• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.3
• /
• pp.401-406
• /
• 2021

Recently, mean-teacher models based on convolutional recurrent neural networks are popularly used in audio event detection. The mean-teacher model is an architecture that consists of two parallel CRNNs and it is possible to train them effectively on the weakly-labelled and unlabeled audio data by using the consistency learning metric at the output of the two neural networks. In this study, we tried to improve the performance of the mean-teacher model by using additional derivative features of the log-mel spectrum. In the audio event detection experiments using the training and test data from the Task 4 of the DCASE 2018/2019 Challenges, we could obtain maximally a 8.1% relative decrease in the ER(Error Rate) in the mean-teacher model using proposed derivative features.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.167-174
• /
• 2021

Recently, deep learning technology is widely used in various computer vision applications, such as object recognition, classification, and image generation. In particular, the deep learning-based super-resolution has been gaining significant performance improvement. Fast super-resolution convolutional neural network (FSRCNN) is a well-known model as a deep learning-based super-resolution algorithm that output image is generated by a deconvolutional layer. In this paper, we propose an FPGA-based convolutional neural networks accelerator that considers parallel computing efficiency. In addition, the proposed method proposes Optimal-FSRCNN, which is modified the structure of FSRCNN. The number of multipliers is compressed by 3.47 times compared to FSRCNN. Moreover, PSNR has similar performance to FSRCNN. We developed a real-time image processing technology that implements on FPGA.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.979-993
• /
• 2021

Acoustic scene classification (ASC) categorizes an audio file based on the environment in which it has been recorded. This has long been studied in the detection and classification of acoustic scenes and events (DCASE). In this study, we considered the problem that ASC faces in real-world applications that the model used should have low-complexity. We compared several models that apply light-weight techniques. First, a base CNN model was proposed using log mel-spectrogram, deltas, and delta-deltas features. Second, depthwise separable convolution, linear bottleneck inverted residual block was applied to the convolutional layer, and Quantization was applied to the models to develop a low-complexity model. The model considering low-complexity was similar or slightly inferior to the performance of the base model, but the model size was significantly reduced from 503 KB to 42.76 KB.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1263-1268
• /
• 2018

In this paper, a video analysis is proposed to implement video surveillance system with deep learning object detection and probabilistic data association filter for tracking multiple objects, and suggests its implementation using GPU. The proposed video analysis technique involves object detection and object tracking sequentially. The deep learning network architecture uses ResNet for object detection and applies probabilistic data association filter for multiple objects tracking. The proposed video analysis technique can be used to detect intruders illegally trespassing any restricted area or to count the number of people entering a specified area. As a results of simulations and experiments, 48 channels of videos can be analyzed at a speed of about 27 fps and real-time video analysis is possible through RTSP protocol.

• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.11
• /
• pp.269-274
• /
• 2018

MMORPG is a role-playing game that tens of thousands of people access it online at the same time. Users connect to the server through the game client and play with their own characters. If the user moves to a management area of another server beyond the area managed by the server, the user information must be transmitted to the server to be moved. In an actual game, the user is required to synchronize the established and the transferred information. In this paper, we propose a migration agent server in the virtual systems. We implement a seamless virtual server using the grid method to experiment with seamless server architecture for virtual systems. We propose a method to minimize the delay and equalize the load when the user moves to another server region in the virtual environment. Migration Agent acts as a cache server to reduce response time, the response time was reduced by 50% in the case of 70,000 people.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.1
• /
• pp.58-71
• /
• 2019

This paper proposes architecture of a cognitive radio engine platform and the allowable frequency channel reasoning method that enables acquisition of the allowable channels for the military tactical network environment. The current military tactical wireless communication system is increasing need to secure a supplementary radio frequency to ensure that multiple wireless networks for different military wireless devices coexist, so that tactical wireless communication between the same or different systems can be operated effectively. This paper presents the allowable frequency channel reasoning method based on cognitive radio engine for realizing DSA(Dynamic Spectrum Access) as an optimal available frequency channel. To this end, a case-based allowable frequency channel reasoning method for cognitive radio devices is proposed through modeling of primary user's traffic status and calculation of channel occupancy probability. Also through the simulation of the performance analysis, changing rate of collision probability between the primary users' occupancy channel and the available channel acquisition information that can be used by the cognitive radio device was analysed.

