• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권6호
• /
• pp.1743-1758
• /
• 2023

Automatic modulation classification is a critical algorithm for non-cooperative communication systems. This paper addresses the challenging problem of closed-set and open-set signal modulation classification in complex channels. We propose a novel approach that incorporates a self-learning filter and center-loss in Deep Residual Shrinking Networks (DRSN) for closed-set modulation classification, and the Opendistance method for open-set modulation classification. Our approach achieves better performance than existing methods in both closed-set and open-set recognition. In closed-set recognition, the self-learning filter and center-loss combination improves recognition performance, with a maximum accuracy of over 92.18%. In open-set recognition, the use of a self-learning filter and center-loss provide an effective feature vector for open-set recognition, and the Opendistance method outperforms SoftMax and OpenMax in F1 scores and mean average accuracy under high openness. Overall, our proposed approach demonstrates promising results for automatic modulation classification, providing better performance in non-cooperative communication systems.

• Korean Chemical Engineering Research
• /
• 제48권4호
• /
• pp.483-489
• /
• 2010

본 연구에서는 화상분석(image analysis)에 기반한 소프트 센서를 설계하고, 이를 색상-질감 특성을 가진 제품의 외관품질 자동분류에 적용하였다. 색상과 질감(texture)을 동시에 가진 화상을 분석하기 위해 다중해상도 다변량 화상분석(Multiresolutional Multivariate Image Analysis, MR-MIA) 기법을 이용하였으며, 자동 분류를 위한 감독 학습법(supervised learning)으로는 Fisher의 판별분석(Fisher's discriminant analysis)을 사용하였다. 잠재변수법의 하나인 Fisher의 판별분석을 사용하였기 때문에, 제품의 외관을 서로 다른 불연속적인 부류로의 분류할 수 있을 뿐 아니라, 연속적인 외관 변화를 일관적이고 정량적으로 추정함은 물론, 외관의 특성 해석 또한 가능하였다. 이 방법은 인조대리석 제조 공정에서 중간 및 최종 제품의 외관 품질을 자동으로 분류하는 데에 성공적으로 적용되었다.

• 전자공학회논문지SC
• /
• 제37권6호
• /
• pp.50-63
• /
• 2000

본 논문에서는 소프트 컴퓨팅 기법을 이용한 새로운 근전도 신호 패턴 분류 방법을 제안한다. 재활 로봇시스템에서 기존에 사용되었던 여러 가지 입력 장치(음성, 레이저 포인터, 키패드, 3차원 입력기 등)에 비해 근전도 신호를 이용한 방식이 가지는 장점을 서술한다. 기존의 근전도 신호 분류 방법의 문제점인 사용자 의존성을 줄이기 위해 제안한 사용자 독립적인 특징 선택 방법에 대해 상술한다. 선택된 특징 집합을 이용하여 퍼지 패턴 분류기 및 퍼지 최대-최소 신경망을 구성하여 학습 전(퍼지 패턴 분류기)과 학습 후(퍼지 최대-최소 신경망)에 각각 83%와 90%의 분류 성공률을 얻어 제안된 방법의 유용성을 확인할 수 있었다.

• 대한치과의사협회지
• /
• 제48권9호
• /
• pp.670-679
• /
• 2010

Esthetic demands for dental treatment are increasing every day. The interdisciplinary relationship of the restorative treatment, periodontal therapy and other treatments such as endodontics, orthodontics and so on is more emphasized nowadays to reconstruct the hard and soft tissue foundation for the esthetic restorative treatment. This article will focus on the periodontal plastic surgery for esthetic restorative treatment. These followings will be discussed. 1. Understand the relationship between teeth and gingival scaffold for esthetics 2. Discuss the classification and treatment of gummy smile 3. Recognize the gingival margin irregularities by gingival recession and how to achieve the harmonic soft tissue margins 4. describe the hard and soft tissue augmentation for ridge augmentation.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2002년도 추계학술발표논문집 (하)
• /
• pp.2375-2378
• /
• 2002

우리 나라의 건설산업은 건설시장 개방으로 외국기업과의 경쟁체제 돌입이 불가피하고, 건설정책과 각종 행정규제의 급변으로 대응전략 수립이 불가피하다. 또한, 건설정보의 지식관리체제의 부재와 건설지식관리시스템의 미구축, 그리고 정보 공유와 공공정보 공개 마인드의 미성숙으로 인하여 적극적인 대처가 불가능한 실정이다. 따라서 본 논문에서는 건설업체, 설계사무소, 감리업체 등이 안고 있는 제반 현황을 고찰하여 건설지식관리시스템 구현방안을 제시 이를 건설지식관리시스템에 적용하여 국내 건설분야 건설지식 맵을 도출하고 환용 실태를 파악하고, 해외 파국에서 지식관리의 활용관련 자료를 분석하여 국제기준 및 국내 건설환경에 적합한 지식 맵 및 분류체계 적용한 시스템 개발 건설업체에 적용하였다.

