Blind Quality Assessment of Tone-Mapped Images Via Analysis of Information, Naturalness, and Structure

ABSTRACT:

High dynamic range (HDR) imaging techniques have been working constantly, actively, and validly in the fault detection and disease diagnosis in the astronomical and medical fields, and currently they have also gained much more attention from digital image processing and computer vision communities. While HDR imaging devices are starting to have friendly prices, HDR display devices are still out of reach of typical consumers. Due to the limited availability of HDR display devices, in most cases tone mapping operators (TMOs) are used to convert HDR images to standard low dynamic range (LDR) images for visualization. But existing TMO scan not work effectively for all kinds of HDR images, with their performance largely depending on brightness, contrast, and structure properties of a scene. To accurately measure and compare the performance of distinct TMOs, in this paper develop an effective and efficient no-reference objective quality metric which can automatically assess LDR images created by different TMOs without access to the original HDR images. Our model is shown to be statistically superior to recent full- and no-reference quality measures on the existing tone-mapped image database and a new relevant database built in this work.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

In our existing the first attempt at developing a lossy, backward compatible HDR video encoding with proper color handling. Since the problem of HDR lossy compression has already been partially addressed, especially for images, we briefly discuss existing solutions from the standpoint of their application to HDR video. The problem of dynamic range compression and expansion arises in many imaging pipelines with constrained bit depth at certain processing stages (only 6 bits per color channel is often used for DVD movies while displays can handle 8 bits/color channel). This may result in the loss of low amplitude signals and false contouring. Bit-depth expansion (BDE) techniques are designed specifi- cally to combat those effects and achieve higher perceived bit depth quality than are physically available. For example imperceptible spatiotemporal noise is added to an image prior to the quantization step in dither techniques When higher bit depth information is not available, low-amplitude details cannot be reconstructed, and processing is focused on removing false contours using adaptive filtering, predictive cancellation, spatial frequency channel coring techniques [Daly and Feng 2004]. All existing BDE and de-contouring techniques are optimized for much lower bit depth expansion than required to accommodate HDR image and video content. Furthermore, storing HDR video using 8-bit encoding with additional spatio-temporal dither is impractical because dither patterns do not compress well. These compression methods however require substantial bit-rates and are not suitable for on-DVD storage or real-time playback. The dynamic range level achieved with analog film and its digital emulation is too low for our purposes. Besides, we argue that the video encoding format should be designed for the capabilities of the human eye rather than analog film or camera characteristics.

DISADVANTAGES OF EXISTING SYSTEM:

Digital emulation is too low for our purposes the problem of HDR lossy compression has already been partially addressed, especially for images

PROPOSED SYSTEM:

In order to better visualize the differences between the high and low dynamic range images, the HDR shop offers an example, as provided Though we cannot easily show the differences between the two images in their darkened and brightened versions are able to show the differences. Some noticeably 2Actually, we distinguished regions are labeled with white rectangles in darkened Images find that the tone-mapped image, due to the limitation of dynamic range, cannot preserve all the information of the original HDR image. It is reasonable to suppose that a good tone-mapped image contains a great amount of information. On this basis, the first consideration of our metric assessing the quality of a tone-mapped image is to straightforwardly estimate the information volume in itself and the intermediate images created by darkening/brightening the original luminance Next, we need to seek for a way to measure the information amount. Information entropy, as an important concept in statistics, is an appropriate criterion. By computing the mean unpredictability of one random signal, entropy represents its disorderly degree. To compare the differences of two probability distributions, other important metrics, such as the Kullback-Leibler (K-L) divergence and its modified symmetric formats are also frequently employed information-theoretic “distances good complement to the existing TMID database, we have proposed a new tone-mapped image database (TMID2015) with up to 16 different TMOs in order to demonstrate the performance accuracy of IQA models across TMOs. Second, we have developed new and efficient features for measuring the information entropy of luminance-changed images. With this type of features, we have derived an effective NR IQA model that is statistically equivalent to the FR TMQI method.

ADVANTAGES OF PROPOSED SYSTEM:

high-accuracy blind quality measure, low computational cost. introduced to reduce the implementation time to a large extent.

MODULES:

1. Gray image

2. Intensity image

3. Bilateral filter

4. RGB components

5. Tone map

6. Image quality assessment

7. SSIM

8. PSNR

MODULES DESCRIPTION:

Gray Image:

In photography and computing, a grayscale or greyscale digital image is an image in which the value of each pixel is a single sample, that is, it carries only intensity information. Images of this sort, also known as black-and-white, are composed exclusively of shades of gray, varying from black at the weakest intensity to white at the strongest. Grayscale images are distinct from one-bit bi-tonal black-and-white images, which in the context of computer imaging are images with only two colors, black and white (also called bilevel or binary images). Grayscale images have many shades of gray in between.

