• Journal of Broadcast Engineering
• /
• v.5 no.1
• /
• pp.50-60
• /
• 2000

In this paper, we propose a new composite feature consists of both color and shape information that are suitable for the task of character image retrieval. This approach extracts shape-based information using Zernike moments from Y image in YCbCr color space. Zernike moments can extract shape-based features that are invariant to rotation, translation, and scaling. We also extract color-based information from the DCT coefficients of Cr and Cb image. This approach is good method reflecting human visual property and is suitable for web application such as large image database system and animation because higher retrieval rate has been achieved using only 36 features. In experiment, this method is applied to 3,834 character images. We confirmed that this approach brought about excellent effect by ANMRR(Average of Normalized, Modified Retrieval Rank), which is used in the evaluation measure of MPEG-7 color descriptor and BEP(Bull's Eye Performance), which is used in evaluation measure of shape descriptor in character image retrieval.

• Journal of Korea Multimedia Society
• /
• v.5 no.4
• /
• pp.415-425
• /
• 2002

In this paper, we present a content-based image retrieval method which is based on the feature extraction in the wavelet transform domain. In order to overcome the drawbacks of the feature vector making up methods which use the global wavelet coefficients in subbands, we utilize the energy value of wavelet coefficients, and the shape-based retrieval of objects is processed by moment which is invariant in translation, scaling, rotation of the objects The proposed methods reduce feature vector size, and make progress performance of classification retrieval which provides fast retrievals times. To offer the abilities of region-based image retrieval, we discussed the image segmentation method which can reduce the effect of an irregular light sources. The image segmentation method uses a region-merging, and candidate regions which are merged were selected by the energy values of high frequency bands in discrete wavelet transform. The region-based image retrieval is executed by using the segmented region information, and the images are retrieved by a color, texture, shape feature vector.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.3
• /
• pp.341-351
• /
• 2003

Digital image watermarking is the method that can protect the copyright of the image by embedding copyright information, which is called watermark. Watermarking must have robustness to intentional or unintentional data changing, called attack. The conventional watermarking schemes are robust to waveform attacks such as image compression, filtering etc. However, they are vulnerable to geometrical attacks such as rotation, scaling, translation, and cropping. Accordingly, this paper proposes new watermarking scheme that is robust to geometrical attacks by using invariant centroid. Invariant centroid is the gravity center of a central area in a gray scale image that remains unchanged even when the image is attacked by RST including cropping and proposed scheme uses invariant centroids of original and inverted image as the template. To make geometrically invariant domain, template and angle compensated Log -Polar Map(LPM) is used. Then Discrete Cosine Transform(DCT) is performed and the watermark is embedded into the DCT coefficients. Futhermore, to prevent a watermarked image from degrading due to interpolation during coordinate system conversion, only the image of the watermark signal is extracted and added to the original image. Experimental results show that the proposed scheme is especially robust to RST attacks including cropping.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.4
• /
• pp.1166-1186
• /
• 2022

Artistic font design has become an integral part of visual media. However, without prior knowledge of the font domain, it is difficult to create distinct font styles. When the number of characters is limited, this task becomes easier (e.g., only Latin characters). However, designing CJK (Chinese, Japanese, and Korean) characters presents a challenge due to the large number of character sets and complexity of the glyph components in these languages. Numerous studies have been conducted on automating the font design process using generative adversarial networks (GANs). Existing methods rely heavily on reference fonts and perform font style conversions between different fonts. Additionally, rather than capturing style information for a target font via multiple style images, most methods do so via a single font image. In this paper, we propose a network architecture for generating multilingual font sets that makes use of geometric structures as content. Additionally, to acquire sufficient style information, we employ multiple style images belonging to a single font style simultaneously to extract global font style-specific information. By utilizing the geometric structural information of content and a few stylized images, our model can generate an entire font set while maintaining the style. Extensive experiments were conducted to demonstrate the proposed model's superiority over several baseline methods. Additionally, we conducted ablation studies to validate our proposed network architecture.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.8
• /
• pp.597-602
• /
• 2016

