• Korean Journal of Radiology
• /
• 제22권3호
• /
• pp.415-424
• /
• 2021

Objective: To determine whether noncontrast computed tomography (NCCT) models based on multivariable, radiomics features, and machine learning (ML) algorithms could further improve the discrimination of early hematoma expansion (HE) in patients with spontaneous intracerebral hemorrhage (sICH). Materials and Methods: We retrospectively reviewed 261 patients with sICH who underwent initial NCCT within 6 hours of ictus and follow-up CT within 24 hours after initial NCCT, between April 2011 and March 2019. The clinical characteristics, imaging signs and radiomics features extracted from the initial NCCT images were used to construct models to discriminate early HE. A clinical-radiologic model was constructed using a multivariate logistic regression (LR) analysis. Radiomics models, a radiomics-radiologic model, and a combined model were constructed in the training cohort (n = 182) and independently verified in the validation cohort (n = 79). Receiver operating characteristic analysis and the area under the curve (AUC) were used to evaluate the discriminative power. Results: The AUC of the clinical-radiologic model for discriminating early HE was 0.766. The AUCs of the radiomics model for discriminating early HE built using the LR algorithm in the training and validation cohorts were 0.926 and 0.850, respectively. The AUCs of the radiomics-radiologic model in the training and validation cohorts were 0.946 and 0.867, respectively. The AUCs of the combined model in the training and validation cohorts were 0.960 and 0.867, respectively. Conclusion: NCCT models based on multivariable, radiomics features and ML algorithm could improve the discrimination of early HE. The combined model was the best recommended model to identify sICH patients at risk of early HE.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 제11회 음성통신 및 신호처리 워크샵 논문집 (SCAS 11권 1호)
• /
• pp.138-141
• /
• 1994

기존의 HMM을 이용한 음성인식기는 대부분 ML 추정에 기초한 Baum-Welch 알고리듬으로 학습되었다. ML학습은 기본적으로 무한한 양의 학습 데이터가 주어지고, 각 모델들이 서로 독립이라는 가정에 기초한다. 하지만 실제적인 학습의 경우에 각 모델들이 서로 독립이라고 보기 어렵고, 학습 데이터의 양도 상당히 제한되어 있어서 인식기의 변별력을 저하시키는 주된 원인이 되고 있다. 본 논문에서는 전통적인 패턴분류기법인 Bayes 결정이론에 따라 최소오차율분류를 위한 MAP 수식화를 유도하고, 그에 기초한 HMM의 변별력 있는 학습 알고리듬을 제안한다. 최소오차율분류를 근사화한 사후확률로 표현된 비용함수를 정의하고, 그 비용함수에 조건부 경사강하법을 적용한다. 제안된 알고리듬을 분류하기 어려운 한국어 단음절 인식에 적용한 결과, 기존의 ML 알고리듬으로 학습한 경우 발생한 오인식 개수의 약 10% 가량이 개선되었다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2007년도 춘계학술발표대회
• /
• pp.397-398
• /
• 2007

본 논문에서는 환경변화에 대해 강인하게 동작하는 음성인식 시스템을 위해 잡음적응 훈련과 변별학습 방식을 결합한 형태의 환경적응 방식을 제안한다. 다중환경 훈련과 잡음제거방식을 결합한 형태인 잡음적응 훈련 방식은 음성인식을 위한 MCE (Minimum Classification Error)의 목적과는 거리가 있고, 음성인식 시스템이 사용되는 모든 환경을 반영하는 것은 현실적으로 어렵다는 점에서 한계가 있다. 이에 잡음적응 훈련방식으로 훈련된 기본 음향모델을 목적환경에서 수집한 소량의 데이터를 이용한 변별학습을 통해 환경적응 모델로 변환함으로써 이러한 단점을 보완할 수 있는 잡음 적응 변별학습을 이용한 훈련방식을 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권1호
• /
• pp.451-465
• /
• 2017

This paper presents a novel discriminative visual tracking algorithm with an adaptive incremental extreme learning machine. The parameters for an adaptive incremental extreme learning machine are initialized at the first frame with a target that is manually assigned. At each frame, the training samples are collected and random Haar-like features are extracted. The proposed tracker updates the overall output weights for each frame, and the updated tracker is used to estimate the new location of the target in the next frame. The adaptive learning rate for the update of the overall output weights is estimated by using the confidence of the predicted target location at the current frame. Our experimental results indicate that the proposed tracker can manage various difficulties and can achieve better performance than other state-of-the-art trackers.

• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.52-60
• /
• 2022

Few-shot learning is a sub-area of machine learning problems, which aims to classify target images that only contain a few labeled samples for training. As a representative few-shot learning method, the Prototypical network has been received much attention due to its simplicity and promising results. However, the Prototypical network uses the sample mean of samples from the same class as the prototypes of that class, which easily results in learning uncharacteristic features in the low-data scenery. In this study, we propose to use local descriptors (i.e., patches along the channel within feature maps) from the same class to explicitly obtain more representative prototypes for Prototypical Network so that significant intra-class feature information can be maintained and thus improving the classification performance on few-shot learning tasks. Experimental results on various benchmark datasets including mini-ImageNet, CUB-200-2011, and tiered-ImageNet show that the proposed method can learn more discriminative intra-class features by the local descriptors and obtain more generic prototype representations under the few-shot setting.

