• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2001.05a
• /
• pp.247-251
• /
• 2001

영상을 처리하는 과정에서 광학시스템과 전기시스템의 특성으로 인해 흐려지고 잡음으로 훼손된 영상을 복원하는 경우에 일반적으로 정칙화 반복복원방법이 사용된다. 기존의 방법은 영상의 국부적인 특성을 고려하지 않고 영상전체에 일률적으로 정칙화 연산자를 사용함으로써 윤곽부분에서는 리플잡음을 초래하고 평면부분에서도 잡음증폭을 피할 수 없으며, 또한 시각적으로 효율적이지 못한 면이 있다. 본 논문에서는 이러한 문제점을 개선하기 위하여, 영상의 국부적인 특성을 고려하여 적응 정칙화 파라메타와 적응 정칙화 연산지를 사용하여 평면영역과 윤곽영역의 방향특성에 따라 적응적으로 처리하는 반복복원방법을 제안한다. 제안한 방법은 기존의 방법과 비교하여 평면영역에서의 잡음 평활화가 개선되고 시각적으로 중요한 윤곽부분 복원에 효율적임을 실험결과를 통해 알 수 있었으며 ISNR 면에서도 우수하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.265-267
• /
• 2002

본 논문에서는 적응적 정규화 자연기울기 학습법과 자연 프루닝(pruning) 방법의 결합을 통하여 일반화 성능이 우수만 신경망을 구성하고자 한다. 먼저 적응적 정규화 자연기울기 학습을 통하여 신경망의 가중치를 최적화 시키고, 자연 프루닝에 의하여 신경망의 구조를 단순화 시킨다. 이러한 모델들 중 최적의 모델은 베이시안 정보 기준에 의해 선택함으로써 일반화 성능이 우수만 신경망을 구성하는 방법을 제안한다 벤치마크 (benchmark) 데이터로 제안하는 방법과 유클리디안(Euclidean) 거리에 기반한 결합 방법과 자연 프루닝만을 적용한 방법을 비교함으로써 우수성을 검증한다.

• Journal of Advanced Information Technology and Convergence
• /
• v.9 no.1
• /
• pp.89-102
• /
• 2019

Collagen can be used in building artificial skin replacements for treatment of burns and towards the reconstruction of bone as well as researching cell behavior and cellular interaction. The strength of collagen in connective tissue rests on the characteristics of collagen fibers. 3D confocal imaging of collagen fibers enables the characterization of their spatial distribution as related to their function. However, the image stacks acquired with confocal laser-scanning microscope does not clearly show the collagen architecture in 3D. Therefore, we developed a new method to reconstruct, visualize and characterize collagen fibers from fluorescence confocal images. First, we exploit the tensor voting framework to extract sparse reliable information about collagen structure in a 3D image and therefore denoise and filter the acquired image stack. We then propose to segment the collagen fibers by defining an energy term based on the Hessian matrix. This energy term is minimized by a min cut-max flow algorithm that allows adaptive regularization. We demonstrate the efficacy of our methods by visualizing reconstructed collagen from specific 3D image stack.

• International Journal of Computer Science & Network Security
• /
• v.23 no.11
• /
• pp.133-141
• /
• 2023

This work take deeper analysis of Adaptive Moment Estimation (Adam) and Adam with Weight Decay (AdamW) implementation in real world text classification problem (DGA Malware Detection). AdamW is introduced by decoupling weight decay from L2 regularization and implemented as improved optimizer. This work introduces a novel implementation of AdamW variant as AdamW+ by further simplifying weight decay implementation in AdamW. DGA malware detection LSTM models results for Adam, AdamW and AdamW+ are evaluated on various DGA families/ groups as multiclass text classification. Proposed AdamW+ optimizer results has shown improvement in all standard performance metrics over Adam and AdamW. Analysis of outcome has shown that novel optimizer has outperformed both Adam and AdamW text classification based problems.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.305-312
• /
• 2023

Physics-Informed Neural Network (PINN) is used to invert bubble size distributions from attenuation losses. By considering a linear system for the bubble population inversion, Adaptive Learned Iterative Shrinkage Thresholding Algorithm (Ada-LISTA), which has been solved linear systems in image processing, is used as a neural network architecture in PINN. Furthermore, a regularization based on the linear system is added to a loss function of PINN and it makes a PINN have better generalization by a solution satisfying the bubble physics. To evaluate an uncertainty of bubble estimation, deep ensemble is adopted. 20 Ada-LISTAs with different initial values are trained using the same training dataset. During test with attenuation losses different from those in the training dataset, the bubble size distribution and corresponding uncertainty are indicated by average and variance of 20 estimations, respectively. Deep ensemble Ada-LISTA demonstrate superior performance in inverting bubble size distributions than the conventional convex optimization solver of CVX.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.326-338
• /
• 2003

