• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.908-909
• /
• 2023

Recently, several generative models have emerged and are being used in various industries. Among them, Cycle GAN is still used in various fields such as style transfer, medical care and autonomous driving. In this paper, we propose two methods to improve the performance of these Cycle GAN model. The ReLU activation function previously used in the generator was changed to Leaky ReLU. And a new loss function is proposed that considers the semantic level rather than focusing only on the pixel level through the VGG feature extractor. The proposed model showed quality improvement on the test set in the art domain, and it can be expected to be applied to other domains in the future to improve performance.

• 대한세포병리학회지
• /
• 제2권2호
• /
• pp.134-141
• /
• 1991

Localized or solitary fibrous tumor (SFT) of the pleura has been classified as a type of mesothelioma, arising from the submesothelial connective tissue cells. The preoperative diagnosis of the tumor at the cytologic or histologic level is very important for the proper handling of the lesion. This preoperative diagnosis is now possible by means of the advance in the transthoracic fine needle aspiration biopsy (FNA) techniques and in the very experience of the cytopathologists. We describe FNA cytologic feature of two cases of SFT arising from the pleura. Cytologic, histologic, immunohistochemical, and electron microscopic characteristics of pleural SFT are discussed. The tumor cells of SFT are spindle or oval in shape with a variable amount of cytoplasm. They are arranged in irregular trabeculae intimately associated with capillaries. A unique cytologic feature observed in this tumor is that thick, eosinophilic, amorphous collagen bundles are scattered between tumor cells.

• 대한원격탐사학회지
• /
• 제3권1호
• /
• pp.25-39
• /
• 1987

The purpose of this research is to develop the Automatic Fingerprint Verfication System by digital computer based on specially in PC level. Fingerprint is used as means of personal identity verification in view of that it has the high reliability and safety. Fingerprint pattern recognition algorithm is constitute of 3 stages, namely of the preprocessing, the feature extraction and the recognition. The preprocessing stage includes smoothing, binarization, thinning and restoration. The feature extraction stage includes the extraction of minutiae and its features. The recognition stage includes the registration and the matching score calculation which measures the similarity between two images. Tests for this study with 325 pairs of fingerprint resulted in 100% of separation which which in turn is turned out to be the reliability of this algorithm.

• 정보과학회 논문지
• /
• 제44권12호
• /
• pp.1290-1295
• /
• 2017

문서의 텍스트를 바탕으로 각 문서가 속한 분류를 찾아내는 문서 분류는 자연어 처리의 기본 분야 중 하나로 주제 분류, 감정 분류 등 다양한 분야에 이용될 수 있다. 문서를 분류하기 위한 신경망 모델은 크게 단어를 기본 단위로 다루는 단어 수준 모델과 문자를 기본 단위로 다루는 문자 수준 모델로 나누어진다. 본 논문에서는 문서를 분류하는 신경망 모델의 성능을 향상시키기 위하여 문자 수준과 단어 수준의 모델을 혼합한 신경망 모델을 제안한다. 제안하는 모델은 각 단어에 대하여 문자 수준의 신경망 모델로 인코딩한 정보와 단어들의 정보를 저장하고 있는 단어 임베딩 행렬의 정보를 결합하여 각 단어에 대한 특징 벡터를 만든다. 추출된 단어들에 대한 특징 벡터를 바탕으로, 주의(attention) 메커니즘을 이용한 순환 신경망을 단어 수준과 문장 수준에 각각 적용하는 계층적 신경망 구조를 통해 문서를 분류한다. 제안한 모델에 대하여 실생활 데이터를 바탕으로 한 실험으로 효용성을 검증한다.

• 대한원격탐사학회지
• /
• 제37권2호
• /
• pp.221-232
• /
• 2021

본 연구는 KOMPSAT-3 및 KOMPSAT-3A호에서 전처리 단계에 따라 구분하여 제공하는 Level 1R 영상과 Level 1G 영상을 이용하여 기준영상의 기하품질에 따른 상호좌표등록 결과 분석을 수행하였다. 기준영상으로 Level 1R 영상 및 1G 영상 각각을 사용하고 대상영상은 Level 1R 영상을 사용하여 상호좌표등록을 수행하였다. 실험을 위해 대전지역에서 촬영된 KOMPSAT-3 및 3A호의 Level 1R, 1G 영상 총 7장을 이용하였다. 상호좌표등록을 수행하기 위해, 우선적으로 특징기반 정합기법인 SURF (Speeded-Up Robust Feature) 기법과 영역기반 정합기법인 위상상관 (Phase Correlation) 기법을 함께 이용한 반복적 정합기법을 통해 두 영상의 기하학적 위치를 개략적으로 일치시켜 주었다. 개략적으로 일치된 영상에서 SURF 기법을 이용하여 정합쌍을 추출하고 Affine 변환모델과 Piecewise Linear 변환모델을 각각 구성하여 상호좌표등록을 수행하였다. 실험결과, 기하오차가 보정된 Level 1G 영상을 기준영상으로 선정하였을 경우, Level 1R 영상을 이용하였을 때보다 상대적으로 많은 수의 정합쌍을 추출하였다. 또한, 기준영상이 Level 1G 영상일 때의 상호좌표등록 RMSE (Root Mean Square Error) 값이 평균 5화소 미만으로 Level 1R 영상을 이용하였을 때보다 더 낮은 것을 확인하였다. 이는 상호좌표등록 수행 시 두 위성영상 간의 초기위치관계가 상호좌표등록 결과에 영향을 끼칠 수 있음을 의미하며, 기준영상의 기하품질이 우수할수록 안정적인 상호좌표등록 정확도를 나타내는 것을 확인하였다.