• 스마트미디어저널
• /
• 제12권9호
• /
• pp.45-59
• /
• 2023

Weakly supervised object localization (WSOL) is a task of localizing an object in an image using only image-level labels. Previous studies have followed the conventional class activation mapping (CAM) pipeline. However, we reveal the current CAM approach suffers from problems which cause original CAM could not capture the complete defects features. This work utilizes a convolutional neural network (CNN) pretrained on image-level labels to generate class activation maps in a multi-scale manner to highlight discriminative regions. Additionally, a vision transformer (ViT) pretrained was treated to produce multi-head attention maps as an auxiliary detector. By integrating the CNN-based CAMs and attention maps, our approach localizes defective regions without requiring bounding box or pixel-level supervision during training. We evaluate our approach on a dataset of apple images with only image-level labels of defect categories. Experiments demonstrate our proposed method aligns with several Object Detection models performance, hold a promise for improving localization.

• 지능정보연구
• /
• 제25권4호
• /
• pp.53-65
• /
• 2019

영상 기반의 보안 시스템의 증가함에 따라 각 용도마다 다른 다양한 객체들에 대한 처리들이 중요해지고 있다. 객체 추적은 객체 인식, 검출과 같은 작업들과 함께 필수적인 작업으로 다뤄진다. 이 객체 추적을 달성하기 위해서 다양한 머신러닝이 적용될 수 있다. 성공적인 분류기로써 전체 에러율 최소화(total-error-rate minimization) 기반의 방법론이 사용될 수 있다. 이 전체 에러율 최소화 기반의 방법론은 오프라인 학습을 기반으로 하고 있다. 객체 추적은 실시간으로 처리하며 갱신해야하는 것이 필수적이므로 온라인 학습(online learning)을 기반으로 하는 것이 적합하다. 온라인 전체 에러율 최소화 방법론이 개발되었지만 점근적으로 재가중되는(approximately reweighted) 작업이 포함되어 에러를 누적시킬 수 있다는 단점이 있다. 본 논문에서는 정확하게 재가중되는(exactly reweighted) 방법론을 제안하면서 온라인 전체 에러율 최소화가 달성되었다. 이 제안된 온라인 학습 방법론을 객체 추적에 적용하여 총 8개의 데이터베이스에서 다른 추적 방법론들 보다 좋은 성능이 달성되었다.

• 대한전자공학회논문지SP
• /
• 제46권3호
• /
• pp.138-142
• /
• 2009

본 논문에서는 보이스 피싱 (Voice Pishing) 예방을 위한 알고리즘을 최소 분류 오차 기법 (Minimum Classification Error)을 기반으로 제한하다. 휴대폰으로 전송되어진 신호를 기반으로 3GPP2 Selectable Mode Vocoder (SMV)의 복호화 과정에서 자동적으로 추출되는 중요 특징벡터를 사용하여 Gaussian Mixture Model (GMM)을 구성하고 이를 기반으로 구해지는 로그(Log) 기반의 우도 (Likelihood)를 사용한 변별적 가중치 학습을 사용하여 보이스 피싱 예방을 위한 검출 알고리즘을 제안하다. 실험 결과 제안된 보이스 피싱 알고리즘이 기존의 방법에 비해 우수한 성능을 보인 것을 알 수 있었다.

• 수산해양교육연구
• /
• 제28권3호
• /
• pp.661-666
• /
• 2016

This study was to explore the living experiences of the disabled adults who were participating in public education Yahak program held at evening class. The study designed in-depth interviews with ten disabled people using a grounded theory approach. Through analyzing process, 34 concepts, 15 subcategories, and eight categories were deduced. In axial coding, casual condition, 'Suffering from unknown cause disabilities' and 'Isolated by social cause', context condition, 'Taking discriminative treat for disabilities' impacted on phenomenon, 'Overcoming their conditions by themselves'. Intervening conditions was 'Taking social supports' and action-interaction condition, 'Enjoying public programs' totally lead to consequence in 'Controlling daily life' and 'Exploring their own social roles'. The periods of process were divided three stages, reflecting disabled situation, formation phase of social relation, and self-developing phase. The core category, 'Trying to be recognized as a member of society' incorporated the relationship between and among all categories and explained the process. The study indicates that social education program for the disabled helped to develop themselves as a member of society. Therefore, we suggest there may be a need for training for professionals who work with disabled people to develop social adaptation.

• 한국컴퓨터정보학회논문지
• /
• 제22권12호
• /
• pp.155-161
• /
• 2017

Recently, as the number of multicultural families and foreigners living in korea increase, the proportion of various types of families and middle-admitted youths is increasing. These youths are less educated than their domestic counterparts, and their conversation time with their family members is relatively weak. Therefore, there is a need for a specialized education system for education and socialization. Immigration background among middle-admitted adolescents, children arrived in korea regardless of their will, with socialization already in the country where they were born, it is a reality that various difficulties are experienced in the socialization of korea society about language, education, emotion and employment. For this reason, some of the migrant background youths are pointed out as a big problem of the multicultural society, which is 18% of the NEET(Not in Education, Employment or Training) classes, which are not educated and are not willing to find jobs or employment. Therefore, in this study, we identified the problems of middle-admitted children of multicultural families as the number of middle-admitted adolescents increased, and suggested the necessary ways for them to achieve rapid socialization and settlement in korea society. For this purpose, we analyzed the problem of education of middle-admitted children as a discriminative approach which is different from general support method for middle-admitted children presented in previous reaearch, since then, we have presented an alternative to carry out realistic, systematic and successful education considering the characteristics of the region centered on the middle-admitted youths of Gwangju city.