Digital Image Inpainting Using Bilateral Filtering

  • سال انتشار: 1395
  • محل انتشار: اولین کنفرانس بین المللی دستاوردهای نوین پژوهشی در مهندسی برق و کامپیوتر
  • کد COI اختصاصی: CBCONF01_0279
  • زبان مقاله: انگلیسی
  • تعداد مشاهده: 923
دانلود فایل این مقاله

نویسندگان

Peyman Mahdavi

M.S. Student of Electrical and Computer engineering department Semnan University Semnan, Iran

Farzin Yaghmaee

Faculty Member of Electrical and Computer engineering faculty Semnan University Semnan, Iran

Vahid K Alilou

PhD. Candidate of Electrical and Computer engineering department Semnan University Semnan, Iran

چکیده

Digital image inpainting refers to the act of repairing damaged parts of an image using its known parts. Exemplar-based inpainting algorithms propagate both image structures and texture. Image structures have an intense impact on ipainting process. Therefore, it is important to detect image structures accurately. The concept of Isophote has been introduced in order to detect the image structures. However, Isophotes of an image have been only referred to its simple edges up to now and for detecting the image structures just a simple gradient function has been used in previous works. It should be noted that it is not necessary for all of the image edges to participate in the inpainting process. Because some of the image edges can mislead the inpainting process. To tackle this problem in this paper a new inpainting algorithm focusing on the linear structures of the image has been introduced. The proposed method is an exemplar-based algorithm based on Criminisi’s work. A Bilateral Filter is a smoothing filter, which preserves main image edges and removes excess edges. In this paper input images have been smoothened by using Bilateral Filtering. Then, the smoothened image has been transformed from RGB color space to YCbCr color space. Afterwards, Isophotes of the image have been extracted by using an edge detector on the Y channel of the smoothened image. The experimental results of the proposed method illustrate that the proposed algorithm preserves fundamental linear structures of the image and the inpainting quality has been significantly improved.

کلیدواژه ها

Digital Image Inpainting; Isophote; Bilateral Filter; Exemplar-Based algorithm; Image Linear Structure

مقالات مرتبط جدید

اطلاعات بیشتر در مورد COI

COI مخفف عبارت CIVILICA Object Identifier به معنی شناسه سیویلیکا برای اسناد است. COI کدی است که مطابق محل انتشار، به مقالات کنفرانسها و ژورنالهای داخل کشور به هنگام نمایه سازی بر روی پایگاه استنادی سیویلیکا اختصاص می یابد.

کد COI به مفهوم کد ملی اسناد نمایه شده در سیویلیکا است و کدی یکتا و ثابت است و به همین دلیل همواره قابلیت استناد و پیگیری دارد.