• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권5호
• /
• pp.1252-1271
• /
• 2013

Automatic image annotation has become an increasingly important research topic owing to its key role in image retrieval. Simultaneously, it is highly challenging when facing to large-scale dataset with large variance. Practical approaches generally rely on similarity measures defined over images and multi-label prediction methods. More specifically, those approaches usually 1) leverage similarity measures predefined or learned by optimizing for ranking or annotation, which might be not adaptive enough to datasets; and 2) predict labels separately without taking the correlation of labels into account. In this paper, we propose a method for image annotation through collaborative similarity metric learning from dataset and modeling the label correlation of the dataset. The similarity metric is learned by simultaneously optimizing the 1) image ranking using structural SVM (SSVM), and 2) image annotation using correlated label propagation, with respect to the similarity metric. The learned similarity metric, fully exploiting the available information of datasets, would improve the two collaborative components, ranking and annotation, and sequentially the retrieval system itself. We evaluated the proposed method on Corel5k, Corel30k and EspGame databases. The results for annotation and retrieval show the competitive performance of the proposed method.

• 한국멀티미디어학회논문지
• /
• 제11권6호
• /
• pp.891-902
• /
• 2008

The software once developed enters in to the $2^{nd}$ phase of its useful operational life, where it performs its functions for which it has been developed. This is an important phase of software life, wherein the developers and users have to assess the performance of the software to prolong its operational life with efficiency, effectiveness and economy. Performance of the software is based on the measurement of the quality characteristics to ascertain that it is meeting or exceeding the needs of customers in terms of its functionality, usability, efficiency, reliability, availability, maintainability, portability etc. These software quality attributes defined under ISO 9126 provide an overall good framework for assessing the performance of the software. This paper describes a Software Performance Assessment (SPA) model using the Goal-Question-Metric (GQM) paradigm which has been applied to the operational software in real context with successful results.

• 한국정보통신학회논문지
• /
• 제24권1호
• /
• pp.130-133
• /
• 2020

Smart pig farm which is based on IoT has been widely adopted by many pig farmers. In order to achieve optimal control of smart pig farm, the relation between environmental conditions and performance metric should be characterized. In this study, the relation between multiple environmental conditions including temperature, humidity and various performance metrics, which are daily gain, feed intake, and MSY, is analyzed based on data obtained from 55 real pig farm. Especially, based on preprocessing of data, various regression based machine learning algorithms are considered. Through performance evaluation, we show that the performance can be predicted with high precision, which can improve the efficiency of management.

• 대한전자공학회논문지TC
• /
• 제41권10호
• /
• pp.45-52
• /
• 2004

본 논문에서는 에러 예측회로를 사용하여 연집에러 입력시 성능저하를 보완한 경판정 비터비 디코더를 제안하였다. 비터비 디코더는 최대유사복호 알고리즘을 사용하므로 랜덤에러 입력시 정정능력이 뛰어나다. 반면에 연집에러 입력시 에러 정정능력이 매우 떨어지는 단점이 있다. 제안하는 에러 예측회로는 비터비 디코더의 연집에러에 대한 에러 정정특성을 향상시키는 기능으로 비터비 디코더에 에러가 입력됨에 따라 path metric값이 증가하는 것을 이용한다. Path metric의 최대값 증가량을 이용하여 연집에러 구간을 예측, 연집에러 구간에 대한 확률 값을 줄여준다. 제안된 알고리즘을 OFDM방식의 IEEE802.11a WLAN에 적용한 비터비 디코더는 AWGN채널에서는 기존의 비터비 디코더와 동일한 성능을 유지하며, 무선 채널 환경인 다중경로 페이딩 채널에서 발생할 수 있는 연집에러에 대하여 15% 개선된 성능을 보였다.

• 한국IT서비스학회지
• /
• 제10권3호
• /
• pp.167-177
• /
• 2011

Code obfuscation is a technique that protects the abstract data contained in a program from malicious reverse engineering and various obfuscation methods have been proposed for obfuscating intention. As the abstract data of control flow about programs is important to clearly understand whole program, many control flow obfuscation transformations have been introduced. Generally, inlining is a compiler optimization which improves the performance of programs by reducing the overhead of calling invocation. In code obfuscation, inlining is used to protect the abstract data of control flow. In this paper, we define new control flow complexity metric based on entropy theory and N-Scope metric, and then apply genetic algorithm to obtain optimal inlining results, based on the defined metric.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제21권2호
• /
• pp.121-125
• /
• 2015

기존의 협업필터링 기반의 추천시스템에 대한 연구는 정확한 평점예측에 집중되면서 추천시스템의 수행시간이 길어지게 되고, 선호아이템을 짧은 시간에 추천해주는 본래의 목적에서 멀어지게 되었다. 본 논문에서는 Preference Difference Metric을 이용하여 평점예측이 아닌 선호 아이템의 분류를 통한 추천을 수행하여 수행시간을 단축하고 정확도를 유지하는 추천 알고리즘을 제안한다.

