• Database Research
• /
• v.34 no.3
• /
• pp.137-147
• /
• 2018

A key to a well-performing human activity recognition (HAR) system through machine learning technique is the availability of a substantial amount of labeled data. Collecting sufficient labeled data is an expensive and time-consuming task. To build a HAR system in a new environment (i.e., the target domain) with very limited labeled data, it is unfavorable to naively exploit the data or trained classifier model from the existing environment (i.e., the source domain) as it is due to the domain difference. While traditional machine learning approaches are unable to address such distribution mismatch, transfer learning approach leverages the utilization of knowledge from existing well-established source domains that help to build an accurate classifier in the target domain. In this work, we propose a transfer learning approach to create an accurate HAR classifier with very limited data through the multitask neural network. The classifier loss function minimization for source and target domain are treated as two different tasks. The knowledge transfer is performed by simultaneously minimizing the loss function of both tasks using a single neural network model. Furthermore, we utilize the unlabeled data in an unsupervised manner to help the model training. The experiment result shows that the proposed work consistently outperforms existing approaches.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06a
• /
• pp.90-91
• /
• 2008

• Korean Journal of Veterinary Research
• /
• v.37 no.3
• /
• pp.547-554
• /
• 1997

Galectin-3 is known as an animal ${\beta}$-galactoside-binding lectin charicterized with S-type carbohydrate recognition domain. It plays a role in growth, adherence and movement of cells. It is, also, related to the cell transformation and metastasis of tumor cells. In this study, we have expressed and purified recombinant human galectin-3 (rHgalectin-3) using E coli system and asialofetuin affinity chromatography for the future development of monoclonal antibody to Hgalectin-3, which is suggested as the tumor marker for the gastric and thyroid gland cancers. Expressed protein was confirmed as the Hgalectin-3 by immunoblot with cross-reactive murine monoclonal antibody. Lectin activity and specificity of purified protein were, also, confirmed by the competitive inhibition with galectin-3 specific carbohydrate, lactose. Like physiological galectin-3, lectin activity of the molecule was not changed in nonreduced condition. Dimer formation, furthermore, was observed at high concentration of the protein even in the reduced condition, which is well known in physiological galectin-3. These results showed purified rHgalectin-3 has the same activity and molecular nature compared to the physiological galectin-3.

• Speech Sciences
• /
• v.10 no.1
• /
• pp.95-105
• /
• 2003

This paper describes a new parameter for voice activity detection which serves as a front-end part for automatic speech recognition systems. The new parameter called run-ratio is derived from the runs test statistic which is used in the statistical test for randomness of a given sequence. The run-ratio parameter has the property that the values of the parameter for the random sequence are about 1. To apply the run-ratio parameter into the voice activity detection method, it is assumed that the samples of an inputted audio signal should be converted to binary sequences of positive and negative values. Then, the silence region in the audio signal can be regarded as random sequences so that their values of the run-ratio would be about 1. The run-ratio for the voiced region has far lower values than 1 and for fricative sounds higher values than 1. Therefore, the parameter can discriminate speech signals from the background sounds by using the newly derived run-ratio parameter. The proposed voice activity detector outperformed the conventional energy-based detector in the sense of error mean and variance, small deviation from true speech boundaries, and low chance of missing real utterances

• Smart Media Journal
• /
• v.1 no.2
• /
• pp.12-22
• /
• 2012

Today we are witnessing an increasingly widespread use of cameras in our lives for video surveillance, robot vision, and mobile phones. This has led to a renewed interest in computer vision in general and an on-going boom in human activity recognition in particular. Although not particularly fancy per se, human gait is inarguably the most common and frequent action. Early on this decade there has been a passing interest in human gait recognition, but it soon declined before we came up with a systematic analysis and understanding of walking motion. This paper presents a set of DBN-based models for the analysis of human gait in sequence of increasing complexity and modeling power. The discussion centers around HMM-based statistical methods capable of modeling the variability and incompleteness of input video signals. Finally a novel idea of extending the discrete state Markov chain with a continuous density function is proposed in order to better characterize the gait direction. The proposed modeling framework allows us to recognize pedestrian up to 91.67% and to elegantly decode out two independent gait components of direction and posture through a sequence of experiments.

