A novel hashing scheme for Depth-image-based-rendering 3D images

详细信息    查看全文

• 作者：Chen Cui^aAuthor Vitae ; Haokun Mao^aAuthor Vitae ; Xiamu Niu^aAuthor Vitae ; Lixian Zhang^b ; ^c ; ^{lixianzhang@hit.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Tasawar Hayat^c ; ^dAuthor Vitae ; Ahmed Alsaedi^d
• 关键词：Depth-image-based rendering (DIBR) ; Image hashing ; Dual-tree complex wavelet transform (DT-CWT) ; Image identification
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：26 May 2016
• 年：2016
• 卷：191
• 期：Complete
• 页码：1-11
• 全文大小：4814 K

文摘

Image hashing as an advanced approach has been widely applied in many areas such as image authentication, and content-based identification and retrieval. Being different from the traditional 2D image hashing scheme, virtual images which are generated in the DIBR process should be considered with the same content as the center image in Depth-image-based rendering (DIBR) 3D image hashing scheme. In this paper, we propose a hashing method for DIBR 3D images by applying three-level Dual-tree complex wavelet transform (DT-CWT) to images and selecting suitable DT-CWT coefficients to form the final hash sequence. Experimental results indicate that the proposed method is robust to a wide range of distortions and attacks. When compared with the current state-of-the-art methods, the proposed method can achieve better identification performances in terms of geometric attacks such as rotation attacks, whilst providing comparable performances under classical distortions such as additive noise, blurring, and compression. Furthermore, the proposed method is also able to ensure that the generated virtual images could be identified with the same content as the corresponding center image due to the characteristic approximate shift invariance of DT-CWT.