Franklin Open最新文献

英文中文

Enhanced hybrid facial emotion detection & classification

Franklin Open

Pub Date : 2024-12-16 DOI: 10.1016/j.fraope.2024.100200

Naim Ajlouni , Adem Özyavaş , Firas Ajlouni , Faruk Takaoğlu , Mustafa Takaoğlu

Emotion Recognition (ER) presents a significant challenge in pattern recognition and is crucial for various Artificial Intelligence (AI) applications, from monitoring children with autism to enhancing video games and human-computer interactions. Facial features are essential for discerning human emotions, motivating this study to improve Facial Emotion detection accuracy, vital for real-time applications. This research explores the use of both full facial features and facial landmarks, suggesting that landmarks offer clearer emotional cues. A Convolutional Neural Network (CNN) is employed for feature extraction, optimized using techniques like Stochastic Gradient Descent (SGD). This study combines upper facial landmarks with full facial features to enhance detection accuracy. An adaptive Genetic Algorithm (GA) optimizes the CNN structure. The dual-input neural network architecture processes a full 48 × 48 facial image through a CNN and upper facial features through another input, which is then concatenated and fed into dense layers with L2 regularization. The model is trained and evaluated using various optimizers to find the best configuration, with performance metrics plotted and compared. This method allows detailed pixel information and focused landmark features to improve emotion detection accuracy. Datasets such as FER2013 and Extended Cohn-Kanade (CK+) are used for training and testing. Results show significant accuracy improvement with the proposed method, further enhanced by a highly optimized model structure and an adaptive SGD optimizer. Integrating adaptive learning and momentum terms minimizes model loss and speeds up convergence.

{"title":"Enhanced hybrid facial emotion detection & classification","authors":"Naim Ajlouni , Adem Özyavaş , Firas Ajlouni , Faruk Takaoğlu , Mustafa Takaoğlu","doi":"10.1016/j.fraope.2024.100200","DOIUrl":"10.1016/j.fraope.2024.100200","url":null,"abstract":"<div><div>Emotion Recognition (ER) presents a significant challenge in pattern recognition and is crucial for various Artificial Intelligence (AI) applications, from monitoring children with autism to enhancing video games and human-computer interactions. Facial features are essential for discerning human emotions, motivating this study to improve Facial Emotion detection accuracy, vital for real-time applications. This research explores the use of both full facial features and facial landmarks, suggesting that landmarks offer clearer emotional cues. A Convolutional Neural Network (CNN) is employed for feature extraction, optimized using techniques like Stochastic Gradient Descent (SGD). This study combines upper facial landmarks with full facial features to enhance detection accuracy. An adaptive Genetic Algorithm (GA) optimizes the CNN structure. The dual-input neural network architecture processes a full 48 × 48 facial image through a CNN and upper facial features through another input, which is then concatenated and fed into dense layers with L2 regularization. The model is trained and evaluated using various optimizers to find the best configuration, with performance metrics plotted and compared. This method allows detailed pixel information and focused landmark features to improve emotion detection accuracy. Datasets such as FER2013 and Extended Cohn-Kanade (CK+) are used for training and testing. Results show significant accuracy improvement with the proposed method, further enhanced by a highly optimized model structure and an adaptive SGD optimizer. Integrating adaptive learning and momentum terms minimizes model loss and speeds up convergence.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"10 ","pages":"Article 100200"},"PeriodicalIF":0.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143102072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Online nonintrusive load monitoring using radial basis neural network with advanced flexible particle swarm optimization algorithm for residential load

