Perceptual Visual Security Index Based on Edge and Texture Similarities

ABSTRACT:

With the development in recent decades of various efficient image encryption algorithms, such as selective encryption, a great demand has arisen for methods of evaluating the visual security of encrypted images. Existing solutions usually adopt well-known metrics of visual quality assessment to measure the quality of encrypted images, but they often exhibit undesired behavior on perceptually encrypted images of low quality. In this paper, we propose a novel visual security index (VSI) based on the human visual system. The proposed VSI evaluates two aspects of the content similarity between plain and encrypted images: the edge similarity extracted via multi-threshold edge detection and the texture similarity measured by means of the co-occurrence matrix. These two components are further integrated to obtain the proposed VSI through adaptive similarity weighting. Extensive experiments were performed on two publicly available image databases. Our experimental results demonstrate that compared with many existing state-of-the-art visual security metrics, the proposed VSI exhibits a better performance and stability on low-quality images.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

The existing solutions suffer many limitations. On the one hand, traditional cryptographic metrics are sometimes too strong for the security goals of image encryption; for example, in selective encryption, it is difficult or sometimes even unacceptable to require that the histogram of an encrypted image be fairly flat. On the other hand, although many metrics used in image processing are often applied in image encryption analysis, they may be inconsistent with the concept of security strength. Specifically, an assessment of poor visual quality in an encrypted image may not indicate a high security strength. For example, sometimes an image with a lower PSNR value may be even more recognizable than one with a higher PSNR value; please refer to Fig. 1 for details. The reason for this inconsistency is that most visual quality metrics are not designed for the entire visual quality range. Instead, they target and work well for relatively high-quality images but often exhibit poor performance on encrypted images, which are usually low in visual quality.

DISADVANTAGES OF EXISTING SYSTEM:

Edges obtained using a low threshold may contain more false edge information. It can be extremely fast because it selectively encrypts only a small portion of the image data in the spatial or transform domain.

PROPOSED SYSTEM:

The visual security assessment of low-quality images that are encrypted using perceptual encryption. A salient feature of encrypted images of this type is that the skeleton of the entire image is still discernible but the details are unintelligible. Considering the functionality of the human visual system (HVS), we analyze the impacts of edge and texture information on visualization and propose a visual security index (VSI) for measuring the visual security of encrypted images of low quality. This index is calculated based on the edge and texture similarities between the encrypted image and its original version. In other words, our proposed VSI measures the similarity between an encrypted image and a plain image from the perspective of the visual information one can obtain. Our contributions can be summarized as follows: We propose a VSI to measure the visual security of low-quality images encrypted via perceptual encryption. By using edge and texture features, we estimate the content similarity between the plain image and the lowquality encrypted image, which is a significantly different approach from that used in most existing work. We use a multi-threshold method of edge extraction to obtain a better approximation of the skeleton of a low-quality encrypted image. We also design a delicate mechanism to accurately measure the edge similarity between plain and encrypted images. texture similarities. We use adaptive weights, instead of constant weights as in the existing literature, to determine the contributions of the edge and texture similarities. • We present systematic and extensive experiments conducted to evaluate the proposed VSI. Compared with the state-of-the-art metrics, the proposed VSI is found to be more consistent with the HVS, to be stable on a wide range of images, and to demonstrate outstanding performance on low-quality images.

ADVANTAGES OF PROPOSED SYSTEM:

contribution of textures to visual security decreases as the strength of the perceptual encryption increases. VSI should exhibit consistent performance across different ranges of image quality (especially at low and moderate qualities).

Because the encryption (or distortion) strength of different images may vary, consistent performance across different quality ranges enables a VSI to accurately measure visual

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System            : Pentium Dual Core.

• Hard Disk       : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB

• Tool                      : MATLAB R2013A

REFERENCE:

Tao Xiang, Shangwei Guo, and Xiaoguo Li, “Perceptual Visual Security Index Based on Edge and Texture Similarities”, IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. 11, NO. 5, MAY 2016.