• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.110.3-110
• /
• 2002

$\textbullet$ Lane detection by TFALDA $\textbullet$ SVM for large scale data and multiclass classification problem $\textbullet$ TLC Classification $\textbullet$ Lateral offset estimation by IPT $\textbullet$ Lane departure warning by a fuzzy system $\textbullet$ Experimental results by HiLS $\textbullet$ Conclusion

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.5
• /
• pp.79-89
• /
• 2016

In this paper, we proposed the object classification method using genetic and dynamic random forest consisting of optimal combination of unit tree. The random forest can ensure good generalization performance in combination of large amount of trees by assigning the randomization to the training samples and feature selection, etc. allocated to the decision tree as an ensemble classification model which combines with the unit decision tree based on the bagging. However, the random forest is composed of unit trees randomly, so it can show the excellent classification performance only when the sufficient amounts of trees are combined. There is no quantitative measurement method for the number of trees, and there is no choice but to repeat random tree structure continuously. The proposed algorithm is composed of random forest with a combination of optimal tree while maintaining the generalization performance of random forest. To achieve this, the problem of improving the classification performance was assigned to the optimization problem which found the optimal tree combination. For this end, the genetic algorithm methodology was applied. As a result of experiment, we had found out that the proposed algorithm could improve about 3~5% of classification performance in specific cases like common database and self infrared database compare with the existing random forest. In addition, we had shown that the optimal tree combination was decided at 55~60% level from the maximum trees.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.11B
• /
• pp.1285-1296
• /
• 2011

Packet classification is one of the basic and important functions of the Internet routers, and it became more important along with new emerging application programs requiring real-time transmission. Since packet classification should be accomplished in line-speed on each incoming input packet for multiple header fields, it becomes one of the challenges in designing Internet routers. Various packet classification algorithms have been proposed to provide the high-speed packet classification. Hierarchical approach achieves effective packet classification performance by significantly narrowing down the search space whenever a field lookup is completed. However, hierarchical approach involves back-tracking problem. In order to solve the problem, set-pruning trie and grid-of-trie algorithms are proposed. However, the algorithm either causes excessive node duplication or heavy pre-computation. In this paper, we propose a smart set-pruning trie which reduces the number of node duplication in the set-pruning trie by the simple merging of the lower-level tries. Simulation result shows that the proposed trie has the reduced number of copied nodes by 2-8% compared with the set-pruning trie.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.166-172
• /
• 2014

Image classification is an important problem in computer vision. However, it is a very challenging problem due to the variability, ambiguity and scale change that exists in images. In this paper, we propose a method of a hypergraph based modeling can consider the higher-order relationships of semantic attributes of a scene image and apply it to a scene image classification. In order to generate the hypergraph optimized for specific scene category, we propose a novel search method based on a probabilistic subspace method and also propose a method to aggregate the expression values of the member semantic attributes that belongs to the searched subsets based on a linear transformation method via likelihood based estimation. To verify the superiority of the proposed method, we showed that the discrimination power of the feature vector generated by the proposed method is better than existing methods through experiments. And also, in a scene classification experiment, the proposed method shows a competitive classification performance compared with the conventional methods.

• Journal of Korea Multimedia Society
• /
• v.16 no.7
• /
• pp.795-801
• /
• 2013

Recently, support vector machine has been widely used in various application fields due to its superiority of classification performance comparing with decision tree and neural network. Since support vector machine is basically designed for the binary classification problem, output coding method to analyze the classification result of multiclass binary classifier is used for the application of support vector machine into the multiclass problem. However, previous feature selection method for output coding based support vector machine found the features to improve the overall classification accuracy instead of improving each classification accuracy of each classifier. In this paper, we propose the novel feature selection method to find the features for maximizing the classification accuracy of each binary classifier in output coding based support vector machine. Experimental result showed that proposed method significantly improved the classification accuracy comparing with previous feature selection method.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.3
• /
• pp.215-220
• /
• 2010

In this paper, we propose a new texture-based page segmentation and classification method in which table region, background region, image region and text region in a given document image are automatically identified. The proposed method for document images consists of two stages, document segmentation and contents classification. In the first stage, we segment the document image, and then, we classify contents of document in the second stage. The proposed classification method is based on a texture analysis. Each contents in the document are considered as regions with different textures. Thus the problem of classification contents of document can be posed as a texture segmentation and analysis problem. Two-dimensional Gabor filters are used to extract texture features for each of these regions. Our method does not assume any a priori knowledge about content or language of the document. As we can see experiment results, our method gives good performance in document segmentation and contents classification. The proposed system is expected to apply such as multimedia data searching, real-time image processing.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2007.04a
• /
• pp.183-186
• /
• 2007

Hyperspectral remote sensing data contain plenty of information about objects, which makes object classification more precise. Over the past several years, different algorithms for the classification of hyperspectral remote sensing images have been developed. In this study, we proposed method based on absorption band extraction and Gravity type model to solve hyperspectral image classification problem. In contrast to conventional methods that are based on correlation techniques, this method is simple and more effective. The proposed approach was tested to evaluate its effectiveness. The evaluation was done by comparing the results of preexiting SFF(Spectral Feature Fitting) classification method. The evaluation results showed the proposed approach has a good potential in the classification of hyperspectral images.

• Journal of Korea Multimedia Society
• /
• v.18 no.11
• /
• pp.1261-1267
• /
• 2015

The problem of underwater target detection and classification has been attracted a substantial amount of attention and studied from many researchers for both military and non-military purposes. The difficulty is complicate due to various environmental conditions. In this paper, we study classifier ensemble methods for active sonar target classification to improve the classification performance. In general, classifier ensemble method is useful for classifiers whose variances relatively large such as decision trees and neural networks. Bagging, Random selection samples, Random subspace and Rotation forest are selected as classifier ensemble methods. Using the four ensemble methods based on 31 neural network classifiers, the classification tests were carried out and performances were compared.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.3E
• /
• pp.91-97
• /
• 2004

In this paper, we propose a classification system of TV program scenes based on audio information. The system classifies the video scene into six categories of commercials, basketball games, football games, news reports, weather forecasts and music videos. Two type of audio feature set are extracted from each audio frame-timbral features and coefficient domain features which result in 58-dimensional feature vector. In order to reduce the computational complexity of the system, 58-dimensional feature set is further optimized to yield l0-dimensional features through Sequential Forward Selection (SFS) method. This down-sized feature set is finally used to train and classify the given TV program scenes using κ -NN, Gaussian pattern matching algorithm. The classification result of 91.6% reported here shows the promising performance of the video scene classification based on the audio information. Finally, the system stability problem corresponding to different query length is investigated.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2367-2370
• /
• 2003

A fast pattern classification algorithm with Cellular Parallel Processing Network-based dynamic programming is proposed. The Cellular Parallel Processing Networks is an analog parallel processing architecture and the dynamic programming is an efficient computation algorithm for optimization problem. Combining merits of these two technologies, fast Pattern classification with optimization is formed. On such CPPN-based dynamic programming, if exemplars and test patterns are presented as the goals and the start positions, respectively, the optimal paths from test patterns to their closest exemplars are found. Such paths are utilized as aggregating keys for the classification. The pattern classification is performed well regardless of degree of the nonlinearity in class borders.