Hyunsoo Choi

Motion Adaptive Deinterlacing With Modular Neural Networks

In this letter, a motion adaptive deinterlacing algorithm based on modular neural networks is proposed. The proposed method uses different neural networks based on the amount of motion. Modular neural networks were selectively used depending on the differences between the adjacent fields. We also used motion vectors to select optimal input pixels from the adjacent fields. Motion estimation was used to find input blocks for the neural networks with minimum errors. Intra/inter-mode switching was employed to address inaccurate motion estimation problems.

Comparison of various subjective video quality assessment methods

In this paper, we present comparison of three subjective testing methods: the double stimulus continuous quality scale (DSCQS) method, the single stimulus continuous quality evaluation (SSCQE) method and the absolute category rating (ACR) method. The DSCQS method was used for validate objective models in the VQEG Phase II FRTV test. The SSCQE method is chosen to be used in the VQEG RRTV test. The ACR method is chosen to be used in the VQEG Multimedia test. Since a different subjective test method is used in each test, analyses of the three methods will provide helpful information in understanding human perception of video quality.

Neural Network Deinterlacing Using Multiple Fields and Field-MSEs

Generally, deinterlacing algorithms can be either classified as intra methods or inter methods. Intra methods interpolate missing lines by using surrounding pixels in the current field. Inter methods interpolate missing lines by using pixels and the motion information of multiple fields. Neural network deinterlacing that uses multiple fields has been proposed. It provides improved performance compared to existing neural network deinterlacing algorithms that use a single field. However, when adjacent fields are very different, neural network deinterlacing that uses multiple fields may not provide good performance. To address this problem, we propose using field-MSE values as additional inputs. These MSE values can provide helpful information so that the networks can consider field differences in using multiple fields. Experimental results show that the use of the proposed algorithm results in improved performance.

Neural Network Deinterlacing Using Multiple Fields

In this paper, we proposed a deinterlacing algorithm using neural networks for conversion of interlaced videos to progressive videos. The proposed method uses multiple fields: a previous field, a current field, and a next field. Since the proposed algorithm uses multiple fields, the neural network is able to take into account the motion pattern which might exists in adjacent fields. Experimental results demonstrate that the proposed algorithm provides better performances than existing neural network deinterlacing algorithms that uses a single field.

Comparison of subjective video quality assessment methods for multimedia applications

In this paper, we compared two subjective assessment methods DSCQS(Double Stimulus Continuous Quality Scale method) and ACR(Absolute Category Rating). These methods are widely used in order to evaluate video quality for multimedia application. We performed subjective quality tests using DSCQS and ACR methods. The subjective scores obtained by the DSCQS and ACR methods show that these methods are highly correlated in terms of MOS(Mean Opinion Score) and have slightly lower correlation in terms of DMOS(Difference Mean Opinion Score). The results indicate that ACR method is an effective subjective quality assessment method, which shows compatible performance with DSCQS method and can evaluate a larger number of video sequences.

Principal component analysis with pre-emphasis for compression of hyperspectral imagery

In this paper, we propose to use the principal component analysis for the compression of hyperspectral images. When hyperspectral images are compressed using conventional image compression algorithms, discriminant features of original data may be lost during compression process. In order to preserve such discriminant information, we first apply a linear feature extraction method to the original data. Then, we emphasize discriminant features and use the principal component analysis in order to compress the images whose discriminant features are enhanced. Experiments show that the proposed method provides improved classification accuracies than existing compression algorithms.

Video quality measurement for multimedia applications using reduced-reference signals

As more multimedia services have become increasingly available over networks where bandwidth is not always guaranteed, quality monitoring has become an important issue. For instance, quality of experience and quality monitoring have become important problems in internet protocol television applications, since transmission errors may introduce all kinds of additional video quality degradations. In this paper, we present a reduced-reference objective model for video quality measurements in multimedia applications. The proposed method first measures edge degradations that are critical for perceptual video quality and then considers transmission error effects. We compared the proposed method with some existing methods. Independent verifications confirmed that the proposed method showed good performance and consequently it was included in an International Telecommunication Union recommendation. The proposed method can be used to monitor video quality at receivers while requiring minimum usage of additional bandwidth.

No-reference image quality metric based on image classification

In this article, we present a new no-reference (NR) objective image quality metric based on image classification. We also propose a new blocking metric and a new blur metric. Both metrics are NR metrics since they need no information from the original image. The blocking metric was computed by considering that the visibility of horizontal and vertical blocking artifacts can change depending on background luminance levels. When computing the blur metric, we took into account the fact that blurring in edge regions is generally more sensitive to the human visual system. Since different compression standards usually produce different compression artifacts, we classified images into two classes using the proposed blocking metric: one class that contained blocking artifacts and another class that did not contain blocking artifacts. Then, we used different quality metrics based on the classification results. Experimental results show that each metric correlated well with subjective ratings, and the proposed NR image quality metric consistently provided good performance with various types of content and distortions.

Subjective video quality comparison using various displays

Objective video quality measurement has become an important issue, as multimedia services are now widely available over the Internet and other wireless communication media. Traditionally, professional CRT monitors have been used to measure subjective video quality. However, the majority of users have LCD, plasma display panel (PDP), or consumer-graded CRT monitors. We compared the subjective video quality of various TV and LCD PC monitors. Subjective tests were performed with a wide range of video sequences using different monitors, and their correlations were analyzed. Although there were high correlations among the various display monitors, care should be taken in selecting a monitor for certain applications.

Compression of hyperspectral images with 2D wavelet transform using adjacent information and SPIHT slgorithm

In this paper, we propose compression algorithms for hyperspectral images based on the wavelet transform using adjacent band information. In the proposed algorithm, we divide the spectral bands into a number of groups. Then, within each group, we select a representative spectral band which has highest correlations with the remaining spectral bands within the group. The representative image is compressed using the 2-dimensional SPIHT algorithm. A characteristic of the SPIHT algorithm is that it provides information on significant coefficients of transformed images, which have large energies. We explore this characteristic and use the information to encode adjacent band images of the representative image.