Generative models have recently become a prominent research topic in the field of artificial intelligence. Among these models, Generative Adversarial Networks (GAN) have revolutionized the field of deep learning by enabling the production of high-quality synthetic data that is very similar to real-world data. However, the effectiveness of GANs largely depends on the size and quality of training data. In many real-world applications, collecting large amounts of high-quality training data is impractical, time-consuming, and expensive. Accordingly, in recent years, there has been intense interest in the development of GAN models that can work with limited data. These models are particularly useful in scenarios where available data is sparse, such as medical imaging, or in creative applications such as creating new works of art. In this study, we propose a GAN model that can learn from a single training image. Our model is based on the principle of multiple GANs operating sequentially at different scales. At each scale, the GAN learns the features of the training image in different dimensions and transfers them to the next GAN. Samples produced by the GAN at the finest scale are images that have the characteristics of the training image but have different realistic structures. In our model, we utilized a self-attention module to increase the realism and quality of the generated images. Additionally, we used a new scaling method to increase the success of the model. The quantitative and qualitative results we obtained from our experimental studies show that our model performs image generation successfully. In addition, we demonstrated the robustness of our model by testing its success in different image manipulation applications. As a result, our model can successfully produce realistic, high-quality, diverse images from a single training image, providing short training time, memory efficiency, and good training stability. Our model is flexible enough to be used in areas where limited data needs to be worked on.
生成模型最近已成为人工智能领域的一个重要研究课题。在这些模型中，生成对抗网络（GAN）通过生成与真实世界数据非常相似的高质量合成数据，在深度学习领域掀起了一场革命。然而，GAN 的有效性在很大程度上取决于训练数据的规模和质量。在现实世界的许多应用中，收集大量高质量的训练数据是不切实际、耗时且昂贵的。因此，近年来，人们对开发能使用有限数据的 GAN 模型产生了浓厚的兴趣。这些模型在可用数据稀少的场景（如医学成像）或创造性应用（如创作新艺术作品）中尤其有用。在本研究中，我们提出了一种可以从单张训练图像中学习的 GAN 模型。我们的模型基于多个 GAN 在不同尺度上依次运行的原理。在每个尺度上，GAN 从不同维度学习训练图像的特征，并将其传输给下一个 GAN。最细尺度的 GAN 生成的样本是具有训练图像特征但现实结构不同的图像。在我们的模型中，我们使用了一个自我关注模块来提高生成图像的真实性和质量。此外，我们还使用了一种新的缩放方法来提高模型的成功率。实验研究得出的定量和定性结果表明，我们的模型可以成功生成图像。此外，我们还通过测试模型在不同图像处理应用中的成功率，证明了模型的鲁棒性。因此，我们的模型可以成功地从单张训练图像生成逼真、高质量、多样化的图像，训练时间短、内存效率高、训练稳定性好。我们的模型足够灵活，可用于需要处理有限数据的领域。
{"title":"Tek görüntü üretimi için öz-dikkat modüllü koşulsuz üretken bir model","authors":"Eyyüp Yildiz, Erkan Yüksel, Selcuk Sevgen","doi":"10.28948/ngumuh.1367602","DOIUrl":"https://doi.org/10.28948/ngumuh.1367602","url":null,"abstract":"Generative models have recently become a prominent research topic in the field of artificial intelligence. Among these models, Generative Adversarial Networks (GAN) have revolutionized the field of deep learning by enabling the production of high-quality synthetic data that is very similar to real-world data. However, the effectiveness of GANs largely depends on the size and quality of training data. In many real-world applications, collecting large amounts of high-quality training data is impractical, time-consuming, and expensive. Accordingly, in recent years, there has been intense interest in the development of GAN models that can work with limited data. These models are particularly useful in scenarios where available data is sparse, such as medical imaging, or in creative applications such as creating new works of art. In this study, we propose a GAN model that can learn from a single training image. Our model is based on the principle of multiple GANs operating sequentially at different scales. At each scale, the GAN learns the features of the training image in different dimensions and transfers them to the next GAN. Samples produced by the GAN at the finest scale are images that have the characteristics of the training image but have different realistic structures. In our model, we utilized a self-attention module to increase the realism and quality of the generated images. Additionally, we used a new scaling method to increase the success of the model. The quantitative and qualitative results we obtained from our experimental studies show that our model performs image generation successfully. In addition, we demonstrated the robustness of our model by testing its success in different image manipulation applications. As a result, our model can successfully produce realistic, high-quality, diverse images from a single training image, providing short training time, memory efficiency, and good training stability. Our model is flexible enough to be used in areas where limited data needs to be worked on.","PeriodicalId":508079,"journal":{"name":"Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi","volume":"13 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139272091","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}
There are many materials used to stabilize or reinforcement low strength soils. While additives such as fly ash, lime, tuff, silica fume are preferred in stabilization works, fibers are used in reinforcement works. The use of fiber, which has emerged as an alternative to traditional methods, has become widespread. The most preferred fiber types in soil reinforcement are glass, polypropylene and basalt fibers. Basalt fiber, which is produced from basalt rock and widely distributed in nature, was chosen as a reinforcement material in this study, which is increasingly used due to its high strength, economic, environmentally friendly, natural and many other superior properties. The aim of this study is to examine the effects of basalt fiber ratio on strength in low plasticity kaolin clay reinforced with basalt fiber. For this purpose, unconfined pressure tests were carried out on cylindrical samples prepared by adding 6 mm long basalt fiber at different rates to clay and compressing it at optimum water content. In the samples reinforced with basalt fiber, the maximum strength was obtained in the sample with 2% basalt fiber ratio. It was determined that the strength decreased in samples with higher basalt fiber content.
