Pub Date : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934759
Narimen Hafsi, H. Hachimi, D. Benterki
For a long time, optimization has been part of our lives and the most recent literature shows a tremendous increase of the number of articles using Revolutionary algorithms in particular Firefly algorithm (FA) and Genetic algorithm. This tendency can be observed nearly in all areas of Computer Sciences and Engineering domain. Some of them are hybridized with other techniques to discover better performance. In addition, literatures found that most of the cases that used (FA) and (GA) techniques have outperformed compare to other metaheuristic algorithms. And because of the extraordinary impact of the COVID-19 pandemic on society and business as a whole, the pandemic generated an increase in the number and range of cybercriminal attacks due to the extensive use of computer networks. As result, new risks have arisen, and improving the speed and accuracy of security mechanisms has become a critical need. The aim of this article is to give the main mechanisme of those approachs and their application alone and hybrided to solve cybercrime problems.
{"title":"A New Hybrid Genetic with Firefly Algorithm for Solving Cyber-Criminal's Attacks","authors":"Narimen Hafsi, H. Hachimi, D. Benterki","doi":"10.1109/ICOA55659.2022.9934759","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934759","url":null,"abstract":"For a long time, optimization has been part of our lives and the most recent literature shows a tremendous increase of the number of articles using Revolutionary algorithms in particular Firefly algorithm (FA) and Genetic algorithm. This tendency can be observed nearly in all areas of Computer Sciences and Engineering domain. Some of them are hybridized with other techniques to discover better performance. In addition, literatures found that most of the cases that used (FA) and (GA) techniques have outperformed compare to other metaheuristic algorithms. And because of the extraordinary impact of the COVID-19 pandemic on society and business as a whole, the pandemic generated an increase in the number and range of cybercriminal attacks due to the extensive use of computer networks. As result, new risks have arisen, and improving the speed and accuracy of security mechanisms has become a critical need. The aim of this article is to give the main mechanisme of those approachs and their application alone and hybrided to solve cybercrime problems.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125811641","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934349
Abdelmjid Benmerrous, L. S. Chadli, A. Moujahid, M. Elomari, S. Melliani
The purpose of this work is to establishes the existence and uniqueness of the Schrödinger problem solution in the extended Colombeau algebra $G_{e}$. Then we look at the association notion in conjunction with the classic solution.
{"title":"Solution of Schrödinger type Problem in Extended Colombeau Algebras","authors":"Abdelmjid Benmerrous, L. S. Chadli, A. Moujahid, M. Elomari, S. Melliani","doi":"10.1109/ICOA55659.2022.9934349","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934349","url":null,"abstract":"The purpose of this work is to establishes the existence and uniqueness of the Schrödinger problem solution in the extended Colombeau algebra $G_{e}$. Then we look at the association notion in conjunction with the classic solution.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116449339","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934503
Oumaima Bouakline, Y. El Merabet, Kenza Khomsi
With the continuous development of the economy and its industrial activities, air pollution has become a serious problem. Therefore, it is absolutely necessary to develop a very accurate air quality forecasting model. In This paper, ten years of records of air pollution parameters and meteorological observations were used to forecast one-daily ahead of PM10 (particulate matters with a diameter less than $10 mumathrm{m}$) for two stations in Casablanca city, Morocco. Recurrent deep learning models namely: Long short-term memory (LSTM), Recurrent Neural Network (RNN), and Gated Recurrent Unit (GRU) are proposed. All of these nonlinear models were tuned using the genetic algorithm (GA) technique, which performed well. Among various combinations of predictors, the EFS (Exhaustive feature selection) method selected the best combination of predictors based on statistical scores mainly MSE. The analysis of the three prediction results shows approximately a similar performance. Interestingly, good scores were observed in terms of Pearson correlation coefficient (r), coefficient of determination (R), mean absolute error (MAE), and root mean squared error (RMSE), allowing decision-makers to anticipate the PM10 ground-level accurately.
