In today’s era of escalating digital threats and the growing need for safeguarding sensitive information, this research strives to advance the field of information concealment by introducing a pioneering steganography methodology. Our approach goes beyond the conventional boundaries of image security by seamlessly integrating classical image processing techniques with the cutting-edge realm of quantum encoding. The foundation of our technique lies in the meticulous identification of distinctive features within the cover image, a crucial step achieved through the utilization of SIFT (Scale-Invariant Feature Transform). These identified key points are further organized into coherent clusters employing the K-means clustering algorithm, forming a structured basis for our covert communication process. The core innovation of this research resides in the transformation of the concealed message into a NEQR (Novel Enhanced Quantum Representation) code, a quantum encoding framework that leverages the power of quantum circuits. This transformative step ensures not only the secrecy but also the integrity of the hidden information, making it highly resistant to even the most sophisticated decryption attempts. The strategic placement of the quantum circuit representing the concealed message at the centroids of the clusters generated by the K-means algorithm conceals it within the cover image seamlessly. This fusion of classical image processing and quantum encoding results in an unprecedented level of security for the embedded information, rendering it virtually impervious to unauthorized access. Empirical findings from extensive experimentation affirm the robustness and efficacy of our proposed strategy.
{"title":"Quantum Steganography: Hiding Secret Messages in Images using Quantum Circuits and SIFT","authors":"Hassan Jameel Azooz, Khawla Ben Salah, Monji Kherallah, Mohamed Saber Naceur","doi":"10.14569/ijacsa.2023.01410107","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.01410107","url":null,"abstract":"In today’s era of escalating digital threats and the growing need for safeguarding sensitive information, this research strives to advance the field of information concealment by introducing a pioneering steganography methodology. Our approach goes beyond the conventional boundaries of image security by seamlessly integrating classical image processing techniques with the cutting-edge realm of quantum encoding. The foundation of our technique lies in the meticulous identification of distinctive features within the cover image, a crucial step achieved through the utilization of SIFT (Scale-Invariant Feature Transform). These identified key points are further organized into coherent clusters employing the K-means clustering algorithm, forming a structured basis for our covert communication process. The core innovation of this research resides in the transformation of the concealed message into a NEQR (Novel Enhanced Quantum Representation) code, a quantum encoding framework that leverages the power of quantum circuits. This transformative step ensures not only the secrecy but also the integrity of the hidden information, making it highly resistant to even the most sophisticated decryption attempts. The strategic placement of the quantum circuit representing the concealed message at the centroids of the clusters generated by the K-means algorithm conceals it within the cover image seamlessly. This fusion of classical image processing and quantum encoding results in an unprecedented level of security for the embedded information, rendering it virtually impervious to unauthorized access. Empirical findings from extensive experimentation affirm the robustness and efficacy of our proposed strategy.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135318753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.01410103
Arnab Dey, Samit Biswas, Dac-Nhoung Le
Human-Human Interaction Recognition (H2HIR) is a multidisciplinary field that combines computer vision, deep learning, and psychology. Its primary objective is to decode and understand the intricacies of human-human interactions. H2HIR holds significant importance across various domains as it enables machines to perceive, comprehend, and respond to human social behaviors, gestures, and communication patterns. This study aims to identify human-human interactions from just one frame, i.e. from an image. Diverging from the realm of video-based inter-action recognition, a well-established research domain that relies on the utilization of spatio-temporal information, the complexity of the task escalates significantly when dealing with still images due to the absence of these intrinsic spatio-temporal features. This research introduces a novel deep learning model called AdaptiveDRNet with Multi-level Attention to recognize Human-Human (H2H) interactions. Our proposed method demonstrates outstanding performance on the Human-Human Interaction Image dataset (H2HID), encompassing 4049 meticulously curated images representing fifteen distinct human interactions and on the publicly accessible HII and HIIv2 related benchmark datasets. Notably, our proposed model excels with a validation accuracy of 97.20% in the classification of human-human interaction images, surpassing the performance of EfficientNet, InceptionResNetV2, NASNet Mobile, ConvXNet, ResNet50, and VGG-16 models. H2H interaction recognition’s significance lies in its capacity to enhance communication, improve decision-making, and ultimately contribute to the well-being and efficiency of individuals and society as a whole.
