• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.783-792
• /
• 2022

Versatile Video Coding (VVC) is the latest video coding standard, which had been developed by the Joint Video Experts Team (JVET) of ITU-T Video Coding Experts Group (VCEG) and ISO/IEC Moving Picture Experts Group (MPEG) in 2020. Although VVC can provide powerful coding performance, it requires tremendous computational complexity to determine the optimal block structures during the encoding process. In this paper, we propose a fast ternary tree decision method using two neural networks with 7 nodes as input vector based on the multi-layer perceptron structure, names STH-NN and STV-NN. As a training result of neural network, the STH-NN and STV-NN achieved accuracies of 85% and 91%, respectively. Experimental results show that the proposed method reduces the encoding complexity up to 25% with unnoticeable coding loss compared to the VVC test model (VTM).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.5
• /
• pp.1431-1445
• /
• 2022

We construct a deep neural network model named ECGResNet. This model can diagnosis diseases based on 12-lead ECG data of eight common cardiovascular diseases with a high accuracy. We chose the 16 Blocks of ResNet50 as the main body of the model and added the Squeeze-and-Excitation module to learn the data information between channels adaptively. We modified the first convolutional layer of ResNet50 which has a convolutional kernel of 7 to a superposition of convolutional kernels of 8 and 16 as our feature extraction method. This way allows the model to focus on the overall trend of the ECG signal while also noticing subtle changes. The model further improves the accuracy of cardiovascular and cerebrovascular disease classification by using a fully connected layer that integrates factors such as gender and age. The ECGResNet model adds Dropout layers to both the residual block and SE module of ResNet50, further avoiding the phenomenon of model overfitting. The model was eventually trained using a five-fold cross-validation and Flooding training method, with an accuracy of 95% on the test set and an F1-score of 0.841.We design a new deep neural network, innovate a multi-scale feature extraction method, and apply the SE module to extract features of ECG data.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1911-1923
• /
• 2022

The availability of high-resolution satellite images provides precise information without direct observation of the research target. Korea Multi-Purpose Satellite (KOMPSAT), also known as the Arirang satellite, has been developed and utilized for earth observation. The machine learning model was continuously proven as a good classifier in classifying remotely sensed images. This study aimed to compare the performance of the support vector machine (SVM) model in classifying the land cover of the Delaware River port area on high and medium-resolution images. Three optical images, which are KOMPSAT-2, KOMPSAT-3A, and Sentinel-2B, were classified into six land cover classes, including water, road, vegetation, building, vacant, and shadow. The KOMPSAT images are provided by Korea Aerospace Research Institute (KARI), and the Sentinel-2B image was provided by the European Space Agency (ESA). The training samples were manually digitized for each land cover class and considered the reference image. The predicted images were compared to the actual data to obtain the accuracy assessment using a confusion matrix analysis. In addition, the time-consuming training and classifying were recorded to evaluate the model performance. The results showed that the KOMPSAT-3A image has the highest overall accuracy and followed by KOMPSAT-2 and Sentinel-2B results. On the contrary, the model took a long time to classify the higher-resolution image compared to the lower resolution. For that reason, we can conclude that the SVM model performed better in the higher resolution image with the consequence of the longer time-consuming training and classifying data. Thus, this finding might provide consideration for related researchers when selecting satellite imagery for effective and accurate image classification.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.10
• /
• pp.503-510
• /
• 2016

FsGr model, which has been proposed recently, is an approach of accelerometer-based gesture recognition by applying DTW algorithm in two steps, which improved recognition success rate. In FsGr model, sets of similar gestures will be produced through training phase, in order to define the notion of a set of similar gestures. At the 1st attempt of gesture recognition, if the result turns out to belong to a set of similar gestures, it makes the 2nd recognition attempt to feature-strengthened parts extracted from the set of similar gestures. However, since a same gesture show drastically different characteristics according to physical traits such as body size, age, and sex, FsGr model may not be good enough to apply to multi-user environments. In this paper, we propose FsGrM model that extends FsGr model for multi-user environment and present a program which controls channel and volume of smart TV using FsGrM model.

