Semi-fragile self-recoverable watermarking scheme for face image protection

Li, Chunlei; Yang, Ruimin; Liu, Zhoufeng; Li, Jianjun; Guo, Zhenduo

doi:10.1016/j.compeleceng.2016.01.026

Cited by 10 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [22,23,23], three methods are proposed for the protection of a special region of an image. In all these papers the image is categorized to the region of important (ROI) and the region of background (ROB).…”

Section: Related Workmentioning

confidence: 99%

“…In paper [22], the ROI and ROB are determined manually. But, the papers [23,23] are used for protection of face or faces in the image and the detection of faces are automatically using face detection algorithms. Recovery watermark is calculated from the ROB region and it is embedded into the ROI region.…”

Section: Related Workmentioning

confidence: 99%

“…As seen in the related works of this section, the amount of extracted bits for all regions of the image is equal. Although, some papers such as [22,23,24] have paid attention to the amount of extracted bits for the special ROI regions. Aside from the amount of extracted bits for different regions, the embedding regions is another important problem that is not taken into consideration by the state-of-the-art methods.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Semi-fragile Tamper Detection and Recovery based on Region Categorization and Two-Sided Circular Block Dependency

Soleymani

2018

Preprint

View full text Add to dashboard Cite

This paper presents a new semi-fragile algorithm for image tamper detection and recovery, which is based on region attention and two-sided circular block dependency. This method categorizes the image blocks into three categories according to their texture. In this method, less information is extracted from areas with the smooth texture, and more information is extracted from areas with the rough texture. Also, the extracted information for each type of blocks is embedded in another block with the same type. So, changes in the smooth areas are invisible to Human Visual System. To increase the localization power a two-sided circular block dependency is proposed, which is able to distinguish partially destroyed blocks. Pairwise block dependency and circular block dependency, which are common methods in the block-based tamper detection, are not able to distinguish the partially destroyed blocks. Cubic interpolation is used in order to decrease the blocking effects in the recovery phase. The results of the proposed method for regions with different texture show that the proposed method is superior to non-region-attention based methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation