In the process of image shooting, due to the influence of angle, distance, complex scenes, illumination intensity, and other factors, small targets and occluded targets will inevitably appear in the image. These targets have few effective pixels, few features, and no obvious features, which makes it difficult to extract their effective features and easily leads to false detection, missed detection, and repeated detection, thus affecting the performance of target detection models. To solve this problem, an improved faster region convolutional neural network (RCNN) algorithm integrating the convolutional block attention module (CBAM) and feature pyramid network (FPN) (CF-RCNN) is proposed to improve the detection and recognition accuracy of small-sized, occluded, or truncated objects in complex scenes. Firstly, it incorporates the CBAM attention mechanism in the feature extraction network in combination with the information filtered by spatial and channel attention modules, focusing on local efficient information of the feature image, which improves the detection ability in the face of obscured or truncated objects. Secondly, it introduces the FPN feature pyramid structure, and links high-level and bottom-level feature data to obtain high-resolution and strong semantic data to enhance the detection effect for small-sized objects. Finally, it optimizes non-maximum suppression (NMS) to compensate for the shortcomings of conventional NMS that mistakenly eliminates overlapping detection frames. The experimental results show that the mean average precision (MAP) of target detection of the improved algorithm on PASCAL VOC2012 public datasets is improved to 76.2%, which is 13.9 percentage points higher than those of the commonly used Faster RCNN and other algorithms. It is better than the commonly used small-sample target detection algorithm.
{"title":"Faster RCNN Target Detection Algorithm Integrating CBAM and FPN","authors":"Wenshun Sheng, Xiongfeng Yu, Jiayan Lin, Xin Chen","doi":"10.3390/app13126913","DOIUrl":"https://doi.org/10.3390/app13126913","url":null,"abstract":"In the process of image shooting, due to the influence of angle, distance, complex scenes, illumination intensity, and other factors, small targets and occluded targets will inevitably appear in the image. These targets have few effective pixels, few features, and no obvious features, which makes it difficult to extract their effective features and easily leads to false detection, missed detection, and repeated detection, thus affecting the performance of target detection models. To solve this problem, an improved faster region convolutional neural network (RCNN) algorithm integrating the convolutional block attention module (CBAM) and feature pyramid network (FPN) (CF-RCNN) is proposed to improve the detection and recognition accuracy of small-sized, occluded, or truncated objects in complex scenes. Firstly, it incorporates the CBAM attention mechanism in the feature extraction network in combination with the information filtered by spatial and channel attention modules, focusing on local efficient information of the feature image, which improves the detection ability in the face of obscured or truncated objects. Secondly, it introduces the FPN feature pyramid structure, and links high-level and bottom-level feature data to obtain high-resolution and strong semantic data to enhance the detection effect for small-sized objects. Finally, it optimizes non-maximum suppression (NMS) to compensate for the shortcomings of conventional NMS that mistakenly eliminates overlapping detection frames. The experimental results show that the mean average precision (MAP) of target detection of the improved algorithm on PASCAL VOC2012 public datasets is improved to 76.2%, which is 13.9 percentage points higher than those of the commonly used Faster RCNN and other algorithms. It is better than the commonly used small-sample target detection algorithm.","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","volume":"68 1","pages":"1549-1569"},"PeriodicalIF":2.2,"publicationDate":"2023-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89604886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-20DOI: 10.32604/csse.2023.034172
Yudong Zhang, Muhammad Attique Khan, Ziquan Zhu, Shuihua Wang
(Aim) The COVID-19 has caused 6.26 million deaths and 522.06 million confirmed cases till 17/May/2022. Chest computed tomography is a precise way to help clinicians diagnose COVID-19 patients. (Method) Two datasets are chosen for this study. The multiple-way data augmentation, including speckle noise, random translation, scaling, salt-and-pepper noise, vertical shear, Gamma correction, rotation, Gaussian noise, and horizontal shear, is harnessed to increase the size of the training set. Then, the SqueezeNet (SN) with complex bypass is used to generate SN features. Finally, the extreme learning machine (ELM) is used to serve as the classifier due to its simplicity of usage, quick learning speed, and great generalization performances. The number of hidden neurons in ELM is set to 2000. Ten runs of 10-fold cross-validation are implemented to generate impartial results. (Result) For the 296-image dataset, our SNELM model attains a sensitivity of 96.35 ± 1.50%, a specificity of 96.08 ± 1.05%, a precision of 96.10 ± 1.00%, and an accuracy of 96.22 ± 0.94%. For the 640-image dataset, the SNELM attains a sensitivity of 96.00 ± 1.25%, a specificity of 96.28 ± 1.16%, a precision of 96.28 ± 1.13%, and an accuracy of 96.14 ± 0.96%. (Conclusion) The proposed SNELM model is successful in diagnosing COVID-19. The performances of our model are higher than seven state-of-the-art COVID-19 recognition models.
