• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1015-1033
• /
• 2020

With promising results and enormous capability, deep learning technology has attracted more and more attention to both theoretical research and applications for a variety of image processing and computer vision tasks. In this paper, we investigate 32 research contributions that apply deep learning techniques to the agriculture domain. Different types of deep neural network architectures in agriculture are surveyed and the current state-of-the-art methods are summarized. This paper ends with a discussion of the advantages and disadvantages of deep learning and future research topics. The survey shows that deep learning-based research has superior performance in terms of accuracy, which is beyond the standard machine learning techniques nowadays.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2005.11a
• /
• pp.57-60
• /
• 2005

본 논문에서는 대화형 증강 현실을 위한 간단하고 유용한 방법을 제안한다. 대화형 증강 현실을 구현하기 위한 가장 일반적인 방법은 ARToolkit 라이브러리를 사용하는 것이다. 본 논문에서는 ARToolkit 라이브러리의 기능을 그대로 사용하지만 ARToolkit 라이브러리에서 사용하는 그래픽 마커를 라이브 마커로 대체한다. 따라서, 그래픽 마커를 사용할 때보다 직관적이면서도 편리한 사용자 인터페이스를 제공해 줄 수 있다. 단, 라이브 마커를 사용하기 때문에 ARToolkit 라이브러리의 마커인식(Identification)을 위한 패턴 매칭 기술은 사용될 수 없으므로, 보다 정교한 객체 인식(object recognition) 기술이 요구된다. 본 논문에서는 원근 변환(perspective transformation)에 불변한 특성을 가지는 ART 기술자(Angular Radial Transform descriptor)를 사용한다. ART 기술자에 대한 간단한 성능 검증을 통해 라이브 마커에 적용 가능성을 확인하였고, 다양한 실험을 통해 라이브 마커의 효용성을 확인하였다. 본 논문에서 제안한 라이브 마커는 기존의 ARToolkit 사용자들에게 발전된 인터페이스 개념을 제공함으로써, 다양한 증강 현실의 구현에 응용될 수 있으리라 기대된다.

• Korean Journal of Computational Design and Engineering
• /
• v.3 no.1
• /
• pp.68-85
• /
• 1998

Solid modeling refers to techniques for unambiguous representations of three-dimensional objects. Feature recognition is a sub-discipline focusing on the design and implementation of algorithms for detecting manufacturing information such as holes, slots, etc. in a solid model. Automated feature recognition has been an active research area in stolid modeling for many years, and is considered to be a critical component for CAD/CAM integration. This paper gives a technical overview of the state of the art in feature recognition research. Rather than giving an exhaustive survey, I focus on the three currently dominant feature recognition technologies: graph-based algorithms, volumetric decomposition techniques, and hint-based geometric reasoning. For each approach, I present a detailed description of the algorithms being employed along with some assessments of the technology. I conclude by outlining important open research and development issues.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1888-1889
• /
• 2007

The power source of inverter welder stable power of low voltage and high current. Because if shouldn't be, it is caused to spark between the parent metal and the peak. So that, we designed to be base on high frequency transformer and reactor of DC part. Then, we optimized control of PWM, current rising slant, voltage, current, pulse current and 인버터 out-put voltage. Also we designed PCB for EMI and noises.

• Current Photovoltaic Research
• /
• v.6 no.4
• /
• pp.109-118
• /
• 2018

Silicon (Si) solar cells are the most successful technology which are ruling the present photovoltaic (PV) market. In that essence, multijunction (MJ) solar cells provided a new path to improve the state-of-art efficiencies. There are so many hurdles to grow the MJ III-V materials on Si substrate as Si with other materials often demands similar qualities, so it is needed to realize the prospective of Si tandem solar cells. However, Si tandem solar cells with MJ III-V materials have shown the maximum efficiency of 30 %. This work reviews the development of the III-V/Si solar cells with the synopsis of various growth mechanisms i.e hetero-epitaxy, wafer bonding and mechanical stacking of III-V materials on Si substrate. Theoretical approaches to design efficient tandem cell with an analysis of state-of-art silicon solar cells, sensitivity, difficulties and their probable solutions are discussed in this work. An analytical model which yields the practical efficiency values to design the high efficiency III-V/Si solar cells is described briefly.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.3
• /
• pp.78-87
• /
• 2002

