• Journal of KIISE:Computer Systems and Theory
• /
• v.37 no.5
• /
• pp.314-318
• /
• 2010

This paper proposes an image retrieval method based on the Multiple Region-of-Interest. In the proposed method, the image is segmented into blocks, among which the blocks overlapped with multiple ROIs are selected. The similarity of images is measured using the MPEG-7 dominant color descriptor(DCD) and considering the relative location of the overlapped blocks. The experimental results showed that the proposed method improves the retrieval performance than the previous methods using the global DCD or comparing the blocks at the same position. In addition, the method that considers the relative position of blocks overlapped with the multiple ROIs also showed a better performance than the existing methods.

• Journal of KIISE:Software and Applications
• /
• v.32 no.11
• /
• pp.1013-1020
• /
• 2005

This paper presents an effective and robust leaf image retrieval system based on shape feature. Specifically, we propose an improved MPP algorithm for more effective representation of leaf images and show a new dynamic matching algorithm that basically revises the Nearest Neighbor search to reduce the matching time. In particular, both leaf shape and leaf arrangement can be sketched in the query for better accuracy and efficiency. In the experiment, we compare our proposed method with other methods including Centroid Contour Distance(CCD), Fourier Descriptor, Curvature Scale Space Descriptor(CSSD), Moment Invariants, and MPP. Experimental results on one thousand leaf images show that our approach achieves a better performance than other methods.

• The KIPS Transactions:PartB
• /
• v.15B no.4
• /
• pp.315-322
• /
• 2008

It is important to retrieve effectively the visual media such as pictures and video in the internet, especially to the application areas such as electronic art museum, e-commerce, and internet shopping malls. It is also needed in these areas to have content-based or even semantic-based multimedia retrieval instead of simple keyword-based retrieval. In our earlier research, we proposed a semantic-based visual media retrieval framework for the effective retrieval of the visual media from the internet. It uses visual media metadata and ontology based on the web service to achieve the semantic-based retrieval. In this research, there are more than one visual media service providers and one central service broker. As a preliminary step to the visual media data retrieval, a method is proposed to retrieve the service providers effectively. The method uses the structure of the ontology tree to obtain the providers and their rankings. It also uses the size of sub nodes and child nodes in the tree. It measures the rankings of providers more effectively than previous method. The experimental results show the accuracy of the method while keeping compatible speed against the existing method.

• Journal of the Korean Society for information Management
• /
• v.30 no.3
• /
• pp.23-48
• /
• 2013

This study compared and analyzed 15 CBMR (Content-based Music Retrieval) systems accessible on the web in terms of DB size and type, query type, access point, input and output type, and search functions, with reviewing features of music information and techniques used for transforming or transcribing of music sources, extracting and segmenting melodies, extracting and indexing features of music, and matching algorithms for CBMR systems. Application of text information retrieval techniques such as inverted indexing, N-gram indexing, Boolean search, truncation, keyword and phrase search, normalization, filtering, browsing, exact matching, similarity measure using edit distance, sorting, etc. to enhancing the CBMR; effort for increasing DB size and usability; and problems in extracting melodies, deleting stop notes in queries, and using solfege as pitch information were found as the results of analysis.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.6
• /
• pp.724-732
• /
• 2001

In this paper, a new method of feature extraction of major objects to represent an image using Vector Quantization(VQ) is proposed. The principal features of the image, which are used in a content-based image retrieval system, are color, texture, shape and spatial positions of objects. The representative color and texture features are extracted from the given image using VQ(Vector Quantization) clustering algorithm with a general feature extraction method of color and texture. Since these are used for content-based image retrieval and searched by objects, it is possible to search and retrieve some desirable images regardless of the position, rotation and size of objects. The experimental results show that the representative feature extraction time is much reduced by using VQ, and the highest retrieval rate is given as the weighted values of color and texture are set to 0.5 and 0.5, respectively, and the proposed method provides up to 90% precision and recall rate for 'person'query images.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.119-126
• /
• 2005

We propose a new content-based image retrieval using a representative color histogram and directional pattern histogram that is adaptive to the classification characteristics of the image blocks. In the proposed method the color and pattern feature vectors are extracted according to the characteristics o: the block classification after dividing the image into blocks with a fixed size. First, the divided blocks are classified as either luminance or color blocks depending on the saturation of the block. Thereafter, the color feature vectors are extracted by calculating histograms of the block average luminance co-occurrence for the luminance block and the block average colors for the color blocks. In addition, block directional pattern feature vectors are extracted by calculating histograms after performing the directional gradient classification of the luminance. Experimental results show that the proposed method can outperform the conventional methods as regards the precision and the size of the feature vector dimension.