{"title":"SNELM: SqueezeNet-Guided ELM for COVID-19 Recognition.","authors":"Yudong Zhang, Muhammad Attique Khan, Ziquan Zhu, Shuihua Wang","doi":"10.32604/csse.2023.034172","DOIUrl":"https://doi.org/10.32604/csse.2023.034172","url":null,"abstract":"<p><p>(Aim) The COVID-19 has caused 6.26 million deaths and 522.06 million confirmed cases till 17/May/2022. Chest computed tomography is a precise way to help clinicians diagnose COVID-19 patients. (Method) Two datasets are chosen for this study. The multiple-way data augmentation, including speckle noise, random translation, scaling, salt-and-pepper noise, vertical shear, Gamma correction, rotation, Gaussian noise, and horizontal shear, is harnessed to increase the size of the training set. Then, the SqueezeNet (SN) with complex bypass is used to generate SN features. Finally, the extreme learning machine (ELM) is used to serve as the classifier due to its simplicity of usage, quick learning speed, and great generalization performances. The number of hidden neurons in ELM is set to 2000. Ten runs of 10-fold cross-validation are implemented to generate impartial results. (Result) For the 296-image dataset, our SNELM model attains a sensitivity of 96.35 ± 1.50%, a specificity of 96.08 ± 1.05%, a precision of 96.10 ± 1.00%, and an accuracy of 96.22 ± 0.94%. For the 640-image dataset, the SNELM attains a sensitivity of 96.00 ± 1.25%, a specificity of 96.28 ± 1.16%, a precision of 96.28 ± 1.13%, and an accuracy of 96.14 ± 0.96%. (Conclusion) The proposed SNELM model is successful in diagnosing COVID-19. The performances of our model are higher than seven state-of-the-art COVID-19 recognition models.</p>","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","volume":"46 1","pages":"13-26"},"PeriodicalIF":2.2,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7614503/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9784682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Multi-Objective Genetic Algorithm Based Load Balancing Strategy for Health Monitoring Systems in Fog-Cloud","authors":"Hayder Makki Shakir, Jaber Karimpour, Jafar Razmara","doi":"10.32604/csse.2023.038545","DOIUrl":"https://doi.org/10.32604/csse.2023.038545","url":null,"abstract":"","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","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":"135563761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.32604/csse.2023.033807
R. Ani, O. S. Deepa, B. R. Manju
The drug development process takes a long time since it requires sorting through a large number of inactive compounds from a large collection of compounds chosen for study and choosing just the most pertinent compounds that can bind to a disease protein. The use of virtual screening in pharmaceutical research is growing in popularity. During the early phases of medication research and development, it is crucial. Chemical compound searches are now more narrowly targeted. Because the databases contain more and more ligands, this method needs to be quick and exact. Neural network fingerprints were created more effectively than the well-known Extended Connectivity Fingerprint (ECFP). Only the largest sub-graph is taken into consideration to learn the representation, despite the fact that the conventional graph network generates a better-encoded fingerprint. When using the average or maximum pooling layer, it also contains unrelated data. This article suggested the Graph Convolutional Attention Network (GCAN), a graph neural network with an attention mechanism, to address these problems. Additionally, it makes the nodes or sub-graphs that are used to create the molecular fingerprint more significant. The generated fingerprint is used to classify drugs using ensemble learning. As base classifiers, ensemble stacking is applied to Support Vector Machines (SVM), Random Forest, Nave Bayes, Decision Trees, AdaBoost, and Gradient Boosting. When compared to existing models, the proposed GCAN fingerprint with an ensemble model achieves relatively high accuracy, sensitivity, specificity, and area under the curve. Additionally, it is revealed that our ensemble learning with generated molecular fingerprint yields 91% accuracy, outperforming earlier approaches.
