• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.37-44
• /
• 2021

Automatic classification of brain MRI images play an important role in early diagnosis of brain tumors. In this work, we present a deep learning-based brain tumor classification model in MRI images using ensemble of deep features. In our proposed framework, three different deep features from brain MR image are extracted using three different pre-trained models. After that, the extracted deep features are fed to the classification module. In the classification module, the three different deep features are first fed into the fully-connected layers individually to reduce the dimension of the features. After that, the output features from the fully-connected layers are concatenated and fed into the fully-connected layer to predict the final output. To evaluate our proposed model, we use openly accessible brain MRI dataset from web. Experimental results show that our proposed model outperforms other machine learning-based models.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.279-279
• /
• 2022

최근 이미지분석 기술은 하드웨어 및 소프트웨어 기술의 급격한 발전으로 인해 의학, 생물학, 지리학, 재료공학 등에서 수많은 연구 분야에서 광범위하게 활용되고 있으며, 이미지분석은 다량의 토사에 대하여 입경을 포함한 형상학적 특성을 간편하게 정량화 할 수 있기 때문에 매우 효과적인 분석 방법으로 판단된다. 현재 모래의 입도분석 방법으로는 신뢰성 있는 체가름 시험법(KSF2302) 등이 있으나, 번거로운 처리과정과 많은 시간이 소요된다. 또한 입자형상은 입경이 세립 할수록 직접 측정이 어렵기 때문에, 최근에는 이미지 분석을 이용하는 방법이 시도되고 있다. 본 연구에서는 75㎛ 이상의 조립질 하상 토사 이미지를 취득하여, 입자들의 장·축단 길이, 면적, 둘레, 공칭직경 및 종횡비 등의 형상학적 특성인자를 자동으로 측정하는 프로그램 개발을 수행하였다. 프로그램은 이미지 분석에 특화된 라이브러리인 OpenCV(Open Source Computer Vision)를 적용하였다. 이미지 분석 절차는 크게 이미지 취득, 기하보정, 노이즈제거, 객체추출 및 형상인자 측정 단계로 구성되며, 이미지 취득시 패널의 하단에 Back light를 부착해 시료에 의해 발생되는 음영을 제거하였다. 기하보정은 원근변환(perspective transform)을 적용했으며, 노이즈 제거는 모폴로지 연산과 입자간의 중첩으로 인한 뭉침을 제거하기 위해 watershed 알고리즘을 적용하였다. 최종적으로 객체의 외곽선 추출하여 입자들의 다양한 정보(장축, 단축, 둘레, 면적, 공칭직경, 종횡비)를 산출하고, 분포형으로 제시하였다. 본 연구에서 제안하는 이미지분석을 적용한 토사의 형상학적 특성 측정 방법은 시간과 비용의 측면에서 보다 효율적으로 하상 토사에 대한 다양한 정보를 획득 할 수 있을 것으로 기대한다.

• Journal of KIISE
• /
• v.43 no.12
• /
• pp.1325-1333
• /
• 2016

Software classification has several applications such as copyright infringement detection, malware classification, and software automatic categorization in software repositories. It can be also employed by software filtering systems to prevent the transmission of illegal software. If illegal software is identified by measuring software similarity in software filtering systems, the average number of comparisons can be reduced by shrinking the search space. In this study, we focused on the classification of Windows executables using API call information and machine learning. We evaluated the classification performance of machine learning-based classifier according to the refinement method for API information and machine learning algorithm. The results showed that the classification success rate of SVM (Support Vector Machine) with PolyKernel was higher than other algorithms. Since the API call information can be extracted from binary executables and machine learning-based classifier can identify tampered executables, API call information and machine learning-based software classifiers are suitable for software filtering systems.

• Journal of Digital Contents Society
• /
• v.13 no.1
• /
• pp.21-29
• /
• 2012

In ubiquitous computing environment, rising large scale data exchange through sensor network with sharply growing the internet, the processing of the continuous stream data is required. Therefore there are some mining researches related to the extracting of frequent structures and the efficient query processing of XML stream data. In this paper, we propose a mining method to extract frequent structures of XML stream data in recent window based on the active window sliding using trigger rule. The proposed method is a basic research to control the stream data flow for data mining and continuous query by trigger rules.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.909-912
• /
• 2012

