2358 images of four common and valuable date palm species (Barhi, Sukkari, Ikhlas, and Saqi), which we collected ourselves. We also applied data augmentation techniques to increase the size and diversity of our dataset. Our model, called DPXception (Date Palm Xception), is a lightweight and efficient CNN architecture that we trained and fine-tuned on our dataset. Unlike the original Xception model, our DPXception model utilizes only the first 100 layers of the Xception model for feature extraction (Adapted Xception), making it more lightweight and efficient. We also applied normalization prior to adapted Xception and reduced the model dimensionality by adding an extra global average pooling layer after feature extraction by adapted Xception. Results and discussion: We compared the performance of our model with seven well-known models: Xception, ResNet50, ResNet50V2, InceptionV3, DenseNet201, EfficientNetB4, and EfficientNetV2-S. Our model achieved the highest accuracy (92.9%) and F1-score (93%) among the models, as well as the lowest inference time (0.0513 seconds). We also developed an Android smartphone application that uses our model to classify date palm species from images captured by the smartphone's camera in real time. To the best of our knowledge, this is the first work to provide a public dataset of date palm images and to demonstrate a robust and practical image-based date palm species classification method. This work will open new research directions for more advanced date palm analysis tasks such as gender classification and age estimation. [ABSTRACT FROM AUTHOR]

Stress is the mood of pressure and tension that a person feels. Usually, when the pressure on an individual decrease, the body begins to stabilize the state and calm down. Hence, stress detection in real-time is a critical duty in medical systems. However, acquiring physiological data requires additional equipment and is difficult for users to carry with them at all times. Depending on this problem, it is possible to detect stress through behavioral data. Smartphones are devices that provide various behavioral data that people use constantly throughout the day. In this study, a real-time stress detection system based on soft keyboard typing behaviors was developed with the data obtained from linear acceleration, gravity, gyroscope sensors, and a touchscreen panel of the smartphone. 172 attributes were extracted from the raw sensor data. However, such a high number of dimensions could negatively affect the performance of machine learning algorithms. To address this problem, the number of features was reduced by various techniques such as filter-based methods and standard binary-code chromosome Genetic Algorithm as a contribution to this study. Then, writing behaviors were classified with the commonly used machine learning methods namely, C4.5, kNN, and Bayesian Networks. As a result of the experiments, the best classification was obtained from the kNN method using the features selected by the Genetic Algorithm with a classification accuracy of 89.61% and F-Measure of 0.9052. Another contribution of this study is that a mobile service and a relaxation application were developed for stress detection and to reduce stress levels using the selected feature vector. [ABSTRACT FROM AUTHOR]

IntroductionDate palm species classification is important for various agricultural and economic purposes, but it is challenging to perform based on images of date palms alone. Existing methods rely on fruit characteristics, which may not be always visible or present. In this study, we introduce a new dataset and a new model for image-based date palm species classification.MethodsOur dataset consists of 2358 images of four common and valuable date palm species (Barhi, Sukkari, Ikhlas, and Saqi), which we collected ourselves. We also applied data augmentation techniques to increase the size and diversity of our dataset. Our model, called DPXception (Date Palm Xception), is a lightweight and efficient CNN architecture that we trained and fine-tuned on our dataset. Unlike the original Xception model, our DPXception model utilizes only the first 100 layers of the Xception model for feature extraction (Adapted Xception), making it more lightweight and efficient. We also applied normalization prior to adapted Xception and reduced the model dimensionality by adding an extra global average pooling layer after feature extraction by adapted Xception.Results and discussionWe compared the performance of our model with seven well-known models: Xception, ResNet50, ResNet50V2, InceptionV3, DenseNet201, EfficientNetB4, and EfficientNetV2-S. Our model achieved the highest accuracy (92.9%) and F1-score (93%) among the models, as well as the lowest inference time (0.0513 seconds). We also developed an Android smartphone application that uses our model to classify date palm species from images captured by the smartphone’s camera in real time. To the best of our knowledge, this is the first work to provide a public dataset of date palm images and to demonstrate a robust and practical image-based date palm species classification method. This work will open new research directions for more advanced date palm analysis tasks such as gender classification and age estimation.

