• Journal of Electrical Engineering and Technology
• /
• v.9 no.2
• /
• pp.739-748
• /
• 2014

In this paper, we propose pyramid appearance and global structure action descriptors on both RGB and depth motion history images and a model-free method for human action recognition. In proposed algorithm, we firstly construct motion history image for both RGB and depth channels, at the same time, depth information is employed to filter RGB information, after that, different action descriptors are extracted from depth and RGB MHIs to represent these actions, and then multimodality information collaborative representation and recognition model, in which multi-modality information are put into object function naturally, and information fusion and action recognition also be done together, is proposed to classify human actions. To demonstrate the superiority of the proposed method, we evaluate it on MSR Action3D and DHA datasets, the well-known dataset for human action recognition. Large scale experiment shows our descriptors are robust, stable and efficient, when comparing with the-state-of-the-art algorithms, the performances of our descriptors are better than that of them, further, the performance of combined descriptors is much better than just using sole descriptor. What is more, our proposed model outperforms the state-of-the-art methods on both MSR Action3D and DHA datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Women's Health Nursing
• /
• v.8 no.4
• /
• pp.506-520
• /
• 2002

The purpose of this study is to research the degree of recognition & performance of traditional postpartal care for postpartal women and to provide the basic data for improvement of service in a postpartum care center. The respondents of this study were 100 women of 6 postpartum care centers within a C province from Oct. 20 to Dec. 10, 2000. The instruments of measure were used for collecting data on the degree of recognition & performance of traditional postpartal care developed by the researcher. Data analysis consisted of frequency, percentage, mean, standard deviation, paired t-test, t-test, ANOVA which are calculated by Scheffe test and Cronbach's alpha which is used as a reliance level by using a SPSS-PC+. The results of the study were as follows:1. The average score for the degree of recognition of traditional postpartal care(Sanhujori) for postpartal women was $3.09{\pm}.31$, and they recognized that it was important. The methods which were ranked were as follows; Protecting the body from a harmful state, invigorating the body by the argumentation of heat and avoidance of cold, handling with whole heart, and keeping clean, resting without working, eating well. 2. The average score for the degree of performance of traditional postpartal care (Sanhujori) for postpartal women was $2.81{\pm}.31$, and they performed that it was important, too. The methods which were ranked were as follows; Protecting the body from a harmful state, invigorating the body by the augumentation of heat and avoidance of cold, eating well, handling with whole heart, and keeping clean, resting without working. 3. There were significant differences statistically (paired-t=-8.39, p=.000) of the degree of recognition & performance of traditional postpartal care(Sanhujori) for the postpartal women. The degree of recognition was higher than the degree of performance. So, the recognition of traditional postpartal care (Sanhujori) was higher than the performance of it. 4. There were no statistical differences of the degree of recognition & performance of traditional postpartal care(Sanhujori) among the postpartal women's age, religion, job, educational background, delivery frequency, delivery method or the sex of baby. So, the Characteristics of the respondents were not influenced as far as the degree of recognition & performance of traditional postpartal care(Sanhujori). 5. There were significant differences statistically of the degree of performance of traditional postpartal care(Sanhujori) among the 5 postpartum care centers except 1 postpartum care center(p<.01). So, the recognition of traditional postpartal care(Sanhujori) was higher than the performance of traditional postpartal care(Sanhujori) in the 5 postpartum care centers. But there was performed as good as recognition in only 1 postpartum care center.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1908-1909
• /
• 2011

Hand gesture recognition has been widely used in virtual reality and HCI (Human-Computer-Interaction) system, which is challenging and interesting subject in the vision based area. The existing approaches for vision-driven interactive user interfaces resort to technologies such as head tracking, face and facial expression recognition, eye tracking and gesture recognition. The purpose of this paper is to combine the finite state machine (FSM) and the gesture recognition method, in other to control Windows Media Player, such as: play/pause, next, pervious, and volume up/down.