A macroblock(MB) in MPEG coded domain can have zero, one, or two motion vectors depending on its frame type and prediction direction (forward-, backward-, or hi-directionally). In this paper, we propose a method that converts these motion vectors on MPEG coded domain as a uniform set, independent of the frame type and the direction of prediction, and directly utilizes these re-analyzed motion vectors for understanding video contents. Also, using this frame-type-independent motion vector, we propose novel methods for detecting and tracking moving objects with frame-based detection accuracy on the compressed domain. These algorithms are performed directly from the MPEG bitstreams after VLC decoding with little time consumption. Experimental results show validity and outstanding performance of our methods.

• Journal of Digital Convergence
• /
• v.13 no.8
• /
• pp.457-463
• /
• 2015

The image distortion field has been studied and applied variously in the fields of engineering and art. This research is to express artistic sense by using the engineering technology on image distortion. In order to do this, the paper first examines the aesthetic interpretations on artistic expressions of image distortion. Then, it explains that image distortion using magnetic force can be an artistic expression method and includes user immersion elements. In addition, it demonstrates that image distortion can be used as an artistic emotional expression based on interaction. This research will contribute to the production and development of creative emotional contents through the convergence of scientific technology and art.

• Journal of the Korea Furniture Society
• /
• v.19 no.3
• /
• pp.180-192
• /
• 2008

The historian called Peter Burger divided it into historical avant-garde before and after the first World War (Dadaism, Futurist, Geometric Abstract, etc.) and Neo- avant-garde during the period of forming social foundation that would be ready to accept new forms of art in the USA, the center of capitalism, after the 2nd World War such as pop art and conceptual art including the then emerging avant-garde minimal art. Meanwhile, he interpreted that Neo- avant-garde succeeded the spirit of historical avant-garde. For modem furniture, the avant-garde terminologies are discussed from the aspect of epistemology or symbolism. On the other hand, they would be exclusively used for advertising slogans, however, it is considered as difficult due to vague terms or their characteristics. This study relates the points of historians such as Peter Burger for theoretical proposal and Richard Kostelanetz's point for formal issues, and contracts and arranges the avant-garde terminologies and its characteristics to the meaning of "getting out of the present decadent trend - aiming for future", "transcending the present aesthetic custom - renovation" and "violating the rules already established in the process of entering new areas - freshness of unfairness". The study was meaningful in providing fundamental clues based on the historical discussion and artists for Neo- avant-garde in the 1960s, that is, the mainstream of design at the end of the 20th century in the period when an avant-garde oriented alternative in the 21st century design has been sincerely required overflowed with the fin de siecle phenomenon.

• The KIPS Transactions:PartB
• /
• v.9B no.1
• /
• pp.119-128
• /
• 2002

This paper proposes an efficient recognition method of car license plate from the car images by using both the dynamical thresholding and the neural network with enhanced learning algorithm. The car license plate is extracted by the dynamical thresholding based on the structural features and the density rates. Each characters and numbers from the p]ate is also extracted by the contour tracking algorithm. The enhanced neural network is proposed for recognizing them, which has the algorithm of combining the modified ART1 and the supervised learning method. The proposed method has applied to the real-world car images. The simulation results show that the proposed method has better the extraction rates than the methods with information of the gray brightness and the RGB, respectively. And the proposed method has better recognition performance than the conventional backpropagation neural network.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.5
• /
• pp.937-941
• /
• 2008

People who have hearing difficulties use sign language as the most important communication method, and they can broaden personal relations and manage their everyday lives without inconvenience through sign language. But they suffer from absence of interpolation between normal people and people who have hearing difficulties in increasing video chatting or video communication services by recent growth of internet communication. In this paper, we proposed a sign language recognition method in order to solve such a problem. In the proposed method, regions of two hands are extracted by tracking of two hands using RGB, YUV and HSI color information from a sign language image acquired from a video camera and by removing noise in the segmented images. The extracted regions of two hands are teamed and recognized by ART2 algorithm that is robust for noise and damage. In the experiment by the proposed method and images of finger number from 1 to 10, we verified the proposed method recognize the numbers efficiently.