• Computers and Concrete
• /
• 제4권3호
• /
• pp.243-257
• /
• 2007

A novel method for recognition, characterization, and quantification of deterioration in bridge components and laboratory concrete samples is presented in this paper. The proposed scheme is based on grey level co-occurrence matrix texture analysis using Haar's discrete wavelet transform on concrete imagery. Each image is described by a subset of band-filtered images containing wavelet coefficients, and then reconstructed images are employed in characterizing the texture, using grey level co-occurrence matrices, of the different types and degrees of damage: map-cracking, spalling and steel corrosion. A comparative study was conducted to evaluate the efficiency of the supervised maximum likelihood and unsupervised K-means classification techniques, in order to classify and quantify the deterioration and its extent. Experimental results show both methods are relatively effective in characterizing and quantifying damage; however, the supervised technique produced more accurate results, with overall classification accuracies ranging from 76.8% to 79.1%.

• 한국인터넷방송통신학회논문지
• /
• 제19권4호
• /
• pp.169-174
• /
• 2019

본 연구에서는 설문지를 이용한 데이터 수집과 데이터 마이닝에서 클러스터링 기법으로 군집하여 지도학습을 이용하여 유사성을 판단하고, 성격들의 상관 관계의 적합성을 분석하기 위해 특징 추출 알고리즘들과 지도학습을 이용하는 것을 목표로 진행한다. 연구 수행은 설문조사를 진행 후 그 설문조사를 토대로 모인 데이터들을 정제하고, 오픈 소스 기반의 데이터 마이닝 도구인 WEKA의 클러스터링 기법들을 통해 데이터 세트를 분류하고 지도학습을 이용하여 유사성을 판단한다. 그리고 특징 추출 알고리즘들과 지도학습을 이용하여 성격에 대해 적합한 결과가 나오는지에 대한 적합성을 판단한다. 그 결과 유사성 판단에 가장 정확도 높게 도움을 주는 것은 EM 클러스터링으로 3개의 분류하고 Naïve Bayes 지도학습을 시킨 것이 가장 높은 유사성 분류 결과를 도출하였고, 적합성을 판단하는데 도움이 되도록 특징추출과 지도학습을 수행하였을 때, Big-5 각 성격마다 문항에 추가되고 삭제되는 것에 따라 정확도가 변하는 모습을 찾게 되었고, 각 성격 마다 차이에 대한 분석을 완료하였다.

• 대한공간정보학회지
• /
• 제25권1호
• /
• pp.9-17
• /
• 2017

초분광 영상(hyperspectral imagery)은 주성분분석이나 최소잡음비율 등을 이용하여 자료의 차원과 잡음을 감소시켜 토지피복분류에 사용되는 것이 일반적이다. 최근에는 분광정보와 공간적 특성을 가진 다양한 입력 자료를 이용한 감독분류에 관한 연구가 활발히 진행되고 있다. 본 연구에서는 초분광 영상을 이용한 토지피복분류를 위해 principle component(PC) 밴드와 normalized difference vegetation index(NDVI) 자료를 감독분류의 입력자료로 활용하였다. NDVI 자료는 초분광 영상에서 추출된 PC 밴드가 포함하고 있지 않는 추가적인 정보를 활용하여 식생지역에 대한 토지피복분류 정확도를 높이고자 사용하였으며, morphological filter를 통해 각 밴드의 extended attribute profiles(EAP)를 제작하여 분류를 위한 입력 자료로 사용하였다. 감독분류기법은 random forest 알고리즘을 이용하였으며, EAP를 기반으로 다양한 입력 자료의 적용에 따른 분류정확도를 비교하고자 하였다. 연구지역으로는 두 대상지를 선정하였으며, 영상 내에서 취득한 참조자료를 이용하여 정량적인 평가를 수행하였다. 본 연구에서 제안한 기법의 분류정확도는 85.72%와 91.14%로 다른 입력 자료들을 이용한 경우와 비교하여 가장 높은 분류정확도를 나타냈다. 향후, 초분광 영상을 이용한 토지피복분류의 정확도를 높이기 위한 분류 알고리즘 개발과 대상지역 특성에 맞는 추가 입력자료 개발에 관한 연구가 필요할 것으로 사료된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권1호
• /
• pp.364-380
• /
• 2016

The most popular approach in object classification is based on the bag of visual-words model, which has several fundamental problems that restricting the performance of this method, such as low time efficiency, the synonym and polysemy of visual words, and the lack of spatial information between visual words. In view of this, an object classification based on weakly supervised E2LSH and saliency map weighting is proposed. Firstly, E2LSH (Exact Euclidean Locality Sensitive Hashing) is employed to generate a group of weakly randomized visual dictionary by clustering SIFT features of the training dataset, and the selecting process of hash functions is effectively supervised inspired by the random forest ideas to reduce the randomcity of E2LSH. Secondly, graph-based visual saliency (GBVS) algorithm is applied to detect the saliency map of different images and weight the visual words according to the saliency prior. Finally, saliency map weighted visual language model is carried out to accomplish object classification. Experimental results datasets of Pascal 2007 and Caltech-256 indicate that the distinguishability of objects is effectively improved and our method is superior to the state-of-the-art object classification methods.