The heavy use of machine learning algorithms in safety-critical systems poses serious questions related to safety, security, and predictability issues, requiring novel architectural approaches to guarantee such properties. This paper presents an architecture solution that leverages heterogeneous platforms and virtualization technologies to support AI-powered applications consisting of modules with mixed criticalities and safety requirements. The hypervisor exploits the security features of the Xilinx ZCU104 MPSoCs to create two isolated execution environments: a high performance domain running deep learning algorithms under the Linux operating system and a safety-critical domain running control and monitoring functions under the freeRTOS real-time operating system. The proposed approach is validated by a use case consisting of an unmanned aerial vehicle capable of tracking moving targets using a deep neural network accelerated on the FGPA available on the platform. [ABSTRACT FROM AUTHOR]

In this paper, we address the challenge of ensuring stability in bipedal walking robots and exoskeletons. We explore the feasibility of real-time implementation for the Predicted Step Viability algorithm (PSV), a complex multi-step optimization criterion for planning future steps in bipedal gait. To overcome the high computational cost of the PSV algorithm, we performed an analysis using 11 classification algorithms and a stacking strategy to predict if a step will be stable or not. We generated three datasets of increasing complexity through PSV simulations to evaluate the classification performance. Among the classifiers, k Nearest Neighbors, Support Vector Machine with Radial Basis Function Kernel, Decision Tree, and Random Forest exhibited superior performance. Multi-Layer Perceptron also consistently performed well, while linear-based algorithms showed lower performance. Importantly, the use of stacking did not significantly improve performance. Our results suggest that the feature vector applied with this approach is applicable across various robotic models and datasets, provided that training data is balanced and sufficient points are used. Notably, by leveraging classifiers, we achieved rapid computation of results in less than 1 ms, with minimal computational cost.

The Internet of Things (IoT) facilitates the deployment of sensing devices for data acquisition and transmission across networks and systems, thus enabling innovative real-time applications. Securing these platforms remains paramount due to the high value placed on data privacy. This study proposes a novel real-time authentication method, fundamentally grounded in a One-Time Pad (OTP) concept. Data are encrypted using dynamic encryption, the parameters of which are determined by data from randomly distributed sensors. Encryption keys are dynamically transmitted, allowing for on-the-fly key generation and exchange. This enhances the security and privacy of user data. A lightweight technique for key creation, exchange, and authentication is presented for data collection from sensors used in real-time applications. This protocol ensures data privacy and security throughout the data collection process. The proposed protocol's efficacy is demonstrated through a discussion of how it meets these criteria, and an example illustrating its operation at each tier in the event of desynchronization or an incident. [ABSTRACT FROM AUTHOR]

Currently, remote sensing equipments are evolving toward intelligence and integration, incorporating edge computing techniques to enable real-time responses. One of the key challenges in enhancing downstream decision-making capabilities is the preprocessing step of image dehazing. Existing dehazing methods usually suffer from steep computational costs with densely connected residual modules, as well as difficulties in maintaining visual quality. To tackle these problems, we designed a lightweight atmosphere scattering model based network structure to extract, fuse, and weight multiscale features. Our proposed LFD-Net demonstrates strong interpretability by exploiting the gated fusion module and attention mechanism to realize feature interactions between multilevel representations. The experimental results of LFD-Net on SOTS dataset reach an average frequency per second of 54.41, approximately eight times faster than seven most popular methods with equivalent metrics. After image dehazing by LFD-Net, the performance of object detection is significantly improved. The mean average precision when IoU = 0.5 (mAP@0.5) based on YOLOv5 is improved by 4.73% on DAIR-V2X dataset, which verified the practicability and adaptability of LFD-Net for real-time vision tasks.

This work assesses the effectiveness of heterogeneous computing based on a CUDA implementation for real-time ego-lane detection using a typical low-cost embedded computer. We propose and evaluate a CUDA-optimized algorithm using a heterogeneous approach based on the extraction of features from an aerial perspective image. The method incorporates well-known algorithms optimized to achieve a very efficient solution with high detection rates and combines techniques to enhance markings and remove noise. The CUDA-based solution is compared to an OpenCV library and to a serial CPU implementation. Practical experiments using TuSimple's image datasets were conducted in an NVIDIA's Jetson Nano embedded computer. The algorithm detects up to 97.9% of the ego lanes with an accuracy of 99.0% in the best-evaluated scenario. Furthermore, the CUDA-optimized method performs at rates greater than 300 fps in the Jetson Nano embedded system, speeding up 25 and 140 times the OpenCV and CPU implementations at the same platform, respectively. These results show that more complex algorithms and solutions can be employed for better detection rates while maintaining real-time requirements in a typical low-power embedded computer using a CUDA implementation.

