• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3583-3597
• /
• 2020

In this paper, a parallel deep learning model using a convolutional neural network and a dilated convolutional neural network is proposed to classify Alzheimer's disease with high accuracy in PET/CT images. The developed model consists of two pipelines, a conventional CNN pipeline, and a dilated convolution pipeline. An input image is sent through both pipelines, and at the end of both pipelines, extracted features are concatenated and used for classifying Alzheimer's disease. Complimentary abilities of both networks provide better overall accuracy than single conventional CNNs in the dataset. Moreover, instead of performing binary classification, the proposed model performs three-class classification being Alzheimer's disease, mild cognitive impairment, and normal control. Using the data received from Dong-a University, the model performs classification detecting Alzheimer's disease with an accuracy of up to 95.51%.

• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.4
• /
• pp.83-92
• /
• 2020

A recent study to determine the fall is focused on analyzing fall motions using a recurrent neural network (RNN), and uses a deep learning approach to get good results for detecting human poses in 2D from a mono color image. In this paper, we investigated the improved detection method to estimate the position of the head and shoulder key points and the acceleration of position change using the skeletal key points information extracted using PoseNet from the image obtained from the 2D RGB low-cost camera, and to increase the accuracy of the fall judgment. In particular, we propose a fall detection method based on the characteristics of post-fall posture in the fall motion analysis method and on the velocity of human body skeleton key points change as well as the ratio change of body bounding box's width and height. The public data set was used to extract human skeletal features and to train deep learning, GRU, and as a result of an experiment to find a feature extraction method that can achieve high classification accuracy, the proposed method showed a 99.8% success rate in detecting falls more effectively than the conventional primitive skeletal data use method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.538-561
• /
• 2020

Recommender Systems (RecSys) have a major role in e-commerce for recommending products, which they may like for every user and thus improve their business aspects. Although many types of RecSyss are there in the research field, the state of the art RecSys has focused on finding the user similarity based on sequence (e.g. purchase history, movie-watching history) analyzing and prediction techniques like Recurrent Neural Network in Deep learning. That is RecSys has considered as a sequence prediction problem. However, evaluation of similarities among the customers is challenging while considering temporal aspects, context and multi-component ratings of the item-records in the customer sequences. For addressing this issue, we are proposing a Deep Learning based model which learns customer similarity directly from the sequence to sequence similarity as well as item to item similarity by considering all features of the item, contexts, and rating components using Dynamic Temporal Warping(DTW) distance measure for dynamic temporal matching and 2D-GRU (Two Dimensional-Gated Recurrent Unit) architecture. This will overcome the limitation of non-linearity in the time dimension while measuring the similarity, and the find patterns more accurately and speedily from temporal and spatial contexts. Experiment on the real world movie data set LDOS-CoMoDa demonstrates the efficacy and promising utility of the proposed personalized RecSys architecture.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.5
• /
• pp.728-733
• /
• 2018

Deep learning has been actively studied for predicting protein secondary structure based only on the sequence information of the amino acids constituting the protein. In this paper, we compared the performances of the convolutional neural networks of various structures to predict the protein secondary structure. To investigate the optimal depth of the layer of neural network for the prediction of protein secondary structure, the performance according to the number of layers was investigated. We also applied the structure of GoogLeNet and ResNet which constitute building blocks of many image classification methods. These methods extract various features from input data, and smooth the gradient transmission in the learning process even using the deep layer. These architectures of convolutional neural networks were modified to suit the characteristics of protein data to improve performance.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.403-408
• /
• 2018

Generally, a large data set is required for learning of deep learning. However, since it is not easy to create large data sets, there are a lot of techniques that make small data sets larger through data expansion such as rotation, flipping, and filtering. However, these simple techniques have limitation on extendibility because they are difficult to escape from the features already possessed. In order to solve this problem, we propose a method to acquire new image data by using existing data. This is done by retrieving and acquiring similar images using existing image data as a query of the content-based image retrieval (CBIR). Finally, we compare the performance of the base model with the model using CBIR.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.11
• /
• pp.5546-5559
• /
• 2019

