• Journal of Information Processing Systems
• /
• v.16 no.6
• /
• pp.1343-1358
• /
• 2020

In the stagewise arithmetic orthogonal matching pursuit algorithm, the weak threshold used in sparsity estimation is determined via maximum iterations. Different maximum iterations correspond to different thresholds and affect the performance of the algorithm. To solve this problem, we propose an improved variable weak threshold based on the stagewise arithmetic orthogonal matching pursuit algorithm. Our proposed algorithm uses the residual error value to control the weak threshold. When the residual value decreases, the threshold value continuously increases, so that the atoms contained in the atomic set are closer to the real sparsity value, making it possible to improve the reconstruction accuracy. In addition, we improved the generalized Jaccard coefficient in order to replace the inner product method that is used in the stagewise arithmetic orthogonal matching pursuit algorithm. Our proposed algorithm uses the covariance to replace the joint expectation for two variables based on the generalized Jaccard coefficient. The improved generalized Jaccard coefficient can be used to generate a more accurate calculation of the correlation between the measurement matrixes. In addition, the residual is more accurate, which can reduce the possibility of selecting the wrong atoms. We demonstrate using simulations that the proposed algorithm produces a better reconstruction result in the reconstruction of a one-dimensional signal and two-dimensional image signal.

• The Journal of the Korea Contents Association
• /
• v.10 no.12
• /
• pp.84-90
• /
• 2010

In this paper, we propose a quadratic discriminant analysis based approach for improving the discriminating strength of weak classifiers based on simple Haar-like features that were used in the Viola-Jones object detection framework. Viola and Jones built a strong classifier using a boosted ensemble of weak classifiers. However, their single threshold (or decision boundary) based weak classifier is sub-optimal and too weak for efficient discrimination between object class and background. A quadratic discriminant analysis based approach is presented which leads to hyper-quadric boundary between the object class and background class, thus realizing multiple thresholds based weak classifiers. Experiments carried out for car detection using 1000 positive and 3000 negative images for training, and 500 positive and 500 negative images for testing show that our method yields higher classification performance with fewer classifiers than single threshold based weak classifiers.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.5
• /
• pp.457-464
• /
• 2009

The weak classifier of AdaBoost algorithm is a central classification element that uses a single criterion separating positive and negative learning candidates. Finding the best criterion to separate two feature distributions influences learning capacity of the algorithm. A common way to classify the distributions is to use the mean value of the features. However, positive and negative distributions of Haar-like feature as an image descriptor are hard to classify by a single threshold. The poor classification ability of the single threshold also increases the number of boosting operations, and finally results in a poor classifier. This paper proposes a weak classifier that uses multiple criterions by adding a probabilistic criterion of the positive candidate distribution with the conventional mean classifier: the positive distribution has low variation and the values are closer to the mean while the negative distribution has large variation and values are widely spread. The difference in the variance for the positive and negative distributions is used as an additional criterion. In the learning procedure, we use a new classifier that provides a better classifier between them by selective switching between the mean and standard deviation. We call this new type of combined classifier the "Mixed Weak Classifier". The proposed weak classifier is more robust than the mean classifier alone and decreases the number of boosting operations to be converged.

• Journal of Oral Medicine and Pain
• /
• v.32 no.4
• /
• pp.413-419
• /
• 2007

The aims of this study were to investigate the relationships among several types of thermal pain thresholds, and pressure pain thresholds. This study was designed to examine whether there were associations among different types of pain thresholds, and among different recording sites for each pain threshold measurement. Pain sensitivity thresholds including cold pain threshold (CPT), heat pain threshold (HPT), heat pain tolerance threshold (PTT), and pressure pain threshold (PPT) of 56 subjects with symptoms of temporomandibular disorders were measured on temporal muscle, masseter muscle, TMJ, and tibial areas. Thermal pain thresholds including CPT, HPT, and PTT did not show any gender differences. However, women showed significantly lower PPTs than men on all recording sites. Three thermal pain thresholds including CPT, HPT, and PTT showed weak to high correlations on all the recording sites (r= 0.324 to 0.754, p<0.05). PPTs did not show any significant correlations between each thermal pain threshold. The pain threshold of each recording site showed weak to high correlations in all pain threshold measures (r= 0.284 to 0.878, p<0.05). Our study demonstrated that thermal pain thresholds, and pain tolerance thresholds were significantly correlated, but did not show any correlation between thermal pain thresholds and pressure pain thresholds. There were relatively high correlations among the pain thresholds of different recording sites.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.2
• /
• pp.166-171
• /
• 2013