• 제어로봇시스템학회논문지
• /
• 제11권1호
• /
• pp.50-57
• /
• 2005

We present a simple and effective method of face and facial feature detection under pose variation of user face in complex background for the human-robot interaction. Our approach is a flexible method that can be performed in both color and gray facial image and is also feasible for detecting facial features in quasi real-time. Based on the characteristics of the intensity of neighborhood area of facial features, new directional template for facial feature is defined. From applying this template to input facial image, novel edge-like blob map (EBM) with multiple intensity strengths is constructed. Regardless of color information of input image, using this map and conditions for facial characteristics, we show that the locations of face and its features - i.e., two eyes and a mouth-can be successfully estimated. Without the information of facial area boundary, final candidate face region is determined by both obtained locations of facial features and weighted correlation values with standard facial templates. Experimental results from many color images and well-known gray level face database images authorize the usefulness of proposed algorithm.

• Journal of Information Processing Systems
• /
• 제11권3호
• /
• pp.421-437
• /
• 2015

This paper aims to present a supervised classification algorithm based on data fusion for the segmentation of the textured images. The feature extraction method we used is based on discrete wavelet transform (DWT). In the segmentation stage, the estimated feature vector of each pixel is sent to the support vector machine (SVM) classifier for initial labeling. To obtain a more accurate segmentation result, two strategies based on information fusion were used. We first integrated decision-level fusion strategies by combining decisions made by the SVM classifier within a sliding window. In the second strategy, the fuzzy set theory and rules based on probability theory were used to combine the scores obtained by SVM over a sliding window. Finally, the performance of the proposed segmentation algorithm was demonstrated on a variety of synthetic and real images and showed that the proposed data fusion method improved the classification accuracy compared to applying a SVM classifier. The results revealed that the overall accuracies of SVM classification of textured images is 88%, while our fusion methodology obtained an accuracy of up to 96%, depending on the size of the data base.

• Steel and Composite Structures
• /
• 제29권6호
• /
• pp.703-716
• /
• 2018

Rapid detection of damages in civil engineering structures, in order to assess their possible disorders and as a result produce competent decision making, are crucial to ensure their health and ultimately enhance the level of public safety. In traditional intelligent health monitoring methods, the features are manually extracted depending on prior knowledge and diagnostic expertise. Inspired by the idea of unsupervised feature learning that uses artificial intelligence techniques to learn features from raw data, a two-stage learning method is proposed here for intelligent health monitoring of civil engineering structures. In the first stage, $Nystr{\ddot{o}}m$ method is used for automatic feature extraction from structural vibration signals. In the second stage, Moving Kernel Principal Component Analysis (MKPCA) is employed to classify the health conditions based on the extracted features. In this paper, KPCA has been implemented in a new form as Moving KPCA for effectively segmenting large data and for determining the changes, as data are continuously collected. Numerical results revealed that the proposed health monitoring system has a satisfactory performance for detecting the damage scenarios of a three-story frame aluminum structure. Furthermore, the enhanced version of KPCA methods exhibited a significant improvement in sensitivity, accuracy, and effectiveness over conventional methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권4호
• /
• pp.1066-1079
• /
• 2023

Multi-target tracking based on the detector is a very hot and important research topic in target tracking. It mainly includes two closely related processes, namely target detection and target tracking. Where target detection is responsible for detecting the exact position of the target, while target tracking monitors the temporal and spatial changes of the target. With the improvement of the detector, the tracking performance has reached a new level. The problem that always exists in the research of target tracking is the problem that occurs again after the target is occluded during tracking. Based on this question, this paper proposes a DeepSORT model based on SIFT features to improve ship tracking. Unlike previous feature extraction networks, SIFT algorithm does not require the characteristics of pre-training learning objectives and can be used in ship tracking quickly. At the same time, we improve and test the matching method of our model to find a balance between tracking accuracy and tracking speed. Experiments show that the model can get more ideal results.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.27-29
• /
• 2006

In this paper, we propose a new method of automatic white balance which is one of the image signal processing techniques. Our method is conceptually based on gray world assumption. However, while previous methods generate linear results as multiplying pixel values by a gain, our method generates non-linear results using the feature of B-Spline curves. The two merits of deriving non-linear results are preventing AWB failure from transforming strong color of high level into wrong color and well preserving original contrast of an input image.