Sangwook Baek

Depth perception estimation of various stereoscopic displays.

In this paper, we investigate the relationship between depth perception and several disparity parameters in stereoscopic images. A number of subjective experiments were conducted using various 3D displays, which indicate that depth perception of stereoscopic images is proportional to depth difference and is inversely related to the camera distance. Based on this observation, we developed some formulas to quantify the degree of depth perception of stereoscopic images. The proposed method uses depth differences and the camera distance between the objects and the 3D camera. This method also produces improved depth perception estimation by using non-linear functions whose inputs include a depth difference and a camera distance. The results show that the proposed method provides noticeable improvements in terms of correlation and produces more accurate depth perception estimations of stereoscopic images.

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.

Detection of counterfeit banknotes using multispectral images

Abstract In this paper, we propose counterfeit banknote detection algorithms using low resolution multispectral images. It has become increasingly difficult to detect professionally produced counterfeit banknotes, so more sophisticated features have had to be implemented in banknotes. However, sensors that are capable of reading these counter-fake features are rather expensive. On the other hand, multispectral images can be used to tackle the counterfeit banknote problem. Recently, multispectral sensors have been developed for ATM applications. We developed efficient counterfeit banknote detection algorithms and the proposed algorithms were tested using 20 different denominations of European Euro (EUR), Indian rupee (INR), and US Dollars (USD). The experimental results show that the proposed methods provided 99.8% classification accuracy for genuine banknotes and 100% detection accuracy for counterfeit banknotes.

Banknote simulator for aging and soiling banknotes using Gaussian models and Perlin noise

In this paper, we propose a banknote simulator that generates aged and soiled banknotes. By analyzing the characteristics of circulating banknotes, we developed Gaussian brightness models for gray level changes of circulating banknotes. In addition, the Perlin noise model was used to simulate soiling. The proposed algorithm was tested using US Dollars (USD) and the experimental results show that the proposed method effectively simulated soiled banknote images from new banknote images.

Soiled banknote fitness determination based on morphology and Otsu's thresholding

This paper presents a soiled banknote determination method based on morphology and Otsus thresholding method. Using morphology, a texture image was extracted with selective morphological closing operations, and the soiling level was estimated by averaging the pixel intensities of the texture image. Next, the Otsu threshold was calculated and used to determine whether each pixel was a foreground or background pixel. Then the soiling level was estimated using the difference between the averages of the foreground and background pixels. We used the weighted sum of the two estimated values as a soiling level measure. The proposed method showed good performance on Euro and Russian banknotes.

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.

Document classification for copy-mode decision

We proposed a low-complexity document classification algorithm to select copy mode. The goal is to select a suitable copy mode during copying process. We first analyzed scanned images and classified them into three modes: text, image, mixed modes. To classify images, we used several features, which include pixel density with low brightness, edge length and text line components. Experimental results showed that the proposed algorithm provided about 95% classification accuracy.

Effects of blurring and vertical misalignment on visual fatigue of stereoscopic displays

In this paper, we investigate two error issues in stereo images, which may produce visual fatigue. When two cameras are used to produce 3D video sequences, vertical misalignment can be a problem. Although this problem may not occur in professionally produced 3D programs, it is still a major issue in many low-cost 3D programs. Recently, efforts have been made to produce 3D video programs using smart phones or tablets, which may present the vertical alignment problem. Also, in 2D-3D conversion techniques, the simulated frame may have blur effects, which can also introduce visual fatigue in 3D programs. In this paper, to investigate the relationship between these two errors (vertical misalignment and blurring in one image), we performed a subjective test using simulated 3D video sequences that include stereo video sequences with various vertical misalignments and blurring in a stereo image. We present some analyses along with objective models to predict the degree of visual fatigue from vertical misalignment and blurring.