{"title":"Farklı oranlardaki bazalt fiber takviyesinin kaolinin serbest basınç dayanımına etkisi","authors":"Yasemin ASLAN TOPÇUOĞLU, Zülfü Gürocak","doi":"10.28948/ngumuh.1352665","DOIUrl":"https://doi.org/10.28948/ngumuh.1352665","url":null,"abstract":"There are many materials used to stabilize or reinforcement low strength soils. While additives such as fly ash, lime, tuff, silica fume are preferred in stabilization works, fibers are used in reinforcement works. The use of fiber, which has emerged as an alternative to traditional methods, has become widespread. The most preferred fiber types in soil reinforcement are glass, polypropylene and basalt fibers. Basalt fiber, which is produced from basalt rock and widely distributed in nature, was chosen as a reinforcement material in this study, which is increasingly used due to its high strength, economic, environmentally friendly, natural and many other superior properties. The aim of this study is to examine the effects of basalt fiber ratio on strength in low plasticity kaolin clay reinforced with basalt fiber. For this purpose, unconfined pressure tests were carried out on cylindrical samples prepared by adding 6 mm long basalt fiber at different rates to clay and compressing it at optimum water content. In the samples reinforced with basalt fiber, the maximum strength was obtained in the sample with 2% basalt fiber ratio. It was determined that the strength decreased in samples with higher basalt fiber content.","PeriodicalId":508079,"journal":{"name":"Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi","volume":"55 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139279157","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}
During coal production, a significant amount of fine coal dust is generated. The utilization of these high-ash coals is of great importance for the national economy. This study aims to recover the slime of the Lavvar facility located in the Kütahya-Tunçbilek region by flotation. The experimental parameters studied include the collector dosage, dispersant dosage, frother dosage, flotation time, air flow rate, and solid ratio. Flotation performance was evaluated by determining the ash, calorific value, sulfur, volatile matter, fixed carbon, and yield values of the products obtained at the end of the experiment. As a result of the experimental studies, the highest yield value was obtained using 2000 g/t sodium silicate, 2000 g/t diesel oil, and 100 g/t MIBC with a solid ratio of 10%, a flotation time of 4 minutes, and an air flow rate of 4 L/min, producing marketable coal with 6312 KCal/kg calorific value, 0.95% sulfur, and 22.4% ash content.
{"title":"Flotation of low-rank coal slimes","authors":"V. Önen, Mustafa Kutkan Karaoğlan","doi":"10.28948/ngumuh.1324175","DOIUrl":"https://doi.org/10.28948/ngumuh.1324175","url":null,"abstract":"During coal production, a significant amount of fine coal dust is generated. The utilization of these high-ash coals is of great importance for the national economy. This study aims to recover the slime of the Lavvar facility located in the Kütahya-Tunçbilek region by flotation. The experimental parameters studied include the collector dosage, dispersant dosage, frother dosage, flotation time, air flow rate, and solid ratio. Flotation performance was evaluated by determining the ash, calorific value, sulfur, volatile matter, fixed carbon, and yield values of the products obtained at the end of the experiment. As a result of the experimental studies, the highest yield value was obtained using 2000 g/t sodium silicate, 2000 g/t diesel oil, and 100 g/t MIBC with a solid ratio of 10%, a flotation time of 4 minutes, and an air flow rate of 4 L/min, producing marketable coal with 6312 KCal/kg calorific value, 0.95% sulfur, and 22.4% ash content.","PeriodicalId":508079,"journal":{"name":"Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139281886","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：