• 대한원격탐사학회지
• /
• 제18권3호
• /
• pp.155-162
• /
• 2002

Neural networks have been focused on as a robust classifier for the remotely sensed imagery due to its statistical independency and teaming ability. Also the artificial neural networks have been reported to be more tolerant to noise and missing data. However, unlike the conventional statistical classifiers which use the statistical parameters for the classification, a neural network classifier uses individual training sample in teaming stage. The training performance of a neural network is know to be very sensitive to the discrepancy of the number of the training samples of each class. In this paper, the effect of the population discrepancy of training samples of each class was analyzed with three layered feed forward network. And a method for reducing the effect was proposed and experimented with Landsat TM image. The results showed that the effect of the training sample size discrepancy should be carefully considered for faster and more accurate training of the network. Also, it was found that the proposed method which makes teaming rate as a function of the number of training samples in each class resulted in faster and more accurate training of the network.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.386-389
• /
• 2008

It is important that the training stage of a supervised classification is designed to provide the spectral information. On the design of the training stage of a classification typically calls for the use of a large sample of randomly selected pure pixels in order to characterize the classes. Such guidance is generally made without regard to the specific nature of the application in-hand, including the classifier to be used. An approach to the training of a support vector machine (SVM) classifier that is the opposite of that generally promoted for training set design is suggested. This approach uses a small sample of mixed spectral responses drawn from purposefully selected locations (geographical boundaries) in training. A sample of such data should, however, be easier and cheaper to acquire than that suggested by traditional approaches. In this research, we evaluated them against traditional approaches with high-resolution satellite data. The results proved that it can be used small mixed pixels to derive a classification with similar accuracy using a large number of pure pixels. The approach can also reduce substantial costs in training data acquisition because the sampling locations used are commonly easy to observe.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권7호
• /
• pp.3172-3193
• /
• 2018

Recently, several studies have shown that linear representation based approaches are very effective and efficient for image classification. One of these linear-representation-based approaches is the Collaborative representation (CR) method. The existing algorithms based on CR have two major problems that degrade their classification performance. First problem arises due to the limited number of available training samples. The large variations, caused by illumintion and expression changes, among query and training samples leads to poor classification performance. Second problem occurs when an image is partially noised (contiguous occlusion), as some part of the given image become corrupt the classification performance also degrades. We aim to extend the collaborative representation framework under limited training samples face recognition problem. Our proposed solution will generate virtual samples and intra-class variations from training data to model the variations effectively between query and training samples. For robust classification, the image patches have been utilized to compute representation to address partial occlusion as it leads to more accurate classification results. The proposed method computes representation based on local regions in the images as opposed to CR, which computes representation based on global solution involving entire images. Furthermore, the proposed solution also integrates the locality structure into CR, using Euclidian distance between the query and training samples. Intuitively, if the query sample can be represented by selecting its nearest neighbours, lie on a same linear subspace then the resulting representation will be more discriminate and accurately classify the query sample. Hence our proposed framework model the limited sample face recognition problem into sufficient training samples problem using virtual samples and intra-class variations, generated from training samples that will result in improved classification accuracy as evident from experimental results. Moreover, it compute representation based on local image patches for robust classification and is expected to greatly increase the classification performance for face recognition task.

• 한국응급구조학회지
• /
• 제21권1호
• /
• pp.59-73
• /
• 2017

Purpose: This study aimed to enhance the efficiency of clinical training education by understanding paramedic students' perceptions of their hospital clinical training experiences. Methods: The subjects were 31 third paramedic students who participated in a population survey from June 25 to August 13, 2016. A Q card and Q sample distribution chart were created, and the P sample was selected by Q classification. The collected data were analyzed by factorial analysis using PC QUANL. Results: Four different perceptions were identified from the survey, which explained 44.1% of the variables. The four types were classified as Self-improvement-oriented (Type 1), Training-site avoidant (Type 2), Confidence acquiring (Type 3), and Over-willed (Type 4). Conclusion: Paramedic instructors and clinical training managers may want to consider these four perception types when planning clinical training and education programs to improve job performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권1호
• /
• pp.368-391
• /
• 2018