Franklin Open

Pub Date : 2024-12-14 DOI: 10.1016/j.fraope.2024.100198

Yashar Toopchi

In recent times, there has been a focus on optimizing energy management systems, specifically through load disaggregation and Nonintrusive Load Monitoring (NILM) approaches. The main objective of NILM system design is to determine which appliances are used within a household by evaluating the energy patterns from the primary supply point at the entry. This allows for analyzing appliance-level energy consumption using only smart meter data. These smart meters can assist power utility companies in developing efficient strategies to reduce energy demand from consumers during peak load periods. This paper introduces a new methodology for online load disaggregation using a radial basis function neural network (RBFNN) model. The optimization of the RBFNN model’s parameters is achieved through an advanced flexible particle swarm optimization (AFPSO) algorithm. The RBFNN, in this method, identifies appliance loads and determines their ON or OFF status by analyzing load signatures. In the online process, the back-propagation (BP) algorithm is used to update RBFNN model parameters such as the weighting coefficients and bias values of the output layers, and the nonlinear function parameters of the hidden layers. Compared to traditional PSO, AFPSO demonstrates superior performance by implementing an adaptive procedure through two additional steps. These steps involve updating and screening mechanisms aimed at enhancing PSO performance when it becomes trapped in a local optimum. The proposed NILM design is applied to the AMPds data set, and test results have shown that the technique accurately identifies loads for residential energy monitoring. This makes it suitable for nonintrusive load monitoring (NILM) systems in residential and commercial buildings. Extensive simulations have demonstrated that the proposed online technique achieves better results in load identification compared to other state-of-the-art methods.

{"title":"Online nonintrusive load monitoring using radial basis neural network with advanced flexible particle swarm optimization algorithm for residential load","authors":"Yashar Toopchi","doi":"10.1016/j.fraope.2024.100198","DOIUrl":"10.1016/j.fraope.2024.100198","url":null,"abstract":"<div><div>In recent times, there has been a focus on optimizing energy management systems, specifically through load disaggregation and Nonintrusive Load Monitoring (NILM) approaches. The main objective of NILM system design is to determine which appliances are used within a household by evaluating the energy patterns from the primary supply point at the entry. This allows for analyzing appliance-level energy consumption using only smart meter data. These smart meters can assist power utility companies in developing efficient strategies to reduce energy demand from consumers during peak load periods. This paper introduces a new methodology for online load disaggregation using a radial basis function neural network (RBFNN) model. The optimization of the RBFNN model’s parameters is achieved through an advanced flexible particle swarm optimization (AFPSO) algorithm. The RBFNN, in this method, identifies appliance loads and determines their ON or OFF status by analyzing load signatures. In the online process, the back-propagation (BP) algorithm is used to update RBFNN model parameters such as the weighting coefficients and bias values of the output layers, and the nonlinear function parameters of the hidden layers. Compared to traditional PSO, AFPSO demonstrates superior performance by implementing an adaptive procedure through two additional steps. These steps involve updating and screening mechanisms aimed at enhancing PSO performance when it becomes trapped in a local optimum. The proposed NILM design is applied to the AMPds data set, and test results have shown that the technique accurately identifies loads for residential energy monitoring. This makes it suitable for nonintrusive load monitoring (NILM) systems in residential and commercial buildings. Extensive simulations have demonstrated that the proposed online technique achieves better results in load identification compared to other state-of-the-art methods.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"10 ","pages":"Article 100198"},"PeriodicalIF":0.0,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101576","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Comparative studies on a zooplankton–fish model subjected to infection in zooplankton with varying rates of disease transmission

Franklin Open

Pub Date : 2024-12-11 DOI: 10.1016/j.fraope.2024.100194

Soumita Sen, Suddhyashil Sarkar, Samares Pal

Eco-epidemiological models offer practical demonstrations of non-linear dynamics, enriching the realm of mathematical biology. The importance of viruses in oceanic plankton ecosystems is becoming more widely acknowledged. Infectious diseases have the potential to significantly alter the interaction within ecological systems. In this article we propose and analyze a three compartmental mathematical model of ODEs involving zooplankton and fish, with SI-type disease in the zooplankton population and two different non-linear infection rates. This study looks at how susceptible zooplankton species behave when in the presence of an infected zooplankton population. We further consider fish population as the sole predator of zooplankton population without considering the specific feeding habits of fish when they consume zooplankton. Additionally, it is thought that the zooplankton population that is infected is more likely to be preyed upon than their non-infected counterparts. The models describe when populations stay constant or show cyclical patterns in behavior. Our research shows that both the systems possess five different biologically feasible steady states. Our study also reveals that the susceptible zooplankton class is adversely impacted by the infection rate. The fish population disappears once the natural mortality rate of fish species surpasses a certain critical value. If the fish population is unable to exceed a certain threshold due to predation of infected zooplankton, they will not survive. For a healthy ecosystem, a lower infection rate and increased fish growth attributed to predation of susceptible zooplankton is preferred. However, an increased rate of infection may cause population collapse in the fish population.