• International Journal of Computer Science & Network Security
• /
• 제22권3호
• /
• pp.292-302
• /
• 2022

The radar tomographic imaging is based on the Radar Cross-Section "RCS" of the materials of a shape under examination and investigation. The RCS varies as the conductivity and permittivity of a target, where the target has a different material profile than other background objects in a scene. In this research paper, we use Hierarchical Performance Modeling "HPM" and a framework developed earlier to determine/spot bottleneck(s) for pattern recognition of materials using a combination of the Single Layer Perceptron (SLP) technique and tomographic images in radar systems. HPM provides mathematical equations which create Objective Functions "OFs" to find an average performance metric such as throughput or response time. Herein, response time is used as the performance metric and during the estimation of it, bottlenecks are found with the help of OFs. The obtained results indicate that processing images consumes around 90% of the execution time.

• 전자공학회논문지A
• /
• 제31A권6호
• /
• pp.1-8
• /
• 1994

In this paper, in order to apply the idea MDPSK to TCM, we use signal set expansion and set partition by phase differences. Through this we propose the trellis-coded MDPSK. And the Viterbi decoder containing branch metrics of the squared Euclidean distance of the Lth order phase difference as well as the first order phase difference is introduced in order to improve the bit error rate(BER) in the differential detection of the trellis-coded MDPSK. The proposed Viterbi decoder is conceptually same to the sliding multiple symbol dection method which uses the branch metric with the first and Lth order phase differences. We investigate the performance of the uncoded DQPSK and the trallis-coded D8PSK in additive white Gaussian noise (AWGN) through the Monte Carlo simulation under the two cases of using and not using the Lth order phase difference metric. The study shows that trellis-coded 8DPSK is an attractive scheme for power and bandlimited systems while also improving the BER performance when the Viterbi decoder is employed to the Lth order phase order difference metric. This performance improvement has been obtained without sacrificing the bandwidth or the power efficiency.

• 정보처리학회논문지C
• /
• 제12C권4호
• /
• pp.543-550
• /
• 2005

Ad-hoc 네트워크 환경에서의 많은 라우팅 알고리즘들은 소스 라우팅 기법을 기초로 설계되었으며 밀집된 ad-hoc 환경에서의 최단경로설정은 짧은 기간동안만 유지되는 제약사항이 존재한다 때문에 SSA와 ABR과 같은 라우팅 프로토콜은 링크 안정성을 고려하여 좀더 안정적인 경로를 탐색한다. 본 논문은 Ad-Hoc network 환경에 현실적으로 적합한 SSA 라우팅 알고리즘을 바탕으로 utilization 메트릭을 고려한 새로운 라우팅 기법에 대해 제안한다. 제안된 방식에서는 SSA의 특정 메트릭에 의한 병목현상을 해결하기 위해 연결을 위한 여러 경로들을 누적 utilization에 따라 분산시키도록 하여 네트워크상의 로드 밸런싱을 추구하였다. 성능 분석을 위해 기존의 라우팅 기법인 AODV와 SSA를 참고 비교 대상으로 하였다. ns-2 네트워크 시뮬레이션 결과 선택되어질 수 있는 안정적인 route가 많은 경우 전체 트래픽이 여러 경로들로 분산되어 경로 재설정(route reconstruction) 횟수를 현저히 줄이는 좋은 성능을 보였다

• 시스템엔지니어링학술지
• /
• 제13권1호
• /
• pp.79-88
• /
• 2017

For the PBL contract, it is necessary for the contracting parties to share information regarding the reasonable inventory-level and the cost of its repair parts for the estimated demand. There are various models which can be used for this purpose. Among them, V-METRIC model is considered to be the most efficient and is most frequently applied. However, this model is usually used for optimizing the inventory level of the repair parts of the system under operation. The model uses a time series forecast model to determine the demand rate, which is a mandatory input factor for the model, based on past field data. However, since the system at the deployment stage has no operational performance record, it is necessary to find another alternative to be used as the demand rate of the model application. This research applies the V-METRIC model to find the optimal inventory level and cost estimation for repairable items to meet the target operational availability, which is a key performance indicator, at the time of the PBL contract for the deployment system. This study uses the calculated value based on the allocated MTBF to the system as the demand rate, which is used as input data for the model. Also, we would like to examine changes in inventory level and cost according to the changes in target operational availability and MTBF allocation.