A preliminary evaluation of CNNs’ performance of meteor detection using training data by generative AI towards FPGA implementation

• Item type: SIG Technical Reports(1)
• 公開日: 2024-06-03
• タイトル: A preliminary evaluation of CNNs’ performance of meteor detection using training data by generative AI towards FPGA implementation
• タイトル / 言語: en
• タイトル: A preliminary evaluation of CNNs’ performance of meteor detection using training data by generative AI towards FPGA implementation
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: AI・機械学習
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: University of Tsukuba
• 著者所属: University of Tsukuba
• 著者所属(英): en
• 著者所属(英): University of Tsukuba
• 著者所属(英): en
• 著者所属(英): University of Tsukuba
• 著者名: Yuping, Zheng; Kenji, Kanazawa
• 著者名(英): Yuping, Zheng; Kenji, Kanazawa
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Convolutional Neural Network (CNN) has been widely used for object detection in recent years and has achieved many amazing results. In the paper, we investigate the effectiveness of Convolutional Neural Networks (CNNs) for the task of meteor detection, specifically examining the impact of training these networks with datasets augmented by generative artificial intelligence (AI). Due to the limitations posed by the scarce availability of diverse training data, we use generative artificial intelligence techniques to generate synthetic meteor images as our training dataset. In the research, we aim to compare different CNNs’ effectiveness for meteor detection and enhance detection accuracy and to explore the feasibility of deploying these models on Field-Programmable Gate Arrays (FPGAs) for real-time applications.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Convolutional Neural Network (CNN) has been widely used for object detection in recent years and has achieved many amazing results. In the paper, we investigate the effectiveness of Convolutional Neural Networks (CNNs) for the task of meteor detection, specifically examining the impact of training these networks with datasets augmented by generative artificial intelligence (AI). Due to the limitations posed by the scarce availability of diverse training data, we use generative artificial intelligence techniques to generate synthetic meteor images as our training dataset. In the research, we aim to compare different CNNs’ effectiveness for meteor detection and enhance detection accuracy and to explore the feasibility of deploying these models on Field-Programmable Gate Arrays (FPGAs) for real-time applications.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10096105
• 書誌情報: 研究報告システム・アーキテクチャ（ARC）; 巻 2024-ARC-257, 号 11, p. 1-4, 発行日 2024-06-03
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8574
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会