• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.3
• /
• pp.71-78
• /
• 2011

This paper propose image steganalysis scheme based on independence between parent and child subband on the multi-layer wavelet domain. The proposed method decompose cover and stego images into 12 subbands by applying 3-level Haar UWT(Undecimated Wavelet Transform), analyze statistical independency between parent and child subband. Because this independency is appeared more difference in stego image than in cover image, we can use it as feature to differenciate between cover and stego image. Therefore we extract 72D features by calculation first 3 order statistical moments from joint characteristic function between parent and child subband. Multi-layer perceptron(MLP) is applied as classifier to discriminate between cover and stego image. We test the performance of proposed scheme over various embedding rates by the LSB, SS, BSS embedding method. The proposed scheme outperforms the previous schemes in detection rate to existence of hidden message as well as exactness of discrimination.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.5
• /
• pp.473-478
• /
• 2013

In this paper, we develop the Heavy Rain Advisory Decision Model based on intelligent neuro-fuzzy algorithm RBFNNs by using KLAPS(Korea Local Analysis and Prediction System) Reanalysis data. the prediction ability of existing heavy rainfall forecasting systems is usually affected by the processing techniques of meteorological data. In this study, we introduce the heavy rain forecast method using the pre-processing techniques of meteorological data are in order to improve these drawbacks of conventional system. The pre-processing techniques of meteorological data are designed by using point conversion, cumulative precipitation generation, time series data processing and heavy rain warning extraction methods based on KLAPS data. Finally, the proposed system forecasts cumulative rainfall for six hours after future t(t=1,2,3) hours and offers information to determine heavy rain advisory. The essential parameters of the proposed model such as polynomial order, the number of rules, and fuzzification coefficient are optimized by means of Differential Evolution.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.1
• /
• pp.47-56
• /
• 2010

The use of airborne LiDAR data obtained by airborne laser scanners has increased in the field of spatial information such as building modeling. LiDAR data consist of irregularly distributed 3D coordinates and lack visual and semantic information. Therefore, LiDAR data processing is complicate. This study suggested a method of LiDAR data segmentation using roof surface patches from aerial images. Each segmented patch was modeled by analyzing geometric characteristics of the LiDAR data. The optimal functions could be determined with segmented data that fits various shapes of the roof surfaces as flat and slanted planes, dome and arch types. However, satisfiable segmentation results were not obtained occasionally due to shadow and tonal variation on the images. Therefore, methods to remove unnecessary edges result in incorrect segmentation are required.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.1
• /
• pp.251-258
• /
• 2013

We studied it about ways to detect damage of a tire about randomness ultrasonic signal which occurs repeatedly while rub a tire of driving car and a road surface. The signal randomness is decreased through the preprocess of short-time energy calculation and the average value of coherence function is used by the normalization expression of the signal randomness. The process limit that can be decide on the dominant period of a signal using the coherence threshold is analyzed and the algorithm to decide the dominant period is proposed by setting up the -3dB threshold of the maximum value on the power spectrum.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2002.02a
• /
• pp.438-443
• /
• 2002

