• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.07a
• /
• pp.95-98
• /
• 2012

본 논문에서는 DCT 변환 공간상의 배경휘도와 주파수를 고려한 2차원의 개선된 광-적응 효과(luminance adaptation: LA) JND 모델을 제안한다. 기존의 LA JND 모델은 배경 휘도가 중간점인 회색에 가까울수록 JND가 낮고, 배경 휘도가 어두워지거나 밝아질수록 JND 값이 증가하는 U자형의 1차원 함수형태를 보였다. 그러나 기존 LA JND 모델은 주파수에 따른 영향이 반영되지 않았기 때문에 DCT와 같은 주파수 공간상 JND 모델로는 부정확 한 단점이 있다. 본 논문에서는 주파수와 배경휘도에 따른 2차원 LA JND 모델을 제안한다. 주파수에 따른 LA JND 값을 실제 실험을 통해 획득하였다. 실험 방법은 9가지 크기의 배경 휘도가 다르고 공간적 복잡도가 없는 균일한 영상을 대상으로 $8{\times}8$ 실수형 DCT를 수행한 다음, 15가지 경우의 주파수 크기가 다른 계수들에 대해 사람이 인지 할 때 까지 노이즈를 증가시켜서 JND 값을 찾는 방식을 사용하였다. 실험 결과 4 cpd(cycle per degree) 보다 작은 주파수 대역 에서는 기존의 LA JND 모델과 유사한 결과를 얻었지만 4 cpd보다 큰 주파수 대역에서는 오히려 배경휘도가 작은 값을 가질수록 JND가 감소하는 형태를 보였다. 수행한 실험 결과를 반영하여 주파수가 반영된 2차원 LA JND 모델을 제안한다.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.1
• /
• pp.22-27
• /
• 2015

Transform-domain JND (Just Noticeable Difference)-based for PVC (Perceptual Video Coding) is often performed in quantization processes to effectively remove perceptual redundancy. This study examined the JND-suppression effects on quantized coefficients of transform in HEVC (High Efficiency Video Coding). To reveal the JND-suppression effect in quantization, the properties of the floor functions were used for modeling the quantized coefficients, and a JND-adjustment process in an HEVC-compliant PVC scheme was used to tune the JND values by analyzing the JND suppression effect. In the experimental results, the bitrate reduction decreases slightly, but the PSNR and perceptual quality are improved significantly when the proposed JND adjustment process is applied.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.1080-1099
• /
• 2019

A better tradeoff between robustness and invisibility will be realized by using the just noticeable (JND) model into the quantization-based watermarking scheme. The JND model is usually used to describe the perception characteristics of human visual systems (HVS). According to the research of cognitive science, HVS can adaptively extract the structure features of an image. However, the existing JND models in the watermarking scheme do not consider the structure features. Therefore, a novel JND model is proposed, which includes three aspects: contrast sensitivity function, luminance adaptation, and contrast masking (CM). In this model, the CM effect is modeled by analyzing the direction features and texture complexity, which meets the human visual perception characteristics and matches well with the spread transform dither modulation (STDM) watermarking framework by employing a new method to measure edge intensity. Compared with the other existing JND models, the proposed JND model based on structural regularity is more efficient and applicable in the STDM watermarking scheme. In terms of the experimental results, the proposed scheme performs better than the other watermarking scheme based on the existing JND models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.438-455
• /
• 2024

Recently, many unsupervised learning-based models have emerged for Just Noticeable Difference (JND) estimation, demonstrating remarkable improvements in accuracy. However, these models suffer from a significant drawback is that their heavy reliance on handcrafted priors for guidance. This restricts the information for estimating JND simply extracted from regions that are highly related to handcrafted priors, while information from the rest of the regions is disregarded, thus limiting the accuracy of JND estimation. To address such issue, on the one hand, we extract the information for estimating JND in an Adversarial Complementary Learning (ACoL) way and propose an ACoL-JND network to estimate the JND by comprehensively considering the handcrafted priors-related regions and non-related regions. On the other hand, to make the handcrafted priors richer, we take two additional priors that are highly related to JND modeling into account, i.e., Patterned Masking (PM) and Contrast Masking (CM). Experimental results demonstrate that our proposed model outperforms the existing JND models and achieves state-of-the-art performance in both subjective viewing tests and objective metrics assessments.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.5
• /
• pp.147-154
• /
• 2015

In this paper, we employed a model which defines the sensitivity according to the background luminance, so called JND (Just Noticeable Difference), and applied to the video coding. The proposed method finds out the maximum possible quantization parameter for the current unit based on the threshold of JND model and reduce the bitrate with similar perceptual quality. It selects the higher quantization parameter and reduce the bitrate when the reconstructed signal which is coded with higher quantization parameter is in a range of allowance based on the JND threshold, i.e. the signal has the similar perceptual quality compared to that is coded with the initial quantization parameter. The proposed algorithm was implemented on HM16.0, which is a reference software of the latest video coding standard HEVC (High Efficiency Video Coding) and the coding performance was evaluated. Compared to HM16.0, the proposed algorithm achieved maximum 20.21% and 6.18% of average bitrate reduction with the similar perceptual quality.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.1C
• /
• pp.48-55
• /
• 2009

