• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권8호
• /
• pp.2053-2067
• /
• 2023

This paper proposes two video quality assessment methods based on deep neural network. (i)The first method uses the IQF-CNN (convolution neural network based on image quality features) to build image quality assessment method. The LIVE image database is used to test this method, the experiment show that it is effective. Therefore, this method is extended to the video quality assessment. At first every image frame of video is predicted, next the relationship between different image frames are analyzed by the hysteresis function and different window function to improve the accuracy of video quality assessment. (ii)The second method proposes a video quality assessment method based on convolution neural network (CNN) and gated circular unit network (GRU). First, the spatial features of video frames are extracted using CNN network, next the temporal features of the video frame using GRU network. Finally the extracted temporal and spatial features are analyzed by full connection layer of CNN network to obtain the video quality assessment score. All the above proposed methods are verified on the video databases, and compared with other methods.

• Journal of Information Technology Applications and Management
• /
• 제25권2호
• /
• pp.41-52
• /
• 2018

This paper presents a method for accurate camera self-calibration based on SIFT Feature Detection and image quality assessment. We performed image quality assessment to select high quality images for the camera self-calibration process. We defined high quality images as those that contain little or no blur, and have maximum contrast among images captured within a short period. The image quality assessment includes blur detection and contrast assessment. Blur detection is based on the statistical analysis of energy and standard deviation of high frequency components of the images using Discrete Cosine Transform. Contrast assessment is based on contrast measurement and selection of the high contrast images among some images captured in a short period. Experimental results show little or no distortion in the perspective view of the images. Thus, the suggested method achieves camera self-calibration accuracy of approximately 93%.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2007년도 7th International Meeting on Information Display 제7권2호
• /
• pp.1494-1496
• /
• 2007

This paper deals with image quality analysis considering the impact of psychological factors involved in assessment. The attributes of image quality requirement were partitioned according to the visual perception characteristics and the preference of image quality were obtained by the factor analysis method. The features of image quality which support the subjective preference were identified, The adequacy of image is evidenced to be the top requirement issues to the display image quality improvement.

• 방송공학회논문지
• /
• 제16권2호
• /
• pp.319-330
• /
• 2011

스테레오스코픽 3D 영상의 객관적인 화질 평가를 위해서 지금까지는 왼쪽과 오른쪽 영상에 대해 각각 2D 영상의 화질을 측정, 평가하는 방법을 사용하였다. 하지만 이 방법은 각각의 영상에 대해서 별도로 화질을 평가해야 하는 불편함이 따랐다. 그래서 본 논문에서는 왼쪽과 오른쪽 영상을 중첩하여 만든 하나의 영상을 통하여 스테레오스코픽 3D 영상의 화질을 평가하는 방법을 제안하여, 보다 간편하고 빠르게 스테레오스코픽 3D 영상의 화질을 평가할 수 있도록 하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권2호
• /
• pp.347-365
• /
• 2013

The performance of finger-vein recognition methods is limited by camera optical defocusing, the light-scattering effect of skin, and individual variations in the skin depth, density, and thickness of vascular patterns. Consequently, all of these factors may affect the image quality, but few studies have conducted quality assessments of finger-vein images. Therefore, we developed a new finger-vein recognition method based on image quality assessment. This research is novel compared with previous methods in four respects. First, the vertical cross-sectional profiles are extracted to detect the approximate positions of vein regions in a given finger-vein image. Second, the accurate positions of the vein regions are detected by checking the depth of the vein's profile using various depth thresholds. Third, the quality of the finger-vein image is measured by using the number of detected vein points in relation to the depth thresholds, which allows individual variations of vein density to be considered for quality assessment. Fourth, by assessing the quality of input finger-vein images, inferior-quality images are not used for recognition, thereby enhancing the accuracy of finger-vein recognition. Experiments confirmed that the performance of finger-vein recognition systems that incorporated the proposed quality assessment method was superior to that of previous methods.

