• Korean Journal of Construction Engineering and Management
• /
• v.22 no.1
• /
• pp.106-115
• /
• 2021

Classifying work trades using images can serve an important role in a multitude of advanced applications in construction management and automated progress monitoring. However, images obtained from work sites may not always be clean. Defective images can damage an image classifier's accuracy which gives rise to a needs for a method to enhance a work trade image classifier's performance. We propose a method that uses work dependency information to aid image classifiers. We show that using work dependency can enhance the classifier's performance, especially when a base classifier is not so great in doing its job.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.11
• /
• pp.7957-7963
• /
• 2015

R&D trend for Auxiliary Power Supply(APU) of electric vehicle can be well understood by analyzing patents at home and abroad. Based on this trend analysis, domestic technology development direction is proposed. To get technique trend, patents of Korea, Japan, Europe and America published until February of 2014 are analyzed by WIPS DB. First, power converter and transformer two big category are classified. Power converter can be classified into resonant DC to DC converter and resonant Half bridge inverter; transformer can be classified into high frequency transformer, ferrite transformer and matching transformer. By analyzing R&D trend of different counties, companies and years, specific technology needed to be developed and trend of technology can be accurately grasped.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.169-172
• /
• 2000

본 연구에서는 청각장애인용 통신보조기기에 적용하기 위한 어휘예측 시스템의 기본구조를 제안한다. 통신보조기기의 어휘는 사용자의 환경을 고려한 어휘이므로, 어휘 예측 시스템도 사용자의 환경과 실생활에서 쉽게 이용할 수 있는 방향으로 고안되어야 한다. 따라서 어휘예측 시스템은 사용자의 환경을 정의하고, 중심어휘와 장소별 도메인에서의 어휘를 발췌한다. 발췌된 어휘는 말뭉치와 의미함축의 원리를 이용하여 분류한다. 분류된 어휘는 문법적 지식을 바탕으로 가상 네트워크를 구성한다. 가상네트워크에서의 어휘는 명사, 조사, 동사의 3부분으로 나눈 후 의미함축과 말뭉치로부터 파생된 어휘를 근접한 거리에 위치시킨다. 동일한 네트워크상에서 어휘의 위치는 문법적 연관성, 빈도수 등을 이용하여 정한다. 따라서 본 연구에서는 어휘예측은 명사, 조사, 동사에서 가장 근접한 어휘를 연결하여 간단한 문장을 작성할 수 있는 어휘 예측 시스템의 기본구조를 제안한다.

• KSBB Journal
• /
• v.11 no.5
• /
• pp.586-592
• /
• 1996

Poly(3-hydroxybutyric-co-3-hydroxyvaleric) acid(PHB/HV) copolymer synthesis by Pseudomonas sp. HJ from glucose and cosubstrate was investigated. Taxonomic analysis suggested that Pseudomonas sp. HJ was best marched to Pseudomonas picketti having 78.8% similarity. Pseudomonas sp. HJ produced PHB/HV copolymer containing 60.8 mol% HV and 44.9 mol% HV when supplied with hexadecane and propionic acid as a cosubstrate, respectively. The HV composition in PHB/HV copolymer was controlled by varying the concentration of hexadecane and propionic acid. Propionic acid added after 24 hours of incubation was incorporated as the HV monomer in the PHB/HV copolymer up to 49.6 mol%.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.185-207
• /
• 2022

Recently, with the development of deep learning technology, researches to apply a deep learning algorithm to analyze unstructured data such as text and images are being actively conducted. Text classification has been studied for a long time in academia and industry, and various attempts are being performed to utilize data characteristics to improve classification performance. In particular, a hierarchical relationship of labels has been utilized for hierarchical classification. However, the top-down approach mainly used for hierarchical classification has a limitation that misclassification at a higher level blocks the opportunity for correct classification at a lower level. Therefore, in this study, we propose a methodology for classifying hierarchical data using the autoencoder-based deeply supervised network that high-level classification does not block the low-level classification while considering the hierarchical relationship of labels. The proposed methodology adds a main classifier that predicts a low-level label to the autoencoder's latent variable and an auxiliary classifier that predicts a high-level label to the hidden layer of the autoencoder. As a result of experiments on 22,512 academic papers to evaluate the performance of the proposed methodology, it was confirmed that the proposed model showed superior classification accuracy and F1-score compared to the traditional supervised autoencoder and DNN model.

• Journal of Biomedical Engineering Research
• /
• v.19 no.1
• /
• pp.1-8
• /
• 1998

Clustered microcalcifications on X-ray mammograms are an important sign for early detection of breast cancer. This paper proposes a computer-aided diagnosis method for the detection of clustered microcalcifications and marking their locations on digitized mammograms. The proposed detection method consists of the region of interest (ROI) selection, the film-artifact removal, the surrounding texture analysis method for the detection of clustered microcalcifications, which is based on the second-order histogram in two nested surrounding regions on the current pixel. This paper also describes the effectiveness of the proposed film-artifact removal filter in terms of the classification performance with the receiver operating-characteristics(ROC) analysis. A three-layer backpropagation neural network is employed as a classifier. The appropriate marking for the locations of clustered microcalcifications can be used to alert radiologists to locations of suspicious lesions.

