• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.160-170
• /
• 2010

This paper proposes a Haar Wavelet Feature Detector (HWFD) based on the Haar wavelet transform and average box filter. By decomposing the original image using the Haar wavelet transform, the proposed detector obtains the variance information of the image, making it possible to extract more distinctive features from the original image. For detection of interest points that represent the regions whose variance is the highest among their neighbor regions, we apply the average box filter to evaluate the local variance information and use the integral image technique for fast computation. Due to utilization of the Haar wavelet transform and the average box filter, the proposed detector is robust to illumination change, scale change, and rotation of the image. Experimental results show that even though the proposed method detects fewer interest points, it achieves higher repeatability, higher efficiency and higher matching accuracy compared with the DoG detector and Harris corner detector.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.2
• /
• pp.786-799
• /
• 2021

Road damage detection is important for road maintenance. With the development of deep learning, more and more road damage detection methods have been proposed, such as Fast R-CNN, Faster R-CNN, Mask R-CNN and RetinaNet. However, because shallow and deep layers cannot be extracted at the same time, the existing methods do not perform well in detecting objects with fewer samples. In addition, these methods cannot obtain a highly accurate detecting bounding box. This paper presents a Multi-level Feature Pyramids method based on M2det. Because the feature layer has multi-scale and multi-level architecture, the feature layer containing more information and obvious features can be extracted. Moreover, an attention mechanism is used to improve the accuracy of local boundary boxes in the dataset. Experimental results show that the proposed method is better than the current state-of-the-art methods.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.2E
• /
• pp.26-31
• /
• 1994

A new feature extraction technique utilizing strange attractor and artificial neural network for speaker recognition is presented. Since many signals change their characteristics over long periods of time, simple time-domain processing techniques should e capable of providing useful information of signal features. In many cases, normal time series can be viewed as a dynamical system with a low-dimensional attractor that can be reconstructed from the time series using time delay. The reconstruction of strange attractor is described. In the technique, the raw signal will be reproduced into a geometric three dimensional attractor. Classification decision for speaker recognition is based upon the processing or sets of feature vectors that are derived from the attractor. Three different methods for feature extraction will be discussed. The methods include box-counting dimension, natural measure with regular hexahedron and plank-type box. An artificial neural network is designed for training the feature data generated by the method. The recognition rates are about 82%-96% depending on the extraction method.

• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.149-163
• /
• 2023

Due to the continuous growth of the E-commerce market, the volume of orders that fulfillment centers have to process has increased, and various customer requirements have increased the complexity of order processing. Along with this trend, the operational efficiency of fulfillment centers due to increased labor costs is becoming more important from a corporate management perspective. Using historical performance data as training data, this study focused on real-time box recommendations applicable to packaging areas during fulfillment center shipping. Four types of data, such as product information, order information, packaging information, and delivery information, were applied to the machine learning model through pre-processing and feature-engineering processes. As an input vector, three characteristics were used as product specification information: width, length, and height, the characteristics of the input vector were extracted through a feature engineering process that converts product information from real numbers to an integer system for each section. As a result of comparing the performance of each model, it was confirmed that when the Gradient Boosting model was applied, the prediction was performed with the highest accuracy at 95.2% when the product specification information was converted into integers in 21 sections. This study proposes a machine learning model as a way to reduce the increase in costs and inefficiency of box packaging time caused by incorrect box selection in the fulfillment center, and also proposes a feature engineering method to effectively extract the characteristics of product specification information.

• Journal of KIISE:Software and Applications
• /
• v.33 no.3
• /
• pp.289-300
• /
• 2006

Product line is nowadays well known as a representative method for reuse. In the product line, important assets are components. Although enough concerns were given of the product line, it was not accomplished to structure and specify a product-line component with variability. This paper presents an approach to specify components in the product line. The approach describes the static and dynamic structure of a product-line component and explains the behavior and concurrency of the component. The component information is separately described in the black-box and white-box using the Feature-Oriented Reuse Method(FORM). This research also formalizes the data on a component specification in the form of BNF. The specification is described through careful consideration for many different characteristics of the product-line component, so this paper helps to easily develop the components in the product line and to well comprehend how to apply a method for the product line.

