• Journal of Korean Society of Rural Planning
• /
• v.6 no.2 s.12
• /
• pp.50-58
• /
• 2000

The purposes of the study were to detect and evaluate the historical land use and land cover changes on the Balan watersheds from three thematic mapper (TM) data, which were taken in 1985, 1993, and 1996. The supervised and unsupervised classification methods were adopted to classify five land cover categories: Paddy, upland, forest, residential, and water. The results indicated residential areas increased significantly during the past eleven years, Forest and paddy were converted to the urban areas. Future land cover patterns were forecasted using a Markov chain method, and the simulated land coiler change ratios presented.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.870-876
• /
• 1993

This paper presents a binary clustering network (BCN) and a heuristic algorithm to detect pitch for recognition of keywords in continuous speech. In order to classify nonlinear patterns, BCN separates patterns into binary clusters hierarchically and links same patterns at root level by using the supervised learning and the unsupervised learning. BCN has many desirable properties such as flexibility of dynamic structure, high classification accuracy, short learning time, and short recall time. Pitch Detection algorithm is a heuristic model that can solve the difficulties such as scaling invariance, time warping, time-shift invariance, and redundance. This recognition algorithm has shown recognition rates as high as 95% for speaker-dependent as well as multispeaker-dependent tests.

• Bulletin of the Korean Chemical Society
• /
• v.11 no.5
• /
• pp.450-455
• /
• 1990

Fifty eight obsidian artifacts and four obsidian source samples have been analyzed by instrumental neutron activation analysis. Artifact samples have been classified into classes by unsupervised learning techniques such as eigenvector projection and nonlinear mapping. The source samples have thereafter been connected to the classes by the supervised learning techniques such as SLDA and SIMCA so as to characterize each class by possible source sites. Some difference attributable to different nonlinear mapping techniques and the elemental effects on the separation between classes have been discussed.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.5
• /
• pp.765-776
• /
• 1995

This paper describes a speaker adaptation method to improve the recognition performance of MLP(multiLayer Perceptron) based HMM(Hidden Markov Model) speech recognizer. In this method, we use lst-order linear transformation network to fit data of a new speaker to the MLP. Transformation parameters are adjusted by back-propagating classification error to the transformation network while leaving the MLP classifier fixed. The recognition system is based on semicontinuous HMM's which use the MLP as a fuzzy vector quantizer. The experimental results show that rapid speaker adaptation resulting in high recognition performance can be accomplished by this method. Namely, for supervised adaptation, the error rate is signifecantly reduced from 9.2% for the baseline system to 5.6% after speaker adaptation. And for unsupervised adaptation, the error rate is reduced to 5.1%, without any information from new speakers.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.05a
• /
• pp.599-606
• /
• 2005

Novelty detection involves identifying novel patterns. They are not usually available during training. Even if they are, the data quantity imbalance leads to a low classification accuracy when a supervised learning scheme is employed. Thus, an unsupervised learning scheme is often employed ignoring those few novel patterns. In this paper, we propose two ways to make use of the few available novel patterns. First, a scheme to determine local thresholds for the Self Organizing Map boundary is proposed. Second, a modification of the Learning Vector Quantization learning rule is proposed so that allows one to keep codebook vectors as far from novel patterns as possible. Experimental results are quite promising.

• Journal of KIISE:Software and Applications
• /
• v.34 no.2
• /
• pp.160-168
• /
• 2007

For developing a text classifier using supervised learning, a manually labeled corpus of large size is required. However, it takes a lot of time and human effort. Recently a research paradigm was proposed to use a raw corpus and a small amount of seed information instead of manually labeled corpus. In this paper we introduce an unsupervised learning method that makes it possible to achieve better performance than other related works. The characteristics of our approach is that average mutual information is used to learn representative words and their weights and then update of the weights is done using a technique inspired by the works in information retrieval. By iterating this teaming process it was shown that a high performance system can be developed.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.355-355
• /
• 2011

