• Journal of KIISE
• /
• v.42 no.1
• /
• pp.33-43
• /
• 2015

Recently, patent images are growing importance and thus patent image retrieval is a growing area of research. However, most existing patent image retrieval systems use edges extracted in the images, whose performance is affected by the quality of edge detection in the image pre-processing step. To overcome this disadvantage, we propose a SURF-based patent image retrieval method which uses the morphological characteristics of the images. The proposed method detects SURF interest points with directions and computes regional histograms. We apply the proposed method to a patent image database with 2000 binary images and we show the proposed retrieval system achieves excellent results, even when the images have some loss or degradation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.5
• /
• pp.887-894
• /
• 2006

This paper proposes a new method that detects the intersection points of handwritten Hangul strokes using run-length. The method firstly finds the strokes' width of handwritten Hangul characters using both horizontal and vertical run-lengths, secondly extracts horizontal and vertical strokes of a character utilizing the strokes' width, and finally detects the intersection points of the strokes exploiting horizontal and vertical strokes. The analysis of both the horizontal and the vertical strokes doesn't use the strokes' angles but both the strokes' width and the changes of the run-lengths. The intersection points of the strokes become the candidated parts for phoneme segmentation, which is one of main techniques for off-line handwritten Hangul recognition. The segmented strokes represent the feature for handwritten Hangul recognition.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.6
• /
• pp.1167-1174
• /
• 2006

Text region detection from a natural scene is useful in many applications such as vehicle license plate recognition. Therefore, in this paper, we propose a text region extraction method using pattern histogram of character-edge maps. We create 16 kinds of edge maps from the extracted edges and then, we create the 8 kinds of edge maps which compound 16 kinds of edge maps, and have a character feature. We extract a candidate of text regions using the 8 kinds of character-edge maps. The verification about candidate of text region used pattern histogram of character-edge maps and structural features of text region. Experimental results show that the proposed method extracts a text regions composed of complex background, various font sizes and font colors effectively.

• Journal of Information Processing Systems
• /
• v.14 no.3
• /
• pp.666-679
• /
• 2018

The discrete wavelet transform (DWT) has good multi-resolution decomposition characteristic and its low frequency component contains the basic information of an image. Based on this, a fragile watermarking using the local binary pattern (LBP) and DWT is proposed for image authentication. In this method, the LBP pattern of low frequency wavelet coefficients is adopted as a feature watermark, and it is inserted into the least significant bit (LSB) of the maximum pixel value in each block of host image. To guarantee the safety of the proposed algorithm, the logistic map is applied to encrypt the watermark. In addition, the locations of the maximum pixel values are stored in advance, which will be used to extract watermark on the receiving side. Due to the use of DWT, the watermarked image generated by the proposed scheme has high visual quality. Compared with other state-of-the-art watermarking methods, experimental results manifest that the proposed algorithm not only has lower watermark payloads, but also achieves good performance in tamper identification and localization for various attacks.

• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.669-672
• /
• 2018

캡술내시경은 식도부터 항문까지 소화기관 전체를 한번에 촬영할 수 있는 의료기기로, 한번의 검사에 평균 8~12 시간 정도의 길이와 5만장 이상의 프레임으로 구성된 영상을 생성한다. 그러나 생성된 영상에 대한 분석은 수작업으로 진행되고 있어, 캡술내시경 영상 분석 자동화에 대한 기술적인 수요가 높아지고 있는 추세이다. 이를 위해, 캡슐내시경 영상 분석에 대한 많은 연구가 진행되고 있는데, 본 연구에서는 그 중에서도 폴립 영상에 대한 검출 자동화 연구에 주목하였다. 폴립이란 위장관 내에서 발견될 수 있는 융기성 병변으로, 많은 연구에서 기계학습 혹은 딥러닝 방식을 적용하여 이를 검출하기 위한 연구를 수행하였다. 그러나 캡슐내시경 영상의 특성상, 병번이 있는 영상이 굉장히 적기 때문에 일반적인 딥러닝 방식의 적용으로 좋은 성능을 내기 어렵다. 따라서 본 논문에서는 폴립 검출 컨볼루션 신경망 설계를 위한 멀티 스케일에 대한 원형 검출기법을 결합하여 폴립이 의심되는 영역을 추출해주는 특징 추출 기법으로, 수집한 데이터 150장에 대한 실험한 결과 약 82%의 성능을 보였다.