Face recognition (FR) with a single sample per person (SSPP) is common in real-world face recognition applications. In this scenario, it is hard to predict intra-class variations of query samples by gallery samples due to the lack of sufficient training samples. Inspired by the fact that similar faces have similar intra-class variations, we propose a virtual sample generating algorithm called k nearest neighbors based virtual sample generating (kNNVSG) to enrich intra-class variation information for training samples. Furthermore, in order to use the intra-class variation information of the virtual samples generated by kNNVSG algorithm, we propose image set based multimanifold discriminant learning (ISMMDL) algorithm. For ISMMDL algorithm, it learns a projection matrix for each manifold modeled by the local patches of the images of each class, which aims to minimize the margins of intra-manifold and maximize the margins of inter-manifold simultaneously in low-dimensional feature space. Finally, by comprehensively using kNNVSG and ISMMDL algorithms, we propose k nearest neighbor virtual image set based multimanifold discriminant learning (kNNMMDL) approach for single sample face recognition (SSFR) tasks. Experimental results on AR, Multi-PIE and LFW face datasets demonstrate that our approach has promising abilities for SSFR with expression, illumination and disguise variations.

• 한국컴퓨터정보학회논문지
• /
• 제18권4호
• /
• pp.19-26
• /
• 2013

의사 샘플 신경망은 학습 샘플의 수가 적은 경우 학습된 신경망이 국부 최적해에 빠져 성능이 저하되는 것을 보완하기 위해 기존 샘플들로부터 의사 샘플을 생성하고 이를 통해 해공간을 평탄화 시킴으로써 학습된 신경망의 성능을 향상시킬 수 있는 신경망의 변형이다. 이는 학습 샘플의 양에 관한 문제로 이 논문에서는 이에 더해 학습 샘플의 질을 향상시킴으로써 학습된 신경망의 성능을 더욱 높일 수 있는 방법을 제시하였다. 잡음이 적게 포함된 전형적인 학습 샘플들만이 주어지고 입력 특징 중 출력과 연관성이 높은 특징만을 사용함으로써 학습된 신경망의 성능을 높일 수 있음은 자명하다. 따라서 이 논문에서는 커널밀도 추정을 통해 비전형적인 학습샘플을 제거하고 입력값이 출력값에 미치는 영향을 나타내는 연관성 척도를 사용하여 연관성이 적은 특징을 제거함으로써 의사 샘플 신경망의 성능을 향상시킬 수 있음을 보였다. 제시한 방법의 유효성은 토석류 데이터를 이용한 실험을 통해 확인할 수 있다.

• 한방재활의학과학회지
• /
• 제15권1호
• /
• pp.39-65
• /
• 2005

Objectives : This study was aimed to find out correlation of relation between cardiopulmonary function and body fat. Methods : We studied tendency of change of cardiopulmonary function and body fat for medication of Ephedrae Herba capsule by ergogenic aids with circuit training. We got the results for Exercise stress test and Segmental Bioelectrical Impedence Analysis. Results : 1. Sample Group of Ephedrae Herba medication and Circuit training generally showed the insignificant improvement of Body composition, but Control Group of Placebo and Circuit training significantly(p<.05) showed significant improvement of Body composition. 2. Sample Group of Ephedrae Herba medication and Circuit training generally showed the significant improvement of cardio-pulmonary function. Control Group of Placebo and Circuit training showed insignificant elevation of Cardiopulmonary function. 3. In the case of Sample Group, there wasn't closely correlation relationship of improvement of cardiopulmonary function and body composition, but in the case of Control Group, there was closely correlation relationship of improvement of cardiopulmonary function and body composition. Conclusions : It might be recognized that cardiopulmonary function has the correlation of body composition, and Ephedrae Herba might help the reduction of Body Fat by elevation of Cardiopulmonary function for ergogenic aids and it might be needed further study In various viewpoints.

