• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.7
• /
• pp.3599-3619
• /
• 2019

In human activity recognition system both static and motion information play crucial role for efficient and competitive results. Most of the existing methods are insufficient to extract video features and unable to investigate the level of contribution of both (Static and Motion) components. Our work highlights this problem and proposes Static-Motion fused features descriptor (SMFD), which intelligently leverages both static and motion features in the form of descriptor. First, static features are learned by two-stream 3D convolutional neural network. Second, trajectories are extracted by tracking key points and only those trajectories have been selected which are located in central region of the original video frame in order to to reduce irrelevant background trajectories as well computational complexity. Then, shape and motion descriptors are obtained along with key points by using SIFT flow. Next, cholesky transformation is introduced to fuse static and motion feature vectors to guarantee the equal contribution of all descriptors. Finally, Long Short-Term Memory (LSTM) network is utilized to discover long-term temporal dependencies and final prediction. To confirm the effectiveness of the proposed approach, extensive experiments have been conducted on three well-known datasets i.e. UCF101, HMDB51 and YouTube. Findings shows that the resulting recognition system is on par with state-of-the-art methods.

• Molecules and Cells
• /
• v.42 no.8
• /
• pp.589-596
• /
• 2019

${\beta}Pix$ is a guanine nucleotide exchange factor for the Rho family small GTPases, Rac1 and Cdc42. It is known to regulate focal adhesion dynamics and cell migration. However, the in vivo role of ${\beta}Pix$ is currently not well understood. Here, we report the production and characterization of ${\beta}Pix$-KO mice. Loss of ${\beta}Pix$ results in embryonic lethality accompanied by abnormal developmental features, such as incomplete neural tube closure, impaired axial rotation, and failure of allantois-chorion fusion. We also generated ${\beta}Pix$-KO mouse embryonic fibroblasts (MEFs) to examine ${\beta}Pix$ function in mouse fibroblasts. ${\beta}Pix$-KO MEFs exhibit decreased Rac1 activity, and defects in cell spreading and platelet-derived growth factor (PDGF)-induced ruffle formation and chemotaxis. The average size of focal adhesions is increased in ${\beta}Pix$-KO MEFs. Interestingly, ${\beta}Pix$-KO MEFs showed increased motility in random migration and rapid wound healing with elevated levels of MLC2 phosphorylation. Taken together, our data demonstrate that ${\beta}Pix$ plays essential roles in early embryonic development, cell spreading, and cell migration in fibroblasts.

• Journal of Distribution Science
• /
• v.20 no.11
• /
• pp.121-129
• /
• 2022

Purpose: Artificial intelligence is widely used, particularly in the popular neural network theory called Deep learning. The improvement of computing speed and capability expedited the progress of Deep learning applications. The application of Deep learning in education has various effects and possibilities in creating and managing educational content and services that can replace human cognitive activity. Among Deep learning, Deep fake technology is used to combine and synchronize human faces with voices. This paper will show how to develop e-Learning content videos using those technologies and open-source tools. Research design, data, and methodology: This paper proposes 4 step development process, which is presented step by step on the Google Collab environment with source codes. This technology can produce various video styles. The advantage of this technology is that the characters of the video can be extended to any historical figures, celebrities, or even movie heroes producing immersive videos. Results: Prototypes for each case are also designed, developed, presented, and shared on YouTube for each specific case development. Conclusions: The method and process of creating e-learning video contents from the image, video, and audio files using Deep fake open-source technology was successfully implemented.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.9-16
• /
• 2023

Speech can actively elicit feelings and attitudes by using words. It is important for researchers to identify the emotional content contained in speech signals as well as the sort of emotion that resulted from the speech that was made. In this study, we studied the emotion recognition system using a database in Arabic, especially in the Saudi dialect, the database is from a YouTube channel called Telfaz11, The four emotions that were examined were anger, happiness, sadness, and neutral. In our experiments, we extracted features from audio signals, such as Mel Frequency Cepstral Coefficient (MFCC) and Zero-Crossing Rate (ZCR), then we classified emotions using many classification algorithms such as machine learning algorithms (Support Vector Machine (SVM) and K-Nearest Neighbor (KNN)) and deep learning algorithms such as (Convolution Neural Network (CNN) and Long Short-Term Memory (LSTM)). Our Experiments showed that the MFCC feature extraction method and CNN model obtained the best accuracy result with 95%, proving the effectiveness of this classification system in recognizing Arabic spoken emotions.