Violent attacks have been one of the hot issues in recent years. In the presence of closed-circuit televisions (CCTVs) in smart cities, there is an emerging challenge in apprehending criminals, leading to a need for innovative solutions. In this paper, the propose a model aimed at enhancing real-time emergency response capabilities and swiftly identifying criminals. This initiative aims to foster a safer environment and better manage criminal activity within smart cities. The proposed architecture combines an image-to-image stable diffusion model with violence detection and pose estimation approaches. The diffusion model generates synthetic data while the object detection approach uses YOLO v7 to identify violent objects like baseball bats, knives, and pistols, complemented by MediaPipe for action detection. Further, a long short-term memory (LSTM) network classifies the action attacks involving violent objects. Subsequently, an ensemble consisting of an edge device and the entire proposed model is deployed onto the edge device for real-time data testing using a dash camera. Thus, this study can handle violent attacks and send alerts in emergencies. As a result, our proposed YOLO model achieves a mean average precision (MAP) of 89.5% for violent attack detection, and the LSTM classifier model achieves an accuracy of 88.33% for violent action classification. The results highlight the model’s enhanced capability to accurately detect violent objects, particularly in effectively identifying violence through the implemented artificial intelligence system.

The processing of underwater images can vastly ease the difficulty of underwater robots' tasks and promote ocean exploration development. This paper proposes a fast and efficient underwater image enhancement model based on conditional GAN with good generalization ability using aggregation strategies and concatenate operations to take full advantage of the limited hierarchical features. A sequential network can avoid frequently visiting additional nodes, which is beneficial for speeding up inference and reducing memory consumption. Through the structural re-parameterization approach, we design a dual residual block (DRB) and accordingly construct a hierarchical attention encoder (HAE), which can extract sufficient feature and texture information from different levels of an image, and with 11.52% promotion in GFLOPs. Extensive experiments were carried out on real and artificially synthesized benchmark underwater image datasets, and qualitative and quantitative comparisons with state-of-the-art methods were implemented. The results show that our model produces better images, and has good generalization ability and real-time performance, which is more conducive to the practical application of underwater robot tasks. [ABSTRACT FROM AUTHOR]

This article presents a computer vision-based approach for monitoring water flow at outlet points of a tile drain system. The approach relies only on video capture of events at outlet points, thus a camera can be installed remotely and without contact with water. The algorithm detects, identifies, and tracks flows by motion, shape, and color features and measures flow rate based on a proposed model and two provided dimensions. The software was tested in a laboratory environment with three different target flow rate conditions: 0.312, 0.946, and 1.58 L/s (5, 15, and 25 gal/min). Flow rates reported by the computer vision approach are within 12% of the ground-truth flow rate baseline. The results of this work show that computer vision can be used as a reliable method for monitoring outlet flows from free-standing outlet structures under laboratory conditions. This work opens the possibility of applying computer vision techniques in tile drain monitoring from outlet points with mobile video recording devices in the field. [ABSTRACT FROM AUTHOR]

With the development of mobile payment, the Internet of Things (IoT) and artificial intelligence (AI), smart vending machines, as a kind of unmanned retail, are moving towards a new future. However, the scarcity of data in vending machine scenarios is not conducive to the development of its unmanned services. This paper focuses on using machine learning on small data to detect the placement of the spiral rack indicated by the end of the spiral rack, which is the most crucial factor in causing a product potentially to get stuck in vending machines during the dispensation. To this end, we propose a k-means clustering-based method for splitting small data that is unevenly distributed both in number and in features due to real-world constraints and design a remarkably lightweight convolutional neural network (CNN) as a classifier model for the benefit of real-time application. Our proposal of data splitting along with the CNN is visually interpreted to be effective in that the trained model is robust enough to be unaffected by changes in products and reaches an accuracy of 100 % . We also design a single-board computer-based handheld device and implement the trained model to demonstrate the feasibility of a real-time application. [ABSTRACT FROM AUTHOR]

