• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.1
• /
• pp.76-81
• /
• 1998

This paper presents an on-line scheme for parameter estimation of continuous-time systems, based on the model adjustment technique and the genetic algorithm technique. To deal with the initialisation and unmeasurable signal problems in on-line parameter estimation of continuous-time systems, a discrete-time model is obtained for the linear differential equation model and approximations of unmeasurable states with the observable output and its time-delayed values are obtained for the nonlinear state space model. Noisy observations may affect these approximation processes and degrade the estimation performance. A digital prefilter is therefore incorporated to avoid direct approximations of system derivatives from possible noisy observations. The parameters of both the model and the designed filter are adjusted on-line by a genetic algorithm, A set of simulation works for linear and nonlinear systems is carried out to demonstrate the effectiveness of the proposed method.

• Current Optics and Photonics
• /
• v.2 no.1
• /
• pp.69-78
• /
• 2018

This paper presents a precise edge detection algorithm for the critical dimension (CD) measurement of a Thin-Film Transistor Liquid-Crystal Display (TFT-LCD) pattern. The sigmoid surface function is proposed to model the blurred step edge. This model can simultaneously find the position and geometry of the edge precisely. The nonlinear least squares fitting method (Levenberg-Marquardt method) is used to model the image intensity distribution into the proposed sigmoid blurred edge model. The suggested algorithm is verified by comparing the CD measurement repeatability from high-magnified blurry and noisy TFT-LCD images with those from the previous Laplacian of Gaussian (LoG) based sub-pixel edge detection algorithm and error function fitting method. The proposed fitting-based edge detection algorithm produces more precise results than the previous method. The suggested algorithm can be applied to in-line precision CD measurement for high-resolution display devices.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.1
• /
• pp.44-53
• /
• 2008

An interacting multiple model (IMM) estimation algorithm based on the mixing of the predicted state estimates is proposed in this paper for a right continuous jump-linear system model different from the left-continuous system model used to develop the existing IMM algorithm. The difference lies in the modeling of the mode switching time. Performance of the proposed algorithm is compared numerically with that of the existing IMM algorithm for noisy system identification. Based on the numerical analysis, the proposed algorithm is applied to target tracking with a large sampling period for performance comparison with the existing IMM.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.4
• /
• pp.191-199
• /
• 2008

In this paper, we propose a MWF-PMC noise processing method which enhances the input speech by using Mel-warped Wiener Filtering (MWF) at pre-processing stage and compensates the recognition model by using PMC (Parallel Model Combination) at post-processing stage for speech recognition in noisy environments. The PMC uses the residual noise extracted from the silence region of enhanced speech at pre-processing stage to compensate the clean speech model and thus this method is considered to improve the performance of speech recognition in noisy environments. For recognition experiments we dew.-sampled KLE PBW (Phoneme Balanced Words) 452 word speech data to 8kHz and made 5 different SNR levels of noisy speech, i.e., 0dB. 5dB, 10dB, 15dB and 20dB, by adding Subway, Car and Exhibition noise to clean speech. From the recognition results, we could confirm the effectiveness of the proposed MWF-PMC method by obtaining the improved recognition performances over all compared with the existing combined methods.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.8
• /
• pp.1238-1246
• /
• 1989

In this paper, a texture synthesis technique based on the NCAR(non-causal auto-regressive) model and the pyramid structure is proposed. In order to estimate the NCAR model parameters accurately from a noisy texture, the MAP(maximum a posteriori) estimation technique is also employed. In our approach, since the input texture is decomposed into the Laplacian oyramid planes first and then the NCAR model is applied to each plane, we are able to obtain a good synthesized texture even if the texture exhibits some non-random local structure or non-homogenity. The usrfulness of the proposed method is demonstrated with seveal real textures in the Brodatz album. Finally, the 2-dimensional MAP estimation technique can be used to the image restoration for noisy images as well as a texture image synthesis.

• The Journal of Korea Robotics Society
• /
• v.6 no.2
• /
• pp.189-196
• /
• 2011

This paper presents a method of topological modeling using only low-cost sonar sensors. The proposed method constructs a topological model by extracting sub-regions from the local grid map. The extracted sub-regions are considered as nodes in the topological model, and the corresponding edges are generated according to the connectivity between two sub-regions. A grid confidence for each occupied grid is evaluated to obtain reliable regions in the local grid map by filtering out noisy data. Moreover, a convexity measure is used to extract sub-regions automatically. Through these processes, the topological model is constructed without predefining the number of sub-regions in advance and the proposed method guarantees the convexity of extracted sub-regions. Unlike previous topological modeling methods which are appropriate to the corridor-like environment, the proposed method can give a reliable topological modeling in a home environment even under the noisy sonar data. The performance of the proposed method is verified by experimental results in a real home environment.

• Speech Sciences
• /
• v.10 no.3
• /
• pp.263-277
• /
• 2003

One of major problems in speech recognition is performance degradation due to the mismatch between the training and test environments. Recently, Stereo-based Piecewise LInear Compensation for Environments (SPLICE), which is frame-based bias removal algorithm for cepstral enhancement using stereo training data and noisy speech model as a mixture of Gaussians, was proposed and showed good performance in noisy environments. In this paper, we propose several methods to improve the conventional SPLICE. First we apply Cepstral Mean Subtraction (CMS) as a preprocessor to SPLICE, instead of applying it as a postprocessor. Secondly, to compensate residual distortion after SPLICE processing, two-stage SPLICE is proposed. Thirdly we employ phonetic information for training SPLICE model. According to experiments on the Aurora 2 database, proposed method outperformed the conventional SPLICE and we achieved a 50% decrease in word error rate over the Aurora baseline system.

• Journal of Multimedia Information System
• /
• v.3 no.2
• /
• pp.21-26
• /
• 2016

Our networked world has been growing exponentially fast. The explosion in volume of machine-to-machine (M2M) transactions threatens to exceed the transport capacity of the networks that link them. Therefore, it is quite essential to reconsider the tradeoff between using many data sets versus using good data sets. We focus on this tradeoff in the context of the quality of information aggregated from many sensors in a noisy environment. We start with a basic theoretical model considered in the famous "CEO problem'' in the field of information theory. From a point of view of large deviations, we successfully find a simple statement for the optimal strategies under the limited network capacity condition. Moreover, we propose an open problem for a sensor network scenario and report a numerical result.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.9-15
• /
• 2006

In this paper we propose Wilcoxon test, Median test and Van der Waerden test such as linear rank tests in two-sample location problem for detecting edges effectively in noisy images. These methods are based on detecting image intensity changes between two pixel neighborhoods using an edge-height model to perform effectively on noisy images. The neighborhood size used here is small and its shape is varied adaptively according to edge orientations. We compare and analysis the performance of these statistical edge detectors on both natural images and synthetic images with and without noise.

• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.182-190
• /
• 2021

Deep learning is an advanced technology for large-scale data analysis, with numerous promising cases like image processing, object detection and significantly more. It becomes customarily to use transfer learning and fine-tune a pre-trained CNN model for most image recognition tasks. Having people taking photos and tag themselves provides a valuable resource of in-data. However, these tags and labels might be noisy as people who annotate these images might not be experts. This paper aims to explore the impact of noisy labels on fine-tuning pre-trained CNN models. Such effect is measured on a food recognition task using Food101 as a benchmark. Four pre-trained CNN models are included in this study: InceptionV3, VGG19, MobileNetV2 and DenseNet121. Symmetric label noise will be added with different ratios. In all cases, models based on DenseNet121 outperformed the other models. When noisy labels were introduced to the data, the performance of all models degraded almost linearly with the amount of added noise.