• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.53-65
• /
• 2019

Object tracking is one of important steps to achieve video-based surveillance systems. Object tracking is considered as an essential task similar to object detection and recognition. In order to perform object tracking, various machine learning methods (e.g., least-squares, perceptron and support vector machine) can be applied for different designs of tracking systems. In general, generative methods (e.g., principal component analysis) were utilized due to its simplicity and effectiveness. However, the generative methods were only focused on modeling the target object. Due to this limitation, discriminative methods (e.g., binary classification) were adopted to distinguish the target object and the background. Among the machine learning methods for binary classification, total error rate minimization can be used as one of successful machine learning methods for binary classification. The total error rate minimization can achieve a global minimum due to a quadratic approximation to a step function while other methods (e.g., support vector machine) seek local minima using nonlinear functions (e.g., hinge loss function). Due to this quadratic approximation, the total error rate minimization could obtain appropriate properties in solving optimization problems for binary classification. However, this total error rate minimization was based on a batch mode setting. The batch mode setting can be limited to several applications under offline learning. Due to limited computing resources, offline learning could not handle large scale data sets. Compared to offline learning, online learning can update its solution without storing all training samples in learning process. Due to increment of large scale data sets, online learning becomes one of essential properties for various applications. Since object tracking needs to handle data samples in real time, online learning based total error rate minimization methods are necessary to efficiently address object tracking problems. Due to the need of the online learning, an online learning based total error rate minimization method was developed. However, an approximately reweighted technique was developed. Although the approximation technique is utilized, this online version of the total error rate minimization could achieve good performances in biometric applications. However, this method is assumed that the total error rate minimization can be asymptotically achieved when only the number of training samples is infinite. Although there is the assumption to achieve the total error rate minimization, the approximation issue can continuously accumulate learning errors according to increment of training samples. Due to this reason, the approximated online learning solution can then lead a wrong solution. The wrong solution can make significant errors when it is applied to surveillance systems. In this paper, we propose an exactly reweighted technique to recursively update the solution of the total error rate minimization in online learning manner. Compared to the approximately reweighted online total error rate minimization, an exactly reweighted online total error rate minimization is achieved. The proposed exact online learning method based on the total error rate minimization is then applied to object tracking problems. In our object tracking system, particle filtering is adopted. In particle filtering, our observation model is consisted of both generative and discriminative methods to leverage the advantages between generative and discriminative properties. In our experiments, our proposed object tracking system achieves promising performances on 8 public video sequences over competing object tracking systems. The paired t-test is also reported to evaluate its quality of the results. Our proposed online learning method can be extended under the deep learning architecture which can cover the shallow and deep networks. Moreover, online learning methods, that need the exact reweighting process, can use our proposed reweighting technique. In addition to object tracking, the proposed online learning method can be easily applied to object detection and recognition. Therefore, our proposed methods can contribute to online learning community and object tracking, detection and recognition communities.

• The Journal of Korean Academy of Prosthodontics
• /
• v.29 no.3
• /
• pp.141-170
• /
• 1991

This study was to analyze the displacement and the magnitude and mode of distribution of the stresses in the lower overdenture, the mucous membrane, the abutment teeth and the mandibular supporting bone when various abutment designs were subjected to different loading schemes. For this study, the two-dimensional finite element method was used. The models of overdenture and mandibe with the canine and the second premolar remaining, were fabricated. In the first design, a 1 mm space was prepared between the denture and the dome abutment with the height of 2 mm(OS). In the second design, a contact between the denture and the occlusal third of the dome abutment with the hight of 2 mm was prepared(OC). In the third design, a 0.5 mm space was prepared between the denture and 8 degree tapered cylindrical abutments with the height of 7 mm(TS). In the fourth design, a contact between the denture and the occlusal two thirds of the conical abutments with the height of 7 mm was prepared(TC). In order to represent the same physiological condition as the fixed areas of the mandible under loading schemes, the eight nodes which lie at the mandibular angle, the coronoid process and the mandibular condyle were assumed to be fixed. Each model was loaded with a magnitude of 10 Kgs on the first molar region (P1) and 7 Kgs on the central incisor region (P2) in a vertical direction. The force of 10 Kgs was then applied distributively from the first premolar to the second molar of each motel in a vertical direction (P3). The results were as follows: 1. The vertical load on the central incisor region(P2) produced the higher displacement and stress concentration than that on the posterior region(P1, P3). 2. The case of space between abutment and denture base produced higher displacement than that of contact, and the case of long abutment produced higher displacement than that of short abutment because of low rigidity of denture base. 3. The magnitude of the torque and vertical force to the abutment teeth and the stress distribution to the denture base was higher in the telescope coping than in the overdenture coping. 4. The vertical load on the central incisor region(P2) produced higher equivalent stress in the mandible than that on the posterior region(P1, P3). 5. The case of space between abutment and denture base produced better stress distribution to the farther abutment from the loading point than that of contact. 6. In case of sound abutment teeth, the type of telescope coping can be used, hilt in case of weak abutment, the type of overdenture coping is considered to be favorable generally.