A traffic sign recognition system is crucial for safely operating an autonomous driving car and efficiently managing road facilities. Recent studies on traffic sign recognition tasks show significant advances in terms of accuracy on several benchmarks. However, they lack performance evaluation in driving cars in diverse road environments. In this study, we develop a traffic sign recognition framework for a vehicle to evaluate and compare deep learning-based object detection and tracking models for practical validation. We collect a large-scale highway image set using a camera-installed vehicle for training models, and evaluate the model inference during a test drive in terms of accuracy and processing time. In addition, we propose a novel categorization method for urban road scenes with possible scenarios. The experimental results show that the YOLOv5 detector and strongSORT tracking model result in better performance than other models in terms of accuracy and processing time. Furthermore, we provide an extensive discussion on possible obstacles in traffic sign recognition tasks to facilitate future research through numerous experiments for each road condition. [ABSTRACT FROM AUTHOR]

During the maneuvering of most unmanned aerial vehicles (UAVs), the GPS is one of the sensors used for navigation. However, this kind of sensor cannot handle indoor navigation applications well. Using a camera might be the answer to performing indoor navigation using its coordinate system. In this study, we considered indoor navigation applications using the ZED2 stereo camera for the quadcopter. To use the ZED 2 camera as a navigation sensor, we first transformed its coordinates into the North, East, down (NED) system to enable the drone to understand its position and maintain stability in a particular position. The experiment was performed using a real-time application to confirm the feasibility of this approach for indoor localization. In the real-time application, we commanded the quadcopter to follow triangular and rectangular paths. The results indicated that the quadcopter was able to follow the paths and maintain its stability in specific coordinate positions. [ABSTRACT FROM AUTHOR]

Emotion recognition is a very challenging research field due to its complexity, as individual differences in cognitive–emotional cues involve a wide variety of ways, including language, expressions, and speech. If we use video as the input, we can acquire a plethora of data for analyzing human emotions. In this research, we use features derived from separately pretrained self-supervised learning models to combine text, audio (speech), and visual data modalities. The fusion of features and representation is the biggest challenge in multimodal emotion classification research. Because of the large dimensionality of self-supervised learning characteristics, we present a unique transformer and attention-based fusion method for incorporating multimodal self-supervised learning features that achieved an accuracy of 86.40% for multimodal emotion classification. [ABSTRACT FROM AUTHOR]

It is important for the autonomous system to understand environmental information. For the autonomous system, it is desirable to have a strong generalization ability to deal with different complex environmental information, as well as have high accuracy and quick inference speed. Network ensemble architecture is a good choice to improve network performance. However, it is unsuitable for real-time applications on the autonomous system. To tackle this problem, a new neural network ensemble named partial-shared ensemble network (PSENet) is presented. PSENet changes network ensemble architecture from parallel architecture to scatter architecture and merges multiple component networks together to accelerate the inference speed. To make component networks independent of each other, a training method is designed to train the network ensemble architecture. Experiments on Camvid and CIFAR-10 reveal that PSENet achieves quick inference speed while maintaining the ability of ensemble learning. In the real world, PSENet is deployed on the unmanned system and deals with vision tasks such as semantic segmentation and environmental prediction in different fields. [ABSTRACT FROM AUTHOR]

The process systems domain is undergoing the fourth industrial revolution, which is helping industries digitize and optimize their production techniques. Concurrently, the field of data-based modeling has been expanding, leading to the proposal of many fault detection models. However, the rapid expansion has created gaps in the field. For instance, Smart Manufacturing (SM) methodologies have yet to be incorporated into undergraduate chemical engineering education. Additionally, only a few developed fault detection models have been deployed for real-time usage and practical applications. This study takes a crucial step toward bridging the two mentioned gaps by enabling undergraduate students to learn SM techniques and developing a safe and controlled academic environment for deploying fault detection models. The demonstration is implemented on a shell and tube heat exchanger, taught in a senior year laboratory course, using the Smart Manufacturing Innovation Platform (SMIP). The implementation provides an easily customizable pipeline for SM applications involving human-in-the-loop decision-making on a real-life hardware system. Actual data from heat exchanger equipment is used to train and compare the performances of several state-of-the-art fault detection models, including fully connected, convolutional, and recurrent neural networks. Current work also presents tutorials on deploying models for practical real-time applications using the SMIP. The overall architecture is a plug-and-play package that will motivate students to learn about SM and catalyze their interest in developing and deploying fault detection models using real-world data. • Architecture to motivate undergraduate students to learn aspects of Smart Manufacturing. • The demonstration is implemented on a shell and tube heat exchanger using the Smart Manufacturing Innovation Platform (SMIP). • Fault detection models are deployed to test real-time applications, including fully connected, convolutional, and recurrent neural networks. • Students get first-hand experience of Human-in-the-Loop decision making similar to a control room. [ABSTRACT FROM AUTHOR]

