• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1000-1011
• /
• 2021

In recent years, using Deep Learning methods to apply for medical and biomedical image analysis has seen many advancements. In clinical, using Deep Learning-based approaches for cancer image analysis is one of the key applications for cancer detection and treatment. However, the scarcity and shortage of labeling images make the task of cancer detection and analysis difficult to reach high accuracy. In 2015, the Unet model was introduced and gained much attention from researchers in the field. The success of Unet model is the ability to produce high accuracy with very few input images. Since the development of Unet, there are many variants and modifications of Unet related architecture. This paper proposes a new approach of using Unet++ with pretrained EfficientNet as backbone architecture for breast tumor cell nuclei segmentation and uses the multi-organ transfer learning approach to segment nuclei of breast tumor cells. We attempt to experiment and evaluate the performance of the network on the MonuSeg training dataset and Triple Negative Breast Cancer (TNBC) testing dataset, both are Hematoxylin and Eosin (H & E)-stained images. The results have shown that EfficientUnet++ architecture and the multi-organ transfer learning approach had outperformed other techniques and produced notable accuracy for breast tumor cell nuclei segmentation.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.136-141
• /
• 2022

A very complex task in deep learning such as image classification must be solved with the help of neural networks and activation functions. The backpropagation algorithm advances backward from the output layer towards the input layer, the gradients often get smaller and smaller and approach zero which eventually leaves the weights of the initial or lower layers nearly unchanged, as a result, the gradient descent never converges to the optimum. We propose a two-factor non-saturating activation functions known as Bea-Mish for machine learning applications in deep neural networks. Our method uses two factors, beta (𝛽) and alpha (𝛼), to normalize the area below the boundary in the Mish activation function and we regard these elements as Bea. Bea-Mish provide a clear understanding of the behaviors and conditions governing this regularization term can lead to a more principled approach for constructing better performing activation functions. We evaluate Bea-Mish results against Mish and Swish activation functions in various models and data sets. Empirical results show that our approach (Bea-Mish) outperforms native Mish using SqueezeNet backbone with an average precision (AP50val) of 2.51% in CIFAR-10 and top-1accuracy in ResNet-50 on ImageNet-1k. shows an improvement of 1.20%.

• Journal of Internet Computing and Services
• /
• v.24 no.2
• /
• pp.1-10
• /
• 2023

The Multi-access Edge Computing (MEC) paradigm equips network edge telecommunication infrastructure with cloud computing resources. It seeks to transform the edge into an IT services platform for hosting resource-intensive and delay-stringent services for mobile users, thereby significantly enhancing perceived service quality of experience. However, erratic user mobility impedes seamless service continuity as well as satisfying delay-stringent service requirements, especially as users roam farther away from the serving MEC resource, which deteriorates quality of experience. This work proposes a deep reinforcement learning based service mobility management approach for ensuring seamless migration of service instances along user mobility. The proposed approach focuses on the problem of selecting the optimal MEC resource to host services for high mobility users, thereby reducing service migration rejection rate and enhancing service availability. Efficacy of the proposed approach is confirmed through simulation experiments, where results show that on average, the proposed scheme reduces service delay by 8%, task computing time by 36%, and migration rejection rate by more than 90%, when comparing to a baseline scheme.

• Current Optics and Photonics
• /
• v.7 no.5
• /
• pp.545-556
• /
• 2023

This paper introduces volume-sharing multi-aperture imaging (VMAI), a potential approach proposed for volume reduction in space-borne imagers, with the aim of achieving high-resolution ground spatial imagery using deep learning methods, with reduced volume compared to conventional approaches. As an intermediate step in the VMAI payload development, we present a phase-1 design targeting a 1-meter ground sampling distance (GSD) at 500 km altitude. Although its optical imaging capability does not surpass conventional approaches, it remains attractive for specific applications on small satellite platforms, particularly surveillance missions. The design integrates one wide-field and three narrow-field cameras with volume sharing and no optical interference. Capturing independent images from the four cameras, the payload emulates a large circular aperture to address diffraction and synthesizes high-resolution images using deep learning. Computational simulations validated the VMAI approach, while addressing challenges like lower signal-to-noise (SNR) values resulting from aperture segmentation. Future work will focus on further reducing the volume and refining SNR management.