In image guided surgery, a patient registration process is a critical process for the successful operation, which is required to use pre-operative images such as CT and MRI during operation. Though several patient registration methods have been studied, we concentrate on one method that utilizes 3D surface measurement data in this paper. First, a hand-held 3D surface measurement device measures the surface of the patient, and secondly this data is matched with CT or MRI data using optimization algorithms. However, generally used ICP algorithm is very slow without a proper initial location and also suffers from local minimum problem. Usually, this problem is solved by manually providing the proper initial location before performing ICP. But, it has a disadvantage that an experience user has to perform the method and also takes a long time. In this paper, we propose a method that can accurately find the proper initial location automatically. The proposed method finds the proper initial location for ICP by converting 3D data to 2D curvature images and performing image matching. Curvature features are robust to the rotation, translation, and even some deformation. Also, the proposed method is faster than traditional methods because it performs 2D image matching instead of 3D point cloud matching.

• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.133-136
• /
• 2004

We are going to procedure various view from two frontal image using trifocal tensor. We found that warping is effective to produce synthesized poses of a face with the small number of mesh point of a given image in previous research[1]. For this research, fundamental matrix is important to calculate trifocal tensor. So, in this paper, we investigate two existing algorithms: Hartley's[2] and Kanatani's[3]. As an experimental result, Kenichi Kantani's algorithm has better performance of fundamental matrix than Harley's algorithm. Then we use the fundamental matrix of Kenichi Kantani's algorithm to calculate trifocal tensor. From trifocal tensor we calculate new trifocal tensor with rotation input and translation input and we use warping to produce new virtual views.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1305-1311
• /
• 2022

With the recent rapid development of mobile terminals and personal computers and the advent of neural network technology, real-time face swapping using images has become possible. In particular, the cycle generative adversarial network made it possible to replace faces using uncorrelated image data. In this paper, we propose an input data processing scheme that can improve the quality of face swapping with less training data and time. The proposed scheme can improve the image quality while preserving facial structure and expression information by combining facial landmarks extracted through a pre-trained neural network with major information that affects the structure and expression of the face. Using the blind/referenceless image spatial quality evaluator (BRISQUE) score, which is one of the AI-based non-reference quality metrics, we quantitatively analyze the performance of the proposed scheme and compare it to the conventional schemes. According to the numerical results, the proposed scheme obtained BRISQUE scores improved by about 4.6% to 14.6%, compared to the conventional schemes.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.49-54
• /
• 2009

A fuzzy weighted mean method on a 2D shape recognition system is introduced in this paper. The bispectrum based on third order cumulant is applied to the contour sequence of each image for the extraction of a feature vector. This bispectral feature vector, which is invariant to shape translation, rotation and scale, represents a 2D planar image. However, to obtain the best performance, it should be considered certain criterion on the calculation of weights for the fuzzy weighted mean method. Therefore, a new method to calculate weights using means by differences of feature values and their variances with the maximum distance from differences of feature values. is developed. In the experiments, the recognition results with fifteen dimensional bispectral feature vectors, which are extracted from 11.808 aircraft images based on eight different styles of reference images, are compared and analyzed.

• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.237-240
• /
• 2003

We implement a system for 3D structure reconstruction from multiple 2D images. It uses geometric primitives such as box, wedge, pyramid, etc, each having translation, rotation, and scale parameters. Primitives are marked on input images with GUI (Graphic User Interface). Lines made by projection of primitives onto an image correspond to marked line segments of the image. Error function is defined by disparity between them and is minimized by downhill simplex method. By assigning relationship between models, the number of parameters to solve can be decreased and the resultant models become more accurate To share variables among other models also reduces computational complexity. Experiments using real images have shown that the proposed method successfully reconstructs 3D structure.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.392-395
• /
• 2003

This paper describes a new iris segmentation and recognition method, which is robust to noises. Combining statistical classification and elastic boundary fitting, the iris is first segmented. Then, the localized iris image is smoothed by a convolution with a Gaussian function, down-sampled by a factor of filtered with a Laplacian operator, and quantized using the Lloyd-Max method. Since the quantized output is sensitive to a small shift of the full-resolution iris image, the outputs of the Laplacian operator are computed for all space shifts. The quantized output with maximum entropy is selected as the final feature representation. An appropriate formulation of similarity measure is defined for the classification of the quantized output. Experimentally we showed that the proposed method produces superb performance in iris segmentation and recognition.