{"title":"Comparative studies on a zooplankton–fish model subjected to infection in zooplankton with varying rates of disease transmission","authors":"Soumita Sen, Suddhyashil Sarkar, Samares Pal","doi":"10.1016/j.fraope.2024.100194","DOIUrl":"10.1016/j.fraope.2024.100194","url":null,"abstract":"<div><div>Eco-epidemiological models offer practical demonstrations of non-linear dynamics, enriching the realm of mathematical biology. The importance of viruses in oceanic plankton ecosystems is becoming more widely acknowledged. Infectious diseases have the potential to significantly alter the interaction within ecological systems. In this article we propose and analyze a three compartmental mathematical model of ODEs involving zooplankton and fish, with SI-type disease in the zooplankton population and two different non-linear infection rates. This study looks at how susceptible zooplankton species behave when in the presence of an infected zooplankton population. We further consider fish population as the sole predator of zooplankton population without considering the specific feeding habits of fish when they consume zooplankton. Additionally, it is thought that the zooplankton population that is infected is more likely to be preyed upon than their non-infected counterparts. The models describe when populations stay constant or show cyclical patterns in behavior. Our research shows that both the systems possess five different biologically feasible steady states. Our study also reveals that the susceptible zooplankton class is adversely impacted by the infection rate. The fish population disappears once the natural mortality rate of fish species surpasses a certain critical value. If the fish population is unable to exceed a certain threshold due to predation of infected zooplankton, they will not survive. For a healthy ecosystem, a lower infection rate and increased fish growth attributed to predation of susceptible zooplankton is preferred. However, an increased rate of infection may cause population collapse in the fish population.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"10 ","pages":"Article 100194"},"PeriodicalIF":0.0,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An object detection solution for early detection of taro leaf blight disease in the West African sub-region

Franklin Open

Pub Date : 2024-12-11 DOI: 10.1016/j.fraope.2024.100197

Chidiebere B. Nwaneto, Chika Yinka-Banjo, Ogban Ugot

Taro Leaf Blight (TLB) poses a significant threat to food security and economic stability in West Africa, where taro is a staple crop. This research presents an object detection system utilizing the YOLOv8 deep learning model to detect TLB early in taro plants. The methodology involved developing a unique dataset comprising images of taro leaves at various stages of infection, collected from farms in Nigeria and Ghana. Fine-tuning the YOLOv8 model with this dataset resulted in a notable improvement, achieving an 85.7 % mean Average Precision (mAP) across all classes—a significant enhancement over existing generic plant disease detection models, which typically achieve mAP values of around 70–75 % on similar datasets. This 15–20 % improvement enables more accurate early detection, crucial for timely interventions. The system was subsequently integrated into an Android application, allowing farmers real-time diagnosis and disease management access. Field tests demonstrated the application's effectiveness and user-friendly design, making it a practical tool for early disease intervention. This research highlights the potential of combining deep learning and mobile technology to address agricultural challenges and improve food security in the region.

{"title":"An object detection solution for early detection of taro leaf blight disease in the West African sub-region","authors":"Chidiebere B. Nwaneto, Chika Yinka-Banjo, Ogban Ugot","doi":"10.1016/j.fraope.2024.100197","DOIUrl":"10.1016/j.fraope.2024.100197","url":null,"abstract":"<div><div>Taro Leaf Blight (TLB) poses a significant threat to food security and economic stability in West Africa, where taro is a staple crop. This research presents an object detection system utilizing the YOLOv8 deep learning model to detect TLB early in taro plants. The methodology involved developing a unique dataset comprising images of taro leaves at various stages of infection, collected from farms in Nigeria and Ghana. Fine-tuning the YOLOv8 model with this dataset resulted in a notable improvement, achieving an 85.7 % mean Average Precision (mAP) across all classes—a significant enhancement over existing generic plant disease detection models, which typically achieve mAP values of around 70–75 % on similar datasets. This 15–20 % improvement enables more accurate early detection, crucial for timely interventions. The system was subsequently integrated into an Android application, allowing farmers real-time diagnosis and disease management access. Field tests demonstrated the application's effectiveness and user-friendly design, making it a practical tool for early disease intervention. This research highlights the potential of combining deep learning and mobile technology to address agricultural challenges and improve food security in the region.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"10 ","pages":"Article 100197"},"PeriodicalIF":0.0,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143102071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimizing stand-alone microgrids with lagrange multiplier technique: A cost-effective and sustainable solution for rural electrification

