An Image Reconstruction Algorithm Based on Frequency Domain for Deep Subcooling of Melt Drops,ERICDATA高等教育知識庫
高等教育出版
熱門: 謝傳崇  羅文君  葉俊廷  紀金山  SDGs  簡淑芸  
高等教育出版
首頁 臺灣期刊   學校系所   學協會   民間出版   大陸/海外期刊   政府機關   學校系所   學協會   民間出版   DOI註冊服務
篇名
An Image Reconstruction Algorithm Based on Frequency Domain for Deep Subcooling of Melt Drops
並列篇名
An Image Reconstruction Algorithm Based on Frequency Domain for Deep Subcooling of Melt Drops
作者 Keqing NingZe SuZhihao ZhangGwang-jun Kim
英文摘要
In a space environment, liquid alloys are in a thermodynamically metastable state, which facilitates research on the material structure and thermophysical properties of deep undercooling melt. Limited by the cost and technology of performing experiments in space, researchers developed electrostatic levitation that uses a drop pipe device to simulate the space environment. A high-speed camera was used to capture the falling image of the deep undercooling melt to study the melting and solidification process. Due to the exposure time and hardware limitations of the image acquisition equipment, the image resolution of the deep undercooling melt is low, which is not conducive to studying the thermophysical properties and solidification interface of the melt. Software design methods, such as super-resolution reconstruction, can more accurately reconstruct image contour information and effectively improve the image resolution. The most current deep learning-based superresolution reconstruction algorithms directly perform Ychannel or Y, Cb, and Cr three-channel learning on the reconstructed image. This is insufficient in terms of providing more prior information to solve the superresolution reconstruction. In this study, a single-frame image super resolution reconstruction network that is based on frequency-domain feature learning is proposed. It builds a time–frequency transformation layer at the front end of the neural network and uses the frequency to realize the neural network in the frequency domain. To evaluate the super-resolution reconstruction performance, the proposed algorithm is compared with the current mainstream interpolation, sparse coding, super resolution convolutional neural network, and enhanced single-image super-resolution deep residual algorithms. The proposed algorithm achieves good reconstruction effects on deep undercooled melt images in terms of the objective evaluation and visual perception. At the same time, the peak signal-to-noise ratio and structural similarity index measure achieved results that exceed the aforementioned comparison algorithms.
起訖頁 1273-1285
關鍵詞 Deep undercooling meltSuper resolution reconstructionConvolutional neural networkFrequency domain learningFeature selection
刊名 網際網路技術學刊  
期數 202111 (22:6期)
出版單位 台灣學術網路管理委員會
DOI 10.53106/160792642021112206007  複製DOI
QR Code
該期刊
上一篇
Using PageRank Algorithm to Predict and Prevent Congestion in Software Defined Networking
該期刊
下一篇
Ruzicka Indexed Regressive Homomorphic Ephemeral Key Benaloh Cryptography for Secure Data Aggregation in WSN

高等教育知識庫  閱讀計畫  教育研究月刊  新書優惠  

教師服務
合作出版
期刊徵稿
聯絡高教
高教FB
讀者服務
圖書目錄
教育期刊
訂購服務
活動訊息
數位服務
高等教育知識庫
國際資料庫收錄
投審稿系統
DOI註冊
線上購買
高點網路書店 
元照網路書店
博客來網路書店
教育資源
教育網站
國際教育網站
關於高教
高教簡介
出版授權
合作單位
知識達 知識達 知識達 知識達 知識達 知識達
版權所有‧轉載必究 Copyright2011 高等教育文化事業有限公司  All Rights Reserved
服務信箱:edubook@edubook.com.tw 台北市館前路 26 號 6 樓 Tel:+886-2-23885899 Fax:+886-2-23892500