{"title":"Recognition of Human Interactions in Still Images using AdaptiveDRNet with Multi-level Attention","authors":"Arnab Dey, Samit Biswas, Dac-Nhoung Le","doi":"10.14569/ijacsa.2023.01410103","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.01410103","url":null,"abstract":"Human-Human Interaction Recognition (H2HIR) is a multidisciplinary field that combines computer vision, deep learning, and psychology. Its primary objective is to decode and understand the intricacies of human-human interactions. H2HIR holds significant importance across various domains as it enables machines to perceive, comprehend, and respond to human social behaviors, gestures, and communication patterns. This study aims to identify human-human interactions from just one frame, i.e. from an image. Diverging from the realm of video-based inter-action recognition, a well-established research domain that relies on the utilization of spatio-temporal information, the complexity of the task escalates significantly when dealing with still images due to the absence of these intrinsic spatio-temporal features. This research introduces a novel deep learning model called AdaptiveDRNet with Multi-level Attention to recognize Human-Human (H2H) interactions. Our proposed method demonstrates outstanding performance on the Human-Human Interaction Image dataset (H2HID), encompassing 4049 meticulously curated images representing fifteen distinct human interactions and on the publicly accessible HII and HIIv2 related benchmark datasets. Notably, our proposed model excels with a validation accuracy of 97.20% in the classification of human-human interaction images, surpassing the performance of EfficientNet, InceptionResNetV2, NASNet Mobile, ConvXNet, ResNet50, and VGG-16 models. H2H interaction recognition’s significance lies in its capacity to enhance communication, improve decision-making, and ultimately contribute to the well-being and efficiency of individuals and society as a whole.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"137 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135319209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.01410118
Henghe Zheng, Xiaojing Chen, Xin Liu
With the expanding utilization of cyber-physical structures and communication networks, cyberattacks have become a serious threat in various networks, including the Internet of Things (IoT) sensors. The state estimation algorithms play an important role in defining the present operational scenario of the IoT sensors. The attack of the false data injection (FDI) is the earnest menace for these estimation strategies (adopted by the operators of the IoT sensor) with the injection of the wicked data into the earned mensuration. The real-time recognition of this group of attacks increases the network resilience while it ensures secure network operation. This paper presents a new method for real-time FDI attack detection that uses a state prediction method basis on deep learning along with a new officiousness identification approach with the use of the matrix of the error covariance. The architecture of the presented method, along with its optimal group of meta-parameters, shows a real-time, scalable, effective state prediction method along with a minimal error border. The earned results display that the proposed method performs better than some recent literature about the prediction of the remaining useful life (RUL) with the use of the C-MAPSS dataset. In the following, two types of attacks of the false data injection are modeled, and then, their effectiveness is evaluated by using the proposed method. The earned results show that the attacks of the FDI, even on the low number of the sensors of the IoT, can severely disrupt the prediction of the RUL in all instances. In addition, our proposed model outperforms the FDI attack in terms of accuracy and flexibility.