• Journal of Information Processing Systems
• /
• v.17 no.5
• /
• pp.1020-1033
• /
• 2021

Nowadays, online or mobile social network services (SNS) are very popular and widely spread in our society and daily lives to instantly share, disseminate, and search information. In particular, SNS such as YouTube, Flickr, Facebook, and Amazon allow users to upload billions of images or videos and also provide a number of multimedia information to users. Information retrieval in multimedia-rich SNS is very useful but challenging task. Content-based media retrieval (CBMR) is the process of obtaining the relevant image or video objects for a given query from a collection of information sources. However, CBMR suffers from the dimensionality curse due to inherent high dimensionality features of media data. This paper investigates the effectiveness of the kernel trick in CBMR, specifically, the kernel principal component analysis (KPCA) for dimensionality reduction. KPCA is a nonlinear extension of linear principal component analysis (LPCA) to discovering nonlinear embeddings using the kernel trick. The fundamental idea of KPCA is mapping the input data into a highdimensional feature space through a nonlinear kernel function and then computing the principal components on that mapped space. This paper investigates the potential of KPCA in CBMR for feature extraction or dimensionality reduction. Using the Gaussian kernel in our experiments, we compute the principal components of an image dataset in the transformed space and then we use them as new feature dimensions for the image dataset. Moreover, KPCA can be applied to other many domains including CBMR, where LPCA has been used to extract features and where the nonlinear extension would be effective. Our results from extensive experiments demonstrate that the potential of KPCA is very encouraging compared with LPCA in CBMR.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.8
• /
• pp.2895-2912
• /
• 2014

Content-based image retrieval has been the most important technique for managing huge amount of images. The fundamental yet highly challenging problem in this field is how to measure the content-level similarity based on the low-level image features. The primary difficulties lie in the great variance within images, e.g. background, illumination, viewpoint and pose. Intuitively, an ideal similarity measure should be able to adapt the data distribution, discover and highlight the content-level information, and be robust to those variances. Motivated by these observations, we in this paper propose a probabilistic similarity learning approach. We first model the distribution of low-level image features and derive the free energy kernel (FEK), i.e., similarity measure, based on the distribution. Then, we propose a learning approach for the derived kernel, under the criterion that the kernel outputs high similarity for those images sharing the same class labels and output low similarity for those without the same label. The advantages of the proposed approach, in comparison with previous approaches, are threefold. (1) With the ability inherited from probabilistic models, the similarity measure can well adapt to data distribution. (2) Benefitting from the content-level hidden variables within the probabilistic models, the similarity measure is able to capture content-level cues. (3) It fully exploits class label in the supervised learning procedure. The proposed approach is extensively evaluated on two well-known databases. It achieves highly competitive performance on most experiments, which validates its advantages.

• Proceedings of the KAIS Fall Conference
• /
• 2011.05a
• /
• pp.371-374
• /
• 2011

Recently, retrieval of various video data has become an important issue as more and more multimedia content services are being provided. To effectively deal with video data, a semantic-based retrieval scheme that allows for processing diverse user queries and saving them on the database is required. In this regard, this paper proposes a semantic-based video retrieval system that allows the user to search diverse meanings of video data for electrical safety-related educational purposes by means of automatic annotation processing. If the user inputs a keyword to search video data for electrical safety-related educational purposes, the mobile agent of the proposed system extracts the features of the video data that are afterwards learned in a continuous manner, and detailed information on electrical safety education is saved on the database. The proposed system is designed to enhance video data retrieval efficiency for electrical safety-related educational purposes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.2 no.6
• /
• pp.280-298
• /
• 2008

For efficient content-based image retrieval, diverse visual features such as color, texture, and shape have been widely used. In the case of leaf images, further improvement can be achieved based on the following observations. Most plants have unique shape of leaves that consist of one or more blades. Hence, blade-based matching can be more efficient than whole shape-based matching since the number and shape of blades are very effective to filtering out dissimilar leaves. Guaranteeing rotational invariance is critical for matching accuracy. In this paper, we propose a new shape representation, indexing and matching scheme for leaf image retrieval. For leaf shape representation, we generated a distance curve that is a sequence of distances between the leaf’s center and all the contour points. For matching, we developed a blade-based matching algorithm called rotation invariant - partial dynamic time warping (RI-PDTW). To speed up the matching, we suggest two additional techniques: i) priority queue-based pruning of unnecessary blade sequences for rotational invariance, and ii) lower bound-based pruning of unnecessary partial dynamic time warping (PDTW) calculations. We implemented a prototype system on the GEMINI framework [1][2]. Using experimental results, we showed that our scheme achieves excellent performance compared to competitive schemes.