본 연구는 수도작에서 토양과 식물체의 분광반사특성과 영상처리를 이용한 기계시각 잡초검출 센서를 개발하기 위한 기초연구로서 분류하고자 하는 대상체들의 분광반사율을 조사하여 주요한 파장을 선정하고 선정된 파장을 이용한 판별분석을 통해 각 대상체에 대한 분류 정확도를 중심으로 잡초검출 가능성을 조사하기 위하여 수행하였으며, 실험으로부터 얻은 결론은 다음과 같다. 1. 토양과 식물체를 구분하는데 효과적인 파장은 마른 토양의 경우 680 nm, 배수 토양에 있어서는 810 nm로 선정하였고, 토양을 배제한 후 벼와 잡초를 구분하기에 효과적인 파장은 580, 680 nm로 선정하였다. 2. 토양과 식물체를 구분하기 위한 판별분석 결과 2가지 토양상태 모두 식물체와 완전히 구분 가능한 것으로 나타났다. 벼와 잡초를 구분하기 위한 실험에서, 벼는 98%의 분류정확도로 구분이 가능하였고, 잡초는 83%의 분류정확도로 구분이 되는 것으로 나타났다. 따라서 차후 분광학적 원리를 이용한 센서를 제작할 때 본 연구에서 선택한 주요 파장과 판별함수를 이용하여 장치를 구성하고 알고리즘을 제작한다면 벼, 잡초, 토양을 효과적으로 구분이 가능할 것으로 판단되었다. 3. 컬러 CCD 카메라를 사용하는 경우에 있어 식물체와 토양을 구분하기 위해 3 종의 파장 중 630 nm 파장만의 이용을 고려하여 그 분류성능을 분석한 결과, 식물체와 토양은 소수의 관측치를 제외하고 완전히 구분이 가능했고, 벼와 잡초를 구분한 결과에서는 비교적 높은 분류능력을 가진 것으로 나타나 차후 컬러 CCD 카메라를 이용하여 장치를 구성하는데 좋은 기초가 될 것으로 판단된다. 배양체의 접종작업은 모든 배양실이 인력에 의존하였으며, 배양체를 배지와 분리하여 불필요한 부분을 제거하고 배양작물에 따라 생육정도를 2~3등급으로 구분하여 배양용기의 배지 위에 치상하는 과정으로 수행되었으며, 작업능률은 호접란의 경우 배양병에 25본을 접종하는데 시간당 6병, 심비디움은 원형 플라스크에 25본을 접종하는데 시간당 10병 정도였다. 바. 식물체의 대량증식에 사용되는 플라스크, 배양병, PE용기 등 배양용기의 세척작업은 농원의 1개배양실에서 간이식 세척기, 이 외의 9개배양실은 모두 물에 담겨 두었다가 세제와 브러쉬 등을 사용하여 인력으로 세척하고 있어 생력화 기술개발이 요구되었다.도가 빠를수록 건조속도가 빨라졌으며, 건조에너지도 1,334kcal/kg.water로 비슷하게 소요되었다. 마. 시험구와 대비구의 건감률은 시험구에서 1.08~1.36w.b./h로 나타나 대비구보다 약 9.9~18.3%가 높게 나타났고, 건조에너지는 10.2~14.6%가 절감되었다. 발아율은 열풍온도가 낮을수록 높게 나타났고 시험구가 대비구보다 발아율이 낮게 나타났으며, 동할률 증가량도 원적외선.열풍 복합건조방법이 높게 나타나 이것은 곡물 표면에 원적외선 방사에의한 복사열이 전달되어 열장해를 받았기 때문으로 판단되며, 금후 더 연구하여 적정 열풍온도 및 방사체 크기를 구명해야 할 것이다.으로 보여진다 따라서 옻나무 유래 F는 포유동물의 생식기능에 중요하게 작용하는 것으로 사료된다.된다.정량 분석한 결과이다. 시편의 조성은 33.6 at% U, 66.4 at% O의 결과를 얻었다. 산화물 핵연료의 표면 관찰 및 정량 분석 시험시 시편 표면을

• Journal of Korean Society of Transportation
• /
• v.26 no.3
• /
• pp.41-51
• /
• 2008

As the local bus service diagram after the bus route reform is studied by variable analysis based on traffic card data and income adjustment data, the characteristic of the local bus system is revealed in Busan Metropolitan city. The relationship between traveling length and traveling time is influenced by traveling velocity. In order to keep a headway within 10 minutes, bus service number per minute should be over 0.1013 vehicles. The traveling time of afternoon is generally longer than that of forenoon. Compared with the bus used by a lot people, the deviation of that used by a few people is larger in the all cases of length, headway, time and velocity. According to the analysis of the relationship among card trip number, average income and transfer rate, the relationship between card trip number and average income is expressed as linear function in the general bus and as exponential function in the high-grade & rapid bus. The 1% increase of transfer rate is equal to 6.3 trip/vehicle/day decrease and 4.9 trip/vehicle/day decrease in two bus types respectively. The four effective variables are defined by the discriminant analysis between the profitable routes and the unprofitable; According to discriminant size, bus service number per km, bus via suburb, subway meeting number, bus via university. In order to increase the income when the minibus will be included among public transit transfer system in 2008, it should be necessary to settle the bus network and revitalize the public transit better. In order to decrease the cost, it should be necessary to reorganize the hierarchy between the local bus and the minibus better.