• International Journal of Advanced Culture Technology
• /
• v.12 no.3
• /
• pp.148-158
• /
• 2024

We investigate Autism Spectrum Disorder (ASD), which is typified by deficits in social interaction, repetitive behaviors, limited vocabulary, and cognitive delays. Traditional diagnostic methodologies, reliant on expert evaluations, frequently result in deferred detection and intervention, particularly in South Korea, where there is a dearth of qualified professionals and limited public awareness. In this study, we employ advanced deep learning algorithms to enhance early ASD screening through automated video analysis. Utilizing architectures such as Convolutional Long Short-Term Memory (ConvLSTM), Long-term Recurrent Convolutional Network (LRCN), and Convolutional Neural Networks with Gated Recurrent Units (CNN+GRU), we analyze video data from platforms like YouTube and TikTok to identify stereotypic behaviors (arm flapping, head banging, spinning). Our results indicate that the LRCN model exhibited superior performance with 79.61% accuracy on the augmented platform video dataset and 79.37% on the original SSBD dataset. The ConvLSTM and CNN+GRU models also achieved higher accuracy than the original SSBD dataset. Through this research, we underscore AI's potential in early ASD detection by automating the identification of stereotypic behaviors, thereby enabling timely intervention. We also emphasize the significance of utilizing expanded datasets from social media platform videos in augmenting model accuracy and robustness, thus paving the way for more accessible diagnostic methods.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.21 no.5
• /
• pp.510-517
• /
• 2001

This paper describes an initial study for the application of eddy current pattern recognition approaches to more realistic flaw characterization in steam generator tubes. For this purpose, finite-element model-based theoretical eddy current testing (ECT) signals are simulated from 5 types of OD flaws with the variation in flaw size parameters and testing frequency. In addition, three kinds of software are developed for the convenience in the application of steps in pattern recognition approaches such as feature extraction feature selection and classification by probabilistic neural networks (PNNs). The cross point of the ECT signals simulated from flaws with non-symmetric cross-sections shows the deviation from the origin of the impedance plane. New features taking advantages of this phenomenon are added to complete the feature set with a total of 18 features. Then, classification with PNNs are performed based on this feature set. The PNN classifiers show high performance for the identification of symmetry in the cross-section of a flaw. However, they show very limited success in the interrogation of the sharpness of flaw tips.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.40 no.1
• /
• pp.3-10
• /
• 2014

Objectives: Although nerve growth factor (NGF) could promote the functional regeneration of an injured peripheral nerve, it is very difficult for NGF to sustain the therapeutic dose in the defect due to its short half-life. In this study, we loaded the NGF-bound heparin-conjugated fibrin (HCF) gel in the NGF-delivering implants and analyzed the time-dependent release of NGF and its bioactivity to evaluate the clinical effectiveness. Materials and Methods: NGF solution was made of 1.0 mg of NGF and 1.0 mL of phosphate buffered saline (PBS). Experimental group A consisted of three implants, in which $0.25{\mu}L$ of NGF solution, $0.75{\mu}L$ of HCF, $1.0{\mu}L$ of fibrinogen and $2.0{\mu}L$ of thrombin was injected via apex hole with micropipette and gelated, were put into the centrifuge tube. Three implants of experimental group B were prepared with the mixture of $0.5{\mu}L$ of NGF solution, $0.5{\mu}L$ HCF, $1.0{\mu}L$ of fibrinogen and $2.0{\mu}L$ of thrombin. These six centrifuge tubes were filled with 1.0 mL of PBS and stirred in the water-filled beaker at 50 rpm. At 1, 3, 5, 7, 10, and 14 days, 1.0 mL of solution in each tubes was collected and preserved at $-20^{\circ}C$ with adding same amount of fresh PBS. Enzyme-linked immunosorbent assay (ELISA) was done to determine in vitro release profile of NGF and its bioactivity was evaluated with neural differentiation of pheochromocytoma (PC12) cells. Results: The average concentration of released NGF in the group A and B increased for the first 5 days and then gradually decreased. Almost all of NGF was released during 10 days. Released NGF from two groups could promote neural differentiation and neurite outgrowth of PC12 cells and these bioactivity was maintained over 14 days. Conclusion: Controlled release system using NGF-HCF gel via NGF-delivering implant could be an another vehicle of delivering NGF to promote the nerve regeneration of dental implant related nerve damage.