随着经济和工业活动的不断发展，空气污染已成为一个严重的问题。因此，开发一种非常精确的空气质量预报模型是十分必要的。本文利用10年的空气污染参数记录和气象观测资料，对摩洛哥卡萨布兰卡市两个站点的PM10(直径小于$10 mu mathm {m}$的颗粒物)提前一天进行了预报。提出了长短期记忆(LSTM)、递归神经网络(RNN)和门控递归单元(GRU)等递归深度学习模型。采用遗传算法对非线性模型进行了优化，取得了良好的效果。在各种预测因子组合中，EFS(穷举特征选择)方法根据统计分数(主要是MSE)选择最佳预测因子组合。对三种预测结果的分析表明，三种预测结果的性能大致相似。有趣的是，在Pearson相关系数(r)、决定系数(r)、平均绝对误差(MAE)和均方根误差(RMSE)方面观察到良好的得分，使决策者能够准确地预测PM10地面水平。
{"title":"Deep-Learning models for daily PM10 forecasts using feature selection and genetic algorithm","authors":"Oumaima Bouakline, Y. El Merabet, Kenza Khomsi","doi":"10.1109/ICOA55659.2022.9934503","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934503","url":null,"abstract":"With the continuous development of the economy and its industrial activities, air pollution has become a serious problem. Therefore, it is absolutely necessary to develop a very accurate air quality forecasting model. In This paper, ten years of records of air pollution parameters and meteorological observations were used to forecast one-daily ahead of PM10 (particulate matters with a diameter less than $10 mumathrm{m}$) for two stations in Casablanca city, Morocco. Recurrent deep learning models namely: Long short-term memory (LSTM), Recurrent Neural Network (RNN), and Gated Recurrent Unit (GRU) are proposed. All of these nonlinear models were tuned using the genetic algorithm (GA) technique, which performed well. Among various combinations of predictors, the EFS (Exhaustive feature selection) method selected the best combination of predictors based on statistical scores mainly MSE. The analysis of the three prediction results shows approximately a similar performance. Interestingly, good scores were observed in terms of Pearson correlation coefficient (r), coefficient of determination (R), mean absolute error (MAE), and root mean squared error (RMSE), allowing decision-makers to anticipate the PM10 ground-level accurately.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128880034","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934531
Nouhayla Ait Oussaid, Mourad El Ouali, Sultana Ben Aadi, Khalid Akhlil, Salma Gaou
In the present paper, we present a non-smooth optimization problem for the synaptic depression model. It involves the Laplace operator defined on locally finite graphs and the local Lipschitz function's Clarke subdifferential. Our basic goal is to show the existence of a weak solution to this problem by using a Galerkin-like method involving an exhaustion procedure. On the multivalued nonmonotone and nonconvex part, we assume the so-called Rauch condition, which expresses the nonmonotonicity of the nonlinearities.
{"title":"Nonsmooth Optimization for Synaptic Depression Dynamics","authors":"Nouhayla Ait Oussaid, Mourad El Ouali, Sultana Ben Aadi, Khalid Akhlil, Salma Gaou","doi":"10.1109/ICOA55659.2022.9934531","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934531","url":null,"abstract":"In the present paper, we present a non-smooth optimization problem for the synaptic depression model. It involves the Laplace operator defined on locally finite graphs and the local Lipschitz function's Clarke subdifferential. Our basic goal is to show the existence of a weak solution to this problem by using a Galerkin-like method involving an exhaustion procedure. On the multivalued nonmonotone and nonconvex part, we assume the so-called Rauch condition, which expresses the nonmonotonicity of the nonlinearities.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114426222","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934649
Issam El Khadiri, M. Abouelmajd, M. Zemzami, N. Hmina, M. Lagache, B. AlMangour, A. Bahlaoui, I. Arroub, S. Belhouideg
The integration of lattice design approaches and topology optimization is the next step in realizing optimal lattice designs for Additive Manufacturing (AM). Hence, this study focuses on a topological optimization method that allows to derive lattice structures from topological optimization results suitable for additive manufacturing such that TPMS lattices are designed to replace solid volumes. To find these lattice solutions, we used the intersected lattice. Simulation tests was carried out to verify the superior stiffness properties of the optimized intersected lattice compared to the basic design using a solid topology optimization solution.
{"title":"TPMS Lattice Structure derived using Topology Optimization for the Design of Additive Manufactured Components","authors":"Issam El Khadiri, M. Abouelmajd, M. Zemzami, N. Hmina, M. Lagache, B. AlMangour, A. Bahlaoui, I. Arroub, S. Belhouideg","doi":"10.1109/ICOA55659.2022.9934649","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934649","url":null,"abstract":"The integration of lattice design approaches and topology optimization is the next step in realizing optimal lattice designs for Additive Manufacturing (AM). Hence, this study focuses on a topological optimization method that allows to derive lattice structures from topological optimization results suitable for additive manufacturing such that TPMS lattices are designed to replace solid volumes. To find these lattice solutions, we used the intersected lattice. Simulation tests was carried out to verify the superior stiffness properties of the optimized intersected lattice compared to the basic design using a solid topology optimization solution.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"26 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128645060","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934291
Youssef Kossale, Mohammed Airaj, Aziz Darouichi
With the rise of a new framework known as Generative Adversarial Networks (GANs), generative models have gained considerable amount of attention in the area of unsupervised learning. GANs have been thoroughly studied since their emergence in 2014, leading to an enormous amount of new models and applications built on this said framework. Although despite their success, GANs suffer from some notorious problems during training, hindering further advances in the field. This paper seeks to highlight one of the most encountered problems in GAN training, namely the “Helvetica scenario” as stated by its authors or “mode collapse” as widely known. We will try to provide an overview of this said challenge, what is it, why it occurs, and some suggested workarounds to reduce its impact on training.
