Shadow removal from document image based on background estimation employing selective median filter and black-top-hat transform

被引:0
|
作者
Imahayashi, Sota [1 ]
Mukaida, Mashiho [2 ]
Takeda, Shuma [2 ]
Suetake, Noriaki [2 ]
机构
[1] Univ Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 3058577, Japan
[2] Yamaguchi Univ, 1677-1 Yoshida, Yamaguchi 7538512, Japan
关键词
Document image; Shadow removal method; Morphological operations; Median filter; Background image;
D O I
10.1007/s10043-023-00806-y
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
Recently, smartphone cameras have been increasingly used to take pictures of documents. When a shadow appears in the captured document image, it causes problems such as reduced visibility and difficulty to read the text because the colors of the text and the shadow are similar. For this reason, many methods have been proposed to remove shadows from document images. However, these methods produce artifacts in the borders of the shadow areas, and this makes it difficult to read characters. In this letter, we propose an effective method of removing shadows. With the proposed method, a selective median filter and black-top-hat transform are employed to estimate the background image and remove shadows. The effectiveness of the proposed method is confirmed by various experiments.
引用
收藏
页码:336 / 340
页数:5
相关论文
共 2 条
  • [1] Shadow removal from document image based on background estimation employing selective median filter and black-top-hat transform
    Sota Imahayashi
    Mashiho Mukaida
    Shuma Takeda
    Noriaki Suetake
    Optical Review, 2023, 30 : 336 - 340
  • [2] Improved Degraded Document Image Binarization Using Median Filter for Background Estimation
    Khitas, Mehdi
    Ziet, Lahcene
    Bouguezel, Saad
    ELEKTRONIKA IR ELEKTROTECHNIKA, 2018, 24 (03) : 82 - 87