Bit-depth expansion using Minimum Risk Based Classification

Abstract

Bit-depth expansion is an art of converting low bit-depth image into high bit-depth image. Bit-depth of an image represents the number of bits required to represent an intensity value of the image. Bit-depth expansion is an important field since it directly affects the display quality. In this paper, we propose a novel method for bit-depth expansion which uses Minimum Risk Based Classification to create high bit-depth image. Blurring and other annoying artifacts are lowered in this method. Our method gives better objective (PSNR) and superior visual quality as compared to recently developed bitdepth expansion algorithms.

CNN-Based Suppression of False Contour and Color Distortion in Bit-Depth Enhancement

“…BDE is designed to eliminate as many false contours as possible in low-bit-depth input images when reconstructing bit-depth [ 10 , 11 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ]. Effective bit-depth enhancement algorithms are essential for energy-efficient VIoT based on bit-depth compression, but the existing BDE algorithms either cannot suppress false contours well or lose much detailed information, so they are not suitable for visual IoT scenarios.…”

“…It takes the q bits of the input image as a whole, periodically attaches it to the latest LSB until the bit-depth is not less than p , and then outputs the most significant p bits. MRC [ 11 ] models bit-depth extension as a minimum classification risk problem. MRC generates an error distribution function of all the possible estimation values and accepts the value that minimizes the associated risk.…”

“…There are already several BDE algorithms, including context-independent methods, such as Multiply by Ideal Gain (MIG) [ 10 ] and Minimum Risk-based Classification (MRC) [ 11 ]. This kind of methods do not make use of context information, so they are not good at suppressing false contours.…”

Bit Depth Expansion using Error Distribution

“…또한, 비트 심도 확장은 영 상의 밝기 범위인 다이나믹 레인지(Dyna-mic Range)를 높 여 영상의 표현력을 높여 줄 뿐만 아니라, 고해상도 영상의 표현이 가능하다는 장점이 존재한다 [6][7] [13] . [7] . 본 논문에서는 비트 심도 확장시 발생하 Daly's [2] Adaptive Filter based [2] Minimum Risk based [7] Flooding based [8,9] Daly's [2] Adaptive Filter based [2] Minimum Risk based [7] Flooding based [8,9] …”

“…[7] . 본 논문에서는 비트 심도 확장시 발생하 Daly's [2] Adaptive Filter based [2] Minimum Risk based [7] Flooding based [8,9] Daly's [2] Adaptive Filter based [2] Minimum Risk based [7] Flooding based [8,9] …”