The method proposed in this paper can improve the performance of the Boosting algorithm in machine learning. The proposed Boundary AdaBoost algorithm can make up for the weak points of Normal binary classifier using threshold boundary concepts. The new proposed boundary can be located near the threshold of the binary classifier. The proposed algorithm improves classification in areas where Normal binary classifier is weak. Thus, the optimal boundary final classifier can decrease error rates classified with more reasonable features. Finally, this paper derives the new algorithm's optimal solution, and it demonstrates how classifier accuracy can be improved using the proposed Boundary AdaBoost in a simulation experiment of pedestrian detection using 10-fold cross validation.

• Asian Journal of Business Environment
• /
• v.11 no.1
• /
• pp.17-25
• /
• 2021

Purpose: This study extends previous empirical work on the threshold effects of institutions on the relationship between infrastructure and economic growth. It does so by using three sub-indices of institutions as the threshold variable in place of aggregate index. This is with a view to determining the roles of the sub-indices in the nexus between infrastructure and economic growth. Research design, data and methodology: The analysis is based on a dynamic panel threshold regression model using a panel data set comprising 41 countries in Sub-Saharan Africa over the sample period of 1996-2015. Data are obtained from Ogbaro (2019). Results: The study finds that infrastructure exerts significant positive effects on economic growth below and above the threshold values of the three sub-indices, with higher effects above the threshold values. Results also show that on average, the Sub-Saharan African countries are not able to satisfy any of the threshold conditions, which accounts for their poor growth experience. Conclusion: The study concludes that countries with weak institutions do not benefit maximally from infrastructure development policies. The paper, therefore, recommends that countries in Sub-Saharan Africa need to focus on improving their institutional patterns if they are to reap the optimum benefits from their infrastructure development efforts.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.10 no.3
• /
• pp.123-129
• /
• 1985

This paper describes the determination of threshold value in order to determine the presence of absence of weak signal with SNR of 0 dB in 100kHz bandwidth. As a detection method, it has been used a recent LOC structure fitting for detecting weak signal in stead of a conventional method like Neyman-Peason crtical criterion. The signal for detection is the OOK modulation signal used in data and morse code transmission. The non-Gaussian noise similar to Laplacian type has been chosen in transmission path. As a result of experiment, comparing probability of detection by one critical point with that by two critical points with fixing as arbitrary false alarm probability, we have found that method has been shown to be better than the conventional method.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.53 no.6
• /
• pp.337-341
• /
• 2004

In this paper the effect of chaos induced instability in Brillouin-active fiber-ring sensor is described. The inherent optical feedback by the backscattered Stokes wave in optical fiber leads to instabilities in the form of optical chaos. The paradigm of optical chaos in fiber serves as a test for fundamental study of chaos and its suppression and exploitation in practical application in communication and sensing. At weak power, the nature of the Brillouin instability can occur at before threshold. At strong power, the temporal evolution above threshold is periodic and at higher intensity can become chaotic. The threshold for the Brillouin instability in fiber-ring sensor is much lower than the threshold of the normal Brillouin instability process.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.794-796
• /
• 2006

This study investigated the Brightness Temperature Difference threshold as criterion between aerosols and clouds in conjunction with radiative transfer model. Surface temperature is caused by a significant error over 50% in the BTD threshold. In addition, The BTD threshold contains the uncertainties about 20% due to the surface emissivity and 8% due to the satellite zenith angle. Therefore, we have composed the Look-up table for BTD between 11㎛and 12㎛ according to satellite zenith angle, surface temperature, and surface emissivity. The modified BTD show the enhanced signal, especially over bright surface such as desert in China. However, a weak aerosol signal over Ocean remains in the modified BTD.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2012.01a
• /
• pp.209-211
• /
• 2012

본 논문에서는 AdaBoost 알고리즘에서 약분류기(Weak Classifier)의 성능을 개선하기 위한 임계값 설정 방법을 제안한다. 일반적으로 약분류기에 사용되는 임계값은 특징들의 평균값을 많이 사용하지만 이는 특징들의 분포가 고려되지 않았기 때문에 분별력이 많이 떨어진다. 그러므로 각 특징들의 분포를 고려한 약분류기의 임계값 설정방법을 제안한다. 이는 얼굴에 대한 간단한 학습 및 테스트를 통하여 기존 방법에 비하여 더 나은 성능을 보임을 입증한다.