• ETRI Journal
• /
• 제40권4호
• /
• pp.499-510
• /
• 2018

A vector space based augmented structural kinematic (VSASK) feature descriptor is proposed for human activity recognition. An action descriptor is built by integrating the structural and kinematic properties of the actor using vector space based augmented matrix representation. Using the local or global information separately may not provide sufficient action characteristics. The proposed action descriptor combines both the local (pose) and global (position and velocity) features using augmented matrix schema and thereby increases the robustness of the descriptor. A multiclass support vector machine (SVM) is used to learn each action descriptor for the corresponding activity classification and understanding. The performance of the proposed descriptor is experimentally analyzed using the Weizmann and KTH datasets. The average recognition rate for the Weizmann and KTH datasets is 100% and 99.89%, respectively. The computational time for the proposed descriptor learning is 0.003 seconds, which is an improvement of approximately 1.4% over the existing methods.

• 대한인간공학회지
• /
• 제31권2호
• /
• pp.271-279
• /
• 2012

Objective: This study classified three different emotional states(boredom, pain, and surprise) using physiological signals. Background: Emotion recognition studies have tried to recognize human emotion by using physiological signals. It is important for emotion recognition to apply on human-computer interaction system for emotion detection. Method: 122 college students participated in this experiment. Three different emotional stimuli were presented to participants and physiological signals, i.e., EDA(Electrodermal Activity), SKT(Skin Temperature), PPG(Photoplethysmogram), and ECG (Electrocardiogram) were measured for 1 minute as baseline and for 1~1.5 minutes during emotional state. The obtained signals were analyzed for 30 seconds from the baseline and the emotional state and 27 features were extracted from these signals. Statistical analysis for emotion classification were done by DFA(discriminant function analysis) (SPSS 15.0) by using the difference values subtracting baseline values from the emotional state. Results: The result showed that physiological responses during emotional states were significantly differed as compared to during baseline. Also, an accuracy rate of emotion classification was 84.7%. Conclusion: Our study have identified that emotions were classified by various physiological signals. However, future study is needed to obtain additional signals from other modalities such as facial expression, face temperature, or voice to improve classification rate and to examine the stability and reliability of this result compare with accuracy of emotion classification using other algorithms. Application: This could help emotion recognition studies lead to better chance to recognize various human emotions by using physiological signals as well as is able to be applied on human-computer interaction system for emotion recognition. Also, it can be useful in developing an emotion theory, or profiling emotion-specific physiological responses as well as establishing the basis for emotion recognition system in human-computer interaction.

• 대한임베디드공학회논문지
• /
• 제18권6호
• /
• pp.319-325
• /
• 2023

In this study, we proposed a method for hyperparameter optimization in the building and training of a deep learning model designed to process point cloud data collected by a millimeter-wave radar system. The primary aim of this study is to facilitate the deployment of a baseline model in resource-constrained IoT devices. We evaluated a RadHAR baseline deep learning model trained on a public dataset composed of point clouds representing five distinct human activities. Additionally, we introduced a coarse-to-fine hyperparameter optimization procedure, showing substantial potential to enhance model efficiency without compromising predictive performance. Experimental results show the feasibility of significantly reducing model size without adversely impacting performance. Specifically, the optimized model demonstrated a 3.3% improvement in classification accuracy despite a 16.8% reduction in number of parameters compared th the baseline model. In conclusion, this research offers valuable insights for the development of deep learning models for resource-constrained IoT devices, underscoring the potential of hyperparameter optimization and model size reduction strategies. This work contributes to enhancing the practicality and usability of deep learning models in real-world environments, where high levels of accuracy and efficiency in data processing and classification tasks are required.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅲ
• /
• pp.1279-1282
• /
• 2003

Estimation of human motion states is important enabling technologies for realizing a pervasive computing environment. In this paper, an improved method fur estimating human motion state from accelerometer data is introduced. Our method fur estimating human motion state utilizes various statistics of accelerometer data, such as mean, standard variation, skewness, kurtosis, eccentricity, as features for classification, and therefore is expected to be more robust than other existing methods that rely on only a few simple statistics. A series of experiments fur testing the effectiveness of the proposed method has been performed, and its result is presented.

• International Journal of Contents
• /
• 제8권4호
• /
• pp.56-63
• /
• 2012

In recent decades, analyzing the activities of human brain achieved some accomplishments by using the functional Magnetic Resonance Imaging (fMRI) technique. fMRI data provide a sequence of three-dimensional images related to human brain's activity which can be used to detect instantaneous cognitive states by applying machine learning methods. In this paper, we propose a new approach for distinguishing human's cognitive states such as "observing a picture" versus "reading a sentence" and "reading an affirmative sentence" versus "reading a negative sentence". Since fMRI data are high dimensional (about 100,000 features in each sample), extremely sparse and noisy, feature selection is a very important step for increasing classification accuracy and reducing processing time. We used the Fisher Discriminant Ratio to select the most powerful discriminative features from some Regions of Interest (ROIs). The experimental results showed that our approach achieved the best performance compared to other feature extraction methods with the average accuracy approximately 95.83% for the first study and 99.5% for the second study.