• 대한공간정보학회지
• /
• 제24권1호
• /
• pp.121-128
• /
• 2016

많은 연구들에서 영상자료와 분류 알고리즘 측면에서 분류정확도를 비교하였지만, 참조자료 또는 분석자에 의존하는 훈련샘플에 의한 분류정확도 비교와 관련된 연구는 부족한 실정이다. 본 연구는 감독분류에 있어 훈련샘플로써 지상 분광반사자료의 유용성을 평가하고자 하였다. 이를 위하여 초분광영상과 다중분광영상을 대상으로 영상 수집 훈련샘플과 지상 분광반사자료를 사용하여 분류 정확도를 비교하였다. 그 결과 영상 수집 훈련샘플 사용 시 초분 광영상과 다중분광영상에서 공통적으로 약 90%의 분류정확도를 얻을 수 있었다. 그러나 지상 분광반사자료를 훈련 샘플로 사용하면 초분광영상의 경우 약 10%p, 다중분광영상의 경우 약 20%p의 분류정확도 감소가 발생하였다. 특히 다중분광영상에서 분광반사특성이 유사하게 나타나는 클래스들의 경우 분류정확도가 초분광영상에 비해 매우 낮게 나타났다. 따라서 지상 분광반사자료는 다중분광영상에 적용하는 데에는 한계가 있지만, 초분광영상을 이용한 토지피복분류에 있어 유용한 훈련샘플이 될 수 있다.

• 대한토목학회논문집
• /
• 제26권3D호
• /
• pp.349-352
• /
• 2006

자동차 수의 급증으로 야기되는 교통혼잡, 교통사고, 주차난 등의 많은 문제에 효율적으로 대응하기 위해서는 제한된 인력과 비용을 사용하는 자동차 관리가 필수적인데 이를 위한 많은 연구들이 국내외적으로 현재 진행되고 있다. 현재 진행되고 있는 여러 연구 분야 중에서 특히 자동차의 차량번호판인식 기술은 법규위반 차량 식별, 통행료 징수, 자동차세 징수, 도난 도주 차량 확인 및 주차 관리 등의 많은 분야에 응용되고 있다. 자동차의 차량번호판 문자 인식 문제와 같이 훈련예제 수집 비용이 많이 드는 경우에 제한된 수의 훈련예제를 최대한 활용하여 분류성능을 향상시키기 위한 방안의 하나로, 수집된 훈련예제들로부터 가상의 예제를 생성하고, 생성된 가상예제를 훈련예제로 추가하여 학습하는 여러 연구가 수행된 바 있다. 본 논문에서는 차량번호판 문자 인식의 성능 향상을 위해 수집된 예제들을 적절히 병합하여 가상의 예제를 생성하는 방안에 관해 기술하고, 문자인식 분야에서 일반적으로 많이 사용되는 여러 알고리즘에 대하여 다양한 가상예제 생성방안 및 다양한 생성비율에 따른 실험을 통해 그 효용성을 확인한다.

• ETRI Journal
• /
• 제39권5호
• /
• pp.621-631
• /
• 2017

Abnormal samples are usually difficult to obtain in production systems, resulting in imbalanced training sample sets. Namely, the number of positive samples is far less than the number of negative samples. Traditional Support Vector Machine (SVM)-based anomaly detection algorithms perform poorly for highly imbalanced datasets: the learned classification hyperplane skews toward the positive samples, resulting in a high false-negative rate. This article proposes a new imbalanced SVM (termed ImSVM)-based anomaly detection algorithm, which assigns a different weight for each positive support vector in the decision function. ImSVM adjusts the learned classification hyperplane to make the decision function achieve a maximum GMean measure value on the dataset. The above problem is converted into an unconstrained optimization problem to search the optimal weight vector. Experiments are carried out on both Cloud datasets and Knowledge Discovery and Data Mining datasets to evaluate ImSVM. Highly imbalanced training sample sets are constructed. The experimental results show that ImSVM outperforms over-sampling techniques and several existing imbalanced SVM-based techniques.