CAJ | 학술논문

针对低分辨率文档图像中噪声模型不确定、字符边缘和纹理走向复杂多变的问题，提出Geman＆McClure（G＆M）范数替代L1、 L2范数用于提高算法的鲁棒性，设计了结合双边全变分（BTV）和Hu-ber函数的正则化项，采用Lucas-Kanade光流配准算法，利用字符结构特征的先验信息，使算法在重建过程中更加注重边缘细节与边缘方向信息。实验表明，与L1BTV、L2BTV和无Huber函数的G＆MBTV正则化（下文简称G＆M方法）重建方法相比，文中算法在混合噪声模型下能够显著平滑噪声、锐化边缘、提升文档图像字符的分辨率，字符识别率提高14.69％的同时运算时间缩短了29.34％。
침대저분변솔문당도상중조성모형불학정、자부변연화문리주향복잡다변적문제，제출Geman＆McClure（G＆M）범수체대L1、 L2범수용우제고산법적로봉성，설계료결합쌍변전변분（BTV）화Hu-ber함수적정칙화항，채용Lucas-Kanade광류배준산법，이용자부결구특정적선험신식，사산법재중건과정중경가주중변연세절여변연방향신식。실험표명，여L1BTV、L2BTV화무Huber함수적G＆MBTV정칙화（하문간칭G＆M방법）중건방법상비，문중산법재혼합조성모형하능구현저평활조성、예화변연、제승문당도상자부적분변솔，자부식별솔제고14.69％적동시운산시간축단료29.34％。
It has been a key problem in document image super-resolution which the noise model is uncer-tain, edge and texture of characters tend to complex and changeable .Geman&McClure ( G&M) norm in-stead of L1 and L2 norm was proposed and used to improve the robustness of the algorithm .A regulariza-tion item combining Bilateral Total Variation (BTV) and Huber function was designed with adopting Lu-cas-Kanade light flow registration algorithm which is highly compatible for the proposed algorithm .This algorithm uses full information of the characters structure characteristics to make the algorithm pays more attention to the edge details in the process of reconstruction , applies the edge direction information more efficiently, and promotes information fusion of a series of low resolution images .Finally, experiments cor-responding to L1BTV, L2BTV and G&M BTV regularization algorithm without Huber function were car-ried out, the results show that the algorithm under the G&M norm with combination of BTV and Huber function is much more excellent than other algorithms .In the environment of mixed noise model , docu-ment image super-resolution reconstruction can smooth the noise significantly , sharpen the edge and im-prove the resolution of characters in document image .The recognition rate of characters was improved by 14.69%, and the operation time of the proposed algorithm was shorten by 29.34%at the same time .