• 한국멀티미디어학회논문지
• /
• 제23권12호
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.

• ETRI Journal
• /
• 제42권1호
• /
• pp.78-89
• /
• 2020

Human activity recognition (HAR) has become effective as a computer vision tool for video surveillance systems. In this paper, a novel biometric system that can detect human activities in 3D space is proposed. In order to implement HAR, joint angles obtained using an RGB-depth sensor are used as features. Because HAR is operated in the time domain, angle information is stored using the sliding kernel method. Haar-wavelet transform (HWT) is applied to preserve the information of the features before reducing the data dimension. Dimension reduction using an averaging algorithm is also applied to decrease the computational cost, which provides faster performance while maintaining high accuracy. Before the classification, a proposed thresholding method with inverse HWT is conducted to extract the final feature set. Finally, the K-nearest neighbor (k-NN) algorithm is used to recognize the activity with respect to the given data. The method compares favorably with the results using other machine learning algorithms.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.478-480
• /
• 2022

This paper studies human activity image classification using deep transfer learning techniques focused on the inception convolutional neural networks (InceptionV3) model. For this, we used UFC-101 public datasets containing a group of students' behaviors in mathematics classrooms at a school in Thailand. The video dataset contains Play Sitar, Tai Chi, Walking with Dog, and Student Study (our dataset) classes. The experiment was conducted in three phases. First, it extracts an image frame from the video, and a tag is labeled on the frame. Second, it loads the dataset into the inception V3 with transfer learning for image classification of four classes. Lastly, we evaluate the model's accuracy using precision, recall, F1-Score, and confusion matrix. The outcomes of the classifications for the public and our dataset are 1) Play Sitar (precision = 1.0, recall = 1.0, F1 = 1.0), 2), Tai Chi (precision = 1.0, recall = 1.0, F1 = 1.0), 3) Walking with Dog (precision = 1.0, recall = 1.0, F1 = 1.0), and 4) Student Study (precision = 1.0, recall = 1.0, F1 = 1.0), respectively. The results show that the overall accuracy of the classification rate is 100% which states the model is more powerful for learning UCF-101 and our dataset with higher accuracy.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.395-397
• /
• 2019

조선기술과 항해장비 기술이 발전하고 있지만 여전히 해양사고는 80%이상이 인적과실에서 비롯되고 있다. 인적과실을 저감시켜 해양사고를 절감시키려는 노력은 항해사를 대상으로 면담이나 설문을 시행하는 등 정성적인 연구방식에 많이 의존하고 있어서 객관적인 인적과실의 실체를 규명하는데 제한이 있다. 본 연구에서는 이 같은 단점을 극복하기 위하여 항해사의 항해 업무 수행을 방해하지 않으며 공간적 제한을 극복할 수 있도록 웨어러블 센서를 활용하여 항해사의 동작을 실측하고 상황인지 여부가 항해 수행 동작에 어떤 영향을 미치는지 구분하고자 한다. Full mission ship handling simulator를 활용하여 항해사가 특정한 시나리오를 수행하는 중에 위험성을 가진 장애물을 발견하기 전과 후의 어떤 행동패턴 변화를 보이는지 측정하였다. 구분된 항해 동작 패턴은 항해 위험 상황에서 적절한 조치를 취하고 있는지 여부를 객관적으로 구분하여 인적과실을 절감하는데 활용될 것으로 기대된다.

• 한국인터넷방송통신학회논문지
• /
• 제22권5호
• /
• pp.177-183
• /
• 2022

최근 COVID-19로 인해 홈 트레이닝의 관심도가 증가하고 있다. 이에 따라 HAR(human activity recognition) 기술을 홈 트레이닝에 적용한 연구가 진행되고 있다. 기존 HAR 분야의 논문에서는 동적인 자세보다는 앉기, 일어서기와 같은 정적인 자세들을 분석한다. 본 논문은 동적인 운동 자세를 분석하여 사용자의 운동 자세 정확도를 보여주는 딥러닝 모델을 제안한다. AI hub의 피트니스 이미지를 blaze pose를 사용하여 사람의 자세 데이터를 분석한다. 3개의 딥러닝 모델: RNN(recurrnet neural networks), LSTM(long short-term memory networks), CNN(convolution neural networks)에 대하여 실험을 진행한다. RNN, LSTM, CNN 모델의 f1-score는 각각 0.49, 0.87, 0.98로 CNN 모델이 가장 적합하다는 것을 확인하였다. 이후 연구로는, 다양한 학습 데이터를 사용하여 더 많은 운동 자세를 분석할 예정이다.