{"title":"Mode Collapse in Generative Adversarial Networks: An Overview","authors":"Youssef Kossale, Mohammed Airaj, Aziz Darouichi","doi":"10.1109/ICOA55659.2022.9934291","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934291","url":null,"abstract":"With the rise of a new framework known as Generative Adversarial Networks (GANs), generative models have gained considerable amount of attention in the area of unsupervised learning. GANs have been thoroughly studied since their emergence in 2014, leading to an enormous amount of new models and applications built on this said framework. Although despite their success, GANs suffer from some notorious problems during training, hindering further advances in the field. This paper seeks to highlight one of the most encountered problems in GAN training, namely the “Helvetica scenario” as stated by its authors or “mode collapse” as widely known. We will try to provide an overview of this said challenge, what is it, why it occurs, and some suggested workarounds to reduce its impact on training.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"214 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133770452","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934141
El Attaoui Anas, Norelislam El Hami
This study presents two population-based, nature-inspired optimization paradigms, named “Harris Hawks Optimization” HHO and “Ant Colony Optimization” ACO. The inspiration of HHO is the collaborative performance and chasing style of Harris' hawks in nature. Otherwise, ACO is inspired by studying the behaviour of real ants. Those two natural motions were scientifically represented to build optimization algorithms. The performance of HHO and ACO optimizers is checked throughout a comparison based on various test functions and an application of a problem called: Minimizing the cost of assigning personnel to a plant.
{"title":"Assignment problem solved by two metaheuristic algorithms ACO and HHO","authors":"El Attaoui Anas, Norelislam El Hami","doi":"10.1109/ICOA55659.2022.9934141","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934141","url":null,"abstract":"This study presents two population-based, nature-inspired optimization paradigms, named “Harris Hawks Optimization” HHO and “Ant Colony Optimization” ACO. The inspiration of HHO is the collaborative performance and chasing style of Harris' hawks in nature. Otherwise, ACO is inspired by studying the behaviour of real ants. Those two natural motions were scientifically represented to build optimization algorithms. The performance of HHO and ACO optimizers is checked throughout a comparison based on various test functions and an application of a problem called: Minimizing the cost of assigning personnel to a plant.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125716284","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934276
Adam El Khaldi, H. Hachimi
In a global context marked by climate change and by peak globalization materializing through highly interconnected and interdependent chains of value. Our subject is naturally aligned in a global dynamic characterized by existential issues and public-level challenges. This motivates us to contribute in this global reflection by proposing concrete and viable responses and measures to safeguard our planet and to perpetuate our heritage for future generations. It is our duty and collective responsibility. Indeed, the balance between environmental, economic and social impacts has become the optimum objective in all contemporary strategies, policies, actions and decisions. This balance is critical in logistics and transport field because it is one of the largest emitters of greenhouse gases directly causing the acceleration of global warming. And also because it's one of the most crucial and sensitive sectors for mankind due to the fact that our survival and comfort depend on it! Our aim is to build a decision aid tool in order to optimally choose a route (also called path) in a multimodal transport network of goods and/or passengers. The system should be as efficient as possible: with a path that causes the least damage and aggression to the environment while being economically and socially beneficial to man. For a multimodal transport network, a mathematical model is established in order to calculate the ecological and socio-economic criteria to be considered‥. Then a multi-objective optimization algorithm is built to find the shortest path by optimizing the defined criteria: An ant colony algorithm is chosen because it is the most optimal and efficient in a complex scenario that takes into account a large number of variable parameters and criteria. Naturally, an implementation on a multimodal transport network is carried out in order to assess the algorithm's performances. Finally, problematic questions are asked in order to incite reflection and explore future perspectives. And because of the subject's richness, it can be used as a starting point for further development and expansion.