• Phonetics and Speech Sciences
• /
• v.2 no.4
• /
• pp.137-143
• /
• 2010

In this paper, a speech endpoint detection algorithm is proposed. The proposed algorithm is a kind of state transition-based ones for speech detection. To reject short-duration acoustic pulses which can be considered noises, it utilizes duration information of all detected pulses. For the optimization of parameters related with pulse lengths and energy threshold to detect speech intervals, an exhaustive search scheme is adopted while speech recognition rates are used as its performance index. Experimental results show that the proposed algorithm outperforms the baseline state-based endpoint detection algorithm. At 5 dB input SNR for the beamforming input, the word recognition accuracies of its outputs were 78.5% for human voice noises and 81.1% for music noises.

• Journal of Digital Contents Society
• /
• v.18 no.3
• /
• pp.559-565
• /
• 2017

A lot of studies have been conducted to recognize the motion state or behavior of the user using the acceleration sensor built in the smartphone. In this paper, we applied the neural networks to the 3-axis acceleration information of smartphone to study human behavior. There are performance issues in applying time series data to neural networks. We proposed a multi-temporal neural networks which have trained three neural networks with different time windows for feature extraction and uses the output of these neural networks as input to the new neural network. The proposed method showed better performance than other methods like SVM, AdaBoot and IBk classifier for real acceleration data.

• Journal of Advanced Navigation Technology
• /
• v.25 no.5
• /
• pp.426-432
• /
• 2021

Continuous wave (CW) Doppler radar has the advantage of being able to solve the privacy problem unlike camera and obtains signals in a non-contact manner. Therefore, this paper proposes a human activity recognition (HAR) system using CW Doppler radar, and presents the hardware design and implementation results for acceleration. CW Doppler radar measures signals for continuous operation of human. In order to obtain a single motion spectrogram from continuous signals, an algorithm for counting the number of movements is proposed. In addition, in order to minimize the computational complexity and memory usage, binarized neural network (BNN) was used to classify human motions, and the accuracy of 94% was shown. To accelerate the complex operations of BNN, the FPGA-based BNN accelerator was designed and implemented. The proposed HAR system was implemented using 7,673 logics, 12,105 registers, 10,211 combinational ALUTs, and 18.7 Kb of block memory. As a result of performance evaluation, the operation speed was improved by 99.97% compared to the software implementation.

• International Journal of Advanced Culture Technology
• /
• v.4 no.1
• /
• pp.28-30
• /
• 2016

Due to the high increment in usage and built-in advanced technology of smartphones, human activity recognition relying on smartphone sensor data has become a focused research area. In order to reduce noise of collected data, most of previous studies assume that smartphones are fixed at certain positions. This strategy is impractical for real life applications. To overcome this issue, we here investigate a framework that allows detecting the status of a traveller as idle or moving regardless the position and the direction of smartphones. The application of our work is to estimate the total energy consumption of a traveller during a trip. A number of experiments have been carried out to show the effectiveness of our framework when travellers are not only walking but also using primitive vehicles like motorbikes.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.245-250
• /
• 2008

최근 동작 및 행동 인식에 대한 연구가 활발하다. 특히, 센서가 소형화되고 저렴해지면서 그 활용을 위한 관심이 증가하고 있다. 기존의 많은 행동 인식 연구에서 사용되어 온 정적 분류 기술 기반 동작 인식 방법은 연속적인 데이터 분류 기술에 비해 유연성 및 활용성이 부족할 수 있다. 본 논문에서는 연속적인 데이터의 패턴 분류 및 인식에 효과적인 확률적 추론 기법인 은닉 마르코프 모델(Hidden Markov Model)과 사전 지식 없이도 자동 학습이 가능하며 의미 깊은 궤적 패턴을 클러스터링하고 효과적인 양자화가 가능한 자기구성지도(Self Organizing Map)를 이용한 동작 인식 기술을 소개한다. 또한, 그 유용성을 입증하기 위해 실제 가속도 센서를 이용하여 다양한 동작에 대한 데이터를 수집하고 분류 성능을 분석 및 평가한다. 실험에서는 실제 가속도 센서를 통해 수집된 숫자를 그리는 동작의 성능 평가 결과를 보이고, 행동 인식기 별 성능과 전체 인식기별 성능을 비교한다.