• 한국정보과학회논문지:소프트웨어및응용
• /
• 제33권12호
• /
• pp.999-1007
• /
• 2006

역전파 학습 방법은 속도가 느리고, 지역 최소점이나 고원에 빠져 수렴에 실패하는 경우가 많다고 알려져 있다. 이제까지 알려진 역전파의 대체 방법들은 수렴 속도와 변수에 따른 수렴의 안정성 사이에서 불균형이라는 대가를 치루고 있다. 기존의 전통적인 역전파에서 발생하는 위와 같은 문제점 중, 특히 지역 최소점을 탈피하는 기능을 추가하여 적은 저장 공간으로 안정성이 보장되면서도 빠른 수렴속도를 유지하는 알고리즘을 제안한다. 이 방법은 전체 신경망을 은닉층-출력층(hidden to output)을 의미하는 상위 연결(upper connections)과 입력층-은닉층(input to hidden)을 의미하는 하위 연결(lower connections) 2개로 분리하여 번갈아 훈련을 시키는 분리 학습방법을 적용한다. 본 논문에서 제안하는 알고리즘은 다양한 classification 문제에 적용한 실험 결과에서 보듯이 전통적인 역전파 및 기타 개선된 알고리즘에 비해 계산량이 적고, 성능이 매우 좋으며 높은 신뢰성을 보장한다.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2006년도 제26회 춘계학술대회논문집
• /
• pp.209-212
• /
• 2006

본 논문에서는 항공기용 터보 축 엔진의 결함 진단 알고리즘을 개발하지 위해 Support Vector Machine(SVM)과 인공신경망(ANN)을 이용하였다. SVM을 이용하여 결함 위치를 판별한 후 인공신경망이 선택적으로 학습하는 분할 학습 알고리즘(SLA)을 제안하였으며 이를 고도 변화에 따른 가스 터빈 엔진의 결함 진단에 적용하여 분류 속도 및 예측 정확률 개선 가능성을 확인하였다.

• Journal of KIEE
• /
• 제10권1호
• /
• pp.7-15
• /
• 2000

Source separation is a statistical method, the goal of which is to separate the linear instantaneous mixtures of statistically independent sources without resorting to any prior knowledge. This paper addresses a source separation algorithm which is able to separate the mixtures of sub- and super-Gaussian sources. The nonlinear function in the proposed algorithm is derived from the generalized Gaussian distribution that is a set of distributions parameterized by a real positive number (Gaussian exponent). Based on the relationship between the kurtosis and the Gaussian exponent, we present a simple and efficient way of selecting proper nonlinear functions for source separation. Useful behavior of the proposed method is demonstrated by computer simulations.

• 한국추진공학회지
• /
• 제11권1호
• /
• pp.43-50
• /
• 2007

본 논문에서는 고도 변화만을 고려한 탈설계 영역에서 항공기용 터보 축 엔진의 결함 진단을 위해 지지 벡터 장치(SVM)과 인공신경망(ANN)을 Hybrid로 사용한 분할 학습 알고리즘을 사용하였다. 지상 정지 상태에서보다 학습 데이터와 테스트 데이터 수가 크게 증가하지만, 분할 학습 알고리즘을 이용한 가스터빈 엔진의 결함 진단이 고도 변화를 고려한 탈설계 영역에서도 높은 결함 예측 정확성을 가짐을 확인하였다.

• 한국추진공학회지
• /
• 제10권2호
• /
• pp.102-109
• /
• 2006

본 논문에서 항공기용 터보 축 엔진의 결함 진단 알고리즘을 개발하기 위해 Support Vector Machine(SVM)과 인공신경망(ANN)을 이용하였다. 신경망을 이용한 시스템은 비선형성이 과도한 데이터를 학습할 때 지역 최소점(Local Minima)에 빠져 분류 정확률이 낮아질 수 있다. 이러한 위험성을 보안하기 위해 SVM에 의한 ANN의 분할 학습 알고리즘(SLA)을 제안하였다. 이것은 SVM을 이용하여 결함 위치를 판별 한 후 신경망이 선택적으로 학습을 하는 방법으로 학습 데이터의 비선형성을 줄여 분류 정확률을 높이기 때문에 신경망을 단독으로 사용할 때보다 개선된 성능을 보여주었다.