• Journal of Chest Surgery
• /
• v.34 no.2
• /
• pp.162-166
• /
• 2001

배경: 흉선암은 드문 질환으로 치료경과 및 예후가 침윤성 흉선종에 비해 나쁜 것으로 알려져 있으며 병기의 분류나 병기에 따른 치료방법이 아직 표준화 되어있지 않아 치료에 어려운 점이 있다. 이에 본교실에서 흉선암으로 진단되었던 환자들의 피료방법 및 성적을 분석하여 그 결과를 보고하고자 한다. 대상 및 방법: 계명대학교 동산의료원 흉부외과학교실에서는 1984년 8월에서 흉선암으로 진단되었던 8례의 환자를 대상으로 의무기록을 참고하여 병기에 따른 치방법료 및 예후 등을 후향적으로 분석하였다. 결과: 연령은 23세에서 67세까지로 평균 46세였으며 전흉부통증이 주증상이었다. 조직학적으로는 임파상피양암(lymphoepithelioma-like carcinoma)이 2례, 편평상피암(squamous cell carcinoma)이 2례, 기저세포암(basaloid carcinoma)이 1례, 혼합형(mixed type)이 3례 있었다. 임상적 병기분류는 Masoka의 분류법을 사용하였으며 제I기 2례, 제II기 4례, 제III기 1례, 제IVAr기가 1례 있었다. 4례의 환자에서는 종양의 완전적출이 가능했으며 3례에서는 고식적 수술을 시행하였다. 1례의 환자는 주위조직으로의 침윤과 심낭에 퍼져있어 조직생검만을 시행하였다. 전례에서 보조적 항암치료를 받았고 술후 병기가 제III기 이상이거나 종양의 절제가 불완전했던 5례의 환자에서는 보조적 방사선치료를 병행하였다. 이들 중 5례에서 술후 보조적 항암치료 및 방사선치료를 받고 현재까지 생존해있다. 평균추적기간은 55.3$\pm$64.6 개원이었고, 3례의 환자는 사망하였으며 4례의 환자는 종양의 재발증거 없이 생존해있다. 결론: 종양의 조기진단과 완전종양적출후 적극적인 보조적 항암치료 및 방사선치료가 흉선암을 치료하는데 도움이 될 것으로 사료된다.

• Journal of Software Assessment and Valuation
• /
• v.15 no.1
• /
• pp.63-70
• /
• 2019

These days, as Interest in Image recognition with deep learning is increasing, there has been a lot of research in image recognition using deep learning. In this study, we propose a system for classifying rocks through rock images of 18 types of rock(6 types of igneous, 6 types of metamorphic, 6 types of sedimentary rock) which are addressed in the high school curriculum, using CNN model based on Tensorflow, deep learning open source framework. As a result, we developed a classifier to distinguish rocks by learning the images of rocks and confirmed the classification performance of rock classifier. Finally, through the mobile application implemented, students can use the application as a learning tool in classroom or on-site experience.

• Smart Media Journal
• /
• v.8 no.4
• /
• pp.33-37
• /
• 2019

Because the ECG signals are time-series data acquired as time elapses, it is important to obtain comparative data the same in size as the enrolled data every time. This paper suggests a network model of GAN (Generative Adversarial Networks) based on an auxiliary classifier to generate synthetic ECG signals which may address the different data size issues. The Cosine similarity and Cross-correlation are used to examine the similarity of synthetic ECG signals. The analysis shows that the Average Cosine similarity was 0.991 and the Average Euclidean distance similarity based on cross-correlation was 0.25: such results indicate that data size difference issue can be resolved while the generated synthetic ECG signals, similar to real ECG signals, can create synthetic data even when the registered data are not the same as the comparative data in size.

• Journal of IKEEE
• /
• v.26 no.2
• /
• pp.176-185
• /
• 2022

Data augmentation is general approach to solve overfitting of machine learning models by applying various data transformations and distortions to dataset. However, when data augmentation is applied in GAN-based model, which is deep learning image generation model, data transformation and distortion are reflected in the generated image, then the generated image quality decrease. To prevent this problem called augmentation leak, we propose a scheme that can prevent augmentation leak regardless of the type and number of augmentations. Specifically, we analyze the conditions of augmentation leak occurrence by type and implement auxiliary augmentation task classifier that can prevent augmentation leak. Through experiments, we show that the proposed technique prevents augmentation leak in the GAN model, and as a result improves the quality of the generated image. We also demonstrate the superiority of the proposed scheme through ablation study and comparison with other representative augmentation leak prevention technique.