• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1220-1223
• /
• 2023

폐기물을 처리하는데 있어 배출과 수거에 대한 프로세스 자동화를 위해 폐기물 객체 유사도 판별이 요구된다. 이를 위해 본 연구에서는 폐기물 데이터셋에서 SIFT(Scale-Invariant Feature Transform)와 HSV(Hue, Saturation, Value)기반으로 두 이미지의 공통된 특징을 추출해 융합하고, 기계학습을 통해 이미지 객체 간의 유사도를 측정하는 모델을 제안한다. 실험을 위해 수집된 폐기물 데이터셋 81,072 장을 활용하여 이미지를 학습시키고, 전통적인 임계치 기반 유사도 측정과 본 논문에서 제시하는 유사도 측정을 비교하여 성능을 확인하였다. 임계치 기반 측정에서 SIFT 와 HSV 는 각각 0.82, 0.89(Acc)가 측정되었고, 본 논문에서 제시한 특징 추출 방법을 사용한 기계학습의 성능은 DT(Decision Tree)와 SVM(Support Vector Machine) 모두 0.93 (Acc)로 4%의 정확도가 향상되었다.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.63-69
• /
• 2021

When a person sees a sentence and understands the sentence, the person understands the sentence by reminiscent of the main word in the sentence as an image. Text-to-image is what allows computers to do this associative process. The previous deep learning-based text-to-image model extracts text features using Convolutional Neural Network (CNN)-Long Short Term Memory (LSTM) and bi-directional LSTM, and generates an image by inputting it to the GAN. The previous text-to-image model uses basic embedding in text feature extraction, and it takes a long time to train because images are generated using several modules. Therefore, in this research, we propose a method of extracting features by using the attention mechanism, which has improved performance in the natural language processing field, for sentence embedding, and generating an image by inputting the extracted features into the GAN. As a result of the experiment, the inception score was higher than that of the model used in the previous study, and when judged with the naked eye, an image that expresses the features well in the input sentence was created. In addition, even when a long sentence is input, an image that expresses the sentence well was created.

• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.234-242
• /
• 2019

In this paper, we propose a deep neural network which extracts SIFT feature points by determining whether the center pixel of a cropped image is a SIFT feature point. The data set of this network consists of a DIV2K dataset cut into $33{\times}33$ size and uses RGB image unlike SIFT which uses black and white image. The ground truth consists of the RobHess SIFT features extracted by setting the octave (scale) to 0, the sigma to 1.6, and the intervals to 3. Based on the VGG-16, we construct an increasingly deep network of 13 to 23 and 33 convolution layers, and experiment with changing the method of increasing the image scale. The result of using the sigmoid function as the activation function of the output layer is compared with the result using the softmax function. Experimental results show that the proposed network not only has more than 99% extraction accuracy but also has high extraction repeatability for distorted images.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.63-65
• /
• 2007

본 연구에서는 비디오 영상데이터로부터 인간의 행동패턴의 인식기술 및 상황인식 기법을 소개하고 이를 활용한 실용적 응용으로서 지능형 감시시스템을 제안한다. 순차적 영상신호에서 형태기반의 정적 특징과 목표물의 움직임 요소를 측정한 동적 특징을 결합한 형태의 특징 표현 및 추출기법과 행동패턴 및 상황패턴에 대한 인식 모델을 제시하고 구현한다. 모듈구조의 시스템에서 영상처리 모듈과 패턴인식 모듈은 특징추출 및 인식과정을 수행하며, 감시영상에 대한 상황판단 기능은 데이터베이스 모듈과 연동하여 효과적인 검색기능과 경보기능 등을 지원한다. 이러한 기능은 기존의 시스템에서 운영자의 지속적인 감시작업과 상황판단 작업을 보조 또는 대행하여 수행할 수 있을 뿐만 아니라 데이터저장 공간을 획기적으로 줄이고 부수적으로 효율적인 영상의 조회기능 및 추적기능 등의 유용한 인터페이스를 지원한다.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.115-121
• /
• 2022