Facial action units (AUs) have recently drawn increased attention because they can be used to recognize facial expressions. A variety of methods have been designed for frontal-view AU detection, but few have been able to handle multi-view face images. In this paper we propose a method for multi-view facial AU detection using a fused multilayer, multi-task, and multi-label deep learning network. The network can complete two tasks: AU detection and facial view detection. AU detection is a multi-label problem and facial view detection is a single-label problem. A residual network and multilayer fusion are applied to obtain more representative features. Our method is effective and performs well. The F1 score on FERA 2017 is 13.1% higher than the baseline. The facial view recognition accuracy is 0.991. This shows that our multi-task, multi-label model could achieve good performance on the two tasks.

• Journal of Advanced Navigation Technology
• /
• v.24 no.2
• /
• pp.148-154
• /
• 2020

Wrinkles are one of the main features of skin aging. Conventional image processing-based wrinkle detection is difficult to effectively cope with various skin images. In particular, Wrinkle extraction performance is significantly decreased when the wrinkles are not strong and similar to the surrounding skin. In this paper, deep learning is applied to extract wrinkles from microscopic skin images. In general, the microscope image is equipped with a wide-angle lens, so the brightness at the boundary area of the image is dark. In this paper, to solve this problem, the brightness of the skin image is estimated and corrected. In addition, We apply the structure of semantic segmentation network suitable for wrinkle extraction. The proposed method obtained an accuracy of 99.6% in test experiments on skin images collected in our laboratory.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.2
• /
• pp.60.3-61
• /
• 2016

Galaxy morphology involves complex effects from both secular and non-secular evolution of galaxies. Although it is a final product of galaxy evolution, it gives a clue to the processes that the a galaxy has gone through. Galaxy clusters are the sites where the most massive galaxies are found, and thus the most dramatic merger histories are embedded. Our deep imaging program (${\mu}{\sim}28\;mag\;arcsec^{-2}$), KASI-Yonsei Deep Imaging Survey for Clusters (KYDISC), targets 14 Abell clusters at z = 0.016 - 0.14 using IMACS/Magellan telescope and MegaCam/CFHT to investigate cluster galaxies especially on low surface brightness features related to galaxy interactions. We visually classify galaxy morphology based on criteria related to secular or merger related evolution and find that the morphological mixture of galaxies varies considerably from cluster to cluster. Moreover it depends on the characteristics (e.g. cluster mass) of cluster itself which implies that environmental effects in cluster scale is also an important factor to the evolution of galaxies together with intrinsic (secular) and galaxy merger. Our deep imaging survey for morphological inspection of cluster galaxies with low surface brightness is expected to be a useful basis to understand the nature of cluster galaxies and their internal/external evolutionary path.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.5
• /
• pp.171-177
• /
• 2018

Recently there are many image datasets which has variety of data class and point to extract general features. But in order to this variety data class and point, deep learning model trained this dataset has not good performance in heterogeneous data feature local area. In this paper, we propose the structure which use sub-category and openset object detection methods to train more robust model, named multi-branch tree using ASSL. By using this structure, we can have more robust object detection deep learning model in heterogeneous data feature environment.

• Journal of Korean Neurosurgical Society
• /
• v.57 no.3
• /
• pp.152-158
• /
• 2015

Objective : The purpose of this study to develop new deep-brain stimulation system for long-term use in animals, in order to develop a variety of neural prostheses. Methods : Our system has two distinguished features, which are the fully implanted system having wearable wireless power transfer and ability to change the parameter of stimulus parameter. It is useful for obtaining a variety of data from a long-term experiment. Results : To validate our system, we performed pre-clinical test in Parkinson's disease-rat models for 4 weeks. Through the in vivo test, we observed the possibility of not only long-term implantation and stability, but also free movement of animals. We confirmed that the electrical stimulation neither caused any side effect nor damaged the electrodes. Conclusion : We proved possibility of our system to conduct the long-term pre-clinical test in variety of parameter, which is available for development of neural prostheses.