{"title":"Ligand Based Virtual Screening of Molecular Compounds in Drug Discovery Using GCAN Fingerprint and Ensemble Machine Learning Algorithm","authors":"R. Ani, O. S. Deepa, B. R. Manju","doi":"10.32604/csse.2023.033807","DOIUrl":"https://doi.org/10.32604/csse.2023.033807","url":null,"abstract":"The drug development process takes a long time since it requires sorting through a large number of inactive compounds from a large collection of compounds chosen for study and choosing just the most pertinent compounds that can bind to a disease protein. The use of virtual screening in pharmaceutical research is growing in popularity. During the early phases of medication research and development, it is crucial. Chemical compound searches are now more narrowly targeted. Because the databases contain more and more ligands, this method needs to be quick and exact. Neural network fingerprints were created more effectively than the well-known Extended Connectivity Fingerprint (ECFP). Only the largest sub-graph is taken into consideration to learn the representation, despite the fact that the conventional graph network generates a better-encoded fingerprint. When using the average or maximum pooling layer, it also contains unrelated data. This article suggested the Graph Convolutional Attention Network (GCAN), a graph neural network with an attention mechanism, to address these problems. Additionally, it makes the nodes or sub-graphs that are used to create the molecular fingerprint more significant. The generated fingerprint is used to classify drugs using ensemble learning. As base classifiers, ensemble stacking is applied to Support Vector Machines (SVM), Random Forest, Nave Bayes, Decision Trees, AdaBoost, and Gradient Boosting. When compared to existing models, the proposed GCAN fingerprint with an ensemble model achieves relatively high accuracy, sensitivity, specificity, and area under the curve. Additionally, it is revealed that our ensemble learning with generated molecular fingerprint yields 91% accuracy, outperforming earlier approaches.","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","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":"135563771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.32604/csse.2023.036320
J. S. Manjaly, T. Subbulakshmi
The default scheduler of Apache Hadoop demonstrates operational inefficiencies when connecting external sources and processing transformation jobs. This paper has proposed a novel scheduler for enhancement of the performance of the Hadoop Yet Another Resource Negotiator (YARN) scheduler, called the Adaptive Node and Container Aware Scheduler (ANACRAC), that aligns cluster resources to the demands of the applications in the real world. The approach performs to leverage the user-provided configurations as a unique design to apportion nodes, or containers within the nodes, to application thresholds. Additionally, it provides the flexibility to the applications for selecting and choosing which node’s resources they want to manage and adds limits to prevent threshold breaches by adding additional jobs as needed. Node or container awareness can be utilized individually or in combination to increase efficiency. On top of this, the resource availability within the node and containers can also be investigated. This paper also focuses on the elasticity of the containers and self-adaptiveness depending on the job type. The results proved that 15%–20% performance improvement was achieved compared with the node and container awareness feature of the ANACRAC. It has been validated that this ANACRAC scheduler demonstrates a 70%–90% performance improvement compared with the default Fair scheduler. Experimental results also demonstrated the success of the enhancement and a performance improvement in the range of 60% to 200% when applications were connected with external interfaces and high workloads.
{"title":"Performance Improvement through Novel Adaptive Node and Container Aware Scheduler with Resource Availability Control in Hadoop YARN","authors":"J. S. Manjaly, T. Subbulakshmi","doi":"10.32604/csse.2023.036320","DOIUrl":"https://doi.org/10.32604/csse.2023.036320","url":null,"abstract":"The default scheduler of Apache Hadoop demonstrates operational inefficiencies when connecting external sources and processing transformation jobs. This paper has proposed a novel scheduler for enhancement of the performance of the Hadoop Yet Another Resource Negotiator (YARN) scheduler, called the Adaptive Node and Container Aware Scheduler (ANACRAC), that aligns cluster resources to the demands of the applications in the real world. The approach performs to leverage the user-provided configurations as a unique design to apportion nodes, or containers within the nodes, to application thresholds. Additionally, it provides the flexibility to the applications for selecting and choosing which node’s resources they want to manage and adds limits to prevent threshold breaches by adding additional jobs as needed. Node or container awareness can be utilized individually or in combination to increase efficiency. On top of this, the resource availability within the node and containers can also be investigated. This paper also focuses on the elasticity of the containers and self-adaptiveness depending on the job type. The results proved that 15%–20% performance improvement was achieved compared with the node and container awareness feature of the ANACRAC. It has been validated that this ANACRAC scheduler demonstrates a 70%–90% performance improvement compared with the default Fair scheduler. Experimental results also demonstrated the success of the enhancement and a performance improvement in the range of 60% to 200% when applications were connected with external interfaces and high workloads.","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","volume":"48 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":"135563782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.32604/csse.2023.041011
Sarah M. Alhammad, Doaa Sami Khafaga, Aya Y. Hamed, Osama El-Koumy, Ehab R. Mohamed, Khalid M. Hosny
Face mask detection has several applications, including real-time surveillance, biometrics, etc. Identifying face masks is also helpful for crowd control and ensuring people wear them publicly. With monitoring personnel, it is impossible to ensure that people wear face masks; automated systems are a much superior option for face mask detection and monitoring. This paper introduces a simple and efficient approach for masked face detection. The architecture of the proposed approach is very straightforward; it combines deep learning and local binary patterns to extract features and classify them as masked or unmasked. The proposed system requires hardware with minimal power consumption compared to state-of-the-art deep learning algorithms. Our proposed system maintains two steps. At first, this work extracted the local features of an image by using a local binary pattern descriptor, and then we used deep learning to extract global features. The proposed approach has achieved excellent accuracy and high performance. The performance of the proposed method was tested on three benchmark datasets: the real-world masked faces dataset (RMFD), the simulated masked faces dataset (SMFD), and labeled faces in the wild (LFW). Performance metrics for the proposed technique were measured in terms of accuracy, precision, recall, and F1-score. Results indicated the efficiency of the proposed technique, providing accuracies of 99.86%, 99.98%, and 100% for RMFD, SMFD, and LFW, respectively. Moreover, the proposed method outperformed state-of-the-art deep learning methods in the recent bibliography for the same problem under study and on the same evaluation datasets.
