• 대한원격탐사학회지
• /
• 제37권4호
• /
• pp.719-731
• /
• 2021

This study proposes a two-stage hybrid classification model for crop classification using multi-temporal remote sensing images; the model combines feature embedding by using an autoencoder (AE) with a convolutional neural network (CNN) classifier to fully utilize features including informative temporal and spatial signatures. Long short-term memory (LSTM)-based AE (LAE) is fine-tuned using class label information to extract latent features that contain less noise and useful temporal signatures. The CNN classifier is then applied to effectively account for the spatial characteristics of the extracted latent features. A crop classification experiment with multi-temporal unmanned aerial vehicle images is conducted to illustrate the potential application of the proposed hybrid model. The classification performance of the proposed model is compared with various combinations of conventional deep learning models (CNN, LSTM, and convolutional LSTM) and different inputs (original multi-temporal images and features from stacked AE). From the crop classification experiment, the best classification accuracy was achieved by the proposed model that utilized the latent features by fine-tuned LAE as input for the CNN classifier. The latent features that contain useful temporal signatures and are less noisy could increase the class separability between crops with similar spectral signatures, thereby leading to superior classification accuracy. The experimental results demonstrate the importance of effective feature extraction and the potential of the proposed classification model for crop classification using multi-temporal remote sensing images.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제11권8호
• /
• pp.315-324
• /
• 2022

이 논문에서는 자연어로 구성된 수학 문장제 문제 자동 풀이하기 위한 Transformer 기반의 생성 모델인 KoEPT를 제안한다. 수학 문장제 문제는 일상 상황을 수학적 형식으로 표현한 자연어 문제이다. 문장제 문제 풀이 기술은 함축된 논리를 인공지능이 파악해야 한다는 요구사항을 지녀 최근 인공지능의 언어 이해 능력을 증진하기 위해 국내외에서 다양하게 연구되고 있다. 한국어의 경우 문제를 유형으로 분류하여 풀이하는 기법들이 주로 시도되었으나, 이러한 기법은 다양한 수식을 포괄하여 분류 난도가 높은 데이터셋에 적용하기 어렵다는 한계가 있다. 본 논문은 이에 대해 '식' 토큰과 포인터 네트워크를 사용하는 KoEPT 모델을 사용했다. 이 모델의 성능을 측정하기 위해 현존하는 한국어 수학 문장제 문제 데이터셋인 IL, CC, ALG514의 분류 난도를 측정한 후 5겹 교차 검증 기법을 사용하여 KoEPT의 성능을 평가하였다. 평가에 사용된 한국어 데이터셋들에 대하여, KoEPT는 CC에서는 기존 최고 성능과 대등한 99.1%, IL과 ALG514에서 각각 89.3%, 80.5%로 새로운 최고 성능을 얻었다. 뿐만 아니라 평가 결과 KoEPT는 분류 난도가 높은 데이터셋에 대해 상대적으로 개선된 성능을 보였다. KoEPT가 분류 난도의 영향을 덜 받으며 좋은 성능을 얻게 된 이유를 '식' 토큰과 포인터 네트워크 때문이라는 것을 ablation study를 통해서 밝혔다.

• 제어로봇시스템학회논문지
• /
• 제6권8호
• /
• pp.710-718
• /
• 2000

In this paper we described an approach to automation of visual inspection of solder joint defects of SMC(Surface Mounted Components) on PCBs(Printed Circuit Board) by using neural network and fuzzy rule-based classification method. Inherently the surface of the solder joints is curved tiny and specular reflective it induces difficulty of taking good image of the solder joints. And the shape of the solder joints tends to greatly vary with the soldering condition and the shapes are not identical to each other even though the solder joints belong to a set of the same soldering quality. This problem makes it difficult to classify the solder joints according to their qualities. Neural network and fuzzy rule-based classification method is proposed to effi-ciently make human-like classification criteria of the solder joint shapes. The performance of the proposed approach is tested on numerous samples of commercial computer PCB boards and compared with the results of the human inspector performance and the conventional Kohonen network.

• Journal of Electrical Engineering and Technology
• /
• 제10권4호
• /
• pp.1877-1885
• /
• 2015

This paper presents a new Nearest-Neighbors based weighted representation for images and weighted K-Nearest-Neighbors (WKNN) classifier to improve the precision of image classification using the Bag of Visual Words (BOVW) based models. Scale-invariant feature transform (SIFT) features are firstly extracted from images. Then, the K-means++ algorithm is adopted in place of the conventional K-means algorithm to generate a more effective visual dictionary. Furthermore, the histogram of visual words becomes more expressive by utilizing the proposed weighted vector quantization (WVQ). Finally, WKNN classifier is applied to enhance the properties of the classification task between images in which similar levels of background noise are present. Average precision and absolute change degree are calculated to assess the classification performance and the stability of K-means++ algorithm, respectively. Experimental results on three diverse datasets: Caltech-101, Caltech-256 and PASCAL VOC 2011 show that the proposed WVQ method and WKNN method further improve the performance of classification.

• 인터넷정보학회논문지
• /
• 제12권3호
• /
• pp.89-99
• /
• 2011

인터넷에 기반한 응용 프로그램의 종류와 네트워크 대역폭이 증가하면서 페이로드 시그니처 기반 트래픽 분류 시스템에서 처리하는 데이터의 양이 급격하게 증가하고 있다. 대용량 트래픽 데이터에 대한 처리 속도를 향상시키기 위한 방법으로 다양한 패턴 매칭 알고리즘이 제안되고 있다. 하지만 비약적으로 늘어나는 시그니처의 수와 트래픽 양에 비해 패턴 매칭 알고리즘의 성능 향상 속도는 한정적이고, 입력데이터의 특성에 의존적인 성능을 나타낸다. 따라서 본 논문에서는 분류 시스템의 입력 데이터로 제공되는 트래픽 데이터와 시그니처의 탐색 공간을 최적화할 수 있는 분류, 시스템 구조를 제안한다. 또한 제안하는 분류 시스템을 학내 망에서 발생하는 대용량의 트래픽에 실시간으로 적용하여 그 타당성을 증명한다.

