Fast SSIM Index for Color Images Employing Reduced-Reference Evaluation

Bhateja, Vikrant; Srivastava, Aastha; Kalsi, Aseem

doi:10.1007/978-3-319-02931-3_51

Cited by 29 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image reconstruction is also possible by using this concept. Future working in this domain will require deployment of image quality assessment (IQA) tools for performance monitoring and benchmarking the restoration quality [17]- [22].…”

Section: Resultsmentioning

confidence: 99%

An Innovative Approach to Show the Hidden Surface by Using Image Inpainting Technique

Sharma¹,

Agarwal²

2015

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Abstract. The research study presented in this paper, focuses on the problems and lacunas of existing image inpainting techniques and shows how proposed approach will prove to be a curing syrup to diseased concepts, available so far for image inpainting. Since, the paper is highlighting image inpainting technique's drawbacks so the former part discusses what actually image inpainting technique means and where this great revolutionary need have its implementations and applications in the real world. Further, the purpose of image inpainting with various existing and latest algorithms/methods which are available so far, to inpaint an image are highlighted as a part of literature survey. The prime focus is to discuss, the innovative approach of the authors to remove disadvantage of existing image inpainting techniques i.e. if an object is small in size and is hidden behind a bigger object then by available inpainting techniques it is next to impossible to generate the image of hidden object as if bigger front object is selected as target region, then whole object along with the hidden object (behind bigger object) will also be removed during the time of object removal phase of inpainting. So, in final phase of paper, various descriptive images and live examples, methodology of whole proposed technology and self-proposed algorithms are discussed to remove this lacuna of the available inpainting techniques. Besides all, the resultant image will have the bigger front object getting transparent and only hidden smaller object as visible on the background image by implementation of proposed concept.

show abstract

Section: Resultsmentioning

confidence: 99%

An Innovative Approach to Show the Hidden Surface by Using Image Inpainting Technique

Sharma¹,

Agarwal²

2015

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

show abstract

“…These are extracted from the subbands decomposed earlier using (1), (5) and (6). In this paper, IQA is performed using SSIM [15], reduced reference SSIM (RR SSIM) [37] and reduced reference fast SSIM (RR FSSIM) [29] along with ISIM. These are compared against the mean opinion score (MOS) obtained by the subjective analysis of images in mentioned databases [31,32].…”

Section: Resultsmentioning

confidence: 99%

“…The computation of standard deviation is time consuming as it requires a large number of addition and multiplication operations. In order to reduce the computational load in the estimation of standard deviation; gradient magnitude approach [29], [36] is applied. This is approximated as follows:…”

Section: Computation Of Global Distortion Measure (Q)mentioning

confidence: 99%

“…Later, Ma et al [25] proposed a reduced reference measure based on re-organization of discrete cosine transform (DCT) subbands using GGD. Few algorithms designed for color images performed well but only on fixed distortion types [26][27][28][29]. With the presented review of various IQA measures; reduced reference IQA has been highlighted as a reliable methodology not just for performance evaluation of algorithms rather as a tool to monitor QoS over faulty transmission channels in multimedia communication networks.…”

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Image similarity metric (ISIM): a reduced reference image quality assessment approach

Bhateja

Kalsi

Srivastava

2015

CSIT

Self Cite

View full text Add to dashboard Cite

Reduced reference image quality assessment (IQA) has been highlighted as a reliable methodology for the purpose of quality-of-service (QoS) monitoring in network visual communication applications; as it requires only selective features of the reference image where complete image cannot be accessed (as reference for IQA). The reduced reference IQA problem has been constrained towards extraction of appropriate features (of reference and transformed images) sensitive to various distortion types along with due coherence with human perception. This paper presents an improved reduced reference IQA approach named as 'ISIM: Image Similarity Metric' based on natural image statistics. The proposed IQA approach involves sub-band decomposition of acquired images for features extraction using wavelets. The extracted features are then used to obtain a global distortion measure (Q) as a function of luminance, contrast and divergence parameters. The computational complexity during features extraction is reduced by employing combination of integral image and gradient magnitude approaches. This distortion measure (Q) is then used to formulate ISIM metric as a linear regression with full-reference SSIM. The validation of the proposed metric is carried out by evaluating the image quality for various types of distortions from CSIQ and IVC databases. Simulation results yield a healthy correlation of proposed ISIM metric and the subjective human opinions along with minimal computational complexity in comparison to other state-of-art reduced reference IQA measures.

show abstract

“…Mathematically, PSNR can be expressed as: where: MAX I is the maximum possible pixel value of the image and MSE is the mean squared error. SSIM measures the structural and perceptual similarity between the original image and the filtered image [36]. It is represented as: Table I and II respectively for differing noise variance levels: 1, 3 Table I & II).…”

Section: B Performance Evaluationmentioning

confidence: 99%

Rayleigh based bilateral filtering for flash and no-flash image pairs

Sinha¹,

Bhateja²,

Sharma³

2015

2015 2nd International Conference on Signal Processing and Integrated Networks (SPIN)

Self Cite

View full text Add to dashboard Cite

Digital photography provides a quick and facile means to capture a pair of images i.e. flash and no-flash image. Digital images are often superimposed by noises at the acquisition stage depending upon the availability of lighting. Image denoising is a key technology to reduce the noise levels from the images while preserving the fine details and structures. Predominantly, conventional Bilateral filter is employed, providing better noise reduction capabilities with low noise density while preserving the edges. The spatial processing of pixels is reckoned by using a Gaussian function (symmetric and smoothly decaying function). This paper introduces a Bilateral filter which incorporates a Rayleigh distribution function in its domain filter to smoothen out the image pairs more effectively. This filter has been implemented on both the images (flash and no-flash) in the present work. Simulations are carried out on the images contaminated with different levels of Gaussian noise and are evaluated on the basis of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) as image quality parameters. Performance of the proposed filter has shown significant results when compared to the conventional Bilateral filter.

show abstract

Fast SSIM Index for Color Images Employing Reduced-Reference Evaluation

Cited by 29 publications

References 27 publications

An Innovative Approach to Show the Hidden Surface by Using Image Inpainting Technique

An Innovative Approach to Show the Hidden Surface by Using Image Inpainting Technique

Image similarity metric (ISIM): a reduced reference image quality assessment approach

Rayleigh based bilateral filtering for flash and no-flash image pairs

Contact Info

Product

Resources

About