{"title":"Identification of the False Data Injection Cyberattacks on the Internet of Things by using Deep Learning","authors":"Henghe Zheng, Xiaojing Chen, Xin Liu","doi":"10.14569/ijacsa.2023.01410118","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.01410118","url":null,"abstract":"With the expanding utilization of cyber-physical structures and communication networks, cyberattacks have become a serious threat in various networks, including the Internet of Things (IoT) sensors. The state estimation algorithms play an important role in defining the present operational scenario of the IoT sensors. The attack of the false data injection (FDI) is the earnest menace for these estimation strategies (adopted by the operators of the IoT sensor) with the injection of the wicked data into the earned mensuration. The real-time recognition of this group of attacks increases the network resilience while it ensures secure network operation. This paper presents a new method for real-time FDI attack detection that uses a state prediction method basis on deep learning along with a new officiousness identification approach with the use of the matrix of the error covariance. The architecture of the presented method, along with its optimal group of meta-parameters, shows a real-time, scalable, effective state prediction method along with a minimal error border. The earned results display that the proposed method performs better than some recent literature about the prediction of the remaining useful life (RUL) with the use of the C-MAPSS dataset. In the following, two types of attacks of the false data injection are modeled, and then, their effectiveness is evaluated by using the proposed method. The earned results show that the attacks of the FDI, even on the low number of the sensors of the IoT, can severely disrupt the prediction of the RUL in all instances. In addition, our proposed model outperforms the FDI attack in terms of accuracy and flexibility.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"137 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135319222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.0141061
Ramya Shree H P, Minavathi -, Dinesh M S
Nuclei Segmentation is the preliminary step towards the task of medical image analysis. Nowadays, there exists several deep learning-based techniques based on Convolutional Neural Networks (CNNs) for the task of nuclei segmentation. In this study, we present a neural network for semantic segmentation. This network harnesses the strengths in both residual learning and U-Net methodologies, thereby amplifying cell segmentation performance. This hybrid approach also facilitates the creation of network with diminished parameter requirement. The network incorporates residual units contributes to a smoother training process and mitigate the issue of vanishing gradients. Our model is tested on a microscopy image dataset which is publicly available from the 2018 Data Science Bowl grand challenge and assessed against U-Net and several other state-of-the-art deep learning approaches designed for nuclei segmentation. Our proposed approach showcases a notable improvement in average Intersection over Union (IoU) gain compared to prevailing state-of-the-art techniques, by exhibiting a significant margin of 1.1% and 5.8% higher gains over the original U-Net. Our model also excels across various key indicators, including accuracy, precision, recall and dice-coefficient. The outcomes underscore the potential of our proposed approach as a promising nuclei segmentation method for microscopy image analysis.
{"title":"An Automatic Nuclei Segmentation on Microscopic Images using Deep Residual U-Net","authors":"Ramya Shree H P, Minavathi -, Dinesh M S","doi":"10.14569/ijacsa.2023.0141061","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.0141061","url":null,"abstract":"Nuclei Segmentation is the preliminary step towards the task of medical image analysis. Nowadays, there exists several deep learning-based techniques based on Convolutional Neural Networks (CNNs) for the task of nuclei segmentation. In this study, we present a neural network for semantic segmentation. This network harnesses the strengths in both residual learning and U-Net methodologies, thereby amplifying cell segmentation performance. This hybrid approach also facilitates the creation of network with diminished parameter requirement. The network incorporates residual units contributes to a smoother training process and mitigate the issue of vanishing gradients. Our model is tested on a microscopy image dataset which is publicly available from the 2018 Data Science Bowl grand challenge and assessed against U-Net and several other state-of-the-art deep learning approaches designed for nuclei segmentation. Our proposed approach showcases a notable improvement in average Intersection over Union (IoU) gain compared to prevailing state-of-the-art techniques, by exhibiting a significant margin of 1.1% and 5.8% higher gains over the original U-Net. Our model also excels across various key indicators, including accuracy, precision, recall and dice-coefficient. The outcomes underscore the potential of our proposed approach as a promising nuclei segmentation method for microscopy image analysis.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135319645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.0140138
Adari Ramesh, C K Subbaraya, G K Ravi Kumar
In recent decades, mobile health (m-health) applications have gained significant attention in the healthcare sector due to their increased support during critical cases like cardiac disease, spinal cord problems, and brain injuries. Also, m-health services are considered more valuable, mainly where facilities are deficient. In addition, it supports wired and advanced wireless technologies for data transmission and communication. In this work, an Artificial Intelligence (AI)-based deep learning model is implemented to predict healthcare data, where the data handling is performed to improve the dynamic prediction performance. It includes the working modules of data collection, normalization, AI-based classification, and decision-making. Here, the m-health data are obtained from the smart devices through the service providers, which comprises the health information related to blood pressure, heart rate, glucose level, etc. The main contribution of this paper is to accurately predict Cardio Vascular Disease (CVD) from the patient dataset stored in cloud using the AI-based m-health system. After obtaining the data, preprocessing can be performed for noise reduction and normalization because prediction performance highly depends on data quality. Consequently, we use the Gorilla Troop Optimization Algorithm (GTOA) to select the most relevant functions for classifier training and testing. Classify his CVD type according to a selected set of features using bidirectional long-term memory (Bi-LSTM). Moreover, the proposed AI-based prediction model's performance is validated and compared using different measures.
