Seok-Lyong Lee

Similarity search for multidimensional data sequences

Time series data, which are a series of one dimensional real numbers, have been studied in various database applications. We extend the traditional similarity search methods on time series data to support a multidimensional data sequence, such as a video stream. We investigate the problem of retrieving similar multidimensional data sequences from a large database. To prune irrelevant sequences in a database, we introduce correct and efficient similarity functions. Both data sequences and query sequences are partitioned into subsequences, and each of them is represented by a Minimum Bounding Rectangle (MBR). The query processing is based upon these MBRs, instead of scanning data elements of entire sequences. Our method is designed: (1) to select candidate sequences in a database, and (2) to find the subsequences of a selected sequence, each of which falls under the given threshold. The latter is of special importance in the case of retrieving subsequences from large and complex sequences such as video. By using it, we do not need to browse the whole of the selected video stream, but just browse the sub-streams to find a scene we want. We have performed an extensive experiment on synthetic, as well as real data sequences (a collection of TV news, dramas, and documentary videos) to evaluate our proposed method. The experiment demonstrates that 73-94 percent of irrelevant sequences are pruned using the proposed method, resulting in 16-28 times faster response time compared with that of the sequential search.

Texture feature extraction based on a uniformity estimation method for local brightness and structure in chest CT images

Texture feature is one of most important feature analysis methods in the computer-aided diagnosis (CAD) systems for disease diagnosis. In this paper, we propose a Uniformity Estimation Method (UEM) for local brightness and structure to detect the pathological change in the chest CT images. Based on the characteristics of the chest CT images, we extract texture features by proposing an extension of rotation invariant LBP (ELBP(riu4)) and the gradient orientation difference so as to represent a uniform pattern of the brightness and structure in the image. The utilization of the ELBP(riu4) and the gradient orientation difference allows us to extract rotation invariant texture features in multiple directions. Beyond this, we propose to employ the integral image technique to speed up the texture feature computation of the spatial gray level dependent method (SGLDM).

Multiple object tracking with partial occlusion handling using salient feature points

Abstract Handling occlusion has been a challenging task in object tracking. In this paper, we propose a multiple object tracking method in the presence of partial occlusion using salient feature points. We first extract the prominent feature points from each target object, and then use a particle filter-based approach to track the feature points in image sequences based on various attributes such as location, velocity and other descriptors. We then detect and revise the feature points that have been tracked incorrectly. The main idea is that, even if some feature points are not successfully tracked due to occlusion or poor imaging condition, the other correctly tracked features can collectively perform the corrections on their behalf. Finally, we track the objects using the correctly tracked feature points through a Hough-like approach, and the object bounding boxes are updated using the relative locations of these feature points. Experimental results demonstrate that our method is proficient in providing accurate human tracking as well as appropriate occlusion handling, compared to the existing methods.

An effective defect inspection system for polarized film images using image segmentation and template matching techniques

In this paper, we present an effective defect inspection system that identifies film defects and determines their types in order to produce polarized films for TFT-LCD (thin film transistor - liquid crystal display). The proposed system is designed and implemented to find defects from polarized film images using image segmentation techniques and to determine defect types through the image analysis of detected defects using template matching techniques. We extract features of the defects such as shape and texture, and compare them to the features of referential defect images stored in a template database. Experimental results using the proposed system show that it identifies defects of test images effectively (Recall=1.00, Precision=0.95) and efficiently (Average response time=0.64s), and also achieves a high correctness in determining the types (Recall=0.95, Precision=0.96) for five classes of defects. In addition the experiment shows that our system is fairly robust with respect to the rotational transformation, achieving the desirable property of the rotation invariance.

A visual shape descriptor using sectors and shape context of contour lines

This paper describes a visual shape descriptor based on the sectors and shape context of contour lines to represent the image local features used for image matching. The proposed descriptor consists of two-component feature vectors. First, the local region is separated into sectors and their gradient magnitude and orientation values are extracted; a feature vector is then constructed from these values. Second, local shape features are obtained using the shape context of contour lines. Another feature vector is then constructed from these contour lines. The proposed approach calculates the local shape feature without needing to consider the edges. This can overcome the difficulty associated with textured images and images with ill-defined edges. The combination of two-component feature vectors makes the proposed descriptor more robust to image scale changes, illumination variations and noise. The proposed visual shape descriptor outperformed other descriptors in terms of the matching accuracy: 14.525% better than SIFT, 21% better than PCA-SIFT, 11.86% better than GLOH, and 25.66% better than the shape context.

