Sungwook Youn

Hybrid Compression of Hyperspectral Images Based on PCA With Pre-Encoding Discriminant Information

It has been shown that image compression based on principal component analysis (PCA) provides good compression efficiency for hyperspectral images. However, PCA might fail to capture all the discriminant information of hyperspectral images, since features that are important for classification tasks may not be high in signal energy. To deal with this problem, we propose a hybrid compression method for hyperspectral images with pre-encoding discriminant information. A feature extraction method is first applied to the original images, producing a set of feature vectors that are used to generate feature images and then residual images by subtracting the feature-reconstructed images from the original ones. Both feature images and residual images are compressed and transmitted. Experiments on data from the Airborne Visible/Infrared Imaging Spectrometer sensor indicate that the proposed method provides better compression efficiency with improved classification accuracy than conventional compression methods.

Efficient multi-currency classification of CIS banknotes

In this paper, we propose a fast and efficient algorithm to classify multi-national banknote images using size information and multi-template correlation matching. Since different banknotes have different sizes, this information was considered to be an important characteristic. Using the size information, we generated a size map to group the banknotes. Then, we determined the discriminant areas of each banknote that have high correlations among the same kind of banknote and low correlations with different kinds of banknotes. Post-processing was applied to handle degradations such as writing, aging, etc. The algorithm was tested using 55 banknotes of 30 different denominations from five countries: KRW, USD, EUR, CNY, and RUB. The experimental results showed 100% classification accuracy for unsoiled banknotes and 99.8% classification accuracy for soiled banknotes. The average processing time was about 4.83ms per banknote.

Towards standardized 3DTV QoE assessment: cross-lab study on display technology and viewing environment parameters

Subjective assessment of Quality of Experience in stereoscopic 3D requires new guidelines for the environmental setup as existing standards such as ITU-R BT.500 may no longer be appropriate. A first step is to perform cross-lab experiments in different viewing conditions on the same video sequences. Three international labs performed Absolute Category Rating studies on a freely available video database containing degradations that are mainly related to video quality degradations. Different conditions have been used in the labs: Passive polarized displays, active shutter displays, differences in viewing distance, the number of parallel viewers, and the voting device. Implicit variations were introduced due to the three different languages in Sweden, South Korea, and France. Although the obtained Mean Opinion Scores are comparable, slight differences occur in function of the video degradations and the viewing distance. An analysis on the statistical differences obtained between the MOS of the video sequences revealed that obtaining an equivalent number of differences may require more observers in some viewing conditions. It was also seen that the alignment of the meaning of the attributes used in Absolute Category Rating in different languages may be beneficial. Statistical analysis was performed showing influence of the viewing distance on votes and MOS results.

Edge Detection for Hyperspectral Images Using the Bhattacharyya Distance

In this paper, we proposed a new edge detection algorithm for hyper spectral images using the Bhattacharyya distance. First, the principal component analysis is applied to hyper spectral images and then dominant eigenimages are selected. To compute the Bhattacharyya distance, four block pairs of each pixel are extracted: up-down, left-right, diagonal-left down and diagonal-right-down. From each pair of blocks, we compute the Bhattacharyya distance, which was used as edge information. Experiments show promising results compared to the conventional Sobel filter.

Perceptual Video Quality Comparison of Various 3D Video Formats and Displays

Abstract As 3D imaging and display technologies have advanced, several 3D video formats have been developed and different kinds of 3D displays have been made available in the market. There have been efforts to standardize the signaling and data format for 3D content delivery. In this regard, the quality of experience (QoE) of 3D programs has emerged as an important issue. In order to investigate the impact of different 3D representation formats and different display technologies on perceptual video quality, a number of subjective video quality tests were conducted using various 3D representation formats and different stereoscopic displays. In particular, the side-by-side (SBS), top–bottom (TB) and frame-sequential formats were tested using active displays and passive displays. The experimental results were analyzed and it is observed that the relative perceptual 3D video quality was consistent across the 3D video formats and displays.

Paper check image quality enhancement with Moire reduction

Paper checks may have complex background features including fine lines and patterns, which make forgery more difficult. Also, halftoning techniques are used to produce continuous tones and to prevent copies with void pantograph features. When these kinds of checks are scanned, Moire patterns may occur. These patterns make it difficult for customers to examine the scanned check images on ATM (Automated Teller Machine) displays. They also can decrease the classification accuracy of check recognition systems. In this paper, we propose an algorithm to enhance the perceptual quality of scanned check images by reducing the Moire patterns. The proposed algorithm consists of foreground extraction, Moire detection and Moire removal. Subjective image quality assessment was performed to evaluate the degree of improvement. Experimental results show that the proposed algorithm improves perceptual quality while maintaining check recognition accuracy.

Effects of compression on classification performance and discriminant information preservation in remotely sensed data

A number of compression methods have been used to compress hyperspectral images. However, these methods may fail to retain all the discriminant characteristics of hyperspectral images since some discriminant features may not be high in signal energy. Also, it has been reported that compression may improve classification performance in some cases. In this paper, we investigate these problems, and analyze potential discriminant information loss and classification performance improvement. We perform some experiments using various compression methods. We examine this phenomenon and its implication.

Objective perceptual quality assesment and depth perception of 3D media

In this paper, we investigate the applicability of existing objective perceptual video quality measurement methods, which were developed for 2D video, to measure the perceptual quality of 3D video programs. We tested two methods: a reduced-reference HDTV model (ITU-T Rec. J.342) and a hybrid NR method (ITU-T Rec. J.343). As more mobile devices support autostereoscopic displays, crosstalk emerges as an important issue. We also investigate the effects of crosstalk on depth perception. Experimental results show promising performance and it appears that 2D video quality measurement methods can be used for 3D contents with potential improvements by taking into account 3D characteristics.

Two-stage compression of hyperspectral images with enhanced classification performance

Most compression methods for hyperspectral images have been optimized to minimize mean squared errors. However, this kind of compression method may not retain all discriminant information, which is important if hyperspectral images are to be used to distinguish among classes. In this paper, we propose a two-stage compression method for hyperspectral images with encoding residual discriminant information. In the proposed method, we first apply a compression method to hyperspectral images, producing compressed image data. From the compressed image data, we produce reconstructed images. Then we generate residual images by subtracting the reconstructed images from the original images. We also apply a feature extraction method to the original images, which produces a set of feature vectors. By applying these feature vectors to the residual images, we generate discriminant feature images which provide the discriminant information missed by the compression method. In the proposed method, these discriminant feature images are also encoded. Experiments with AVIRIS data show that the proposed method provides better compression efficiency and improved classification accuracy than other compression methods.

Remote logo detection using angle-distance histograms

Among all the various computer vision applications, automatic logo recognition has drawn great interest from industry as well as various academic institutions. In this paper, we propose an angle-distance map, which we used to develop a robust logo detection algorithm. The proposed angle-distance histogram is invariant against scale and rotation. The proposed method first used shape information and color characteristics to find the candidate regions and then applied the angle-distance histogram. Experiments show that the proposed method detected logos of various sizes and orientations.