• Journal of the Society of Disaster Information
• /
• v.15 no.1
• /
• pp.121-132
• /
• 2019

Purpose: This study aims to analyze the meaning of architectural space of Alvar Aalto, which is based on architectural theories of periodic flows and the form of the space with the various elements applied to the works. Method: It analyzes the interrelationships of light, form and material in triangular composition by synthesizing the architectural vocabulary of Alvar Aalto from the viewpoint of modernism. Results: The architectural works of Alvar Aalto are centered on the combination of light, form, material, and space diversity, emphasizing social awareness and linking the human scale with the building environment. Conclusion: The structure of the Alvar Aalto is a synonym of nature and culture, society and individuals, standardization and diversity, universality and locality, intellectual and emotional, scientific and human psychological. It is possible to apply reciprocity elements to architectural spaces in a comprehensive manner while being different from each other in terms of rationality and immediacy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.10
• /
• pp.1353-1358
• /
• 2021

IoT (Internet of Things) emerges from various technologies such as communications, micro processors and embedded system and so on. The IoT has also been used to UAV (Unmanned Aerial Vehicle) system. In manned aircraft, a pilot and co-pilot should control FCS (Flight Control System) with FBW(Fly By Wire) system for flight operation. In contrast, the flight operation in UAV system is remotely and fully managed by GCS (Ground Control System) almost in real time. To make it possible the communication channel should be necessary between the UAV and the GCS. There are many protocols between two systems. Amongst them, MAVLink (Macro Air Vehicle Link) protocol is representatively used due to its open architecture. MAVLink does not define any securities itself, which results in high vulnerability from external attacks. This paper proposes the method to enhance data security in GCS network by applying cryptographic methods to the MAVLink messages in order to support safe UAV operations.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.2
• /
• pp.89-97
• /
• 2021

Blockchain is designed to collect and store the data recorded on the network in one block unit, and is connected and stored back and forth, and its form is similar to how the blocks are connected, so it is called a blockchain. Many companies are trying to popularize blockchain-based services at home and abroad, and blockchains are used in various industries. This study introduces the technical characteristics of the blockchain and deals with application services utilizing the blockchain. Introducing 5 types of blockchain architecture and core technologies and introducing blockchain application services that are used in payment services, blockchain service networks, blockchain real estate platforms, identity verification, cryptocurrency, diamond distribution path tracking, and blog information recording. do. It is expected to increase the understanding of the blockchain and provide usefulness in future blockchain research and service development.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.34 no.9
• /
• pp.9-20
• /
• 2018

There are many large and small spaces in the interior of the building, and these spaces are arranged and connected to form a systematic spatial structure. A structure is a collection of several parts to form a whole. In other words, the spatial structure in architecture can be seen as a whole organized and organized as individual unit spaces are gathered together. Therefore, in order to understand the spatial structure, we first need to define the unit spaces that form part, how they are interconnected and arranged, and then understand how and how these unit spaces are organized to form a whole. The main purpose of this study is to study the structural system of space based on the shape information of space on architectural plane. This means interpreting the process and method of how the unit spaces defined as a certain shape on the architectural plane are organized step by step, integrated into a higher level, and eventually integrated into one whole. In this paper, the shape and layout of the unit space are identified in the architectural plan, the connection relation is defined, and expressed in the network form. And suggests a new methodology for interpreting the organizational process in which the following spaces are integrated as a whole. This new methodology is based on human perceptual characteristics. When people recognize an object, they recognize the object partly and completely. We want to explain the relationship between parts of space and the whole according to their characteristics.