An automated cell tracking system is automatically to analyze and track changes of cell behaviors in time-lapse cell images acquired from microscope in the cell culture. In this paper, we proposed and developed an ellipse fitting based algorithm for separating very small size overlapping cells in a cell image consisted of thousands or ten thousands cells. We were extracted contours of clusters and divided them into line segments and then produced their fitted ellipses for each line segment. By experimentations, our algorithm was separated clusters with average 91% precision for two overlapping cells and average 84% precision for three overlapping cells respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.4 no.1
• /
• pp.47-53
• /
• 1999

In medical field, the computer has been used in the automatic processing of data derived in hospital. the automation of diagonal devices, and processing of medical digital images. In this paper, we classify red blood cell into 16 class including normal cell to the automation of blood analysis to diagnose disease. First, using UNL Fourier and invariant moment algorithm, we extract features of red blood cell from blood cell image and then construct multi-layer backpropagation neural network to recognize. We proof that the system can give support to blood analyzer through blood sample analysis of 10 patients.

• Journal of Korea Society of Industrial Information Systems
• /
• v.4 no.4
• /
• pp.40-46
• /
• 1999

This paper introduce a Koran-Japanese machine translation system which is a module in the spoken language interpreting system It is implemented based on the TDMT(Transfre Driven Machine Translation). We define a new unit of translation so called TOKEN. The TOKEN-based translation method resolves nonstructural feature in Korean sentences and increases the quaity of translating results. In our system, we get rid of useless effort for traditional parsing by performing semi-parsing. The semi-parser makes the dependency tree which has minimum information needed generating module. We constructed the generation dictionaries by using the corpus obtained from ETRI spoken language database. Our system was tested with 600 utterances which is collected from travel planning domain The success-ratio of our system is 87% on restricted testing environment and 71% on unrestricted testing environment.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.4
• /
• pp.1002-1013
• /
• 1996

The segmentation of characters is an important step in the automatic recognition of handwritten text. This paper proposes the segmenting method of off-line handwritten Hangul. The suggested approach is based on the structural characteristics of Hangul. The first step extracts the local features. connected component and strokes from the imput word. In the second step we identify the class of strokes. The third segmenting step specifies WRC(White Run Column) before consonant or horizontal vowel. If the segment is longer than threshold, the system estimates segmenting columns using the consonant-vowel information and column features, and then finds a cornered boundary along the strokes within the estimated segmenting columns.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.4
• /
• pp.833-838
• /
• 2003

This paper presents novel methods of recognizing license plates of passing vehicles outdo(n. In particular, the proposed method is much robust for inclined plates caused by the changes of camera placement. To acquire fine images of quickly passing vehicles under a wide range of illumination conditions, we developed a sensing system having superb characteristics. We expanded the dynamic range and eliminated the blurring of images of fast moving vehicles by synthesizing a pair of synchronized images with different intensities. furthermore, to extend the flexibility of the positioning of the TV camera, we propose a recognition algorithm that can be applied to inclined plates. The performance of the integrated system was investigated on real images of vehicles captured under various illumination conditions. The recognition rates of over 99% (conventional plates) and over 97% (highly inclined plates) shows that the developed system is effective for license plate recognition.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06b
• /
• pp.261-264
• /
• 2008

본 논문에서는 영상의 중요한 특징을 강조하는 점의 분포를 가지는 자동화된 점묘화(stippling) 제작 방법을 제시한다. 예술가의 점묘화 일러스트 작품을 살펴보면 영상의 특징을 강조하는 방향성이 있는 점들을 사용해서 회화적인 느낌을 살림과 동시에 사물의 형태를 좀 더 명백히 파악할 수 있게 해준다. 하지만 컴퓨터 그래픽스 분야에서 연구된 기존 점묘화 기법 알고리즘은 입력 영상의 특징적인 형태를 고려하지 않고 색조에 따른 점의 밀도 변화만으로 사물을 표현하기 때문에 사물의 형태가 제대로 드러나지 않는 단점이 있다. 본 방법에서는 점의 분포가 대상의 형태를 반영하며 분포되게 하는 알고리즘을 적용하여 사물의 특징적인 형태를 강조한다. 이를 위해 영상의 특징선으로부터 추출한 특징 흐름(feature flow)을 따라 점을 배치시키는 방법을 사용한다. 그리고 입력 영상의 색조(tone)를 점묘화에 반영하기 위해 점의 크기가 입력 영상의 색조에 따라 자동으로 결정되도록 한다.