• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.179-181
• /
• 2023

본 논문에서는 GAN 기반의 영상 생성 방법론을 이용해 delayed PET 영상을 생성하는 연구를 수행하였다. PET은 양전자를 방출하는 방사성 동위원소를 표지한 방사성의약품의 체내 분포를 시각화함으로서 암 세포 진단에 이용되는 의료영상 기법이다. 하지만 PET의 스캔 과정에서 방사성의약품이 체내에 분포하는 데에 걸리는 시간이 오래 걸린다는 문제점이 존재한다. 따라서 본 연구에서는 방사성의약품이 충분히 분포되지 않은 상태에서 얻은 PET 영상을 통해 목표로 하는 충분히 시간이 지난 후에 얻은 PET 영상을 생성하는 모델을 GAN (generative adversarial network)에 기반한 image-to-image translation(I2I)를 통해 수행했다. 특히, 생성 전후의 영상 간의 영상 쌍을 고려한 paired I2I인 Pix2pix와 이를 고려하지 않은 unpaired I2I인 CycleGAN 두 가지의 방법론을 비교하였다. 연구 결과, Pix2pix에 기반해 생성한 delayed PET 영상이 CycleGAN을 통해 생성한 영상에 비해 영상 품질이 좋음을 확인했으며, 또한 실제 획득한 ground-truth delayed PET 영상과의 유사도 또한 더 높음을 확인할 수 있었다. 결과적으로, 딥러닝에 기반해 early PET을 통해 delayed PET을 생성할 수 있었으며, paired I2I를 적용할 경우 보다 높은 성능을 기대할 수 있었다. 이를 통해 PET 영상 획득 과정에서 방사성의약품의 체내 분포에 소요되는 시간을 딥러닝 모델을 통해 줄여 PET 이미징 과정의 시간적 비용을 절감하는 데에 크게 기여할 수 있을 것으로 기대된다.

• Smart Media Journal
• /
• v.8 no.2
• /
• pp.46-57
• /
• 2019

In recent research, deep learning-based methods have achieved state-of-the-art performance in various computer vision tasks. However, these methods are commonly supervised, and require huge amounts of annotated data to train. Acquisition of data demands an additional costly effort, particularly for the tasks where it becomes challenging to obtain large amounts of data considering the time constraints and the requirement of professional human diligence. In this paper, we present a data level synthetic sampling solution to learn from small and imbalanced data sets using Generative Adversarial Networks (GANs). The reason for using GANs are the challenges posed in various fields to manage with the small datasets and fluctuating amounts of samples per class. As a result, we present an approach that can improve learning with respect to data distributions, reducing the partiality introduced by class imbalance and hence shifting the classification decision boundary towards more accurate results. Our novel method is demonstrated on a small dataset of 2789 tomato plant disease images, highly corrupted with class imbalance in 9 disease categories. Moreover, we evaluate our results in terms of different metrics and compare the quality of these results for distinct classes.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.375-378
• /
• 2000

In this paper, we propose a novo] content-based image retrieval system using both Hidden Markov Model(HMM) and an improved chain code. The Gaussian Mixture Model(GMM) is applied to statistically model a color information of the image, and Deterministic Annealing EM(DAEM) algorithm is employed to estimate the parameters of GMM. This result is used to segment the given image. We use an improved chain code, which is invariant to rotation, translation and scale, to extract the feature vectors of the shape for each image in the database. These are stored together in the database with each HMM whose parameters (A, B, $\pi$) are estimated by Baum-Welch algorithm. With respect to feature vector obtained in the same way from the query image, a occurring probability of each image is computed by using the forward algorithm of HMM. We use these probabilities for the image retrieval and present the highest similarity images based on these probabilities.

• Archives of design research
• /
• v.19 no.4 s.66
• /
• pp.145-154
• /
• 2006

In today's brand apartment ads, the image of high-tech apartment seeking consumers' convenience thanks to the introduction of ubiquitous and the image of nature-friendly apartment stressing consumers' well-being life have become a main stream. However, it seems that there lacks differentiation in expressing ad images, except some brands. This may also be related to the matter of providing effective communication, which is the ultimate goal of advertisement. Therefore, to suggest the effective communication methods of brand apartment ads, this study semiologically translated first the images thai recent brand apartment ads present, and then in order to review whether these images are properly delivered to consumers, this study conducted the image survey for brand apartment ads with 123 subjects, who are the group of design professionals. As a result, the survey showed that, in the semiological translation of ad expression, meaning sharing is made when ad concept and ad image are semiologically consistent, and it showed that the realization of successful ad communication between advertisers and consumers is possible, which starts from this kind of meaning sharing. Based on the results of this study, we could know that the introduction of semiology into ad planning can be a logical judgment criterion to provide effective ad communication.