Franklin Open

Pub Date : 2024-12-11 DOI: 10.1016/j.fraope.2024.100199

Rasha Elazab, Ahmed Mamdouh Ewais, Maged Ahmed Abu-Adma

This paper addresses the challenge of optimally sizing and planning stand-alone microgrids in remote areas, focusing on generating sources using a novel algorithm based on the Lagrange multiplier optimization technique. The study centers on the Gulf of Aqaba, Egypt, where five configurations of PV and wind micro plants with diesel generation are evaluated. Detailed cost analysis includes wind turbines, PV modules, and diesel generators. Motivated by the need for cost-effective, reliable, and environmentally friendly microgrids, the proposed algorithm is benchmarked against the widely used Hybrid Optimization Model for Energy Renewable HOMER software. Key criteria for analysis include economic benefits, environmental impacts, and system reliability. Results highlight the superior performance of the proposed optimization technique. It achieves up to 22 % lower net present costs (NPC) and up to 13 % lower cost of electricity (COE) compared to HOMER. Additionally, the algorithm demonstrates significant environmental benefits, with emissions reductions of up to 25 % for carbon dioxide and substantial decreases in carbon monoxide and nitrogen oxides. Reliability is also enhanced, with the proposed schemes showing higher excess energy and renewable energy contributions. The study justifies the use of the Lagrange multiplier optimization technique as a superior approach for planning and sizing stand-alone microgrids, offering significant economic and environmental advantages. This work provides a comprehensive framework for developing sustainable energy solutions in isolated regions.

{"title":"Optimizing stand-alone microgrids with lagrange multiplier technique: A cost-effective and sustainable solution for rural electrification","authors":"Rasha Elazab, Ahmed Mamdouh Ewais, Maged Ahmed Abu-Adma","doi":"10.1016/j.fraope.2024.100199","DOIUrl":"10.1016/j.fraope.2024.100199","url":null,"abstract":"<div><div>This paper addresses the challenge of optimally sizing and planning stand-alone microgrids in remote areas, focusing on generating sources using a novel algorithm based on the Lagrange multiplier optimization technique. The study centers on the Gulf of Aqaba, Egypt, where five configurations of PV and wind micro plants with diesel generation are evaluated. Detailed cost analysis includes wind turbines, PV modules, and diesel generators. Motivated by the need for cost-effective, reliable, and environmentally friendly microgrids, the proposed algorithm is benchmarked against the widely used Hybrid Optimization Model for Energy Renewable HOMER software. Key criteria for analysis include economic benefits, environmental impacts, and system reliability. Results highlight the superior performance of the proposed optimization technique. It achieves up to 22 % lower net present costs (NPC) and up to 13 % lower cost of electricity (COE) compared to HOMER. Additionally, the algorithm demonstrates significant environmental benefits, with emissions reductions of up to 25 % for carbon dioxide and substantial decreases in carbon monoxide and nitrogen oxides. Reliability is also enhanced, with the proposed schemes showing higher excess energy and renewable energy contributions. The study justifies the use of the Lagrange multiplier optimization technique as a superior approach for planning and sizing stand-alone microgrids, offering significant economic and environmental advantages. This work provides a comprehensive framework for developing sustainable energy solutions in isolated regions.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"10 ","pages":"Article 100199"},"PeriodicalIF":0.0,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy conservation and techno-environmental analysis in natural gas liquefaction with single and dual-mixed refrigerants: A comparison

Franklin Open

Pub Date : 2024-12-09 DOI: 10.1016/j.fraope.2024.100196

Edose Osagie , Joseph Akpan , Wilson Ekpotu , Gabriel Umoh , Joseph Akintola , Queen Moses , Philemon Udom