• Journal of Satellite, Information and Communications
• /
• v.11 no.4
• /
• pp.1-9
• /
• 2016

This paper was studied in two ways with respect to the information received from the video signal separation technique of PPI Scop radar device. The proposed technique consists in generating an image signal through the video signal separation and synthesis, symbol generation, the residual image signal generation process. This technology can greatly improve the operating convenience with improved ease of discrimination, screen readability for the operator in analyzing radar information. The first proposed method was constructed for high-speed FPGA-based information processing systems for high speed operation stability of the system. The second proposed method was implemented intelligent algorithms and a software algorithm function curve associated resources.This was required to meet the constraints on the radar information, analysis system. Existing radar systems have not the frame data analysis unit image. However, this study was designed to image data stored in the frame-by-frame analysis of radar images with express information MPEG4 video. Key research content is to highlight the key observations expresses the target, the object-specific monitoring information to the positive image processing algorithm and the function curve delays. For high-definition video, high-speed to implement data analysis and expressing a variety of information was applied to the ARM Processor Support in Pro ASIC3.

• Tunnel and Underground Space
• /
• v.15 no.1 s.54
• /
• pp.47-54
• /
• 2005

One of the standard procedures of discontinuity survey is the joint set identification from the population of field orientation data. Discontinuity set identification is fundamental to rock engineering tasks such as rock mass classification, discrete element analysis, key block analysis. and discrete fracture network modeling. Conventionally, manual method using contour plot had been widely used for this task, but this method has some short-comings such as yielding subjective identification results, manual operations, and so on. In this study, the method of discontinuity set identification using genetic algorithm was introduced, but slightly modified to handle the orientation data. Finally, based on the genetic algorithm, we developed a FORTRAN program, Genetic Algorithm based Clustering(GAC) and applied it to two different discontinuity data sets. Genetic Algorithm based Clustering(GAC) was proved to be a fast and efficient method for the discontinuity set identification task. In addition, fitness function based on variance showed more efficient performance in finding the optimal number of clusters when compared with Davis - Bouldin index.

• Biomedical Science Letters
• /
• v.1 no.1
• /
• pp.81-88
• /
• 1995

RDW and MCV are thought to be the highly sensitive blood cell parameters in the differentiation of iron deficiency anemias. Through the medical records of 227 anemic patients and the physical checking results of 143 healthy persons in a General Hospital during the recent five years, the authors evaluated various blood cell parameters including RDW and MCV. Iron deficiency anemia, aplastic anemia and other anemias associated with chronic disease were shown as the three major causes of anemias in Korea. In the patients of iron deficiency anemia MCV was very low(62.9$\pm$13.7fl), while RDW was very high(19.3$\pm$4.8) showing much lower MCV and much higher RDW in severe IDA compared with in mild IDA. To differentiate iron deficiency anemias form other anemias, a discriminant function was developed from some blood cell parameters like MCV, MCH, MCHC, RDW and platelets(D.F.=0.26-0.012MCV -0.130MCH +0.073MCHC +0.052RDW+0.003PLT).

• Smart Media Journal
• /
• v.10 no.4
• /
• pp.45-54
• /
• 2021

The skin is the body's first line of defense against external infection. When a skin disease strikes, the skin's protective role is compromised, necessitating quick diagnosis and treatment. Recently, as artificial intelligence has advanced, research for technical applications has been done in a variety of sectors, including dermatology, to reduce the rate of misdiagnosis and obtain quick treatment using artificial intelligence. Although previous studies have diagnosed skin diseases with low incidence, this paper proposes a method to classify common illnesses such as warts and corns using a convolutional neural network. The data set used consists of 3 classes and 2,515 images, but there is a problem of lack of training data and class imbalance. We analyzed the performance using a deep metric loss function and a cross-entropy loss function to train the model. When comparing that in terms of accuracy, recall, F1 score, and accuracy, the former performed better.