In scalable video coding, MCTF plays an important role for time-scalability and SNR-scalability. But there is image quality decreasing as MCTF level is increased because time interval of each frame is extended so that is hard to find suitable motion vector. In this paper, we propose an algorithm to prevent image quality from decreasing with unsuitable motion vector during MCTF update process using JND. We adapt JND to find errors within blocks of image and set a threshold which is used to add high frequency components during update process. We can overcome time-gap between frames and achieve better image quality through the proposed algorithm.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.213-216
• /
• 2018

본 논문에서는 사람의 인지 시각 특성 중 하나인 JND(Just Noticeable Difference)를 이용한 인지 비디오 부호화 기법을 제안한다. JND 기반 인지 부호화 방법은 사람의 인지 시각 특성을 이용해 시각적으로 인지가 잘 되지 않는 인지 신호를 제거함으로 부호화 효율을 높이는 방법이다. 제안된 방법은 기존 수학적 모델 기반의 JND 기법이 아닌 최근 각광 받고 있는 데이터 중심(data-driven) 모델링 방법인 심층 신경망 기반 JND 모델 생성 기법을 제안한다. 제안된 심층 신경망 기반 JND 모델은 비디오 부호화 과정에서 입력 영상에 대한 전처리를 통해 입력 영상의 인지 중복(perceptual redundancy)를 제거하는 역할을 수행한다. 부호화 실험에서 제안된 방법은 동일하거나 유사한 인지화질을 유지한 상태에서 평균 16.86 %의 부호화 비트를 감소 시켰다.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.135-140
• /
• 2008

최근 관련 기술의 발전과 디지털 컨버전스의 가속화로 모바일 제품이 소형화, 다기능화 되어 가고 있다. 이러한 경향에 따라, 다양한 기능을 지원하기 위한 높은 사양의 디스플레이가 요구되고 있으며, 많은 업체들이 고화질, 고해상도 디스플레이의 개발에 경쟁적으로 매진하고 있다. 그러나 사용자의 인지적 특성을 고려하지 않은 고해상도 경쟁은 생산비용만 높이는 결과를 초래할 수 있다. 본 연구는 디스플레이의 설계 요소별 선호도를 분석하고, 사람이 탐지할 수 있는 두 자극 간의 최소한의 차이역(difference threshold)인 JND(Just Noticeable Difference)을 활용하여, 설계 요소의 인지적 특성을 파악하는 것을 목적으로 한다. 이를 위하여 모바일 제품에 주로 사용되는 TFT-LCD 를 평가 대상으로, 30 명의 피실험자를 대상으로 JND 측정실험을 수행하였으며, 실험결과를 바탕으로 디스플레이에 대한 주요 설계변수들의 특성을 파악하였다. 이 연구결과는 사용자의 선호도를 고려한 모바일 제품의 디스플레이 설계지침으로 활용될 수 있을 것이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.8
• /
• pp.3935-3949
• /
• 2017

In this paper, a novel multiple description image coding (MDC) scheme is proposed, which is based on the characteristics of the human visual model. Due to the inherent characteristics of human vision, the human eye can only perceive the change of the specific thresholds, that is, the just noticeable difference (JND) thresholds. Therefore, JND model is applied to improve MDC syetem. This paper calculates the DCT coefficients firstly, and then they are compared with the JND thresholds. The data that is less than the JND thresholds can be neglected, which will improve the coding efficiency. Compared with other existing methods, the experimental results of the proposed method are superior.

• Journal of Broadcast Engineering
• /
• v.21 no.6
• /
• pp.929-943
• /
• 2016

This paper proposes the rate control algorithm based on the S-JND (Saliency-Just Noticeable Difference) model for considering perceptual visual quality. The proposed rate control algorithm employs the S-JND model to simultaneously reflect human visual sensitivity and human visual attention for considering characteristics of human visual system. During allocating bits for CTU (Coding Tree Unit) level in a rate control, the bit allocation model calculates the S-JND threshold of each CTU in a picture. The threshold of each CTU is used for adaptively allocating a proper number of bits; thus, the proposed bit allocation model can improve perceptual visual quality. For performance evaluation of the proposed algorithm, the proposed algorithm was implemented on HM 16.9 and tested for sequences in Class B and Class C under the CTC (Common Test Condition) RA (Random Access), Low-delay B and Low-delay P case. Experimental results show that the proposed method reduces the bit-rate of 2.3%, and improves BD-PSNR of 0.07dB and bit-rate accuracy of 0.06% on average. We achieved MOS improvement of 0.03 with the proposed method, compared with the conventional method based on DSCQS (Double Stimulus Continuous Quality Scale).