Developing operational performance metrics using image comparison metrics and the concept of degradation space

Sayood

2002

J. Electron. Imaging

530

Abstract. In this work we comprehensively categorize image quality measures, extend measures defined for gray scale images to their multispectral case, and propose novel image quality measures. They are categorized into pixel difference-based, correlation-based, edge-based, spectral-based, context-based and human visual system (HVS)-based measures. Furthermore we compare these measures statistically for still image compression applications. The statistical behavior of the measures and their sensitivity to coding artifacts are investigated via analysis of variance techniques. Their similarities or differences are illustrated by plotting their Kohonen maps. Measures that give consistent scores across an image class and that are sensitive to coding artifacts are pointed out. It was found that measures based on the phase spectrum, the multiresolution distance or the HVS filtered mean square error are computationally simple and are more responsive to coding artifacts. We also demonstrate the utility of combining selected quality metrics in building a steganalysis tool.

“…24 Note that there could be more than one fundamental metric. Similarly, the metric that responds adequately to all sorts of distortion effects is denoted as the ''global metric.''…”

Section: Anova Resultsmentioning

confidence: 99%

“…An alternative use of image quality metrics is in inverse mapping from metrics to the nature of distortions. 24 In other words, given the image quality metrics, one tries to reconstruct the distortions ͑e.g., the amount of blur, noise, etc., in distortion coordinates͒ that could have resulted in the measured metric values.…”

Section: Introductionmentioning

confidence: 99%

Statistical evaluation of image quality measures

Sayood

2002

J. Electron. Imaging

530

“…Image quality measurement continues to be the subject of intensive research and experimentation [5,10,11]. Objective measures are also utilized in performance prediction of vision algorithms against quality loss due to sensor inadequacy or compression artifacts [8]. The interest in developing objective measures for assessing multimedia data lies in the fact that subjective measurements are costly, timeconsuming and not easily reproducible.…”

Section: Choice Of Image Quality Measuresmentioning

confidence: 99%

<title>Steganalysis of watermarking techniques using image quality metrics</title>

Memon²,

2001

SPIE Proceedings

In this paper, we present techniques for steganalysis of images that have been potentially subjected to a watermarking algorithm. Our hypothesis is that a particular watermarking scheme leaves statistical evidence or structure that can be exploited for detection with the aid of proper selection of image features and multivariate regression analysis. We use some sophisticated image quality metrics as the feature set to distinguish between watermarked and unwatermarked images. To identify specific quality measures, which provide the best discriminative power, we use analysis of variance (ANOVA) techniques. The multivariate regression analysis is used on the selected quality metrics to build the optimal classifier using images and their blurred versions. The idea behind blurring is that the distance between an unwatermarked image and its blurred version is less than the distance between a watermarked image and its blurred version. Simulation results with a specific feature set and a well-known and commercially available watermarking technique indicates that our approach is able to accurately distinguish between watermarked and unwatermarked images.

“…Objective quality measures have been utilized in coding artifact evaluation, performance prediction of vision algorithms, quality loss due to sensor inadequacy etc. [13]. In this paper, however, we want to exploit image quality measures, not as predictors of subjective image quality or algorithmic performance, but specifically as a steganalysis tool, that is, as features in detecting watermarks or hidden messages.…”

Section: Choice Of Image Quality Measuresmentioning

confidence: 99%

“…Thus an image can be divided into blocks of size , say 32 32, and block wise spectral distortions can be computed. Let the DFT of the th block of the th band image be (12) where and , or in the magnitude-phase form (13) Then the following measures can be defined in the transform domain over the th block (14) (15) (16) with the relative weighting factor of the magnitude and phase spectra. Among possible rank order operations on the block spectral differences the median has proven useful.…”

Section: Spectral Measuresmentioning

confidence: 99%

Steganalysis using image quality metrics

Memon²,

IEEE Trans. on Image Process.

2003

461

218

Abstract-We present techniques for steganalysis of images that have been potentially subjected to steganographic algorithms, both within the passive warden and active warden frameworks. Our hypothesis is that steganographic schemes leave statistical evidence that can be exploited for detection with the aid of image quality features and multivariate regression analysis. To this effect image quality metrics have been identified based on the analysis of variance (ANOVA) technique as feature sets to distinguish between cover-images and stego-images. The classifier between cover and stego-images is built using multivariate regression on the selected quality metrics and is trained based on an estimate of the original image. Simulation results with the chosen feature set and wellknown watermarking and steganographic techniques indicate that our approach is able with reasonable accuracy to distinguish between cover and stego images.