The need to meet energy demand and produce cleaner energy has propped up the drive to explore natural gas options and, more recently, liquefied natural gas. The thermal efficiency of a liquefaction process is very important when considering the energy requirement. The liquefaction technology by the refrigerant is one of the most common technologies becoming popular among researchers for its flexible features. This study compares the energy performance of the dual mixed refrigerant (PRICO DMR) and single mixed refrigerant (PRICO SMR) propane pre-cool liquefaction technology using a process simulator by carrying out sensitivity analysis on the process parameters to see its impact on the refrigerant flow rate, power consumption, and the specific power. The Honeywell UNISIM R451, software process simulator, was used to simulate the PRICO SMR and DMR processes with natural gas supplied at 55 °C, 61 bar, and 150 MMSCFD as the base condition. From the simulation, the ‘UA’ for the Heat Exchangers used with mixed refrigerant, the SMR was 36470 kW/°C, and the DMR was 5172 and 4612 kW/°C. The result shows that at the base condition, the refrigerant flow rate, power consumption, and specific power for the DMR and SMR was 404.9 MMSCFD, 96.8 MW, 0.409 kWhr/kg-LNG, and 507 MMSCFD, 144.5 MW, and 1.261 kWhr/kg-LNG, respectively. With sensitivity analysis, the simulation results showed that for both the SMR and DMR processes at any temperature, increasing pressure increases the refrigerant flow rate and power consumption but decreases the specific power. In addition, the DMR provided a potential CO₂ emissions potentials saving by an average factor of 2.953 and a better savings in energy demand with the specific power reaching 67.7 % and refrigerant consumption than the SMR process.

{"title":"Energy conservation and techno-environmental analysis in natural gas liquefaction with single and dual-mixed refrigerants: A comparison","authors":"Edose Osagie , Joseph Akpan , Wilson Ekpotu , Gabriel Umoh , Joseph Akintola , Queen Moses , Philemon Udom","doi":"10.1016/j.fraope.2024.100196","DOIUrl":"10.1016/j.fraope.2024.100196","url":null,"abstract":"<div><div>The need to meet energy demand and produce cleaner energy has propped up the drive to explore natural gas options and, more recently, liquefied natural gas. The thermal efficiency of a liquefaction process is very important when considering the energy requirement. The liquefaction technology by the refrigerant is one of the most common technologies becoming popular among researchers for its flexible features. This study compares the energy performance of the dual mixed refrigerant (PRICO DMR) and single mixed refrigerant (PRICO SMR) propane pre-cool liquefaction technology using a process simulator by carrying out sensitivity analysis on the process parameters to see its impact on the refrigerant flow rate, power consumption, and the specific power. The Honeywell UNISIM R451, software process simulator, was used to simulate the PRICO SMR and DMR processes with natural gas supplied at 55 °C, 61 bar, and 150 MMSCFD as the base condition. From the simulation, the ‘UA’ for the Heat Exchangers used with mixed refrigerant, the SMR was 36470 kW/°C, and the DMR was 5172 and 4612 kW/°C. The result shows that at the base condition, the refrigerant flow rate, power consumption, and specific power for the DMR and SMR was 404.9 MMSCFD, 96.8 MW, 0.409 kWhr/kg-LNG, and 507 MMSCFD, 144.5 MW, and 1.261 kWhr/kg-LNG, respectively. With sensitivity analysis, the simulation results showed that for both the SMR and DMR processes at any temperature, increasing pressure increases the refrigerant flow rate and power consumption but decreases the specific power. In addition, the DMR provided a potential CO<sub>2</sub> emissions potentials saving by an average factor of 2.953 and a better savings in energy demand with the specific power reaching 67.7 % and refrigerant consumption than the SMR process.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"10 ","pages":"Article 100196"},"PeriodicalIF":0.0,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Erratum to “Event-triggering-based H∞ load frequency control for multi-area cyber–physical power system under DoS attacks” [Franklin Open Volume 3, June 2023, 100012]

Franklin Open

Pub Date : 2024-12-01 DOI: 10.1016/j.fraope.2024.100124

Xingyue Liu , Kaibo Shi , Kun Zhou , Shiping Wen , Yiqian Tang , Lin Tang

引用次数: 0

Erratum to “Identify and classify pests in the agricultural sector using metaheuristics deep learning approach” [Franklin Open Volume 3, June 2023, 100024]

Franklin Open

Pub Date : 2024-12-01 DOI: 10.1016/j.fraope.2024.100122

Atul B. Kathole , Jayashree Katti , Savita Lonare , Gulbakshee Dharmale

引用次数: 0

Semantic communication-based convolutional neural network for enhanced image classification 基于语义通信的卷积神经网络增强图像分类