In this paper, we propose a music classification system according to user emotions using Electroencephalogram (EEG) features that appear when listening to music. In the proposed system, the relationship between the emotional EEG features extracted from EEG signals and the auditory features extracted from music signals is learned through a deep regression neural network. The proposed system based on the regression model automatically generates EEG features mapped to the auditory characteristics of the input music, and automatically classifies music by applying these features to an attention-based deep neural network. The experimental results suggest the music classification accuracy of the proposed automatic music classification framework.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.122-131
• /
• 2011

A widely used single image-based 3D face reconstruction method, 3D morphable shape model, reconstructs an accurate 3D facial shape when 2D facial feature points are correctly extracted from an input face image. However, in the case that a user's cooperation is not available such as a real-time 3D face reconstruction system, this method can be vulnerable to the errors of automatic facial feature point extraction. In order to solve this problem, we automatically classify extracted facial feature points into two groups, erroneous and correct ones, and then reconstruct a 3D facial shape by using only the correctly extracted facial feature points. The experimental results showed that the 3D reconstruction performance of the proposed method was remarkably improved compared to that of the previous method which does not consider the errors of automatic facial feature point extraction.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.2 s.308
• /
• pp.82-86
• /
• 2006

The design of a pattern recognition system generally involves the three aspects: preprocessing, feature extraction, and decision making. Among them, a feature extraction method determines an appropriate subspace of dimensionality in the original feature space of dimensionality so that it can reduce the complexity of the system and help to improve successful recognition rates. Linear transforms, such as principal component analysis, factor analysis, and linear discriminant analysis have been widely used in pattern recognition for feature extraction. This paper shows that singular value decomposition (SVD) can be applied usefully in feature extraction stage of pattern recognition. As an application, a remote sensing problem is applied to verify the usefulness of SVD. The experimental result indicates that the feature extraction using SVD can improve the recognition rate about 25% compared with that of PCA.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.1
• /
• pp.19-25
• /
• 2001

A novel approach for the feature extraction of handwritten characters is proposed by using spatial filtering neural networks with 4 layers. The proposed system first removes rough pixels which are easy to occur in handwritten characters. The system then extracts and removes the boundary information which have no influence on characters recognition. Finally, The system extracts feature information and removes the noises from feature information. The spatial filters adapted in the system correspond to the receptive fields of ganglion cells in retina and simple cells in visual cortex. With PE2 Hangul database, we perform experiments extracting features of handwritten characters recognition. It will be shown that the network can extract feature informations from handwritten characters successfully.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.418-421
• /
• 2011

인간의 감성은 개인이 생활을 통하여 갖게 되는 자신의 기준에 의하여 동일한 외부자극에 대해서 다양하게 나타난다. 대부분 우리는 인간이 수행하는 각각의 행동들을 통해 문제 해결을 위한 감성의 변화와 생각, 사고의 절차 등을 인지할 수 있기 때문에 감성은 인간의 작업 능력과 정보 분석 및 해결 등의 문제들과 밀접하게 관계가 있다. 본 논문에서는 이미지의 내재 정보에서 특징점을 추출하고 추출한 특징점을 SVM 알고리즘을 이용하여 학습시킨 후 결과 클래스를 러셀 평면 좌표계에 맵핑함으로써 이미지의 감성 정보를 추출하는 연구를 진행하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.11
• /
• pp.427-432
• /
• 2019

In this paper, we propose three approaches to modeling Android malware. The first method involves human security experts for meticulously selecting feature sets. With the second approach, we choose 300 features with the highest importance among the top 99% features in terms of occurrence rate. The third approach is to combine multiple models and identify malware through weighted voting. In addition, we applied a novel method of eliminating permission information which used to be regarded as a critical factor for distinguishing malware. With our carefully generated feature sets and the weighted voting by the ensemble algorithm, we were able to reach the highest malware detection accuracy of 97.8%. We also verified that discarding the permission information lead to the improvement in terms of false positive and false negative rates.