• Journal of Internet of Things and Convergence
• /
• v.8 no.2
• /
• pp.79-84
• /
• 2022

This paper presents an enhancement method of document restoration capability which is robust for security, loss, and contamination, It is based on two methods, that is, encryption and DnCNN(DeNoise Convolution Neural Network). In order to implement this encryption method, a mathematical model is applied as a spatial frequency transfer function used in optics of 2D image information. Then a method is proposed with optical interference patterns as encryption using spatial frequency transfer functions and using mathematical variables of spatial frequency transfer functions as ciphers. In addition, by applying the DnCNN method which is bsed on deep learning technique, the restoration capability is enhanced by removing noise. With an experimental evaluation, with 65% information loss, by applying Pre-Training DnCNN Deep Learning, the peak signal-to-noise ratio (PSNR) shows 11% or more superior in compared to that of the spatial frequency transfer function only. In addition, it is confirmed that the characteristic of CC(Correlation Coefficient) is enhanced by 16% or more.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1935-1941
• /
• 2021

In order to develop an algorithm using deep learning, which has been recently applied to various fields, it is necessary to have rich, high-quality learning data. In this paper, we propose an acquisition system for biological signals that simultaneously collects bio-signal data such as optical videos, thermal videos, and voices, which are mainly used in developing deep learning algorithms and useful in derivation of information, and transmit them to the server. To increase the portability of the collector, it was made based on Raspberry Pi, and the collected data is transmitted to the server through the wireless Internet. To enable simultaneous data collection from multiple collectors, an ID for login was assigned to each subject, and this was reflected in the database to facilitate data management. By presenting an example of biological data collection for fatigue measurement, we prove the application of the proposed acquisition system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1807-1822
• /
• 2023

Innovation and rapid increased functionality in user friendly smartphones has encouraged shutterbugs to have picturesque image macros while in work environment or during travel. Formal signboards are placed with marketing objectives and are enriched with text for attracting people. Extracting and recognition of the text from natural images is an emerging research issue and needs consideration. When compared to conventional optical character recognition (OCR), the complex background, implicit noise, lighting, and orientation of these scenic text photos make this problem more difficult. Arabic language text scene extraction and recognition adds a number of complications and difficulties. The method described in this paper uses a two-phase methodology to extract Arabic text and word boundaries awareness from scenic images with varying text orientations. The first stage uses a convolution autoencoder, and the second uses Arabic Character Segmentation (ACS), which is followed by traditional two-layer neural networks for recognition. This study presents the way that how can an Arabic training and synthetic dataset be created for exemplify the superimposed text in different scene images. For this purpose a dataset of size 10K of cropped images has been created in the detection phase wherein Arabic text was found and 127k Arabic character dataset for the recognition phase. The phase-1 labels were generated from an Arabic corpus of quotes and sentences, which consists of 15kquotes and sentences. This study ensures that Arabic Word Awareness Region Detection (AWARD) approach with high flexibility in identifying complex Arabic text scene images, such as texts that are arbitrarily oriented, curved, or deformed, is used to detect these texts. Our research after experimentations shows that the system has a 91.8% word segmentation accuracy and a 94.2% character recognition accuracy. We believe in the future that the researchers will excel in the field of image processing while treating text images to improve or reduce noise by processing scene images in any language by enhancing the functionality of VGG-16 based model using Neural Networks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.938-958
• /
• 2024

Unsupervised Video Object Segmentation (UVOS) is a highly challenging problem in computer vision as the annotation of the target object in the testing video is unknown at all. The main difficulty is to effectively handle the complicated and changeable motion state of the target object and the confusion of similar background objects in video sequence. In this paper, we propose a novel deep Dual-stream Co-enhanced Network (DC-Net) for UVOS via bidirectional motion cues refinement and multi-level feature aggregation, which can fully take advantage of motion cues and effectively integrate different level features to produce high-quality segmentation mask. DC-Net is a dual-stream architecture where the two streams are co-enhanced by each other. One is a motion stream with a Motion-cues Refine Module (MRM), which learns from bidirectional optical flow images and produces fine-grained and complete distinctive motion saliency map, and the other is an appearance stream with a Multi-level Feature Aggregation Module (MFAM) and a Context Attention Module (CAM) which are designed to integrate the different level features effectively. Specifically, the motion saliency map obtained by the motion stream is fused with each stage of the decoder in the appearance stream to improve the segmentation, and in turn the segmentation loss in the appearance stream feeds back into the motion stream to enhance the motion refinement. Experimental results on three datasets (Davis2016, VideoSD, SegTrack-v2) demonstrate that DC-Net has achieved comparable results with some state-of-the-art methods.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.12 no.2
• /
• pp.55-65
• /
• 2010

