• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.3
• /
• pp.209-219
• /
• 2019

Satellite imagery occurs geometric and radiometric errors due to external environmental factors at the acquired time, which in turn causes false-alarm in change detection. These errors should be eliminated by geometric and radiometric corrections. In this study, we propose a methodology that automatically and simultaneously performs geometric and radiometric corrections by using the SURF (Speeded-Up Robust Feature) algorithm and the mask filter. The MPs (Matching Points), which show invariant properties between multi-temporal imagery, extracted through the SURF algorithm are used for automatic geometric correction. Using the properties of the extracted MPs, PIFs (Pseudo Invariant Features) used for relative radiometric correction are selected. Subsequently, secondary PIFs are extracted by generated mask filters around the selected PIFs. After performing automatic using the extracted MPs, we could confirm that geometric and radiometric errors are eliminated as the result of performing the relative radiometric correction using PIFs in geo-rectified images.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1252-1255
• /
• 2002

In this paper, face recognition algorithm, by using skin color information of HSI color coordinate collected from face images, elliptical mask, fratures of face including eyes, nose and mouth, and geometrical feature vectors of face and facial angles, is proposed. The proposed algorithm improved face region extraction efficacy by using HSI information relatively similar to human's visual system along with color tone information about skin colors of face, elliptical mask and intensity information. Moreover, it improved face recognition efficacy with using feature information of eyes, nose and mouth, and Θ1(ACRED), Θ2(AMRED) and Θ 3(ANRED), which are geometrical face angles of face. In the proposed algorithm, it enables exact face reading by using color tone information, elliptical mask, brightness information and structural characteristic angle together, not like using only brightness information in existing algorithm. Moreover, it uses structural related value of characteristics and certain vectors together for the recognition method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.12
• /
• pp.984-993
• /
• 2003

In Gravure printing process, there are a lot of printing errors caused by expansion and contraction of printing materials and difficulty of printing of small letters, accordingly we cannot detect those errors with eyes. In this paper, we describe the algorithm which can detect small errors automatically in Gravure printing process and a real-time detection system adopting the algorithm. We present the Perimetric Mask algorithm that can eliminate tiny errors occurring near the contour of printing objects to achieve accurate inspection, and also construct an algorithm utilizing symmetry transform which can emphasize tiny errors to make a robust inspection system. We have made a system running in real-time and verified the efficiency of the algorithm.

• International journal of advanced smart convergence
• /
• v.8 no.3
• /
• pp.131-137
• /
• 2019

This study proposes a method of learning and recognizing the characteristics that are the classification criteria of Hangul using Mask R-CNN, one of the deep learning techniques, to recognize and classify atypical Hangul characters. The atypical characters on the Hangul signboard have a lot of deformed and colorful shapes beyond the general characters. Therefore, in order to recognize the Hangul signboard character, it is necessary to learn a separate atypical Hangul character rather than the existing formulaic one. We selected the Hangul character '닭' as sample data and constructed 5,383 Hangul image data sets and used them for learning and verifying the deep learning model. The accuracy of the results of analyzing the performance of the learning model using the test set constructed to verify the reliability of the learning model was about 92.65% (the area detection rate). Therefore we confirmed that the proposed method is very useful for Hangul signboard character recognition, and we plan to extend it to various Hangul data.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.88-90
• /
• 2021

본 논문은 화학 문서 내에 존재하는 표를 검출하는 문제를 다룬다. 우선 문서에서 표가 있을 만한 영역만을 남긴 후, 객체 검출 분야에서 좋은 성능을 보이는 Cascade Mask R-CNN을 이용하여 화학 문서 내 표 검출을 수행하였다. 더 나아가 감마 보정과 스캔 잡음을 이용하여 학습 데이터를 증강함으로써 다양한 스타일의 표들을 강인하게 검출할 수 있도록 하였다. 합성 화학문서와 실제 화학 문서에 대해 제안한 방법을 적용하여 표 검출 성능을 측정하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.130-132
• /
• 2012

The use of visual media has been increased by development of contemporary society. To use these information of image, there are various methods of image processing. Edge detection which is one of those is technique to detect dramatically changing part of image brightness. Existing methods detect edge through mask which is composited by constant values. Because existing methods do not consider factor as location, direction of pixel in image, performance of edge detecting in insufficient. Therefore, an algorithm which is using variable vector for the variation of brightness in mask of $3{\times}3$ pixels is proposed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.9
• /
• pp.1865-1870
• /
• 2011

In this paper, we propose an effective method that automatically detects flaws in air deck by using non-destructive testing. First, Gamma correlation transform, 7 ${\times}$ 7 and 13 ${\times}$ 13 Sobel mask apply to the image of air deck acquired non-destructive testing in order to detect the edge of the image. Second, the edge detection area is smoothed and corrected by mean binarization method. Finally, the region of flaws in air deck is detected by a labeling method after removing the noise by the erosion and the dilation operation. In experimental results, we showed that the proposed detection method is effective in air deck.

• Journal of the Korean Society for Precision Engineering
• /
• v.17 no.3
• /
• pp.58-67
• /
• 2000

Electron gun heaters are used to heat a cathode in video(TV) monitors. Major defects of the electron gun heaters include dimensional inaccuracy and pollution with dirty materials. In this paper, to save the labor and time being taken to inspect the heaters, a machine vision system is considered. For the system, a new algorithm is developed to measure the 9 different dimensions of each heater and to detect polluted defects. The algorithm consists of three stages. In the first stage, the center of the heater image is obtained and then its boundary detection is performed. For the efficient boundary detection, a mask called the sum mask is used. In the second stage of the algorithm, a set of fiducial points are determined on the boundary image. Finally, using the fiducial points specified dimensions are measured and the amount of polluted area is computed in the third stage. The performance of the algorithm is evaluated for a set of real specimens. The results indicate that measurements obtained by the algorithm satisfy the tolerance limits fur most of the dimensions and the algorithm detects the polluted defects successfully.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.201-205
• /
• 2004

The digital color image can be distorted by noise for a transmission or other elements of system. It happens to vague of a boundary side in the division of a color image object, especially, boundary side of an input color image is very important because it can be determined to the division and detection element in pattern recognition. Therefore it is boundary part In this paper, it detects the optimal edge with applying this color image to WCNN algorithm, after it does level up a boundary side of a color image by using the adaptive morphology as the threshold of an input color image. Also, it is used not a conventional fixed mask edge detection method but variable mask method which is cal led a variable BBM. It is confirmed by simulation that the proposed algorithm can be got the batter result edge at the place of closing to each edges and having smoothly curved line.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.4
• /
• pp.473-480
• /
• 2004

This paper presents a new edge detection algorithm in color image. The proposed Adaptive morphological Wavelet-CNN algorithm is divided into two parts : The Adaptive morpholog and WCNN(Wavelet Cellular Neural Networks). It detects the optimal edge with applying this color image to WCNN algorithm, after it does level up a boundary side of a color image by using the adaptive morphology as the threshold of an input color image. Also, it is used not a conventional fixed mask edge detection method but variable mask method which is called a variable BBM. Finally, to show the feasibility of the proposed algorithm, this paper provides by simulation that the color image consists of 30.