• 대한원격탐사학회지
• /
• 제37권4호
• /
• pp.719-731
• /
• 2021

This study proposes a two-stage hybrid classification model for crop classification using multi-temporal remote sensing images; the model combines feature embedding by using an autoencoder (AE) with a convolutional neural network (CNN) classifier to fully utilize features including informative temporal and spatial signatures. Long short-term memory (LSTM)-based AE (LAE) is fine-tuned using class label information to extract latent features that contain less noise and useful temporal signatures. The CNN classifier is then applied to effectively account for the spatial characteristics of the extracted latent features. A crop classification experiment with multi-temporal unmanned aerial vehicle images is conducted to illustrate the potential application of the proposed hybrid model. The classification performance of the proposed model is compared with various combinations of conventional deep learning models (CNN, LSTM, and convolutional LSTM) and different inputs (original multi-temporal images and features from stacked AE). From the crop classification experiment, the best classification accuracy was achieved by the proposed model that utilized the latent features by fine-tuned LAE as input for the CNN classifier. The latent features that contain useful temporal signatures and are less noisy could increase the class separability between crops with similar spectral signatures, thereby leading to superior classification accuracy. The experimental results demonstrate the importance of effective feature extraction and the potential of the proposed classification model for crop classification using multi-temporal remote sensing images.

• 대한원격탐사학회지
• /
• 제35권5_1호
• /
• pp.681-692
• /
• 2019

이 논문에서는 작물 분류를 목적으로 작물의 시공간 특징을 고려할 수 있는 딥러닝 모델 2D convolution with bidirectional long short-term memory(2DCBLSTM)을 제안하였다. 제안 모델은 우선 작물의 공간 특징을 추출하기 위해 2차원의 합성곱 연산자를 적용하고, 추출된 공간 특징을 시간 특징을 고려할 수 있는 양방향 LSTM 모델의 입력 자료로 이용한다. 제안 모델의 분류 성능을 평가하기 위해 안반덕에서 수집된 다중시기 무인기 영상을 이용한 밭작물 구분 사례 연구를 수행하였다. 비교를 목적으로 기존 딥러닝 모델인 2차원의 공간 특징을 이용하는 2D convolutional neural network(CNN), 시간 특징을 이용하는 LSTM과 3차원의 시공간 특징을 이용하는 3D CNN을 적용하였다. 하이퍼 파라미터의 영향 분석을 통해, 시공간 특징을 이용함으로써 작물의 오분류 양상을 현저히 줄일 수 있었으며, 제안 모델이 공간 특징이나 시간 특징만을 고려하는 기존 딥러닝 모델에 비해 가장 우수한 분류 정확도를 나타냈다. 따라서 이 연구에서 제안된 모델은 작물의 시공간 특징을 고려할 수 있기 때문에 작물 분류에 효과적으로 적용될 수 있을 것으로 기대된다.

• Journal of Information Processing Systems
• /
• 제19권1호
• /
• pp.130-138
• /
• 2023

The traffic flow in an urban area is affected by the date, weather, and regional traffic flow. The existing methods are weak to model the dynamic road network features, which results in inadequate long-term prediction performance. To solve the problems regarding insufficient capacity for dynamic modeling of road network structures and insufficient mining of dynamic spatio-temporal features. In this study, we propose a novel traffic flow prediction framework called shared spatio-temporal attention convolution optimization network (SSTACON). The shared spatio-temporal attention convolution layer shares a spatio-temporal attention structure, that is designed to extract dynamic spatio-temporal features from historical traffic conditions. Subsequently, the graph optimization module is used to model the dynamic road network structure. The experimental evaluation conducted on two datasets shows that the proposed method outperforms state-of-the-art methods at all time intervals.