• 대한공간정보학회지
• /
• 제24권1호
• /
• pp.121-128
• /
• 2016

많은 연구들에서 영상자료와 분류 알고리즘 측면에서 분류정확도를 비교하였지만, 참조자료 또는 분석자에 의존하는 훈련샘플에 의한 분류정확도 비교와 관련된 연구는 부족한 실정이다. 본 연구는 감독분류에 있어 훈련샘플로써 지상 분광반사자료의 유용성을 평가하고자 하였다. 이를 위하여 초분광영상과 다중분광영상을 대상으로 영상 수집 훈련샘플과 지상 분광반사자료를 사용하여 분류 정확도를 비교하였다. 그 결과 영상 수집 훈련샘플 사용 시 초분 광영상과 다중분광영상에서 공통적으로 약 90%의 분류정확도를 얻을 수 있었다. 그러나 지상 분광반사자료를 훈련 샘플로 사용하면 초분광영상의 경우 약 10%p, 다중분광영상의 경우 약 20%p의 분류정확도 감소가 발생하였다. 특히 다중분광영상에서 분광반사특성이 유사하게 나타나는 클래스들의 경우 분류정확도가 초분광영상에 비해 매우 낮게 나타났다. 따라서 지상 분광반사자료는 다중분광영상에 적용하는 데에는 한계가 있지만, 초분광영상을 이용한 토지피복분류에 있어 유용한 훈련샘플이 될 수 있다.

• 한국환경복원기술학회지
• /
• 제18권2호
• /
• pp.89-104
• /
• 2015

In order to effectively manage forested areas in South Korea on a national scale, using remotely sensed data is considered most suitable. In this study, utilizing Land coverage maps and Forest type maps of national geographic information instead of collecting field data was tested for conducting supervised classification on SPOT-5 and KOMPSAT-2 imagery focusing on forested areas. Supervised classification were conducted in two ways: analysing a whole area around the study site and/or only forested areas around the study site, using Support Vector Machine. The overall accuracy for the classification on the whole area ranged from 54.9% to 68.9% with kappa coefficients of over 0.4, which meant the supervised classification was in general considered moderate because of sub-classifying forested areas into three categories (i.e. hardwood, conifer, mixed forests). Compared to this, the overall accuracy for forested areas were better for sub-classification of forested areas probably due to less distraction in the classification. To further improve the overall accuracy, it is needed to gain individual imagery rather than mosaic imagery to use more spetral bands and select more suitable conditions such as seasonal timing. It is also necessary to obtain precise and accurate training data for sub-classifying forested areas. This new approach can be considered as a basis of developing an excellent analysis manner for understanding and managing forest landscape.

• Journal of Computing Science and Engineering
• /
• 제5권2호
• /
• pp.150-160
• /
• 2011

Automatic text classification has a long history and many studies have been conducted in this field. In particular, many machine learning algorithms and information retrieval techniques have been applied to text classification tasks. Even though much technical progress has been made in text classification, there is still room for improvement in text classification. In this paper, we will discuss remaining issues in improving text classification. In this paper, three improvement issues are presented including automatic training data generation, noisy data treatment and term weighting and indexing, and four actual studies and their empirical results for those issues are introduced. First, the semi-supervised learning technique is applied to text classification to efficiently create training data. For effective noisy data treatment, a noisy data reduction method and a robust text classifier from noisy data are developed as a solution. Finally, the term weighting and indexing technique is revised by reflecting the importance of sentences into term weight calculation using summarization techniques.

• 한국경영과학회지
• /
• 제33권3호
• /
• pp.93-105
• /
• 2008

Semi-supervised learning uses a small amount of labeled data to predict labels of unlabeled data as well as to improve clustering performance, whereas unsupervised learning analyzes only unlabeled data for clustering purpose. We propose a new clustering-based semi-supervised learning method by reflecting the initial predicted labels of unlabeled data on the objective function. The initial prediction should be done in terms of a discrete probability distribution through a classification method using labeled data. As a result, clusters are formed and labels of unlabeled data are predicted according to the Information of labeled data in the same cluster. We evaluate and compare the performance of the proposed method in terms of classification errors through numerical experiments with blinded labeled data.

• IEIE Transactions on Smart Processing and Computing
• /
• 제3권4호
• /
• pp.234-239
• /
• 2014

This paper presents an empirical study on selecting a small amount of useful unlabeled data to improve the classification accuracy of semi-supervised learning algorithms. In particular, a hybrid method of unifying the simply recycled selection method and the incrementally-reinforced selection method was considered and evaluated empirically. The experimental results, which were obtained from well-known benchmark data sets using semi-supervised support vector machines, demonstrated that the hybrid method works better than the traditional ones in terms of the classification accuracy.

• 대한원격탐사학회지
• /
• 제40권2호
• /
• pp.179-190
• /
• 2024

This study aims to compare supervised classification methods with phenology-based approaches, specifically pixel-based and segment-based methods, for accurate crop mapping in agricultural landscapes. We utilized Sentinel-2A imagery, which provides multispectral data for accurate crop mapping. 31 normalized difference vegetation index (NDVI) images were calculated from the Sentinel-2A data. Next, we employed phenology-based approaches to extract valuable information from the NDVI time series. A set of 10 phenology metrics was extracted from the NDVI data. For the supervised classification, we employed the maximum likelihood (MaxLike) algorithm. For the phenology-based approaches, we implemented both pixel-based and segment-based methods. The results indicate that phenology-based approaches outperformed the MaxLike algorithm in regions with frequent rainfall and cloudy conditions. The segment-based phenology approach demonstrated the highest kappa coefficient of 0.85, indicating a high level of agreement with the ground truth data. The pixel-based phenology approach also achieved a commendable kappa coefficient of 0.81, indicating its effectiveness in accurately classifying the crop types. On the other hand, the supervised classification method (MaxLike) yielded a lower kappa coefficient of 0.74. Our study suggests that segment-based phenology mapping is a suitable approach for regions like South Korea, where continuous cloud-free satellite images are scarce. However, establishing precise classification thresholds remains challenging due to the lack of adequately sampled NDVI data. Despite this limitation, the phenology-based approach demonstrates its potential in crop classification, particularly in regions with varying weather patterns.