• The Journal of Korean Academy of Prosthodontics
• /
• v.28 no.1
• /
• pp.63-94
• /
• 1990

The purpose of this study was to analyze the displacement and the magnitude and the mode of distribution of the stresses in the lower overdenture, the mucous membrane, the abutment tooth and the mandibular supporting bone when various denture base materials, such as acrylic resin and 0.5mm metal base, and various denture base designs were subjected to different loading schemes. For this study, the two-dimensional finite element method was used. Mandibular arch models, with only canine remaining, were fabricated. In the first denture base design, a space, approximately 1mm thick, was prepared between the denture and the dome abutment. In the second denture base design, contact between the denture and the dome abutment was eliminated except the contact of the occlusal third of the abutment. In order to represent the same physiological condition as the fixed areas of the mandible under loading schemes, the eight nodes which lie at the mandibular angle region, the coronoid process and the mandibular condyle were assumed to be fixed. Each model was loaded with a magnitude of 10 kgs on the first molar region(P1) and 7 kgs on the central incisal region (P2) in a vertical direction. Then the force of 10 kgs was applied distributively from the first premolar to the second molar of each model in a vertical direction(P3). The results were as follows. : 1. When the testing vertical loads were given to the selected points of the overdenture, the overdenture showed the rotatory phenomenon, as well as sinking and the displacements of alveolar ridge, abutment and lower border of mandible under the metal base overdenture were less than those under the acrylic resin overdenture. 2. The maximum principal stresses(the maximum tensile stresses) being considered, high tensile stresses occured at the buccal shelf area, the posterior region of the ridge crest and the anterior border region of the mandibular ramus. 3. The minimum principal stresses(the maximum compressive stresses) being considered, high compressive stresses occured at the inferior and posterior border region of the mandible, the mandibular angle and the posterior border region of the mandibular ramus. 4. The vertical load on the central incisal region(P2) produced higher equivalent stress in the mandible than that on any other region(P1, P3) because of the long lever arm distance from the fixed points to the loading point. 5. Higher equivalent stresses were distributed throughout the metal base overdenture than the resin base overdenture under the same loading condition. 6. The case of occlusal third contact of the abutment to the denture produced higher equivalent stresses in the abutment, the mandibular area around the abutment and the overdenture than the case of a 1mm space between the denture and the abutment. 7. Without regard to overdenture base materials and designs, the amounts and distribution patterns of equivalent stresses under the same loading condition were similar in the mucous membrane.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.3
• /
• pp.185-192
• /
• 2023

In this study, fluctuating wind velocity for time history analysis is simulated by a single variate, single-dimensional random process using the KBC2022 spectrum about across-wind direction. This study analyzed and obtained the inelastic dynamic response for structures modeled as a single-degree-of-freedom system. It is assumed that the wind response is excellent in the primary mode, the change in vibration owing to plasticization is minor, along-wind vibration and across-wind vibration are independent, and the effect of torsional vibration is small. The numerical results, obtained by the Newmark-𝛽 method, shows the time-history responses and trends of maximum displacements. As a result of analyzing the inelastic dynamic response of the structure with the second stiffness ratio(𝛼) and yield displacement ratio (𝛽) as variables, it is identified that as the yield displacement ratio (𝛽) increases when the second stiffness ratio is constant, the maximum displacement ratio decreases, then reaches a minimum value, and then increases. When the stiffness ratio is greater than 0.5, there is a yield point ratio at which the maximum displacement ratio is less than 1, indicating that the maximum deformation is reduced compared to the elastically designed building even if the inelastic behavior is permitted in the inelastic wind design.