{"title":"Fast and Accurate Detection of Masked Faces Using CNNs and LBPs","authors":"Sarah M. Alhammad, Doaa Sami Khafaga, Aya Y. Hamed, Osama El-Koumy, Ehab R. Mohamed, Khalid M. Hosny","doi":"10.32604/csse.2023.041011","DOIUrl":"https://doi.org/10.32604/csse.2023.041011","url":null,"abstract":"Face mask detection has several applications, including real-time surveillance, biometrics, etc. Identifying face masks is also helpful for crowd control and ensuring people wear them publicly. With monitoring personnel, it is impossible to ensure that people wear face masks; automated systems are a much superior option for face mask detection and monitoring. This paper introduces a simple and efficient approach for masked face detection. The architecture of the proposed approach is very straightforward; it combines deep learning and local binary patterns to extract features and classify them as masked or unmasked. The proposed system requires hardware with minimal power consumption compared to state-of-the-art deep learning algorithms. Our proposed system maintains two steps. At first, this work extracted the local features of an image by using a local binary pattern descriptor, and then we used deep learning to extract global features. The proposed approach has achieved excellent accuracy and high performance. The performance of the proposed method was tested on three benchmark datasets: the real-world masked faces dataset (RMFD), the simulated masked faces dataset (SMFD), and labeled faces in the wild (LFW). Performance metrics for the proposed technique were measured in terms of accuracy, precision, recall, and F1-score. Results indicated the efficiency of the proposed technique, providing accuracies of 99.86%, 99.98%, and 100% for RMFD, SMFD, and LFW, respectively. Moreover, the proposed method outperformed state-of-the-art deep learning methods in the recent bibliography for the same problem under study and on the same evaluation datasets.","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","volume":"10 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":"135563956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.32604/csse.2023.036455
Mohammad Yamin, Sarah B. Basahel, Saleh Bajaba, Mona Abusurrah, E. Lydia
{"title":"Leveraging Retinal Fundus Images with Deep Learning for Diabetic Retinopathy Grading and Classification","authors":"Mohammad Yamin, Sarah B. Basahel, Saleh Bajaba, Mona Abusurrah, E. Lydia","doi":"10.32604/csse.2023.036455","DOIUrl":"https://doi.org/10.32604/csse.2023.036455","url":null,"abstract":"","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","volume":"120 1","pages":"1901-1916"},"PeriodicalIF":2.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89019370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.32604/csse.2023.034509
Babangida Isyaku, K. AbuBakar, W. Nagmeldin, Abdelzahir Abdelmaboud, Faisal Saeed, Fuad A. Ghaleb
{"title":"Reliable Failure Restoration with Bayesian Congestion Aware for Software Defined Networks","authors":"Babangida Isyaku, K. AbuBakar, W. Nagmeldin, Abdelzahir Abdelmaboud, Faisal Saeed, Fuad A. Ghaleb","doi":"10.32604/csse.2023.034509","DOIUrl":"https://doi.org/10.32604/csse.2023.034509","url":null,"abstract":"","PeriodicalId":50634,"journal":{"name":"Computer Systems Science and Engineering","volume":"5 1","pages":"3729-3748"},"PeriodicalIF":2.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90506370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：