• Toxicological Research
• /
• v.14 no.3
• /
• pp.357-363
• /
• 1998

The teratogenic potential of the anticonvulsant drug phenytoin (PHT) has been well documented both in the human and in the experimental animals. However there are few reports on the effects of PHT on embryonic development in rats in vitro. The present study was performed to evaluate the teratogenic effects of PHT using whole-embryo culture system in rats. Sprague-Dawley rat embryos were explanted on gestational day (GD) 9.5 and cultured for 48 hrs in the immediately centrifuged and heat-inactivated rat serum containing 0,25,50, or $100{\mu}g$ PHT/mL. At the end of culture period the embryos were scored for morphological development according to the procedure of Van Maele-Fabry, and their total protein contents were determined. At 100 ${\mu}$g/mL of culture medium. PHT caused significant reduction in developmental score and protein content of embryos and a high incidence morphological abnormalities (100%). Characteristic malformations included altered yolk and embryonic circulation, craniofacial hypoplasia, neural tube schisis, branchial arch defects, abnormal ratation, and limb bud hypoplasia, among others. There were no adverse effects on embryonic growth and development at concentrations of 25 and 50 ${\mu}$g /mL of culture medium. The results indicated that the dysmorphogenic effect of PHT on cultured embryos is due to a direct interference with embryonic development.

• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.11 no.2
• /
• pp.178-184
• /
• 2000

Laryngeal evoked EMG is the objective and quantitative method to measure the innervation of laryngeal muscle. If there is a mobility disorder of vocal cords, the cause and location of neural lesion co be understood by the laryngeal evoked EMG and if there is a vocal cord paralysis, the degree of recovery and the policy of treatment can be determined by it. Recently, the studies of reinnervation after recurrent laryngeal nerve injury have been actively carried out. Laryngeal evoked EMC is useful to these studies. The aim of study is to know whether noninvasive methods for stimulating the recurrent laryngeal nerve and for recording of compound action potential(CAP) using surface electrode are as useful as the invasive method using needle electrode. We obtained EMG of laryngeal muscle by various stimulating and recording methods : 1) Direct nerve stimulation by placing nerve cuff electrode made out of silastic tube and platinum wire and recording by insertion of hook wire electrode into posterior cricoarytenoid(PCA) and thyroarytenoid(TA) muscles, respectively. 2) Recording of compound action potential by surface electrode after stimulation of recurrent laryngeal nerve by the insertion of 27 gauge of needle electrode. 3) Recording of compound action potential by surface electrode after stimulating the recurrent laryngeal nerve by transcutaneous blunt rod electrode at tracheoesophageal groove. The amplitude, duration and latency of the CAP evoked by recurrent laryngeal nerve stimulation were compared among the three groups. The amplitude of CAP was smallest in the group recorded from posterior cricoarytenoid and hyroarytenoid muscle, and that recorded by surface electrode after stimulation by needle electrode was largest. The difference in amplitude between the group by hook wire recording and the two groups by surface electrode recording was significant statistically. There is no significant difference in duration and latency among three groups. Since the waveform of CAP from all three methods has similar duration, latency, we concluded that noninvasive method is a useful as invasive methods.

• The Journal of Korean Physical Therapy
• /
• v.13 no.2
• /
• pp.433-444
• /
• 2001

This review highlights the ontogenesis and the differentiation of motor neuron in spinal cord, and introduce the survival motor neuron(SMN) which is associated with growth and survival of motor neurons. The differentiation of floor plate cells and motor neurons in the vertebrate neural tube appears to be induced by signals from the notochord. This signal is Sonic hedgehog(Shh). The early development of motor neurons involves the inductive action of Shh. The SMN gene is essential for embryonic viability. SMN mRNA is also expressed in virtually all cell types in spinal cord, including large motor neurons. The SMN protein is involved in RNA processing and during early embryonic development is necessary fer cell survival. Two SMN genes are present in 5q 13 in humans: the telomeric gene(SMNt), which is the SMA-determining gene, and the centromeric analog gene(SMNc). The majority of transcripts from the SMNt gene are full length but, major transcripts of the SMNc gene have a high degrees of alternative splicing and tend to have little or no exon 7. The SMN is involved in the RNA processing(the biogenesis of snRNPs and pre-mRNA splicing), the anti-apoptotic effects, and regulating gene expression.