The present study purposed to compare formative expression between the fashion of the end of 19th century - early 20th century and that of the 21st century which were influenced by art nouveau and art deco. For this purpose, we analyzed contents in literature, Internet, etc.,. According to the results of this study, the formative expression of art nouveau were divided into continuous movement, optical illusion and decorative expression. Continuous movement was used to express fancy and soft images in the end of 19th century, and to express images giving the feeling of speed and tension in the 21st century. Optical illusion was expressed by the movement the thin and soft materials with the body movement and the movement of excessive frills adornment. And in movement produced using thin and soft materials and by the bias cutting method in the 21st century. Decorative expression was expressed as fantasy and decadent image in the end of 19th century and as mild passionate and modern image with color combination in the 21st century. The formative expression of art deco were divided into visual simplicity, contrastive loudness, and functional expression. Visual simplicity was expressed as a boyish image in the early of 20th century, but as a complex expression combining simplicity and femininity in the 21st century. Contrastive loudness was expressed by using colors both in the early of 20th century and 21st century. Functional expression was reinforced through the lightening of design and materials in the early of 20th century, but through using bias cutting high tech materials in the 21st century.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.42 no.5 s.335
• /
• pp.23-28
• /
• 2005

Recently, new concept, NP(Network Processor) was emerged into communication systems to cope with the various service requirements from Internet users. NP is an unique promising technique to capable of implementing of the packet processing in wire-speed and providing the flexibility for supporting the newly network services, having satisfied with implementation using hardware and software respectively in past, This paper deals with the implementation techniques and evaluation results of the line card capable to do packet forwarding function with packet processing power of wire-speed and applicable to various physical interfaces. There are several interfaces of POS, Gigabit ethernet and EPON in E-OLT(EPON Optical Line Terminal) system of PATH(Photonic Access To Home) network. Therefore, the E-OLT's packet forwarding engine have to support various subscriber's interface in wire speed. Our system is implemented the subscriber's card in daughter board and the setup procedure is done by system firmware based on the module's identifier acquired from installed physical board.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.5
• /
• pp.925-933
• /
• 2003

Network is expected to be developed into optical Internet network collected IP layer and optical layer, but GMPLS is risen at the transitional evolution stage because of the present technology level. GMPLS that MPLS is extended and generalized is able to support not only the packet switching device but also the devices which perform switching in time, wavelength, and space domain. To implement the common control plane to these various switching types, GMPLS extends the existing MPLS signaling and routing protocol. In this paper, we describe the overview of GMPLS technology, and then we will refer to the OSPF(Open Shortest Path First), which was used to exchange the status information of link, as the plan of routing extension to exchange the information of various link type, bandwidth, link protection type etc. And also, we describe the definition of new protocol, so called, LMP that is a signaling protocol for solving complex problem which manages hundreds and thousands of links between two nodes. And we will examine and analyze the plan of signaling protocol extension to apply signaling protocol RSVP-TE(Resource Reservation Protocol) for traffic engineering in MPLS to network, and the message objects and formats associated with modified RSVP.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.2
• /
• pp.372-378
• /
• 2015

Smoke generated from business establishments and factories will not only cause air pollution but also have a significant impact on the human body. Generally, the most common method for measuring the turbidity of the plume generated from the stack is a method of observing by the transmissometer mounted in the chimney or Method 9 from the US EPA(Environmental Protection Agency) which is a visual method of determining plume turbid emitted from stationary sources. However, these methods need a lot of cost to build and maintain. In this paper, we build a plume turbidity measurement module programs using a digital camera. We design and implement a module which acquires the pictures of the plume using a digital camera and measures the turbidity of it using the DOM(Digital Optical Method). In addition, we demonstrate the excellence by comparing and analyzing implemented module and other methods.

• Journal of Biomedical Engineering Research
• /
• v.24 no.5
• /
• pp.411-419
• /
• 2003

This paper aims at developing 'gyro-mouse' which provides decent and comfortable human-computer interface that supports the usage of such software as an internet-browser in PC for the people paralyzed in upper limbs. This interface operates on information collected from head movement to get the cursor control. The interface is composed of two modules. One is hardware module in which the head horizontal and vertical angular velocities are detected and transmitted into PC. The other is a PC software that translates the received data into movement and click signals of the mouse. The ANN (artificial neural network) learns the quick nodding pattern of each user as click input so that it can provide user-friendly interface. The performance of the system was evaluated by three indices that are click recognition rate. error in cursor position control. and click rate of the moving target box. The performance result of the gyro-mouse was compared with that of the optical-mouse to assess the efficiency of the gyro-mouse. The average click recognition rate was 93%, average error in cursor position control was 1.4∼5 times of optical mouse. and the click rate with 50 pixels target box was 40%(30 clicks/min) to that of optical mouse. The click rate increased monotonously with the number of trial from 35% to 44%. The suggested system is expected to provide a new possibility to communicate with the society.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.1 no.1
• /
• pp.56-62
• /
• 2006

Currently, Most of the Backhual and Submarine's networks are being installed using DWDM and EDFA to deal with dramatically increasing internet traffic. In a conventional type optical system, the so-called 3R repeater, fiber non- linearity was not significant factor but it is one of the most important consideration when designing new DWDM network based on the EDFAs. To minimize the signal pulse distortion in long-haul optical network, Zero-dispersion fiber is indispensable but causes the fiber non-linearity. Therefore this conflicting problems should be managed to transmit a high-speed capacity traffic. In this paper, key technologies in DWDM submarine optic network were found out for a high capacity traffic and introduced the dispersion management way for solving the conflicting problems between Zero-dispersion and non-linearity. Lastly designed a dispersion map in section between Korea and Hong-kong.