• Journal of Intelligence and Information Systems
• /
• v.16 no.2
• /
• pp.95-108
• /
• 2010

In this paper, we present a hybrid neural network model for dynamic hand gesture recognition. The model consists of two modules, feature extraction module and pattern classification module. We first propose a modified CNN(convolutional Neural Network) a pattern recognition model for the feature extraction module. Then we introduce a weighted fuzzy min-max(WFMM) neural network for the pattern classification module. The data representation proposed in this research is a spatiotemporal template which is based on the motion information of the target object. To minimize the influence caused by the spatial and temporal variation of the feature points, we extend the receptive field of the CNN model to a three-dimensional structure. We discuss the learning capability of the WFMM neural networks in which the weight concept is added to represent the frequency factor in training pattern set. The model can overcome the performance degradation which may be caused by the hyperbox contraction process of conventional FMM neural networks. From the experimental results of human action recognition and dynamic hand gesture recognition for remote-control electric home appliances, the validity of the proposed models is discussed.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.8
• /
• pp.31-37
• /
• 2021

In this paper, we propose a method to effectively search by automatically classifying scenes in which specific images such as pitching or swing appear in baseball game images using a CNN(Convolution Neural Network) model. In addition, we propose a video scene search system that links the classification results of specific motions and game records. In order to test the efficiency of the proposed system, an experiment was conducted to classify the Korean professional baseball game videos from 2018 to 2019 by specific scenes. In an experiment to classify pitching scenes in baseball game images, the accuracy was about 90% for each game. And in the video scene search experiment linking the game record by extracting the scoreboard included in the game video, the accuracy was about 80% for each game. It is expected that the results of this study can be used effectively to establish strategies for improving performance by systematically analyzing past game images in Korean professional baseball games.

• Design Convergence Study
• /
• v.14 no.2
• /
• pp.67-87
• /
• 2015

With the development of cultural industries and technology, digital video marketing has grown immensely and 'Book Trailer' are now one of the main methods of promoting a title in publishing industry. The demands are growing, but the development lacks basic theoretical foundation and systematic structure. At this crucial moment, basic theoretical research on Book Trailer is necessary to establish a solid foundation for the medium to expand. In this study, I have focused on defining the basic concepts of Book Trailers and analyzing the classification of Book Trailers. In sorting the type of Book Trailers, content and formal appearance were used as standards. As a result, I was able to classify the content genre into message type, conflict type, character type using the elements of story. The formal appearance genre was classified into still photography type, motion typography type, interview type, story movie type and animation type.

• Geophysics and Geophysical Exploration
• /
• v.10 no.3
• /
• pp.183-190
• /
• 2007

The site effects of seismic stations were evaluated by conducting a simultaneous inversion of the stochastic point-source ground-motion model (STGM model; Boore, 2003) parameters based on the accumulated dataset of horizontal shear-wave Fourier spectra. A model parameter $K_0$ and frequency-dependent site amplification function A(f) were used to express the site effects. Once after a H/V ratio of the Fourier spectra was used as an initial estimate of A(f) for the inversion, the final A(f) which is considered to be the result of combined effect of the crustal amplification and loca lsite effects was calculated by averaging the log residuals at the site from the inversion and adding the mean log residual to the H/V ratio. The seismic stations were classified into five classes according to $logA_{1-10}^{max}$(f), the maximum level of the site amplification function in the range of 1 Hz < f < 10 Hz, i.e., A: $logA_{1-10}^{max}$(f) < 0.2, B: 0.2 $\leq$ $logA_{1-10}^{max}$(f) < 0.4, C: 0.4 $\leq$ $logA_{1-10}^{max}$(f) < 0.6, D: 0.6 $\leq$ $logA_{1-10}^{max}$(f) < 0.8, E: 0.8 $\leq$ $logA_{1-10}^{max}$(f). Implication of the classified result was supported by observing a shift of the dominant frequency of average A(f) for each classified stations as the class changes. Change of site classes after moving seismic stations to a better site condition was successfully described by the result of the station classification. In addition, the observed PGA (Peak Ground Acceleration)-values for two recent moderate earthquakes were well classified according to the proposed station classes.