• Transactions of Materials Processing
• /
• v.7 no.4
• /
• pp.393-399
• /
• 1998

For minimizing wall thickness thinning of circular shells, a new stamping technology, the deep draw-ing process combined with ironing is approached and investigated. The design requirements for the deep drawing shells are to keep the optimum wall thickness with max. 10 percent thickness thinning of the initial blank thickness, to make uniform thickness strain distribution for the wall of circular shell and to improve the shape accuracy for the roundness and concentricity. In order to check the validity and effectiveness of proposed work, a sample process design is applied to a circular shell needed for a 4multi-stepped deep drawing. Through experiments, the variations of the thickness strain distribution in each drawing process are observed. Also a series of experiments are performed to investigate optimum process variables such as the geometry of tooling, radius and drawing rate. In particular, the advantage of current approach with ironing is shown in contrast to the conventional deep drawing process. From the results of proposed method, the optimum value of process variables are obtained, which contribute more uniform thickness strain distribution and better quality in the drawn product.

• Genomics & Informatics
• /
• v.17 no.3
• /
• pp.30.1-30.4
• /
• 2019

Neuroblastoma is a major cause of cancer death in early childhood, and its timely and correct diagnosis is critical. Gene expression datasets have recently been considered as a powerful tool for cancer diagnosis and subtype classification. However, no attempts have yet been made to apply deep learning using gene expression to neuroblastoma classification, although deep learning has been applied to cancer diagnosis using image data. Taking the International Neuroblastoma Staging System stages as multiple classes, we designed a deep neural network using the gene expression patterns and stages of neuroblastoma patients. Despite a small patient population (n = 280), stage 1 and 4 patients were well distinguished. If it is possible to replicate this approach in a larger population, deep learning could play an important role in neuroblastoma staging.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.1
• /
• pp.210-218
• /
• 2021

CNN (Convolutional Neural Network), a type of deep learning algorithm, is a type of artificial neural network used to analyze visual images. In deep learning, it is classified as a deep neural network and is most commonly used for visual image analysis. Accordingly, an AI autonomous driving model was constructed through real-time image processing, and a crosswalk image of a road was used as an obstacle. In this paper, we proposed a low-cost model that can actually implement autonomous driving based on the CNN model. The most well-known deep neural network technique for autonomous driving is investigated and an end-to-end model is applied. In particular, it was shown that training and self-driving on a simulated road is possible through a practical approach to realizing lane detection and keeping.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.192-201
• /
• 2007

This paper is a study on constructing a natural language interface to database, concentrating on generating textual answers. TGEN, a system that generates textual answer from query result tables is presented. The TGEN architecture guarantees its portability across domains. A combination of a frame-based approach and natural language generation techniques in the TGEN provides text fluency and text flexibility. The implementation result shows that this approach is feasible while a deep NLG approach is still far to be reached.

• Journal of Korea Multimedia Society
• /
• v.22 no.12
• /
• pp.1376-1384
• /
• 2019

The cosmetic and behavioral aspects of aging have become increasingly evident over the years. Physical aging in people can easily be observed on their face, posture, voice, and gait. In contrast, bone aging only becomes apparent once significant bone degeneration manifests through degenerative bone diseases. Therefore, a more accurate and timely assessment of bone aging is needed so that the determinants and its mechanisms can be more effectively identified and ultimately optimized. This study proposed a deep learning approach to assess the bone age of an adult using whole-body bone scintigraphy. The proposed approach uses multiple inputs deep neural network architectures using a loss function, called mean-variance loss. The data set was collected from Chonnam National University Hwasun Hospital. The experiment results show the effectiveness of the proposed method with a mean absolute error of 3.40 years.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.454-475
• /
• 2018

This paper presents a novel Android malware classification model planned to classify and categorize Android malicious code at Drebin dataset. The amount of malicious mobile application targeting Android based smartphones has increased rapidly. In this paper, Restricted Boltzmann Machine and Deep Belief Network are used to classify malware into families of Android application. A texture-fingerprint based approach is proposed to extract or detect the feature of malware content. A malware has a unique "image texture" in feature spatial relations. The method uses information on texture image extracted from malicious or benign code, which are mapped to uncompressed gray-scale according to the texture image-based approach. By studying and extracting the implicit features of the API call from a large number of training samples, we get the original dynamic activity features sets. In order to improve the accuracy of classification algorithm on the features selection, on the basis of which, it combines the implicit features of the texture image and API call in malicious code, to train Restricted Boltzmann Machine and Back Propagation. In an evaluation with different malware and benign samples, the experimental results suggest that the usability of this method---using Deep Belief Network to classify Android malware by their texture images and API calls, it detects more than 94% of the malware with few false alarms. Which is higher than shallow machine learning algorithm clearly.