The efficient utilization of energy is a crucial consideration for plug-in hybrid electric buses (PHEB). However, achieving the optimal energy management strategy (EMS) for PHEB necessitates the harmonious optimization of both driving conditions and battery status. Consequently, an innovative EMS that integrates the data-driven prediction of the driving cycles and battery health status. Explicitly, the bi-directional long short-term memory algorithm is employed to reconstruct and train the health status data of power batteries, enabling the accurate estimation of their State of Health (SOH) values. Moreover, a database of historical driving behavior is meticulously collected through a real-world PHEB platform, and the driving cycles for a certain while are predicted with the clustering algorithm. Then, this invaluable information on the battery status and driving cycles is seamlessly introduced into an adaptive-network-based fuzzy inference system (ANFIS) to regulate of the energy conversion factor in the following equivalent consumption minimization strategy (ECMS). Besides, the offline global optimization of equivalent fuel consumption with the Pontryagin's minimum principle (PMP) is preliminarily conducted to generate the pre-established energy conversion factors, facilitating convenient access to the engine and motor's reasonable power distribution range. Finally, simulations and experiments have validated the effectiveness of the proposed EMS, demonstrating a 23.57 % reduction in battery SOC depletion and a 10.53 % decrease in fuel consumption compared to traditional ECMS methods. More importantly, the proposed method has been validated for its executability in embedded devices through real-world bench experiments. This research holds significant reference value for energy utilization and practical implementation in the PHEB, thereby contributing to the advancement of knowledge in this field. • A real-time energy management strategy for a PHEV is proposed by incorporating the offline global PMP optimization and online ECMS execution. • The equivalent energy conversion factor for the EMS fully considers the battery health status and driving cycle recognition. • The driving cycles are well recognized with the Bi-LSTM algorithm which has enhanced the accurate future driving information prediction. • The real-world experimental test is implemented for the practical application. [ABSTRACT FROM AUTHOR]

Conductive hydrogel represents an important development direction in modern/future electronic industry. However, successfully achieving high performance hydrogels which are capable of performing a real-world task still remain extremely challenging because most conventional approaches of constructing hydrogels are difficult to make hydrogels achieve the integration of multiple performances like robust mechanical properties, excellent conductivity and stable electrochemical performance. Here, we develop a simple and universal two-layer hydrogen bond networks (TLHBNs) strategy by which seven types of PVA based hydrogels with simultaneously robust mechanical properties (stress, strain, toughness, elasticity, fatigue resistance, notch insensitivity, anti-damage), excellent conductivity (7.2− 12.2 S m−1, 11.7− 19.8 S m−1, 10.5− 21.6 S m−1), reversible/stable network structure and constantly sensitive signal output capability are achieved via sophisticatedly layer-by-layer introducing TLHBNs to PVA hydrogel. Benefiting from the extraordinary performances, one hydrogel is so strong that it can load as heavy as 80 kg of a boy volunteer. Meanwhile, the hydrogels, assembled into flexible sensors, satisfactorily fulfills practical tasks (human joint motions and bionic skin) upon handling real-world application. This strategy, enabling extraordinary performances while using a generic and facile designing concept, opens up a new insight into the design of high-performance PVA hydrogel and beyond. • Engineer PVA/polyethyleneimine/sodium carboxylate hydrogel sensors. • Robust mechanical properties, excellent conductivity and practical application. • Stable signal output, high sensitivity and insensitivity against breakage. • Two-layer hydrogen bond networks responsible to high performances. • Human motion monitoring, bionic skin applications and loading 80 kg a boy volunteer. [ABSTRACT FROM AUTHOR]