• 로봇학회논문지
• /
• 제7권2호
• /
• pp.92-100
• /
• 2012

A map of complex environment can be generated using a robot carrying sensors. However, representation of environments directly using the integration of sensor data tells only spatial existence. In order to execute high-level applications, robots need semantic knowledge of the environments. This research investigates the design of a system for recognizing objects in 3D point clouds of urban environments. The proposed system is decomposed into five steps: sequential LIDAR scan, point classification, ground detection and elimination, segmentation, and object classification. This method could classify the various objects in urban environment, such as cars, trees, buildings, posts, etc. The simple methods minimizing time-consuming process are developed to guarantee real-time performance and to perform data classification on-the-fly as data is being acquired. To evaluate performance of the proposed methods, computation time and recognition rate are analyzed. Experimental results demonstrate that the proposed algorithm has efficiency in fast understanding the semantic knowledge of a dynamic urban environment.

• 전자공학회논문지
• /
• 제50권2호
• /
• pp.185-192
• /
• 2013

본 논문에서는 MFCC를 이용한 GMM 기반의 음성과 혼합 신호 분류 알고리즘을 MPEG의 표준 코덱인 USAC에 적용하였다. 효과적인 패턴 인식을 위해 GMM을 이용하였고, EM알고리즘을 사용하여 최적의 GMM 파라미터를 추출하였다. 제안하는 분류 알고리즘은 두 가지 중요한 부분으로 나뉜다. 첫째는 GMM을 통해 최적의 파라미터를 추출하는 것 이고, 두 번째는 MFCC 값을 이용한 패턴인식을 통해 음성/혼합 신호를 분류하였다. 제안된 알고리즘의 성능을 평가한 결과 MFCC를 이용한 GMM 기반의 제안된 방법이 기존 USAC의 방법보다 우수한 음성/혼합 신호 분류 성능을 보였다.

• 응용통계연구
• /
• 제27권5호
• /
• pp.773-786
• /
• 2014

본 연구는 ROC 곡선에서 형성되는 면적 형태로 나타나는 분류정확도기준인 오분류율곱(multiplication of false rates; MFR)를 제안한다. MFR 기준과 다른 기준로부터 구한 최적분류점의 분류성과에 대하여 비교 분석한다. 다양한 분포함수에 대하여 최적분류점을 구하고 이에 대응하는 FNR과 FPR을 비교하면서 MFR의 특징과 장점을 유도한다. 일반적인 비용함수를 바탕으로 분류점에 대한 비용비율을 다양한 분류기준을 이용하여 구한다. 비용곡선에 대한 비용비율의 관계를 정리하여 MFR 기준의 장점을 탐색한다. MFR 기준의 정의를 다차원 ROC 분석으로 확장하고 다차원의 다른 분류기준과의 관계를 설명하면서 토론한다.

• 산업공학
• /
• 제22권3호
• /
• pp.252-262
• /
• 2009

It is crucial to prevent low yields in the semiconductor industry. Since many factors affect variation in yield and they are deeply related, preventing low yield is difficult. There have been substantial researches in the field of yield prediction. Many researchers had used the statistical methods. Many studies have shown that artificial neural network (ANN) achieved better performance than traditional statistical methods. However, despite ANN's superior performance some problems such as over-fitting and poor explanatory power arise. In order to overcome these limitations, a relatively new machine learning technique, support vector machine (SVM), is introduced to classify the yield. SVM is simple enough to be analyzed mathematically, and it leads to high performances in practical applications. This study presents a new efficient classification methodology, Stepwise-SVM (SSVM), for detecting high and low yields. SSVM is step-by-step adjustment of parameters to be precisely the classification for actual high and low yield lot. The objective of this paper is to examine the feasibility of SVM and SSVM in the yield classification. The experimental results show that SVM and SSVM provides a promising alternative to yield classification for the field data.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권14호
• /
• pp.5593-5596
• /
• 2014

Background: The objective of this study was to determine a diagnostic classification scheme using a decision tree based model. Materials and Methods: The study was conducted as a retrospective case-control study in Imam Khomeini hospital in Tehran during 2001 to 2009. Data, including demographic and clinical-pathological characteristics, were uniformly collected from 624 females, 312 of them were referred with positive diagnosis of breast cancer (cases) and 312 healthy women (controls). The decision tree was implemented to develop a diagnostic classification scheme using CART 6.0 Software. The AUC (area under curve), was measured as the overall performance of diagnostic classification of the decision tree. Results: Five variables as main risk factors of breast cancer and six subgroups as high risk were identified. The results indicated that increasing age, low age at menarche, single and divorced statues, irregular menarche pattern and family history of breast cancer are the important diagnostic factors in Iranian breast cancer patients. The sensitivity and specificity of the analysis were 66% and 86.9% respectively. The high AUC (0.82) also showed an excellent classification and diagnostic performance of the model. Conclusions: Decision tree based model appears to be suitable for identifying risk factors and high or low risk subgroups. It can also assists clinicians in making a decision, since it can identify underlying prognostic relationships and understanding the model is very explicit.