{"title":"Optimized management of green supply chains by the use of Ant Colonies multi-objective algorithm: The integration of the economic, environmental and social impacts of multimodal transport","authors":"Adam El Khaldi, H. Hachimi","doi":"10.1109/ICOA55659.2022.9934276","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934276","url":null,"abstract":"In a global context marked by climate change and by peak globalization materializing through highly interconnected and interdependent chains of value. Our subject is naturally aligned in a global dynamic characterized by existential issues and public-level challenges. This motivates us to contribute in this global reflection by proposing concrete and viable responses and measures to safeguard our planet and to perpetuate our heritage for future generations. It is our duty and collective responsibility. Indeed, the balance between environmental, economic and social impacts has become the optimum objective in all contemporary strategies, policies, actions and decisions. This balance is critical in logistics and transport field because it is one of the largest emitters of greenhouse gases directly causing the acceleration of global warming. And also because it's one of the most crucial and sensitive sectors for mankind due to the fact that our survival and comfort depend on it! Our aim is to build a decision aid tool in order to optimally choose a route (also called path) in a multimodal transport network of goods and/or passengers. The system should be as efficient as possible: with a path that causes the least damage and aggression to the environment while being economically and socially beneficial to man. For a multimodal transport network, a mathematical model is established in order to calculate the ecological and socio-economic criteria to be considered‥. Then a multi-objective optimization algorithm is built to find the shortest path by optimizing the defined criteria: An ant colony algorithm is chosen because it is the most optimal and efficient in a complex scenario that takes into account a large number of variable parameters and criteria. Naturally, an implementation on a multimodal transport network is carried out in order to assess the algorithm's performances. Finally, problematic questions are asked in order to incite reflection and explore future perspectives. And because of the subject's richness, it can be used as a starting point for further development and expansion.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123858789","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934723
Malak Dargham, Hanaa Hachimi, M. Boutalline
The rise of robot writers could have a major impact on marketing and communication. AI is capable of writing marketing content, thanks to the advances in natural language generation (NLG). The software not only mimics human speech, but it can also use data and language to narrate events in real time.
{"title":"How AI is automating writing: The rise of robot writers","authors":"Malak Dargham, Hanaa Hachimi, M. Boutalline","doi":"10.1109/ICOA55659.2022.9934723","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934723","url":null,"abstract":"The rise of robot writers could have a major impact on marketing and communication. AI is capable of writing marketing content, thanks to the advances in natural language generation (NLG). The software not only mimics human speech, but it can also use data and language to narrate events in real time.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129594529","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 : 2022-10-06DOI: 10.1109/ICOA55659.2022.9934130
El Abassi Fouzia, Darouichi Aziz, Ouaarab Aziz
Deep learning is a subset of machine learning that encompasses a variety of neural network architectures used to perform diverse computer vision tasks such as medical image classification and segmentation, which are time-consuming, effortful, delicate, and extremely tedious for doctors. The high variability of shape, location, size and texture of the medical images as well as the noise and parasites that degrade the image quality present a big problem for the segmentation process, therefore, a various segmentation methods based on deep learning have been proposed in the literature to fully automated the segmentation process. At the same time, the large number of hyperparameters of a deep learning algorithm in general and of a convolutional neural network in particular presents a problem when developing an automatic segmentation system with an appropriate structure and hyperparameters. Metaheuristics are approximate optimization methods to solve this type of problems. In this study, we review the most used and efficient segmentation methods based on deep learning for medical images segmentation, their optimization with metaheuristics as well as we compared three deep CNN encoder-decoder architectures, namely FCN, SegNet and Unet. These architectures trained and tested on MRI (Magnetic resonance imaging) images in order to study each of those architectures, compare them and finally choose the most efficient model.
{"title":"Images Segmentation using Deep Learning Algorithms and Metaheuristics","authors":"El Abassi Fouzia, Darouichi Aziz, Ouaarab Aziz","doi":"10.1109/ICOA55659.2022.9934130","DOIUrl":"https://doi.org/10.1109/ICOA55659.2022.9934130","url":null,"abstract":"Deep learning is a subset of machine learning that encompasses a variety of neural network architectures used to perform diverse computer vision tasks such as medical image classification and segmentation, which are time-consuming, effortful, delicate, and extremely tedious for doctors. The high variability of shape, location, size and texture of the medical images as well as the noise and parasites that degrade the image quality present a big problem for the segmentation process, therefore, a various segmentation methods based on deep learning have been proposed in the literature to fully automated the segmentation process. At the same time, the large number of hyperparameters of a deep learning algorithm in general and of a convolutional neural network in particular presents a problem when developing an automatic segmentation system with an appropriate structure and hyperparameters. Metaheuristics are approximate optimization methods to solve this type of problems. In this study, we review the most used and efficient segmentation methods based on deep learning for medical images segmentation, their optimization with metaheuristics as well as we compared three deep CNN encoder-decoder architectures, namely FCN, SegNet and Unet. These architectures trained and tested on MRI (Magnetic resonance imaging) images in order to study each of those architectures, compare them and finally choose the most efficient model.","PeriodicalId":345017,"journal":{"name":"2022 8th International Conference on Optimization and Applications (ICOA)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115764536","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：