The objective of a neural network design and model selection is to construct an optimal network with a good generalization performance. However, training data include noises, and the number of training data is not sufficient, which results in the difference between the true probability distribution and the empirical one. The difference makes the teaming parameters to over-fit only to training data and to deviate from the true distribution of data, which is called the overfitting phenomenon. The overfilled neural network shows good approximations for the training data, but gives bad predictions to untrained new data. As the complexity of the neural network increases, this overfitting phenomenon also becomes more severe. In this paper, by taking statistical viewpoint, we proposed an integrative process for neural network design and model selection method in order to improve generalization performance. At first, by using the natural gradient learning with adaptive regularization, we try to obtain optimal parameters that are not overfilled to training data with fast convergence. By adopting the natural pruning to the obtained optimal parameters, we generate several candidates of network model with different sizes. Finally, we select an optimal model among candidate models based on the Bayesian Information Criteria. Through the computer simulation on benchmark problems, we confirm the generalization and structure optimization performance of the proposed integrative process of teaming and model selection.

• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.136-147
• /
• 2003

To achieve an optimal performance for a given problem, we need an integrative process of the parameter optimization via learning and the structure optimization via model selection. In this paper, we propose an efficient optimization method for improving generalization performance by considering the property of each sub-method and by combining them with common theoretical properties. First, weight parameters are optimized by natural gradient teaming with adaptive regularization, which uses a diverse error function. Second, the network structure is optimized by eliminating unnecessary parameters with natural pruning. Through iterating these processes, candidate models are constructed and evaluated based on the Bayesian Information Criterion so that an optimal one is finally selected. Through computational experiments on benchmark problems, we confirm the weight parameter and structure optimization performance of the proposed method.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.421-436
• /
• 2023

This paper presents a vibration displacement measurement and damage identification method for a space truss structure from its vibration videos. Features from Accelerated Segment Test (FAST) algorithm is combined with adaptive threshold strategy to detect the feature points of high quality within the Region of Interest (ROI), around each node of the truss structure. Then these points are tracked by Kanade-Lucas-Tomasi (KLT) algorithm along the video frame sequences to obtain the vibration displacement time histories. For some cases with the image plane not parallel to the truss structural plane, the scale factors cannot be applied directly. Therefore, these videos are processed with homography transformation. After scale factor adaptation, tracking results are expressed in physical units and compared with ground truth data. The main operational frequencies and the corresponding mode shapes are identified by using Subspace Stochastic Identification (SSI) from the obtained vibration displacement responses and compared with ground truth data. Structural damages are quantified by elemental stiffness reductions. A Bayesian inference-based objective function is constructed based on natural frequencies to identify the damage by model updating. The Success-History based Adaptive Differential Evolution with Linear Population Size Reduction (L-SHADE) is applied to minimise the objective function by tuning the damage parameter of each element. The locations and severities of damage in each case are then identified. The accuracy and effectiveness are verified by comparison of the identified results with the ground truth data.

• Journal of Biomedical Engineering Research
• /
• v.38 no.5
• /
• pp.219-226
• /
• 2017

Penalized-likelihood (PL) reconstruction methods for transmission tomography are known to provide improved image quality for reduced dose level by efficiently smoothing out noise while preserving edges. Unfortunately, however, most of the edge-preserving penalty functions used in conventional PL methods contain at least one free parameter which controls the shape of a non-quadratic penalty function to adjust the sensitivity of edge preservation. In this work, to avoid difficulties in finding a proper value of the free parameter involved in a non-quadratic penalty function, we propose a new adaptive method of space-variant smoothing with a simple quadratic penalty function. In this method, the smoothing parameter is adaptively selected for each pixel location at each iteration by using the image roughness measured by a pixel-wise standard deviation image calculated from the previous iteration. The experimental results demonstrate that our new method not only preserves edges, but also suppresses noise well in monotonic regions without requiring additional processes to select free parameters that may otherwise be included in a non-quadratic penalty function.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.1
• /
• pp.95-101
• /
• 2007

Recently, two phase scheme for removing impulse noise in images is proposed. This algorithms first detect the noise candidates based on the adaptive median filter, and then apply optimizing techniques recursively only to those noise candidates to restore image. Thus the noise detector with high accuracy is important role on this algorithm, In this paper, novel noise detector is proposed, which can detect impose noise with high accuracy while reducing the probability of false detecting image details as impulses. And the method for reducing computational cost of regularization phase is presented also.