• International Journal of Computer Science & Network Security
• /
• v.23 no.12
• /
• pp.101-106
• /
• 2023

Chronic illnesses are among the most common serious problems affecting human health. Early diagnosis of chronic diseases can assist to avoid or mitigate their consequences, potentially decreasing mortality rates. Using machine learning algorithms to identify risk factors is an exciting strategy. The issue with existing feature selection approaches is that each method provides a distinct set of properties that affect model correctness, and present methods cannot perform well on huge multidimensional datasets. We would like to introduce a novel model that contains a feature selection approach that selects optimal characteristics from big multidimensional data sets to provide reliable predictions of chronic illnesses without sacrificing data uniqueness.[1] To ensure the success of our proposed model, we employed balanced classes by employing hybrid balanced class sampling methods on the original dataset, as well as methods for data pre-processing and data transformation, to provide credible data for the training model. We ran and assessed our model on datasets with binary and multivalued classifications. We have used multiple datasets (Parkinson, arrythmia, breast cancer, kidney, diabetes). Suitable features are selected by using the Hybrid feature model consists of Lassocv, decision tree, random forest, gradient boosting,Adaboost, stochastic gradient descent and done voting of attributes which are common output from these methods.Accuracy of original dataset before applying framework is recorded and evaluated against reduced data set of attributes accuracy. The results are shown separately to provide comparisons. Based on the result analysis, we can conclude that our proposed model produced the highest accuracy on multi valued class datasets than on binary class attributes.[1]

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.536-543
• /
• 2023

Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

• The Journal of Society for e-Business Studies
• /
• v.26 no.1
• /
• pp.29-41
• /
• 2021

Recently, there are increasing attempts to utilize deep learning methodology in the fashion industry. Accordingly, research dealing with various fashion-related problems have been proposed, and superior performances have been achieved. However, the studies for fashion style classification have not reflected the characteristics of the fashion style that one outfit can include multiple styles simultaneously. Therefore, we aim to solve the multi-label classification problem by utilizing the dependencies between the styles. A multi-label recognition model based on a graph convolution network is applied to detect and explore fashion styles' dependencies. Furthermore, we accelerate model training and improve the model's performance through transfer learning. The proposed model was verified by a dataset collected from social network services and outperformed baselines.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.45-55
• /
• 2022

Recently, large-scale language models (LPLM) have been shown state-of-the-art performances in various tasks of natural language processing including intent classification. However, fine-tuning LPLM requires much computational cost for training and inference which is not appropriate for dialog system. In this paper, we propose compressed intent classification model for multi-agent in low-resource like CPU. Our method consists of two stages. First, we trained sentence encoder from LPLM then compressed it through knowledge distillation. Second, we trained agent-specific adapter for intent classification. The results of three intent classification datasets show that our method achieved 98% of the accuracy of LPLM with only 21% size of it.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.6-10
• /
• 2023

High-potential data can be predicted and stored in the cache to prevent cache misses, thus reducing the processor's request and wait times. As a result, the processor can work non-stop, hiding memory latency. By utilizing the temporal/spatial locality of memory access, the prefetcher introduced to improve the performance of these computers predicts the following memory address will be accessed. We propose a prefetcher that applies the GRU model, which is advantageous for handling time series data. Display the currently accessed address in binary and use it as training data to train the Gated Recurrent Unit model based on the difference (delta) between consecutive memory accesses. Finally, using a GRU model with learned memory access patterns, the proposed data prefetcher predicts the memory address to be accessed next. We have compared the model with the multi-layer perceptron, but our prefetcher showed better results than the Multi-Layer Perceptron.

• Structural Engineering and Mechanics
• /
• v.71 no.4
• /
• pp.329-339
• /
• 2019

Assessment of failure probability, especially for a complex structure, requires a considerable number of calls to the numerical model. Reliability methods have been developed to decrease the computational time. In this approach, the original numerical model is replaced by a surrogate model which is usually explicit and much faster to evaluate. The current paper proposed an efficient reliability method based on Monte Carlo simulation (MCS) and multi-gene genetic programming (MGGP) as a robust variant of genetic programming (GP). GP has been applied in different fields; however, its application to structural reliability has not been tested. The current study investigated the performance of MGGP as a surrogate model in structural reliability problems and compares it with other surrogate models. An adaptive Metropolis algorithm is utilized to obtain the training data with which to build the MGGP model. The failure probability is estimated by combining MCS and MGGP. The efficiency and accuracy of the proposed method were investigated with the help of five numerical examples.