• Asia Marketing Journal
• /
• 제16권1호
• /
• pp.1-12
• /
• 2014

The goal-gradient hypothesis states that the tendency to approach a goal increases with the increasing proximity of the goal (Hull 1932). It was initially supported with an evidence of animal experiments and since then, several papers have investigated the goal-gradient hypothesis in humans. Although there are some evidences related to the goal-gradient hypothesis in human behaviors, none of previous studies can properly explain its underlying mechanism, and what's more, they were not able to suggest useful managerial applications in human behaviors. From these perspectives, this work points out that there are some theoretical weaknesses to apply the goal-gradient hypothesis into the complicated human decision-making behaviors and proposes an alternative theoretical mechanismthat underlies the goal-gradient hypothesis in human. Finally, it offers insights into managerial implications of the goal-gradient hypothesis in the marketing field. This study focuses on the changes in motivations for achieving goals, in terms of how approaches to goals vary according to temporal distance from those goals. Specifically, the temporal construal theory (Liberman and Trope 1998) is considered as the underlying mechanism of the goal-gradient in that the temporal construal theory argues how the temporal distance from a goal makes people change their associated values regarding to that goal. According to the temporal construal theory, the value of distant future outcomes (near future outcomes) is construed on the basis of abstract and central features (concrete and peripheral features), and it argues that distant future situations are construed on a higher level than near future situations. This means that the value associated with the high-level construal is enhanced over delay, whereas the value associated with the low-level construal is discounted over delay. Our propositions suggest that the goal-gradient behavior in human can be motivated by the different aspects or characteristics of the goal as time changes based on the temporal construal theory. Thus, the following propositions are proposed. P 1-1: If the goal is far away, consumers put more value on the central features that are more associated with the desirability of the goal. P 1-2: If the goal is far away, consumers put more effort into accomplishing the goal that has more central features, regardless of its peripheral features. P 2-1: If a goal is near, consumers put more value on the peripheral features that are more associated with the feasibility of the goal. P 2-2: If a goal is near, consumers put more effort into accomplishing the goal that has more peripheral features, regardless of its central features. We hope to provide sufficient managerial implications for the companies as our research aims to show how consumers react differently as they progress toward the goal. Proposed propositions may provide guidance for companies developing a loyalty program, enabling them to understand what kinds of benefits or services they should provide or emphasize to consumers in loyalty programs on the basis of the time-dependent changes in outcome values (such as gifts, reward coupons). The effects of temporal distance from a goal should inform companies' marketing activities and help themto determine where emphasis should be placed in designing the benefits of their loyalty program.

• 정보처리학회논문지B
• /
• 제18B권6호
• /
• pp.365-374
• /
• 2011

최근 인터넷, IPTV/SMART TV, 소셜 네트워크 (social network)와 같은 정보 유통 채널의 다양화로 유해 비디오 분류 및 차단 기술 연구에 대한 요구가 높아가고 있으나, 현재까지는 비디오에 대한 유해성을 판단하는 연구는 부족한 실정이다. 기존 유해 이미지 분류 연구에서는 이미지에서의 피부 영역의 비율이나 Bag of Visual Words (BoVW)와 같은 공간적 특징들 (spatial features)을 이용하고 있다. 그러나, 비디오에서는 공간적 특징 이외에도 모션 반복성 특징이나 시간적 상관성 (temporal correlation)과 같은 시간적 특징들 (temporal features)을 추가적으로 이용하여 유해성을 판단할 수 있다. 기존의 유해 비디오 분류 연구에서는 공간적 특징과 시간적 특징들에서 하나의 특징만을 사용하거나 두 개의 특징들을 단순히 결정 단계에서 데이터 융합하여 사용하고 있다. 일반적으로 결정 단계 데이터 융합 방법은 특징 단계 데이터 융합 방법보다 높은 성능을 가지지 못한다. 본 논문에서는 기존의 유해 비디오 분류 연구에서 사용되고 있는 공간적 특징과 시간적 특징들을 특징 단계 융합 방법을 이용하여 융합하여 유해 비디오를 분류하는 방법을 제안한다. 실험에서는 사용되는 특징이 늘어남에 따른 분류 성능 변화와 데이터 융합 방법의 변화에 따른 분류 성능 변화를 보였다. 공간적 특징만을 이용하였을 때에는 92.25%의 유해 비디오 분류 성능을 보이는데 반해, 모션 반복성 특징을 이용하고 특징 단계 데이터 융합 방법을 이용하게 되면 96%의 향상된 분류 성능을 보였다.

• 한국음향학회지
• /
• 제29권2호
• /
• pp.133-140
• /
• 2010

연속 심음신호로부터 추출한 새로운 시간영역에서의 특징들을 추가하여 심장질환 분류의 성능을 개선한다. 기존에 사용되고 있는 켑스트럼 영역 특징인 멜주파수 켑스트럼 계수 (MFCC)에 심음 포락선, 심잡음 확률벡터, 심잡음 진폭값 변동으로 구성된 새로운 3종류의 시간영역 특징을 추가한다. 심장 질환 분류 및 검출 실험에서, 시간영역 특징의 분류 정확도에 대한 기여도를 평가하고 순차적 특징선택 방식을 이용하여 시간영역 특징을 선택한다. 선택된 특징들은 다층 퍼셉트론(MLP), support rector machine (SVM), extreme learning machine (ELM)와 같은 신경회로망 패턴 분류기에 대하여 의미있고 일관되게 분류 정확도를 개선함을 보여준다.

