Effective Document Image Rectification via a Deep Learning Framework
學年 112
學期 2
出版（發表）日期 2024-02-01
作品名稱 Effective Document Image Rectification via a Deep Learning Framework
作品名稱（其他語言）
著者 Hsiau Wen Lin; Hwei-Jen Lin; Yihjia Tsai; Yoshimasa Tokuyama; Chou-Wei Kong
單位
出版者
著錄名稱、卷期、頁數 Int. J. Pattern Recognit. Artif. Intell. 38(2)，2351023:1-2351023:17 (2024)
摘要 This paper proposes an efficient method for rectifying distorted document images via deep learning, ultimately improving the legibility of graphics and text in documents. The framework comprises two interconnected UNets, working in tandem to predict a 3D coordinate map and a forward map for the input distorted document image, respectively. At the beginning of the process, a page mask is predicted and used as input to both U-Nets to help improve the performance of their tasks. In the last step, the predicted forward map is transformed into a corresponding backward map, which is utilized to rectify the distorted image. The experimental results not only reveal that the predicted page masks and 3D coordinate maps significantly enhance the accuracy of predicting forward maps for subsequent rectification but also demonstrate satisfactory results both globally and locally.
關鍵字 Document image rectification;deep learning;UNet;skip connection;convolutional neural network;forward map;backward map
語言 en_US
ISSN 17936381 02180014
期刊性質 國外
收錄於 SCI SSCI EI
產學合作
通訊作者
審稿制度 否
國別 TWN
公開徵稿
出版型式 ,電子版,紙本
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/126783 )