• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.2
• /
• pp.65-72
• /
• 2021

Recently, attacks on the network environment have been rapidly escalating and intelligent. Thus, the signature-based network intrusion detection system is becoming clear about its limitations. To solve these problems, research on machine learning-based intrusion detection systems is being conducted in many ways, but two problems are encountered to use machine learning for intrusion detection. The first is to find important features associated with learning for real-time detection, and the second is the imbalance of data used in learning. This problem is fatal because the performance of machine learning algorithms is data-dependent. In this paper, we propose the HSF-DNN, a network intrusion detection model based on a deep neural network to solve the problems presented above. The proposed HFS-DNN was learned through the NSL-KDD data set and performs performance comparisons with existing classification models. Experiments have confirmed that the proposed Hybrid Feature Selection algorithm does not degrade performance, and in an experiment between learning models that solved the imbalance problem, the model proposed in this paper showed the best performance.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.31-39
• /
• 2023

In this paper, we propose an Internal/External Knowledge Distillation (IEKD), which utilizes both external correlations between feature maps of heterogeneous models and internal correlations between feature maps of the same model for transferring knowledge from a teacher model to a student model. To achieve this, we transform feature maps into a sequence format and extract new feature maps suitable for knowledge distillation by considering internal and external correlations through a transformer. We can learn both internal and external correlations by distilling the extracted feature maps and improve the accuracy of the student model by utilizing the extracted feature maps with feature matching. To demonstrate the effectiveness of our proposed knowledge distillation method, we achieved 76.23% Top-1 image classification accuracy on the CIFAR-100 dataset with the "ResNet-32×4/VGG-8" teacher and student combination and outperformed the state-of-the-art KD methods.

• 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.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.385-398
• /
• 2021

The complexity of deep learning models affects the real-time performance of gesture recognition, thereby limiting the application of gesture recognition algorithms in actual scenarios. Hence, a residual learning neural network based on a deep convolutional neural network is proposed. First, small convolution kernels are used to extract the local details of gesture images. Subsequently, a shallow residual structure is built to share weights, thereby avoiding gradient disappearance or gradient explosion as the network layer deepens; consequently, the difficulty of model optimisation is simplified. Additional convolutional neural networks are used to accelerate the refinement of deep abstract features based on the spatial importance of the gesture feature distribution. Finally, a fully connected cascade softmax classifier is used to complete the gesture recognition. Compared with the dense connection multiplexing feature information network, the proposed algorithm is optimised in feature multiplexing to avoid performance fluctuations caused by feature redundancy. Experimental results from the ISOGD gesture dataset and Gesture dataset prove that the proposed algorithm affords a fast convergence speed and high accuracy.

• Journal of Korean Society for Geospatial Information Science
• /
• v.11 no.4 s.27
• /
• pp.35-43
• /
• 2003

Currently, the government has been driving numerous projects to build the e-government which can enable limitless access and utilization of the information through the accomplishment of the real time based various administrative services. In water resource field, a project to generate digital river thematic maps has been undergoing as a part of the computerization projects. As a partial results, the RIMGIS project has been completed and generation of the various river thematic maps has been required to fully utilize the DB built from RIMGIS project. For the effective generation of the thematic maps, a data model needs to be developed. A data model has been developed in this study to provide more efficient method to generate the thematic maps utilizing existing DB. The data model proposed from this study has defined the relationships between core feature data and framework Data along with relationships among data elements to represent the rivers in the real world more accurately. The core feature data and framework layers have been defined based on the survey of the domestic and foreign case studies along with requirement analysis of the users in the water resource field. The proposed core feature data has been defined based on the minimum unit of 'class', and the relationship between classes has been established based on the ArcGIS Hydro Data Model for the integrated processing of the river information. The proposed spatial data model can be judged to contribute establishing more efficient generation methodology of the river thematic maps.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.96.3-96
• /
• 2002

A scheme of clarifying fault feature in the residual is given for model-based fault detection and diagnosis of control systems. It is based on the residual generation using a robust filter and the noise suppresion in test statistics of the residual by multi-scale discrete wavelet transform. By clarifying the fault feature in the residual, the difficulties of existing model based approaches via adopting a threshold can be overcomed and it has advantage of taking the false alarm and missed detection into acount at the same time, which can make the fault detection and diagnosis easy and correct. To show the effectiveness of our approach, the simulation results are illustrated for a linear syste...

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.27 no.3
• /
• pp.59-65
• /
• 2004

This paper explores an image segmentation and representation method using Vector Quantization(VQ) on color and texture for content-based image retrieval system. The basic idea is a transformation from the raw pixel data to a small set of image regions which are coherent in color and texture space. These schemes are used for object-based image retrieval. Features for image retrieval are three color features from HSV color model and five texture features from Gray-level co-occurrence matrices. Once the feature extraction scheme is performed in the image, 8-dimensional feature vectors represent each pixel in the image. VQ algorithm is used to cluster each pixel data into groups. A representative feature table based on the dominant groups is obtained and used to retrieve similar images according to object within the image. The proposed method can retrieve similar images even in the case that the objects are translated, scaled, and rotated.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2463-2466
• /
• 2003

This paper studied the pattern recognition algorithm and feature parameters for emotion recognition. In this paper, KNN algorithm was used as the pattern matching technique for comparison, and also VQ and GMM were used lot speaker and context independent recognition. The speech parameters used as the feature are pitch, energy, MFCC and their first and second derivatives. Experimental results showed that emotion recognizer using MFCC and their derivatives as a feature showed better performance than that using the Pitch and energy Parameters. For pattern recognition algorithm, GMM based emotion recognizer was superior to KNN and VQ based recognizer

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1989-2011
• /
• 2019

Vehicle detection based on aerial images is an interesting and challenging research topic. Most of the traditional vehicle detection methods are based on the sliding window search algorithm, but these methods are not sufficient for the extraction of object features, and accompanied with heavy computational costs. Recent studies have shown that convolutional neural network algorithm has made a significant progress in computer vision, especially Faster R-CNN. However, this algorithm mainly detects objects in natural scenes, it is not suitable for detecting small object in aerial view. In this paper, an accurate and effective vehicle detection algorithm based on Faster R-CNN is proposed. Our method fuse a hyperactive feature map network with Eltwise model and Concat model, which is more conducive to the extraction of small object features. Moreover, setting suitable anchor boxes based on the size of the object is used in our model, which also effectively improves the performance of the detection. We evaluate the detection performance of our method on the Munich dataset and our collected dataset, with improvements in accuracy and effectivity compared with other methods. Our model achieves 82.2% in recall rate and 90.2% accuracy rate on Munich dataset, which has increased by 2.5 and 1.3 percentage points respectively over the state-of-the-art methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.12
• /
• pp.5685-5700
• /
• 2018

Cyber intelligence, surveillance, and reconnaissance (ISR) has become more important than traditional military ISR. An agent used in cyber ISR resides in an enemy's networks and continually collects valuable information. Thus, this agent should be able to determine what is, and is not, useful in a short amount of time. Moreover, the agent should maintain a classification rate that is high enough to select useful data from the enemy's network. Traditional feature selection algorithms cannot comply with these requirements. Consequently, in this paper, we propose an effective hybrid feature selection method derived from the filter and wrapper methods. We illustrate the design of the proposed model and the experimental results of the performance comparison between the proposed model and the existing model.