토지피복의 분류는 토지표면의 물리적인 지표면의 상태를 나타내는 자료로 환경, 행정, 수자원, 재해 등 다방면으로 이용되고 있다. 특히 수자원과 관련하여 식생의 증산과 토양의 증발을 통칭하는 증발산과 유출, 토양수분 등과 연관되어 있다. 광범위한 토지피복의 산정에는 경제성 및 주기성 등의 장점으로 인하여 인공위성 영상을 이용하는 기법이 적합하다. 위성영상분류법은 훈련지역의 선정 여부에 따라 감독분류와 무감독 분류로 나누어지며 각각의 알고리즘의 특성에 따라 더욱 세분화된다. 본 연구에서는 Landsat-TM (Thematic Mapper) 영상을 이용하여 감독 분류와 무감독 분류를 각각 적용하여 한강유역의 토지피복을 수역, 시가, 나지 습지, 초지, 산림, 농지의 7가지 부분으로 대분류로 산정하고 비교하였다. 두 경우의 정확도는 각각 91.6%, 90.9%의 비슷한 정확도를 나타내었으며, 세부적으로 우리나라의 대부분의 면적에 분포하는 산림, 농지, 시가, 수역의 정확도가 높게 나타났다. 또한 각 항목별로 정확도를 비교하였을 때 감독분류가 무감독분류에 비해 다소 정확한 것을 확인할 수 있었다. 추후 외부자료를 도입하면 비교적 낮은 정확도를 나타낸 초지, 습지, 나지의 정확도를 보완할 수 있을 것이다.

• Korean Journal of Remote Sensing
• /
• v.27 no.3
• /
• pp.315-328
• /
• 2011

Landsat data incorporated with additional bands-normalized difference vegetation index (NDVI) and band ratios were used to assess the extent and rate of deforestation in the Gunung Mutis Nature Reserve (GMNR), a mountainous tropical forest in Eastern of Indonesia. Hybrid classification was chosen as the classification approach. In this approach, the unsupervised classification-iterative self-organizing data analysis (ISODATA) was used to create signature files and training data set. A statistical separability measurement-transformed divergence (TD) was used to identify the combination of bands that showed the highest distinction between the land cover classes in training data set. Supervised classification-maximum likelihood classification (MLC) was performed using selected bands and the training data set. Post-classification smoothing and accuracy assessment were applied to classified image. Post-classification comparison was used to assess the extent of deforestation, of which the rate of deforestation was calculated by the formula suggested by Food Agriculture Organization (FAO). The results of two periods of deforestation assessment showed that the extent of deforestation during 1989-1999 was 720.72 ha, 0.80% of annual rate of deforestation, and its extent of deforestation during 1999-2009 was 1,059.12 ha, 1.31% of annual rate of deforestation. Such results are important for the GMNR authority to establish strategies, plans and actions for combating deforestation.

• Journal of Forest and Environmental Science
• /
• v.36 no.3
• /
• pp.199-206
• /
• 2020

Satchari National Park is one of the most biodiverse forest in Bangladesh and home of many endangered flora and fauna. 206 tons of CO₂ per hectare is sequestrated in this national park every year which helps to mitigate climate issues. As people living near the area are dependent on this forest, degradation has become a regular phenomenon destroying the forest biodiversity by altering its forest cover. So, it is important to map land cover quickly and accurately for the sustainable management of Satchari National Park. The main objective of this study was to obtain information on land cover change using remote sensing data. Combination of unsupervised NDVI classification and supervised classification using maximum likelihood is followed in this study to find out land cover map. The analysis showed that the land cover is gradually converting from one land use type to another. Dense forest becoming degraded forest or bare land. Although it was slowed down by the establishment of 'National Park' on the study site, forecasting shows that it is not enough to mitigate forest degradation. Legal steps and proper management strategies should be taken to mitigate causes of degradation such as illegal felling.

• International Journal of Knowledge Content Development & Technology
• /
• v.7 no.1
• /
• pp.57-78
• /
• 2017

There are several statistical classification algorithms available for land use/land cover classification. However, each has a certain bias or compromise. Some methods like the parallel piped approach in supervised classification, cannot classify continuous regions within a feature. On the other hand, while unsupervised classification method takes maximum advantage of spectral variability in an image, the maximally separable clusters in spectral space may not do much for our perception of important classes in a given study area. In this research, the output of an ANN algorithm was compared with the Possibilistic c-Means an improvement of the fuzzy c-Means on both moderate resolutions Landsat8 and a high resolution Formosat 2 images. The Formosat 2 image comes with an 8m spectral resolution on the multispectral data. This multispectral image data was resampled to 10m in order to maintain a uniform ratio of 1:3 against Landsat 8 image. Six classes were chosen for analysis including: Dense forest, eucalyptus, water, grassland, wheat and riverine sand. Using a standard false color composite (FCC), the six features reflected differently in the infrared region with wheat producing the brightest pixel values. Signature collection per class was therefore easily obtained for all classifications. The output of both ANN and FCM, were analyzed separately for accuracy and an error matrix generated to assess the quality and accuracy of the classification algorithms. When you compare the results of the two methods on a per-class-basis, ANN had a crisper output compared to PCM which yielded clusters with pixels especially on the moderate resolution Landsat 8 imagery.