Object Removal Algorithm by using distributed boundary based on Seam Carving
碩士 === 國立臺灣科技大學 === 電子工程系 === 106 === Object removal is widely used to remove strangers from photographic portraits, or logos from images or videos, i.e., the process of removing target object(s) from digital images. This article proposes a seam carving based method to remove a target object. First,...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2018
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/7rgfk5 |
| Summary: | 碩士 === 國立臺灣科技大學 === 電子工程系 === 106 === Object removal is widely used to remove strangers from photographic portraits,
or logos from images or videos, i.e., the process of removing target object(s) from
digital images. This article proposes a seam carving based method to remove a target
object. First, we build the distributed boundary for the removing and inserting stages
to guide the seams to be found. Then we find all unnoticeable inserting seams to
enlarge the image, and find removing seams one by one to remove the object. After all
damaged areas have been removed, we adjust the luminance in a visually reasonable
way. Compared to previous proposals, the proposed method has good visual quality
and provides different resolutions for different devices. Experimental results
demonstrated that the proposed approach can remove the target object(s) without
destroying other object edges and adjust luminance in a visually meaningful manner.
|
|---|