Automatic evaluation of cardiac hypertrophy using cardiothoracic area ratio in chest radiograph images

To evaluate the cardiac hypertrophy from chest radiograph images, radiologists usually examine the cardiothoracic ratio (frequently called CTR) which is a standard diagnostic index. The CTR is computed by the maximum transverse diameter of the heart shadow divided by the maximum transverse diameter of right and left lung boundaries. In this paper, we present a method to evaluate the cardiac hypertrophy by comparing the area of heart with that of lung, instead of the cardiothoracic ratio to get more desirable diagnostic results. We introduce a new index, a cardiothoracic area ratio (CTAR), which is computed by dividing the area of heart region by the area of lung region of specific interest. We first segment a chest region of interest in a radiograph image and then automatically compute the traditional CTR and the CTAR to evaluate the cardiac hypertrophy. And finally, we provide the visual presentation of those ratios on the chest radiograph image. The experimental results using a set of radiograph images show that the proposed method can be used effectively for determining the cardiac hypertrophy in a real-time diagnostic environment. It provides the higher discrimination power than the CTR to identify hypertrophied hearts by recognizing the heart enlargement. It also can be used together with the traditional CTR as a complementary measure when it is difficult to determine abnormalities by the CTR, reducing the rate of wrong diagnosis.

Hyper-rectangle based segmentation and clustering of large video data sets

Video information processing has been one of great challenging areas in the database community since it needs huge amount of storage space and processing power. In this paper, we investigate the problem of clustering large video data sets that are collections of video clips as foundational work for the subsequent processing such as video retrieval. A video clip, a sequence of video frames, is represented by a multidimensional data sequence, which is partitioned into video segments considering temporal relationship among frames, and then similar segments of the clip are grouped into video clusters. We present the effective video segmentation and clustering algorithm which guarantees the clustering quality to such an extent that satisfies predefined conditions, and show its effectiveness via experiments on various video data sets.

On the effective clustering of multidimensional data sequences

In this paper, we investigate the problem of clustering multidimensional data sequences such as video streams. Each sequence is represented by a small number of hyper-rectangular clusters for subsequent indexing and similarity search processing. We present a linear clustering algorithm that guarantees the predefined level of clustering quality, and show its effectiveness via experiments on various video data sets.

Distributed similarity search algorithm in distributed heterogeneous multimedia databases

The collection fusion problem in multimedia databases is concerned with the merging of results retrieved by content based retrieval from distributed heterogeneous multimedia databases in order to find the most similar objects to a query object. We propose distributed similarity search algorithms, two heuristic algorithms and an algorithm using the linear regression, to solve this problem. To our knowledge, these algorithms are the first research results in the area of distributed content based retrieval for heterogeneous multimedia databases.

Distance approximation techniques to reduce the dimensionality for multimedia databases

Recently, databases have been used to store multimedia data such as images, maps, video clips, and music clips. In order to search them, they should be represented by various features, which are composed of high-dimensional vectors. As a result, the dimensionality of data is increased considerably, which causes ‘the curse of dimensionality’. The increase of data dimensionality causes poor performance of index structures. To overcome the problem, the research on the dimensionality reduction has been conducted. However, some reduction methods do not guarantee no false dismissal, while others incur high computational cost. This paper proposes dimensionality reduction techniques that guarantee no false dismissal while providing efficiency considerable by approximating distances with a few values. To provide the no false dismissal property, approximated distances should always be smaller than original distances. The Cauchy–Schwarz inequality and two trigonometrical equations are used as well as the dimension partitioning technique is applied to approximate distances in such a way to reduce the difference between the approximated distance and the original distance. As a result, the proposed techniques reduce the candidate set of a query result for efficient query processing.