• Journal of information and communication convergence engineering
• /
• v.9 no.5
• /
• pp.607-609
• /
• 2011

In this work, we present solutions of two problems. First, the representation of pattern recognition problem in the standard $T_{nml}$ table of the algorithm estimate calculation was considered. Second, the problem of finding the model, consisting of the optimal parameters of an algorithm was considered. Such procedure is carried out by the selection optimal values of the parameters of extreme algorithms. This serves to reduce the number of calculations in the algorithms of estimate calculation and to increase the quality of recognition process. The algorithmic data base of the developed system was based on mathematical apparatus of pattern recognition.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.4
• /
• pp.1657-1662
• /
• 2017

Facial expression recognition systems using video devices have emerged as an important component of natural human-machine interfaces which contribute to various practical applications such as security systems, behavioral science and clinical practices. In this work, we present a new method to analyze, represent and recognize human facial expressions using a sequence of facial images. Under our proposed facial expression recognition framework, the overall procedure includes: accurate face detection to remove background and noise effects from the raw image sequences and align each image using vertex mask generation. Furthermore, these features are reduced by principal component analysis. Finally, these augmented features are trained and tested using Hidden Markov Model (HMM). The experimental evaluation demonstrated the proposed approach over two public datasets such as Cohn-Kanade and AT&T datasets of facial expression videos that achieved expression recognition results as 96.75% and 96.92%. Besides, the recognition results show the superiority of the proposed approach over the state of the art methods.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.3
• /
• pp.197-201
• /
• 2006

In the implementation of a smart home, activity recognition technology using simple sensors is very important. In this paper, we propose a new activity recognition method based on Bayesian network (BN). The structure of the BN is learned by K2 algorithm and is composed of sensor nodes, activity nodes and time node whose state is quantized with reasonable interval. In the proposed method, the BN has less complexity and provides better activity recognition rate than the previous method.

• Phonetics and Speech Sciences
• /
• v.9 no.2
• /
• pp.95-101
• /
• 2017

This paper proposes a new method to train Deep Neural Network (DNN)-based acoustic models for speech recognition of native and foreign speakers. The proposed method consists of determining multi-set state clusters with various acoustic properties, training a DNN-based acoustic model, and recognizing speech based on the model. In the proposed method, hidden nodes of DNN are shared, but output nodes are separated to accommodate different acoustic properties for native and foreign speech. In an English speech recognition task for speakers of Korean and English respectively, the proposed method is shown to slightly improve recognition accuracy compared to the conventional multi-condition training method.

• Journal of Korea Multimedia Society
• /
• v.24 no.3
• /
• pp.423-430
• /
• 2021

Aiming Graph convolutional networks (GCNs) have achieved outstanding performances on skeleton-based action recognition. However, several problems remain in existing GCN-based methods, and the problem of low recognition rate caused by single input data information has not been effectively solved. In this article, we propose a Dual-stream fusion method that combines video data and skeleton data. The two networks respectively identify skeleton data and video data and fuse the probabilities of the two outputs to achieve the effect of information fusion. Experiments on two large dataset, Kinetics and NTU-RGBC+D Human Action Dataset, illustrate that our proposed method achieves state-of-the-art. Compared with the traditional method, the recognition accuracy is improved better.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.4
• /
• pp.229-237
• /
• 2022

Motion recognition is very useful for implementing an intuitive HMI (Human-Machine Interface). In particular, hands are the body parts that can move most precisely with relatively small portion of energy. Thus hand motion has been used as an efficient communication interface with other persons or machines. In this paper, we design and implement a light-weight ANN (Artificial Neural Network)-based hand motion recognition using a state-of-the-art flex sensor. The proposed design consists of data collection from a wearable flex sensor, preprocessing filters, and a light-weight NN (Neural Network) classifier. For verifying the performance and functionality of the proposed design, we implement it on a low-end embedded device. Finally, our experiments and prototype implementation demonstrate that the accuracy of the proposed hand motion recognition achieves up to 98.7%.