Franklin Open

Pub Date : 2024-12-01 DOI: 10.1016/j.fraope.2024.100192

Nivine Guler , Zied Ben Hazem

In AI-IoT environments, the traditional centralized cloud computing approach leads to high network transmission volumes and communication delays, negatively affecting intelligent task performance. This study addresses these issues by introducing an innovative semantic communication system model for intelligent tasks, leveraging deep learning techniques. The research focuses on image classification tasks constrained by bandwidth and delay in AI-IoT scenarios. The proposed model features a tailored semantic communication network architecture, where image feature maps are extracted on IoT devices. These semantic relations are then compressed based on the extracted feature maps to reduce power consumption on IoT devices and mitigate communication transmission pressures. Simulations and comparative analyses of various network performance metrics show that the proposed semantic communication system improves image classification accuracy by 90%% at low signal-to-noise ratios compared to traditional methods. With an 80%% compression ratio, the classification accuracy loss is minimal—within 2%%—when the signal-to-noise ratio exceeds 0. Additionally, at a signal-to-noise ratio of 20, the semantic compression transmission scheme enhances classification accuracy by 30%% compared to random compression. Moreover, the proposed system outperforms traditional approaches in execution time by approximately 80%%.

在AI-IoT环境下，传统的集中式云计算方式导致网络传输量大，通信延迟，对智能任务性能产生负面影响。本研究通过引入一种创新的智能任务语义通信系统模型，利用深度学习技术来解决这些问题。重点研究AI-IoT场景下受带宽和时延约束的图像分类任务。提出的模型具有定制的语义通信网络架构，其中在物联网设备上提取图像特征映射。然后根据提取的特征映射压缩这些语义关系，以降低物联网设备的功耗并减轻通信传输压力。各种网络性能指标的仿真和对比分析表明，与传统方法相比，所提出的语义通信系统在低信噪比下将图像分类精度提高了90%。当压缩比为80%时，当信噪比超过0时，分类精度损失最小，在2%以内。此外，在信噪比为20的情况下，与随机压缩相比，语义压缩传输方案的分类准确率提高了30%。此外，所提出的系统在执行时间上比传统方法高出约80%。

{"title":"Semantic communication-based convolutional neural network for enhanced image classification","authors":"Nivine Guler , Zied Ben Hazem","doi":"10.1016/j.fraope.2024.100192","DOIUrl":"10.1016/j.fraope.2024.100192","url":null,"abstract":"<div><div>In AI-IoT environments, the traditional centralized cloud computing approach leads to high network transmission volumes and communication delays, negatively affecting intelligent task performance. This study addresses these issues by introducing an innovative semantic communication system model for intelligent tasks, leveraging deep learning techniques. The research focuses on image classification tasks constrained by bandwidth and delay in AI-IoT scenarios. The proposed model features a tailored semantic communication network architecture, where image feature maps are extracted on IoT devices. These semantic relations are then compressed based on the extracted feature maps to reduce power consumption on IoT devices and mitigate communication transmission pressures. Simulations and comparative analyses of various network performance metrics show that the proposed semantic communication system improves image classification accuracy by 90%% at low signal-to-noise ratios compared to traditional methods. With an 80%% compression ratio, the classification accuracy loss is minimal—within 2%%—when the signal-to-noise ratio exceeds 0. Additionally, at a signal-to-noise ratio of 20, the semantic compression transmission scheme enhances classification accuracy by 30%% compared to random compression. Moreover, the proposed system outperforms traditional approaches in execution time by approximately 80%%.</div></div>","PeriodicalId":100554,"journal":{"name":"Franklin Open","volume":"9 ","pages":"Article 100192"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142743112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Identifiability in networks of nonlinear dynamical systems with linear and/or nonlinear couplings

Franklin Open

Pub Date : 2024-12-01 DOI: 10.1016/j.fraope.2024.100195

Nathalie Verdière

The identifiability study of dynamical systems is a property that ensures the uniqueness of parameters with respect to the model’s measurement(s). Several methods exist, but for nonlinear differential equations, these methods are often limited by the size of the systems. Some recent work on network identifiability has been published, but strong constraints on the system’s linearities and coupling functions are still imposed. Unfortunately, in fields like neuroscience, such restrictions are no longer applicable due to the complex dynamics of the neurons and their interactions. This paper aims to present a method for studying identifiability in networks composed of linear and/or nonlinear systems with linear and/or nonlinear coupling functions. Based on the observation of certain variables of interest of some nodes, it determines which subsystems are identifiable. Additionally, the method outlines the paths and steps required to identify these subsystems. It has been automated by an algorithm described in this paper, implemented in Maple and applied to an example in neuroscience, a neural network of the C. elegans worm.

引用次数: 0

首页上一页

下一页尾页

类型

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

Franklin Open

全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.

﹀