{"title":"AI based Dynamic Prediction Model for Mobile Health Application System","authors":"Adari Ramesh, C K Subbaraya, G K Ravi Kumar","doi":"10.14569/ijacsa.2023.0140138","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.0140138","url":null,"abstract":"In recent decades, mobile health (m-health) applications have gained significant attention in the healthcare sector due to their increased support during critical cases like cardiac disease, spinal cord problems, and brain injuries. Also, m-health services are considered more valuable, mainly where facilities are deficient. In addition, it supports wired and advanced wireless technologies for data transmission and communication. In this work, an Artificial Intelligence (AI)-based deep learning model is implemented to predict healthcare data, where the data handling is performed to improve the dynamic prediction performance. It includes the working modules of data collection, normalization, AI-based classification, and decision-making. Here, the m-health data are obtained from the smart devices through the service providers, which comprises the health information related to blood pressure, heart rate, glucose level, etc. The main contribution of this paper is to accurately predict Cardio Vascular Disease (CVD) from the patient dataset stored in cloud using the AI-based m-health system. After obtaining the data, preprocessing can be performed for noise reduction and normalization because prediction performance highly depends on data quality. Consequently, we use the Gorilla Troop Optimization Algorithm (GTOA) to select the most relevant functions for classifier training and testing. Classify his CVD type according to a selected set of features using bidirectional long-term memory (Bi-LSTM). Moreover, the proposed AI-based prediction model's performance is validated and compared using different measures.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135470667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.0140931
Duan Li, Peisen Liu, Yongquan Xia
The brain-computer interface (BCI) based on motor imagery (MI) is a promising technology aimed at assisting individuals with motor impairments in regaining their motor abilities by capturing brain signals during specific tasks. However, non-invasive electroencephalogram (EEG) signals collected using EEG caps often contain large numbers of artifacts. Automatically and effectively removing these artifacts while preserving task-related brain components is a key issue for MI de-coding. Additionally, multi-channel EEG signals encompass temporal, frequency and spatial domain features. Although deep learning has achieved better results in extracting features and de-coding motor imagery EEG (MI-EEG) signals, obtaining a high-performance network on MI that achieves optimal matching of feature extraction, thus classification algorithms is still a challenging issue. In this study, we propose a scheme that combines a novel automatic artifact removal strategy with a spatial attention-based multiscale CNN (SA-MSCNN). This work obtained independent component analysis (ICA) weights from the first subject in the dataset and used K-means clustering to determine the best feature combination, which was then applied to other subjects for artifact removal. Additionally, this work designed an SA-MSCNN which includes multiscale convolution modules capable of extracting information from multiple frequency bands, spatial attention modules weighting spatial information, and separable convolution modules reducing feature information. This work validated the performance of the proposed model using a real-world public dataset, the BCI competition IV dataset 2a. The average accuracy of the method was 79.83%. This work conducted ablation experiments to demonstrate the effectiveness of the proposed artifact removal method and SA-MSCNN network and compared the results with outstanding models and state-of-the-art (SOTA) studies. The results confirm the effectiveness of the proposed method and provide a theoretical and experimental foundation for the development of new MI-BCI systems, which is very useful in helping people with disabilities regain their independence and improve their quality of life.