• Journal of Information Processing Systems
• /
• 제10권3호
• /
• pp.443-458
• /
• 2014

An extreme learning machine (ELM) is a recently proposed learning algorithm for a single-layer feed forward neural network. In this paper we studied the ensemble of ELM by using a bagging algorithm for facial expression recognition (FER). Facial expression analysis is widely used in the behavior interpretation of emotions, for cognitive science, and social interactions. This paper presents a method for FER based on the histogram of orientation gradient (HOG) features using an ELM ensemble. First, the HOG features were extracted from the face image by dividing it into a number of small cells. A bagging algorithm was then used to construct many different bags of training data and each of them was trained by using separate ELMs. To recognize the expression of the input face image, HOG features were fed to each trained ELM and the results were combined by using a majority voting scheme. The ELM ensemble using bagging improves the generalized capability of the network significantly. The two available datasets (JAFFE and CK+) of facial expressions were used to evaluate the performance of the proposed classification system. Even the performance of individual ELM was smaller and the ELM ensemble using a bagging algorithm improved the recognition performance significantly.

• 디지털산업정보학회논문지
• /
• 제16권4호
• /
• pp.19-28
• /
• 2020

Vehicle license plate recognition system is not generalized in Malaysia due to the loose character layout rule and the varying number of characters as well as the mixed capital English characters and italic English words. Because the italic English word is hard to segmentation, a separate method is required to recognize in Malaysian license plate. In this paper, we propose a mixed character level and word level English license plate recognition algorithm using deep learning neural networks. The difference of Gaussian method is used to segment character and word by generating a black and white image with emphasized character strokes and separated touching characters. The proposed deep learning neural networks are implemented on the LPR system at the gate of a building in Kuala-Lumpur for the collection of database and the evaluation of algorithm performance. The evaluation results show that the proposed Malaysian English LPR can be used in commercial market with 98.01% accuracy.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.858-869
• /
• 2023

As one of the pivotal techniques of image restoration, single-channel blind source separation (SCBSS) is capable of converting a visual-only image into multi-source images. However, image degradation often results from multiple mixing methods. Therefore, this paper introduces an innovative SCBSS algorithm to effectively separate source images from a composite image in various mixed modes. The cornerstone of this approach is a novel triple generative adversarial network (TriGAN), designed based on dual learning principles. The TriGAN redefines the discriminator's function to optimize the separation process. Extensive experiments have demonstrated the algorithm's capability to distinctly separate source images from a composite image in diverse mixed modes and to facilitate effective image restoration. The effectiveness of the proposed method is quantitatively supported by achieving an average peak signal-to-noise ratio exceeding 30 dB, and the average structural similarity index surpassing 0.95 across multiple datasets.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권2호
• /
• pp.69-77
• /
• 2002

Speaker verification can be classified in two categories, text-dependent speaker verification and text-independent speaker verification. In this paper, we discuss text-dependent speaker verification. Text-dependent speaker verification system determines whether the sound characteristics of the speaker are equal to those of the specific person or not. In this paper we obtain the speaker data using a sound card in various noisy conditions, apply a new Hidden LMS (Least Mean Square) adaptive algorithm to it, and extract LPC (Linear Predictive Coding)-cepstrum coefficients as feature vectors. Finally, we use a competitive learning neural network for speaker verification. The proposed hidden LMS adaptive filter using a neural network reduces noise and enhances features in various noisy conditions. We construct a separate neural network for each speaker, which makes it unnecessary to train the whole network for a new added speaker and makes the system expansion easy. We experimentally prove that the proposed method improves the speaker verification performance.

• 한국멀티미디어학회논문지
• /
• 제21권2호
• /
• pp.98-107
• /
• 2018

In this paper, we propose a method to classify the skeletal maturity with a small amount of hand wrist X-ray image using deep learning-based meta-learning. General deep-learning techniques require large amounts of data, but in many cases, these data sets are not available for practical application. Lack of learning data is usually solved through transfer learning using pre-trained models with large data sets. However, transfer learning performance may be degraded due to over fitting for unknown new task with small data, which results in poor generalization capability. In addition, medical images require high cost resources such as a professional manpower and mcuh time to obtain labeled data. Therefore, in this paper, we use meta-learning that can classify using only a small amount of new data by pre-trained models trained with various learning tasks. First, we train the meta-model by using a separate data set composed of various learning tasks. The network learns to classify the bone maturity using the bone maturity data composed of the radiographs of the wrist. Then, we compare the results of the classification using the conventional learning algorithm with the results of the meta learning by the same number of learning data sets.