• Annual Conference of KIPS
• /
• 2015.04a
• /
• pp.776-778
• /
• 2015

논문에서는 DNN(Deep Neural Network)의 dropout 기법을 이용하여 산란계가 고온 스트레스를 받고 있는지 여부를 닭의 울음소리 정보를 통해 탐지하는 방법을 제안한다. 실험에서는 $21^{\circ}C$ 정상 온도에서 100개의 소리 데이터, $35^{\circ}C$ 고온에서 200개의 소리 데이터를 사용한다. 먼저, DNN의 학습을 위해서 취득한 울음소리에서 54개의 소리 특징 정보를 추출한다. 둘째, CFS(Correlation Feature Selection)을 이용하여, 추출된 특징 중 온도 구분을 위한 중요한 특정 10개를 선택한다. 셋째, 선택된 소리특징을 DNN에 적용하여 온도 환경을 구분하는 시스템이다. DNN의 과적합(over-fitting) 영향을 감소시키고, 성능 향상을 위하여 dropout 비율을 조정하여 실험을 진행하였다. 본 연구에서는 실제 계사에서 수집된 소리 정보를 이용하여 모의실험을 수행한 결과 매우 우수한 성능을 보임을 확인하였다.

• The KIPS Transactions:PartB
• /
• v.11B no.3
• /
• pp.257-266
• /
• 2004

This paper proposes caricature generation system that uses shading mechanism that extracts textural features of face. Using this method, we can get more realistic caricature. Since this system If vector-based, the generated character's face has no size limit and constraint. so it is available to transform the shape freely and to apply various facial expressions to 2D face. Moreover, owing to the vector file's advantage, It can be used in mobile environment as small file size This paper presents methods that generate vector-based face, create shade and synthesize the shade with the vector face.

• Korean Journal of Remote Sensing
• /
• v.26 no.3
• /
• pp.325-333
• /
• 2010

Image registration involves overlapping images of an identical region and assigning the data into one coordinate system. Image registration has proved important in remote sensing, enabling registered satellite imagery to be used in various applications such as image fusion, change detection and the generation of digital maps. The image descriptor, which extracts matching points from each image, is necessary for automatic registration of remotely sensed data. Using contrast enhancement algorithms such as histogram equalization and image stretching, the normalized data are applied to the image descriptor. Drawing on the different spectral characteristics of high resolution satellite imagery based on sensor type and acquisition date, the applied normalization method can be used to change the results of matching interest point descriptors. In this paper, the matching points by scale invariant feature transformation (SIFT) are extracted using various contrast enhancement algorithms and injection of Gaussian noise. The results of the extracted matching points are compared with the number of correct matching points and matching rates for each point.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.5
• /
• pp.397-402
• /
• 2019

Recently, with the emergence of autonomous vehicles and the increasing interest in safety, a variety of research has been being actively conducted to precisely estimate the position of a vehicle by fusing sensors. Previously, researches were conducted to determine the location of moving objects using GNSS (Global Navigation Satellite Systems) and/or IMU (Inertial Measurement Unit). However, precise positioning of a moving vehicle has lately been performed by fusing data obtained from various sensors, such as LiDAR (Light Detection and Ranging), on-board vehicle sensors, and cameras. This study is designed to enhance kinematic vehicle positioning performance by using feature-based recognition. Therefore, an analysis of the required precision of the observations obtained from the images has carried out in this study. Velocity and attitude observations, which are assumed to be obtained from images, were generated by simulation. Various magnitudes of errors were added to the generated velocities and attitudes. By applying these observations to the positioning algorithm, the effects of the additional velocity and attitude information on positioning accuracy in GNSS signal blockages were analyzed based on Kalman filter. The results have shown that yaw information with a precision smaller than 0.5 degrees should be used to improve existing positioning algorithms by more than 10%.

• Journal of Korea Multimedia Society
• /
• v.25 no.11
• /
• pp.1547-1556
• /
• 2022

As mechanical facilities are interacting with each other, the failure of some equipment can affect the entire system, so it is necessary to quickly detect and diagnose the abnormality of mechanical equipment. This study proposes a deep learning model that can effectively diagnose abnormalities in rotating machinery and equipment. CNN is widely used for feature extraction and LSTMs are known to be effective in learning sequential information. In LSTM, the number of parameters and learning time increase as the length of input data increases. In this study, we propose a method of segmenting an input segment signal into shorter-length sub-segment signals, sequentially inputting them to CNN through a time-distributed method for extracting features, and inputting them into LSTM. A failure diagnosis test was performed using the vibration data collected from the motor for ventilation equipment installed at the urban railway station. The experiment showed an accuracy of 99.784% in fault diagnosis. It shows that the proposed method is effective in the fault diagnosis of rotating machinery and equipment.