• 한국콘텐츠학회논문지
• /
• 제12권8호
• /
• pp.87-95
• /
• 2012

본 연구는 객관적인 화질 평가 요소들과 선호도를 기반으로 한 주관적 화질 평가 모형을 구축하여 감상자들의 화질에 대한 선호도를 객관적 요소들로 분석할 수 있도록 하였다. 즉 제조사들이 이해하기 쉬운 객관적 화질 평가 요소들을 선정하고, 이와 같은 요소들을 질문하는 문항을 기반으로 하여 사진의 품질을 평가하는 방식으로 주관적 화질 평가 모형을 구축하였다. 또한 화질 평가에 사용되는 실제 사진을 일반인들이 주로 촬영하는 장면인 인물사진으로 선택하여 실험결과의 일반화와 타당성을 추구하였다. 본 실험의 주관적 화질 평가 모형을 통하여 감상자들이 선호하는 화질을 평가하고 그 결과가 최종적인 사진의 선호도에 어떠한 상관관계를 갖으며 영향을 미치는지 분석하였다. 이와 같은 상관관계 분석을 통하여 감상자가 선호하는 화질에 대해 파악하고 화질을 향상시키는 요소를 분석할 수 있었다. 그 결과 선호도와 가장 상관관계가 높은 변수는 색재현력, 다이내믹 레인지, 노이즈, 해상도 순이었으며, 인물 사진 자극의 특성상 색재현력과 선호도가 가장 높은 수준의 상관관계를 보였다. 이러한 결과는 단순 수치화된 객관적 화질 평가 요소들을 언어로 확장시켜, 선호도에 기반을 둔 주관적 화질 평가 모형을 제시함으로써 일반 사용자와 제조사 모두가 쉽게 이해할 수 있는 새로운 방식의 접근이라고 할 수 있다.

• 전기전자학회논문지
• /
• 제16권3호
• /
• pp.182-188
• /
• 2012

본 논문에서는 그래디언트 히스토그램을 기반으로 하는 영상 품질 평가 알고리즘을 제안하였다. 이는 목표 영상의 그래디언트 영상을 히스토그램으로 나타낼 경우 영상의 특성을 잘 나타낸다는 장점을 이용하였다. 제안한 방법에서 영상의 품질은 목표 영상에서 얻어진 그래디언트 히스토그램의 기울기에 의해 평가되고, 그래디언트 히스토그램을 대표하는 선의 기울기는 RANSAC (Random Sample Consensus)에 의해 측정된다. LIVE 영상 품질 평가 데이터베이스를 사용한 실험 결과를 통하여 제안한 알고리즘이 현존하는 다른 알고리즘에 비해 실제 사람의 영상에 대한 평가와 유사하다는 것을 확인할 수 있다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.628-630
• /
• 2003

Synthetic Aperture Radar(SAR) is an active microwave instrument that performs high-resolution observation under almost all weather condition. Research and algorithms have been proposed to process radar signal and to increase the quality of SAR products. In fact, many complicated steps are involved in order to generate a SAR image product. The purpose of this paper is to derive quality assessment procedures and define important test parameters in each procedure inside a SAR processor. Thus those test parameter values indicate the quality of SAR image products and verify the processor's performance. Moreover, required procedures to correct and handle errors which are indicated during the assessment are also presented.

• Journal of Information Processing Systems
• /
• 제13권3호
• /
• pp.448-463
• /
• 2017

Multimedia is a ubiquitous and indispensable part of our daily life and learning such as audio, image, and video. Objective and subjective quality evaluations play an important role in various multimedia applications. Blind image quality assessment (BIQA) is used to indicate the perceptual quality of a distorted image, while its reference image is not considered and used. Blur is one of the common image distortions. In this paper, we propose a novel BIQA index for Gaussian blur distortion based on the fact that images with different blur degree will have different changes through the same blur. We describe this discrimination from three aspects: color, edge, and structure. For color, we adopt color histogram; for edge, we use edge intensity map, and saliency map is used as the weighting function to be consistent with human visual system (HVS); for structure, we use structure tensor and structural similarity (SSIM) index. Numerous experiments based on four benchmark databases show that our proposed index is highly consistent with the subjective quality assessment.

• Journal of Multimedia Information System
• /
• 제9권2호
• /
• pp.75-86
• /
• 2022

The existing image quality assessment (IQA) datasets have a small number of samples. Some methods based on transfer learning or data augmentation cannot make good use of image quality-related features. A No Reference (NR)-IQA method based on multi-task training and quality awareness is proposed. First, single or multiple distortion types and levels are imposed on the original image, and different strategies are used to augment different types of distortion datasets. With the idea of weak supervision, we use the Full Reference (FR)-IQA methods to obtain the pseudo-score label of the generated image. Then, we combine the classification information of the distortion type, level, and the information of the image quality score. The ResNet50 network is trained in the pre-train stage on the augmented dataset to obtain more quality-aware pre-training weights. Finally, the fine-tuning stage training is performed on the target IQA dataset using the quality-aware weights to predicate the final prediction score. Various experiments designed on the synthetic distortions and authentic distortions datasets (LIVE, CSIQ, TID2013, LIVEC, KonIQ-10K) prove that the proposed method can utilize the image quality-related features better than the method using only single-task training. The extracted quality-aware features improve the accuracy of the model.