• Proceedings of the IEEK Conference
• /
• 2000.06d
• /
• pp.114-117
• /
• 2000

In this paper, we proposed robust digital image watermarking based on modulation transfer function (MTF) of human visual system (HVS). Using the proposed method, robust watermarking is possible both in common image processing operations such as cropping and lossy compression and in geometrical transforms such as rotation, scaling, and translation, because it can embed watermark and template signal maximally using MTF of HVS. Experimental results show that the proposed watermarking method is more robust to several common image processing operations and geometrical transforms.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2108-2111
• /
• 2003

This paper introduces a new local feature extraction method and image matching method for the localization and classification of targets. Proposed method is based on the block-by-block projection associated with directional pattern of blocks. Each pattern has its own eigen-vertors called as CEBs(Classified Eigen-Blocks). Also proposed block-based image matching method is robust to translation and occlusion. Performance of proposed feature extraction and matching method is verified by the face localization and FLIR-vehicle-image classification test.

• Journal of the Semiconductor & Display Technology
• /
• v.6 no.2 s.19
• /
• pp.53-57
• /
• 2007

In this paper, a new content-based color image retrieval method is proposed, in which both the color content and the spatial relationship of image have been taken into account. In order to represent the spatial distribution information of image, a disorder matrix, which has the invariance to the rotation and translation of the image content, has been designed. This is based on multi-resolution color-spatial information. We present our algorithm in the following section, and then verified the search results with comparison to other methods, such as color histogram, wavelet histogram, correlogram and wavelet correlogram. Experimental results with various types of images show that the proposed method not only achieves a high image retrieval performance but also improve the retrieval precision.

• The Journal of Bigdata
• /
• v.7 no.1
• /
• pp.173-185
• /
• 2022

Nowadays, Korean webtoons are leading the global digital comic market. Webtoons are being serviced in various languages around the world, and dramas or movies produced with Webtoons' IP (Intellectual Property Rights) have become a big hit, and more and more webtoons are being visualized. However, with the success of these webtoons, the working environment of webtoon creators is emerging as an important issue. According to the 2021 Cartoon User Survey, webtoon creators spend 10.5 hours a day on creative activities on average. Creators have to draw large amount of pictures every week, and competition among webtoons is getting fiercer, and the amount of paintings that creators have to draw per episode is increasing. Therefore, this study proposes to generate webtoon background images using deep learning algorithms and use them for webtoon production. The main character in webtoon is an area that needs much of the originality of the creator, but the background picture is relatively repetitive and does not require originality, so it can be useful for webtoon production if it can create a background picture similar to the creator's drawing style. Background generation uses CycleGAN, which shows good performance in image-to-image translation, and CartoonGAN, which is specialized in the Cartoon style image generation. This deep learning-based image generation is expected to shorten the working hours of creators in an excessive work environment and contribute to the convergence of webtoons and technologies.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1140-1145
• /
• 2003

Image normalization is one of the important areas in pattern recognition. Also, log-polar images are useful in the sense that their image data size is reduced dramatically comparing with conventional images and it is possible to develop faster pattern recognition algorithms. Especially, the log-polar image is very similar with the structure of human eyes. However, there are almost no researches on pattern recognition using the log-polar images while a number of researches on visual tracking have been executed. We propose an image normalization technique of log-polar images using momentums applicable for affine-invariant pattern recognition. We handle basic distortions of an image including translation, rotation, scaling, and skew of a log-polar image. The algorithm is experimented in a PC-based real-time vision system successfully.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.30 no.3 s.246
• /
• pp.215-221
• /
• 2006

A stereoscopic particle tracking velocimetry (SPTV) technique based on the 2-frame hybrid particle tracking velocimetry (PTV) method was developed. The expansion of 2D PTV to SPTV is facilitated by the fact that the PTV method tracks individual particle centroids. To evaluate the performance and measurement accuracy of the present SPTV technique, it was applied to flow images of rigid body translation and synthetic standard images of jet shear flow and impinging jet flow. The data processing routine and measurement uncertainty of the SPTV technique are compared with those of conventional stereoscopic particle image velecimet.y (SPBV). In addition, the centroid translation effect of 2D particle image velocimetry (PIV) is defined and its effect on SPIV measurements is discussed. Compared to the SPIV method, the SPTV technique has inherited merits of concise and precise velocity evaluation procedures and provides better spatial resolution and measurement accuracy.