Improved multi-scale composition based on high-quality candidate region extraction

详细信息    查看全文

• 作者：Yueli Li^a ; Rongfang Bie^b ; Yi Wang^c ; Hao Wu^b ; ^{10112056@bjtu.edu.cn}
• 关键词：Image composition ; Poisson equation ; Candidate region ; Wavelet transform
• 刊名：Optik - International Journal for Light and Electron Optics
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：131
• 期：Complete
• 页码：145-151
• 全文大小：1689 K
• 卷排序：131

文摘

Image composition is the process of first extracting a candidate region from a candidate image and then embedding this region into a target image. Traditional composite methods focus on reducing appearance gaps (boundary, brightness, color, and sharpness) between the candidate region and the target image. However, in the composite process, low-quality candidate region extraction negatively affects the composite results. In several complicated images, the composite results are not realistic, especially for the boundary. This study proposes an innovative algorithm that can solve the drawback of traditional methods We combined wavelet transform and high-quality candidate region extraction model to achieve composition. On one hand, segmentation and improved matting models are used for high-quality candidate region extraction. On the other hand, wavelet transform is used for multi-scale decomposition, which obviously improves the composite results. Experimental results using a large database show the superiority of the proposed method.