Grayscale images are often the result of measuring the intensity of light at each pixel in a single band of the electromagnetic spectrum (e.g. infrared, visible light, ultraviolet, etc.), and in such cases they are monochromatic proper when only a given frequency is captured. But also they can be synthesized from a full color image; see the section about converting to grayscale.

Intensity image:

In grayscale images, it’s depicted by the grey level value at each pixel (e.g., 127 is darker than 220).

In the digital Image processing perception, the intensity of an image could refer to a global measure of that image, such as mean pixel intensity.

Contrast:

Contrast is the difference in luminance or color that makes an object (or its representation in an image or display) distinguishable.

In visual perception of the real world, contrastis determined by the difference in the color and brightness of the object and other objects within the same field of view.

Brightness:

Brightness is a relative term. It depends on your visual perception. Since brightness is a relative term, so brightness can be defined as the amount of energy output by a source of light relative to the source we are comparing it to.

Bilateral filter:

A bilateral filter is a non-linear, edge-preserving and noise-reducing smoothing filter for images. The intensity value at each pixel in an image is replaced by a weighted average of intensity values from nearby pixels. This weight can be based on a Gaussian distribution. Crucially, the weights depend not only on Euclidean distance of pixels, but also on the radiometric differences (e.g. range differences, such as color intensity, depth distance, etc.). This preserves sharp edges by systematically looping through each pixel and adjusting weights to the adjacent pixels accordingly.

RGB COMPONENTS:

The RGB color model is an additive color model in which red, green and blue light are added together in various ways to reproduce a broad array of colors. The name of the model comes from the initials of the three additive primary colors, red, green and blue. The main purpose of the RGB color model is for the sensing, representation and display of images in electronic systems, such as televisions and computers, though it has also been used in conventional photography. Before the electronic age, the RGB color model already had a solid theory behind it, based in human perception of colors.

RGB is a device-dependent color model: different devices detect or reproduce a given RGB value differently, since the color elements (such as phosphors or dyes) and their response to the individual R, G and B levels vary from manufacturer to manufacturer, or even in the same device over time. Thus an RGB value does not define the same color across devices without some kind of color management.

Tone map:

one mapping is a technique used in image processing and computer graphics to map one set of colors to another to approximate the appearance of high-dynamic-range images in a medium that has a more limited dynamic range. Print-outs, CRT or LCD monitors, and projectors all have a limited dynamic range that is inadequate to reproduce the full range of light intensities present in natural scenes. Tone mapping addresses the problem of strong contrast reduction from the scene radiance to the displayable range while preserving the image details and color appearance important to appreciate the original scene content.

Image Quality Assessment:

Image quality is a characteristic of an image that measures the perceived image degradation (typically, compared to an ideal or perfect image). Imaging systems may introduce some amounts of distortion or artifacts in the signal, so the quality assessment is an important problem. There are several techniques and metrics that can be measured objectively and automatically evaluated by a computer program. Therefore, they can be classified as full-reference (FR) methods and no-reference (NR) methods. In FR image quality assessment methods, the quality of a test image is evaluated by comparing it with a reference image that is assumed to have perfect quality. NR metrics try to assess the quality of an image without any reference to the original one.

SSIM:

The structural similarity (SSIM) index is a method for predicting the perceived quality of digital television and cinematic pictures, as well as other kinds of digital images and videos. It was first developed in the Laboratory for Image and Video Engineering (LIVE) at The University of Texas at Austin and in subsequent collaboration with New York University. SSIM is used for measuring the similarity between two images. The SSIM index is a full reference metric; in other words, the measurement or prediction of image quality is based on an initial uncompressed or distortion-free image as reference. SSIM is designed to improve on traditional methods such as peak signal-to-noise ratio (PSNR) and mean squared error (MSE), which have proven to be inconsistent with human visual perception.

PSNR:

Peak signal-to-noise ratio, often abbreviated PSNR, is an engineering term for the ratio between the maximum possible power of a signal and the power of corrupting noise that affects the fidelity of its representation. Because many signals have a very wide dynamic range, PSNR is usually expressed in terms of the logarithmic decibel scale.

PSNR is most commonly used to measure the quality of reconstruction of lossy compression codecs (e.g., for image compression). The signal in this case is the original data, and the noise is the error introduced by compression. When comparing compression codecs, PSNR is an approximation to human perception of reconstruction quality. Although a higher PSNR generally indicates that the reconstruction is of higher quality, in some cases it may not. One has to be extremely careful with the range of validity of this metric; it is only conclusively valid when it is used to compare results from the same codec (or codec type) and same content.

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:

Ke Gu, Shiqi Wang, Member, IEEE, Guangtao Zhai, Member, IEEE, Siwei Ma, Member, IEEE, Xiaokang Yang, Senior Member, IEEE, Weisi Lin, Fellow, IEEE, Wenjun Zhang, Fellow, IEEE, andWen Gao, Fellow, IEEE, “Blind Quality Assessment of Tone-Mapped Images Via Analysis of Information, Naturalness, and Structure”, IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 18, NO. 3, MARCH 2016.