• Journal of the Korea Convergence Society
• /
• v.10 no.3
• /
• pp.1-6
• /
• 2019

In modern society, lifestyle and individuality are important, and personalized lifestyle and patterns are emerging. The number of people with articulation diseases is increasing due to wrong living habits. In addition, as the number of households increases, there is a case where emergency care is not received at the appropriate time. We need information that can be managed by ourselves through accurate analysis according to the individual's condition for health and disease management, and care appropriate to the emergency situation. It is effectively used for classification and prediction of data using CNN in deep learning. CNN differs in accuracy and processing time according to the data features. Therefore, it is necessary to improve processing speed and accuracy for real-time healthcare. In this paper, we propose motion monitoring using Mask R-CNN for articulation disease management. The proposed method uses Mask R-CNN which is superior in accuracy and processing time than CNN. After the user's motion is learned in the neural network, if the user's motion is different from the learned data, the control method can be fed back to the user, the emergency situation can be informed to the guardian, and appropriate methods can be taken according to the situation.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.107-119
• /
• 2023

The technology that recognizes a soldier's motion and movement status has recently attracted large attention as a combination of wearable technology and artificial intelligence, which is expected to upend the paradigm of troop management. The accuracy of state determination should be maintained at a high-end level to make sure of the expected vital functions both in a training situation; an evaluation and solution provision for each individual's motion, and in a combat situation; overall enhancement in managing troops. However, when input data is given as a timer series or sequence, existing feedforward networks would show overt limitations in maximizing classification performance. Since human behavior data (3-axis accelerations and 3-axis angular velocities) handled for military motion recognition requires the process of analyzing its time-dependent characteristics, this study proposes a high-performance data-driven classifier which utilizes the long-short term memory to identify the order dependence of acquired data, learning to classify eight representative military operations (Sitting, Standing, Walking, Running, Ascending, Descending, Low Crawl, and High Crawl). Since the accuracy is highly dependent on a network's learning conditions and variables, manual adjustment may neither be cost-effective nor guarantee optimal results during learning. Therefore, in this study, we optimized hyperparameters using Bayesian optimization for maximized generalization performance. As a result, the final architecture could reduce the error rate by 62.56% compared to the existing network with a similar number of learnable parameters, with the final accuracy of 98.39% for various military operations.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.977-985
• /
• 2020

Aiming at the problem of complex feature extraction and low accuracy in human action recognition, this paper proposed a network structure combining batch normalization algorithm with GoogLeNet network model. Applying Batch Normalization idea in the field of image classification to action recognition field, it improved the algorithm by normalizing the network input training sample by mini-batch. For convolutional network, RGB image was the spatial input, and stacked optical flows was the temporal input. Then, it fused the spatio-temporal networks to get the final action recognition result. It trained and evaluated the architecture on the standard video actions benchmarks of UCF101 and HMDB51, which achieved the accuracy of 93.42% and 67.82%. The results show that the improved convolutional neural network has a significant improvement in improving the recognition rate and has obvious advantages in action recognition.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.281-285
• /
• 2000

This paper proposes context-based coding methods for variable length coding of inter-frame DCT coefficients. The proposed methods classify run-level symbols depending on the preceding coefficients. No extra overhead needs to be transmitted, since the information of the previously transmitted coefficients is used for classification. Two entropy coding methods, arithmetic coding and Huffman coding, are used for the proposed context-based coding. For Huffman coding, there is no complexity increase from the current standards by using the existing inter/intra VLC tables. Experimental results show that the proposed methods give ～ 19% bits gain and ～ 0.8 dB PSNR improvement for adaptive inter/intra VLC table selection, and ～ 37% bits gain and ～ 2.7dB PSNR improvement for arithmetic coding over the current standards, MPEG-4 and H.263. Also, the proposed methods obtain larger gain for small quantizaton parameters and the sequences with fast and complex motion. Therefore, for high quality video coding, the proposed methods have more advantage.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.26-27
• /
• 2014

This paper presents an adaptive interpolation filtering scheme for the High Efficiency Video Coding (HEVC) standard. In regards to interpolation for motion estimation and compensation, the conventional HEVC employs 8-tap and 4-tap filters for luma and chroma samples, respectively. Coefficients in such filters are determined by discrete cosine transform (DCT). In the proposed scheme, boundary strength values are stored after the execution of the deblocking filter. For each block, the sum of boundary strength values is calculated to indicate whether its region is complex or simple. Consequently, based on the region classification, 12-tap and 8-tap interpolation filters are used for complex and simple regions, respectively. This process is applied to luma sample interpolation only. Simulation results show 1.8% average BD-rate reduction compared to the conventional method.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1851-1863
• /
• 2015

– In the paper, to handle the problem of traditional bag-of-features model ignoring the spatial relationship of local features in human action recognition, we proposed a Multiscale Spatial Position Coding under Locality Constraint method. Specifically, to describe this spatial relationship, we proposed a mixed feature combining motion feature and multi-spatial-scale configuration. To utilize temporal information between features, sub spatial-temporal-volumes are built. Next, the pooled features of sub-STVs are obtained via max-pooling method. In classification stage, the Locality-Constrained Group Sparse Representation is adopted to utilize the intrinsic group information of the sub-STV features. The experimental results on the KTH, Weizmann, and UCF sports datasets show that our action recognition system outperforms the classical local ST feature-based recognition systems published recently.