{"title":"A Novel Artifact Removal Strategy and Spatial Attention-based Multiscale CNN for MI Recognition","authors":"Duan Li, Peisen Liu, Yongquan Xia","doi":"10.14569/ijacsa.2023.0140931","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.0140931","url":null,"abstract":"The brain-computer interface (BCI) based on motor imagery (MI) is a promising technology aimed at assisting individuals with motor impairments in regaining their motor abilities by capturing brain signals during specific tasks. However, non-invasive electroencephalogram (EEG) signals collected using EEG caps often contain large numbers of artifacts. Automatically and effectively removing these artifacts while preserving task-related brain components is a key issue for MI de-coding. Additionally, multi-channel EEG signals encompass temporal, frequency and spatial domain features. Although deep learning has achieved better results in extracting features and de-coding motor imagery EEG (MI-EEG) signals, obtaining a high-performance network on MI that achieves optimal matching of feature extraction, thus classification algorithms is still a challenging issue. In this study, we propose a scheme that combines a novel automatic artifact removal strategy with a spatial attention-based multiscale CNN (SA-MSCNN). This work obtained independent component analysis (ICA) weights from the first subject in the dataset and used K-means clustering to determine the best feature combination, which was then applied to other subjects for artifact removal. Additionally, this work designed an SA-MSCNN which includes multiscale convolution modules capable of extracting information from multiple frequency bands, spatial attention modules weighting spatial information, and separable convolution modules reducing feature information. This work validated the performance of the proposed model using a real-world public dataset, the BCI competition IV dataset 2a. The average accuracy of the method was 79.83%. This work conducted ablation experiments to demonstrate the effectiveness of the proposed artifact removal method and SA-MSCNN network and compared the results with outstanding models and state-of-the-art (SOTA) studies. The results confirm the effectiveness of the proposed method and provide a theoretical and experimental foundation for the development of new MI-BCI systems, which is very useful in helping people with disabilities regain their independence and improve their quality of life.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135955484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.01409116
Terefe Feyisa, Seble Hailu
Morphological segmenter is an important component in Amharic natural language processing systems. Despite this fact, Amharic lacks large amount of morphologically segmented corpus. Large amount of corpus is often a requirement to develop neural network-based language technologies. This paper presents an alternative method to generate large amount of morph-segmented corpus for Amharic language. First, a relatively small (138,400 words) morphologically annotated Amharic seed-corpus is manually prepared. The annotation enables to identify prefixes, stem, and suffixes of a given word. Second, a supervised approach is used to create a conditional random field-based seed-model (on the seed-corpus). Applying the seed-model (an unsupervised technique on a large unsegmented raw Amharic words) for prediction, a large corpus size (3,777,283) of segmented words are automatically generated. Third, the newly generated corpus is used to train an Amharic morphological segmenter (based on a supervised neural sequence-to-sequence (seq2seq) approach using character embeddings). Using the seq2seq method, an F-score of 98.65% was measured. Results show an agreement with previous efforts for Arabic language. The work presented here has profound implications for future studies of Ethiopian language technologies and may one day help solve the problem of the digital-divide between resource-rich and under-resourced languages.