• Interdisciplinary Bio Central
• /
• v.4 no.1
• /
• pp.1.1-1.7
• /
• 2012

Background: The F-box proteins represent one of the largest families of proteins in eukaryotes. Apart from being a component of the ubiquitin (Ub)/26 S proteasome pathways, their regulatory roles in other cellular and developmental pathways have also been reported. One interesting feature of the genes encoding the proteins of this particular family is their variable selection patterns across different lineages. This resulted in the presence of lineage specific F-box proteins across different species. Findings: In this study, 48 non-redundant F-box proteins in E. siliculosus have been identified by a homology based approach and classified into three classes based on their variable C-terminal domains. A greater number of the F-box proteins have domains similar to the ones identified in other species. On the other hand, when the proteins having unknown or no C-terminal domain (as predicted by InterProScan) were analyzed, it was found that some of them have the polyglutamine repeats. To gain evolutionary insights on the genes encoding the F-box proteins, their selection patterns were analyzed and a strong positive selection was observed which indicated the adaptation potential of the members of this family. Moreover, four lineage specific F-box genes were found in E. siliculosus with no identified homolog in any other species. Conclusions: This study describes a genome wide in silico analysis of the F-box proteins in E. siliculosus which sheds light on their evolutionary patterns. The results presented in this study provide a strong foundation to select candidate sequences for future functional analysis.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.549-555
• /
• 2023

Jacking force loss management inside the PSC Box girder of a common high-speed railway is a very important feature in girder performance, and requires detailed management during the maintenance of the girder. This study aimed to analyze the timing of re-tension prediction of PSC Box girder based on the reduction level of the packing force inside the girder and the results of the tension loss measured without the train load test. As a result of predicting the timing of re-tension according to the level of tension reduction of the PSC Box Girder, the Jacking Force Loss curve was gently analyzed before the structure reached 17 years after confirmed completion, and 17 years later, it was found that the jacking force loss curve progressed rapidly. The results confirmed that the tension of the structure decreases with the service life increase, but considerably decreases as the structure ages. Therefore, more data and research on tension loss of facilities over 20 years are much required.

• Sijohaknonchong
• /
• v.19 no.1
• /
• pp.235-260
• /
• 2003

Recently, I presented Gagok(歌曲) Collection Yeong-eon永言 to learned circles. Yeong-eon is very similar to Yukdang-version六堂本 CheongGuYeongeon靑丘永言. Compared with Cheong- Yuk, it is the same age or little bit early time of Cheong-Yuk in culture Icon. This paper paid attention to the considerable difference between Yeong-eon and Cheong-Yuk in the way of the distribution of Yisakdaeyap二數大 葉's Woo mode羽調 and Ke-myeon mode界面調. There was the way of gathering ‘real name’+‘namelessness’ in Yisakdaeyap, which is the feature of the 18th century Gagok Collection. I found this way just put on the 19th century Gagok Collection way which is the distribution of Yisakdaeyap's 'Woo mode and Ke-myeon mode' in CheongYuk. Then I proved in this paper that the way of gathering Yisakdaeyap in Cheong Yuk didn't correspond to an actual singing in the early 19th century when 'Woo mode and Ke-myeon mode' was fixed. In case of Yeong-eon, however, it was not written any writers' names at all, when it was researched retroactively, I knew it was distributed evenly both the works of 'real name' and 'namelessness' in Yisakdaeyaps 'Woo mode and Ke-myeon mode'. Consequently, I found Yeong-eon is the good Gagok Collection for an actual singing at that time. In addition, there was discord in the mode or key distribution among many Gagok Collections. I found this issue of the application had kept on make Gagok Collections edit during 2 centuries. Because the actual Gagok enjoyment in the specific time is connected the way of the mode application directly.

• Journal of Korea Multimedia Society
• /
• v.23 no.7
• /
• pp.840-849
• /
• 2020

Walt Disney Feature Animation released high quality films such as 'Frozen', 'Big Hero 6', 'Wreck-it Ralph', 'Zootopia', 'Moana', 'Frozen 2' and not only got high score in box office but showed great CG and visuals. However, making feature animation requires a lot of time, money, and efforts so it is very important to support studios' artists to finish each show within limited budget and time. This paper shows artists' training program such as 'Short Circuit' and 'Bootcamp' that walt disney feature animation provides their artists to improve their creativity and do their jobs artistically and efficiently. Disney's training program not only provides artists various training classes but gives them chances to work on short animation which enhances artistic skills and enable them to work in different departments and experience different tasks. This paper also explains some training cases of CG studios in South Korea and Disney Animations' in-house tools.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1550-1552
• /
• 2005

In this paper, we have designed a Fingerprint Recognition System based on the Embedded LINUX. The fingerprint is captured using the AS-S2 semiconductor sensor. To extract a feature vector we transform the image of the fingerprint into a column vector. The image is row-wise filtered with the low-pass filter of the Haar wavelet. The feature vectors of the different fingerprints are compared by computing with the probabilistic neural network the distance between the target feature vector and the stored feature vectors in advance. The system implemented consists of a server PC based on the LINUX and a client based on the Embedded LINUX. The client is a Tynux box-x board using a PXA-255 CPU. The algorithm is simple and fast in computing and comparing the fingerprints.