Physics-based lithium-ion battery model is rarely used in real-time applications due to its complexity, restricting the development of next-generation refined battery management systems. Fidelity and computation efficiency are still challenging to coordinate for existing Model Order Reduction (MOR) methods, especially in solving either electrochemical reaction rate distribution or coupled multi-physics fields. This paper proposes a novel reduced-order electrochemical model called the Triple Region Transmission-Line Model (TRTLM), aiming at a win-win situation for both fidelity and real-time applicability. The TRTLM stands on a simple triple region transmission-line circuit topology, concisely depicting the inhomogeneous distribution of the electric field. Current and potential distribution are obtained considering the effect of the reaction distribution and concentration field. Two different linearized MIMO state-space models are then established to solve liquid and solid phase concentration fields continuously based on Padé Approximation techniques, achieving less distortion with limited discretization of the electric field. Convenient parameterized governing equations and extreme computation efficiency are both realized. Coupling of electric and concentration field is implemented by solving them alternatively and iteratively. A comparison study of the TRTLM with a Full-Order Model (FOM) and other representative MOR is conducted. The result shows less than 10 mV maximum distortion is reached by the TRTLM, even in extreme conditions, mainly through increasing the fidelity of reaction overpotential and liquid phase polarization. Moreover, hardware validation of the TRTLM is executed under different computation periods (down to 0.01s) based on the dSPACE platform, intuitively proving the real-time applicability of the model. [Display omitted] • A reduced order electrochemical model is proposed with high fidelity and efficiency. • A new topology is constructed to characterize electrical inhomogeneity concisely. • Compared with other reduced order methods in fidelity and computation speed. • A hardware validation is executed under multiple computation periods. [ABSTRACT FROM AUTHOR]

Software defined network (SDN) is a new network architecture proposed by the clean slate research group of Stanford University.The significant feature of this architecture is to decouple the functions of control plane and forwarding plane,and to flexibly forward the network traffic.Based on this,Internet service providers have deployed SDN technology in their backbone network to maximize the utilization of network resources.However,due to the limitation of economic cost and technical conditions,the backbone network of internet service providers must be in the hybrid SDN network for a long time.The studies have shown that single network node failure is inevitable and occurs frequently.Therefore,it is a key scientific problem to study the routing protection me-thod for single network component failure in hybrid SDN networks.In this paper,the route protection method for single network component failure in hybrid SDN network is described,and then two heuristic methods are used to solve the problem.Finally,the proposed heuristic algorithms are tested in real and simulated topologies.The experimental results show that in the traditional backbone network,only a part of the traditional devices need to be upgraded to SDN devices,and the algorithms proposed in this paper can deal with all possible single network node failure cases in the network.

The control of robotic manipulators has become increasingly difficult over recent years due to their high accuracy, performance, speed, and reliability in a variety of applications, such as industry, medicine, research, etc. These serial manipulator systems are extremely complex because their dynamic models include perturbations, parametric variations, coupled nonlinear dynamics, and non-modular dynamics, all of which require robust control for trajectory tracking. This paper compares two control techniques: computational torque control (CTC) and sliding mode control (SMC). In this study, the latter was used for a physical robotic arm with six degrees of freedom (DOF) and online experiments were conducted, which have received little attention in the literature. As a result, the contribution of this work was based on the real-time application of this controller via a self-developing interface. The great resilience of sliding mode controllers to disturbances was also demonstrated in this study. [ABSTRACT FROM AUTHOR]

Facial expression detection is a method to predict human facial emotions. This work is a trending research topic that can be implemented for human-robot interaction. More recently, deep convolutional neural network provides a robust extractor features but tends to be slow in real-time implementations and often requires a large memory and graphics processing units for fast execution. In this article, an efficient CPU-based facial expression detector is proposed using a sequential attention network to improve the baseline performance. The proposed attention network consists of three modules, global representation to capture the global features, channel representation, and dimension representation, which are focused on the channel and using spatial attention to discriminate local features. The efficient partial transfer module is also presented as a light backbone to extract facial features from an image. The entire module is trained and tested on several benchmarks to classify seven facial expressions. As a result, the proposed model reaches an accuracy of 98.18%, 98.75%, 95.63%, and 74.17% on CK+, JAFFE, KDEF, and FER-2013, respectively. It achieves competitive performance when compared to state-of-the-art methods. Lastly, it is integrated with a face detector and runs in real-time without a constraint at 69 frames per second on a CPU. [ABSTRACT FROM AUTHOR]