{"title":"Corpus Generation to Develop Amharic Morphological Segmenter","authors":"Terefe Feyisa, Seble Hailu","doi":"10.14569/ijacsa.2023.01409116","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.01409116","url":null,"abstract":"Morphological segmenter is an important component in Amharic natural language processing systems. Despite this fact, Amharic lacks large amount of morphologically segmented corpus. Large amount of corpus is often a requirement to develop neural network-based language technologies. This paper presents an alternative method to generate large amount of morph-segmented corpus for Amharic language. First, a relatively small (138,400 words) morphologically annotated Amharic seed-corpus is manually prepared. The annotation enables to identify prefixes, stem, and suffixes of a given word. Second, a supervised approach is used to create a conditional random field-based seed-model (on the seed-corpus). Applying the seed-model (an unsupervised technique on a large unsegmented raw Amharic words) for prediction, a large corpus size (3,777,283) of segmented words are automatically generated. Third, the newly generated corpus is used to train an Amharic morphological segmenter (based on a supervised neural sequence-to-sequence (seq2seq) approach using character embeddings). Using the seq2seq method, an F-score of 98.65% was measured. Results show an agreement with previous efforts for Arabic language. The work presented here has profound implications for future studies of Ethiopian language technologies and may one day help solve the problem of the digital-divide between resource-rich and under-resourced languages.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135955493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Smoke is often present in the early stages of a fire. Detecting low smoke concentration and small targets during these early stages can be challenging. This paper proposes an improved smoke detection algorithm that leverages the characteristics of smoke concentration using YOLOv7tiny. The improved algorithm consists of the following components: 1) utilizing the dark channel prior theory to extract smoke concentration characteristics and using the synthesized αRGB image as an input feature to enhance the features of sparse smoke; 2) designing a light-BiFPN multi-scale feature fusion structure to improve the detection performance of small target smoke; 3) using depth separable convolution to replace the original standard convolution and reduce the model parameter quantity. Experimental results on a self-made dataset show that the improved algorithm performs better in detecting sparse smoke and small target smoke, with mAP@0.5 and Recall reaching 94.03% and 95.62% respectively, and the detection FPS increasing to 118.78 frames/s. Moreover, the model parameter quantity decreases to 4.97M. The improved algorithm demonstrates superior performance in the detection of sparse and small smoke in the early stages of a fire.
{"title":"Improved Model for Smoke Detection Based on Concentration Features using YOLOv7tiny","authors":"Yuanpan ZHENG, Liwei Niu, Xinxin GAN, Hui WANG, Boyang XU, Zhenyu WANG","doi":"10.14569/ijacsa.2023.01409114","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.01409114","url":null,"abstract":"Smoke is often present in the early stages of a fire. Detecting low smoke concentration and small targets during these early stages can be challenging. This paper proposes an improved smoke detection algorithm that leverages the characteristics of smoke concentration using YOLOv7tiny. The improved algorithm consists of the following components: 1) utilizing the dark channel prior theory to extract smoke concentration characteristics and using the synthesized αRGB image as an input feature to enhance the features of sparse smoke; 2) designing a light-BiFPN multi-scale feature fusion structure to improve the detection performance of small target smoke; 3) using depth separable convolution to replace the original standard convolution and reduce the model parameter quantity. Experimental results on a self-made dataset show that the improved algorithm performs better in detecting sparse smoke and small target smoke, with mAP@0.5 and Recall reaching 94.03% and 95.62% respectively, and the detection FPS increasing to 118.78 frames/s. Moreover, the model parameter quantity decreases to 4.97M. The improved algorithm demonstrates superior performance in the detection of sparse and small smoke in the early stages of a fire.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135955988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.01409104
Xiaoning Dou, Weijing Chen, Lei Zhu, Yingmei Bai, Yan Li, Xiaoxiao Wu
The smart city concept originated a few years ago as a combination of ideas about how information and communication technologies can improve urban life. With the advent of the digital revolution, many cities globally are investing heavily in designing and implementing smart city solutions and projects. Machine Learning (ML) has evolved into a powerful tool within the smart city sector, enabling efficient resource management, improved infrastructure, and enhanced urban services. This paper discusses the diverse ML algorithms and their potential applications in smart cities, including Artificial Intelligence (AI) and Intelligent Transportation Systems (ITS). The key challenges, opportunities, and directions for adopting ML to make cities smarter and more sustainable are outlined.