• 한국농공학회지
• /
• 제45권2호
• /
• pp.96-106
• /
• 2003

134 Dutch cone (mechanical cone) and 9 piezocone (electronic cone) penetration tests have been performed in the southwestern part of Korea. In general, Dutch cone results may be different from that of piezocone due to the difference in structure of the cones. 6 Dutch cone and piezocone test data which were obtained at the same point respectively, were analyzed and plotted in soil classification chart proposed by Robertson et. al.(1986, 1990). Cone factors of Dutch cone and piezocone test empirically have been determined using laboratory and field vane test results. Using this cone factors, it was shown that there was good correlation between shear strength estimated using cone resistance and that of laboratory test and field vane tests. It was found that there was a good correlation between cone resistance from Dutch cone and that from piezocone. Relationship formula was also suggested. Dutch cone test provides a useful means for stratigraphic profiling in large project and has some advantage over piezocone in particular situations, such as very soft clay ground and dredged area.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2002년도 가을 학술발표회 논문집
• /
• pp.237-244
• /
• 2002

A comparative study of 134 mechanical (Dutch cone) and 9. electronic cone (Piezocone) penetration data from the southern part of Korea has been performed. In general, Dutch cone results may be different from piezocone results due to the difference in structure of the cones. Cone penetrometer test data were analyzed and plotted in soil classification chart proposed by Robertson et. al.(1986,1990) Cone factors of Dutch cone and piezocone test have empirically been determined using laboratory and field vane test results. Using this cone factors, it was shown that there was good correlation between shear strength estimated using cone resistance and that of laboratory test and field vane tests. It was found that there was a good correlation between cone resistance from Dutch cone and that from piezocone. Dutch cone test provides a useful means for stratigraphic profiling in large project and has some advantage over piezocone in particular situations, such as very soft clay ground and dredged area.

• 한국컴퓨터정보학회논문지
• /
• 제23권4호
• /
• pp.147-153
• /
• 2018

In this paper, we propose an efficient WBC 14-Diff classification which performs using the WBC-ResNet-152, a type of CNN model. The main point of view is to use Super-pixel for the segmentation of the image of WBC, and to use ResNet for the classification of WBC. A total of 136,164 blood image samples (224x224) were grouped for image segmentation, training, training verification, and final test performance analysis. Image segmentation using super-pixels have different number of images for each classes, so weighted average was applied and therefore image segmentation error was low at 7.23%. Using the training data-set for training 50 times, and using soft-max classifier, TPR average of 80.3% for the training set of 8,827 images was achieved. Based on this, using verification data-set of 21,437 images, 14-Diff classification TPR average of normal WBCs were at 93.4% and TPR average of abnormal WBCs were at 83.3%. The result and methodology of this research demonstrates the usefulness of artificial intelligence technology in the blood cell image classification field. WBC-ResNet-152 based morphology approach is shown to be meaningful and worthwhile method. And based on stored medical data, in-depth diagnosis and early detection of curable diseases is expected to improve the quality of treatment.

• 한국분말재료학회지
• /
• 제15권4호
• /
• pp.320-330
• /
• 2008

On the base of experience in development of Magnetic Powder Composites, and particularly Soft Magnetic Composites, authors are trying to systematize classification and indicate possible development prospective of Magnetic Nanocomposites (MN) technology and their applications in electrical converters. Clear classification and systematization, at an early stage of any materials and technology development, are essential and lead for better understanding and communication between researchers and industry involved. This concern MN as well and it seems to be the right time to make it at present stage of their development. Presented proposal of classification distinguishes various types of MN by their magnetic properties and area of possible applications. It is not a close set of types, and can be extended due to increase of knowledge concern these nanocomposites.

• 한국측량학회지
• /
• 제30권6_2호
• /
• pp.653-663
• /
• 2012

Using soft classification, it is possible to obtain the land cover proportions from the remotely sensed image. These land cover proportions are then used as input data for a procedure called "super-resolution mapping" to produce the predicted hard land cover layers at higher resolution than the original remotely sensed image. Superresolution mapping can be implemented using a number of algorithms in which the Hopfield Neural Network (HNN) has showed some advantages. The HNN has improved the land cover classification through superresolution mapping greatly with the high resolution data. However, the super-resolution mapping is based on the spatial dependence assumption, therefore it is predicted that the accuracy of resulted land cover classes depends on the relative size of spatial features and the spatial resolution of the remotely sensed image. This research is to evaluate the capability of HNN to implement the super-resolution mapping for SPOT image to create higher resolution land cover classes with different zoom factor.