• 말소리와 음성과학
• /
• 제1권4호
• /
• pp.127-132
• /
• 2009

The feature vectors which are used in conventional speaker recognition (SR) systems may have many correlations between their neighbors. To improve the performance of the SR, many researchers adopted linear transformation method like principal component analysis (PCA). In general, the linear transformation of the feature vectors is based on concatenated form of the static features and their dynamic features. However, the linear transformation which based on both the static features and their dynamic features is more complex than that based on the static features alone due to the high order of the features. To overcome these problems, we propose an efficient method that applies linear transformation and temporal information of the features to reduce complexity and improve the performance in speaker verification (SV). The proposed method first performs a linear transformation by PCA coefficients. The delta parameters for temporal information are then obtained from the transformed features. The proposed method only requires 1/4 in the size of the covariance matrix compared with adding the static and their dynamic features for PCA coefficients. Also, the delta parameters are extracted from the linearly transformed features after the reduction of dimension in the static features. Compared with the PCA and conventional methods in terms of equal error rate (EER) in SV, the proposed method shows better performance while requiring less storage space and complexity.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권10호
• /
• pp.2632-2649
• /
• 2012

Most approaches to human action recognition is limited due to the use of simple action datasets under controlled environments or focus on excessively localized features without sufficiently exploring the spatio-temporal information. This paper proposed a framework for recognizing realistic human actions. Specifically, a new action representation is proposed based on computing a rich set of descriptors from keypoint trajectories. To obtain efficient and compact representations for actions, we develop a feature fusion method to combine spatial-temporal local motion descriptors by the movement of the camera which is detected by the distribution of spatio-temporal interest points in the clips. A new topic model called Markov Semantic Model is proposed for semantic feature selection which relies on the different kinds of dependencies between words produced by "syntactic " and "semantic" constraints. The informative features are selected collaboratively based on the different types of dependencies between words produced by short range and long range constraints. Building on the nonlinear SVMs, we validate this proposed hierarchical framework on several realistic action datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권3호
• /
• pp.1595-1613
• /
• 2017

The IR camera and laser-based IR projector provide an effective solution for real-time collection of moving targets in RGB-D videos. Different from the traditional RGB videos, the captured depth videos are not affected by the illumination variation. In this paper, we propose a novel feature extraction framework to describe human activities based on the above optical video capturing method, namely spatial-temporal texture features for 3D human activity recognition. Spatial-temporal texture feature with depth information is insensitive to illumination and occlusions, and efficient for fine-motion description. The framework of our proposed algorithm begins with video acquisition based on laser projection, video preprocessing with visual background extraction and obtains spatial-temporal key images. Then, the texture features encoded from key images are used to generate discriminative features for human activity information. The experimental results based on the different databases and practical scenarios demonstrate the effectiveness of our proposed algorithm for the large-scale data sets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권9호
• /
• pp.2361-2376
• /
• 2023

The electrocardiogram (ECG) signal is commonly used to screen and diagnose cardiovascular diseases. In recent years, deep neural networks have been regarded as an effective way for automatic ECG disease diagnosis. The convolutional neural network is widely used for ECG signal extraction because it can obtain different levels of information. However, most previous studies adopt single scale convolution filters to extract ECG signal features, ignoring the complementarity between ECG signal features of different scales. In the paper, we propose a dual-scale network with convolution filters of different sizes for 12-lead ECG classification. Our model can extract and fuse ECG signal features of different scales. In addition, different spatial and time periods of the feature map obtained from the 12-lead ECG may have different contributions to ECG classification. Therefore, we add a spatial-temporal attention to each scale sub-network to emphasize the representative local spatial and temporal features. Our approach is evaluated on PTB-XL dataset and achieves 0.9307, 0.8152, and 89.11 on macro-averaged ROC-AUC score, a maximum F1 score, and mean accuracy, respectively. The experiment results have proven that our approach outperforms the baselines.