{"title":"Machine Learning for Smart Cities: A Comprehensive Review of Applications and Opportunities","authors":"Xiaoning Dou, Weijing Chen, Lei Zhu, Yingmei Bai, Yan Li, Xiaoxiao Wu","doi":"10.14569/ijacsa.2023.01409104","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.01409104","url":null,"abstract":"The smart city concept originated a few years ago as a combination of ideas about how information and communication technologies can improve urban life. With the advent of the digital revolution, many cities globally are investing heavily in designing and implementing smart city solutions and projects. Machine Learning (ML) has evolved into a powerful tool within the smart city sector, enabling efficient resource management, improved infrastructure, and enhanced urban services. This paper discusses the diverse ML algorithms and their potential applications in smart cities, including Artificial Intelligence (AI) and Intelligent Transportation Systems (ITS). The key challenges, opportunities, and directions for adopting ML to make cities smarter and more sustainable are outlined.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135956275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.14569/ijacsa.2023.0140966
Ahmed Shawky Elbhrawy, Mohamed A. Belal, Mohamed Sameh Hassanein
Predicting underwriting risk has become a major challenge due to the imbalanced datasets in the field. A real-world imbalanced dataset is used in this work with 12 variables in 30144 cases, where most of the cases were classified as "accepting the insurance request", while a small percentage classified as "refusing insurance". This work developed 55 machine learning (ML) models to predict whether or not to renew policies. The models were developed using the original dataset and four data-level approaches resampling techniques: random oversampling, SMOTE, random undersampling, and hybrid methods with 11 ML algorithms to address the issue of imbalanced data (11 ML× (4 resampling techniques + unbalanced datasets) = 55 ML models). Seven classifier efficiency measures were used to evaluate these 55 models that were developed using 11 ML algorithms: logistic regression (LR), random forest (RF), artificial neural network (ANN), multilayer perceptron (MLP), support vector machine (SVM), naive Bayes (NB), decision tree (DT), XGBoost, k-nearest neighbors (KNN), stochastic gradient boosting (SGB), and AdaBoost. The seven classifier efficiency measures namely are accuracy, sensitivity, specificity, AUC, precision, F1-measure, and kappa. CRISP-DM methodology is utilisied to ensure that studies are conducted in a rigorous and systematic manner. Additionally, RapidMiner software was used to apply the algorithms and analyze the data, which highlighted the potential of ML to improve the accuracy of risk assessment in insurance underwriting. The results showed that all ML classifiers became more effective when using resampling strategies; where Hybrid resampling methods improved the performance of machine learning models on imbalanced data with an accuracy of 0.9967 and kappa statistics of 0.992 for the RF classifier.
{"title":"AIRA-ML: Auto Insurance Risk Assessment-Machine Learning Model using Resampling Methods","authors":"Ahmed Shawky Elbhrawy, Mohamed A. Belal, Mohamed Sameh Hassanein","doi":"10.14569/ijacsa.2023.0140966","DOIUrl":"https://doi.org/10.14569/ijacsa.2023.0140966","url":null,"abstract":"Predicting underwriting risk has become a major challenge due to the imbalanced datasets in the field. A real-world imbalanced dataset is used in this work with 12 variables in 30144 cases, where most of the cases were classified as \"accepting the insurance request\", while a small percentage classified as \"refusing insurance\". This work developed 55 machine learning (ML) models to predict whether or not to renew policies. The models were developed using the original dataset and four data-level approaches resampling techniques: random oversampling, SMOTE, random undersampling, and hybrid methods with 11 ML algorithms to address the issue of imbalanced data (11 ML× (4 resampling techniques + unbalanced datasets) = 55 ML models). Seven classifier efficiency measures were used to evaluate these 55 models that were developed using 11 ML algorithms: logistic regression (LR), random forest (RF), artificial neural network (ANN), multilayer perceptron (MLP), support vector machine (SVM), naive Bayes (NB), decision tree (DT), XGBoost, k-nearest neighbors (KNN), stochastic gradient boosting (SGB), and AdaBoost. The seven classifier efficiency measures namely are accuracy, sensitivity, specificity, AUC, precision, F1-measure, and kappa. CRISP-DM methodology is utilisied to ensure that studies are conducted in a rigorous and systematic manner. Additionally, RapidMiner software was used to apply the algorithms and analyze the data, which highlighted the potential of ML to improve the accuracy of risk assessment in insurance underwriting. The results showed that all ML classifiers became more effective when using resampling strategies; where Hybrid resampling methods improved the performance of machine learning models on imbalanced data with an accuracy of 0.9967 and kappa statistics of 0.992 for the RF classifier.","PeriodicalId":13824,"journal":{"name":"International Journal of Advanced Computer Science and Applications","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135956277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：