• Journal of Korea Multimedia Society
• /
• v.22 no.12
• /
• pp.1347-1355
• /
• 2019

Deep learning approach based on convolution neural network (CNN) has extensively studied in the field of computer vision. However, periocular feature extraction using CNN was not well studied because it is practically impossible to collect large volume of biometric data. This study uses the ResNet model which was trained with the ImageNet dataset. To overcome the problem of insufficient training data, we focused on the training of multi-layer perception (MLP) having simple structure rather than training the CNN having complex structure. It first extracts features using the pretrained ResNet model and reduces the feature dimension by principle component analysis (PCA), then trains a MLP classifier. Experimental results with the public periocular dataset UBIPr show that the proposed method is effective in person authentication using periocular region. Especially it has the advantage which can be directly applied for other biometric traits.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.11b
• /
• pp.118-121
• /
• 2002

The study on shilla monumental inscription has been accomplished by many historians. However, the research on segmentation of monumental inscription image using digital image processing is not sufficient for restoration of the image. Although, many image processing methods have been proposed for region extraction in still image, there is no suitable method for accurate interpretation of monumental inscription image. To distinguish foreground and background region in the image, this paper presents new segmentation algorithm composed of contrast adjustment and median filtering, thresholding and sobel operation, as pre-processing and post-processing. The result show that background and foreground regions are segmented in monumental inscription image.

• Journal of information and communication convergence engineering
• /
• v.15 no.1
• /
• pp.49-52
• /
• 2017

Ganglion cysts are benign soft tissues usually encountered in the wrist. In this paper, we propose a method to extract a ganglion cyst region from ultrasonography images by using image segmentation. The proposed method using the possibilistic c-means (PCM) clustering method is applicable to ganglion cyst extraction. The methods considered in this thesis are fuzzy stretching, median filter, PCM clustering, and connected component labeling. Fuzzy stretching performs well on ultrasonography images and improves the original image. Median filter reduces the speckle noise without decreasing the image sharpness. PCM clustering is used for categorizing pixels into the given cluster centers. Connected component labeling is used for labeling the objects in an image and extracting the cyst region. Further, PCM clustering is more robust in the case of noisy data, and the proposed method can extract a ganglion cyst area with an accuracy of 80% (16 out of 20 images).

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.827-828
• /
• 2008

Blur identification is the first and the most important step of restoring images. Edge region of the image usually conveys important information of blur parameters. In this paper we propose a region-based edge extraction method for estimating point-spread-function (PSF). As a result, the proposed method can detect the starting and the ending points of a step response, and provides the PSF parameters to the restoration process.

• Journal of the Semiconductor & Display Technology
• /
• v.2 no.2
• /
• pp.17-20
• /
• 2003

The current drive in an MOSFET is limited by the intrinsic channel resistance. All the other parasitic elements in a device structure playa significant role and degrade the device performance. These other resistances need to be less than 10%-20% of the channel resistance. To achieve the requirements, we should investigate a methodology of separation and quantification of those resistances. In this paper, we developed the extraction method of resistances using calibrated TCAD simulation. The resistance of the extension region is also partially determined by the formation of a surface accumulation region that forms under the gate in the tail region of the extension profile.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.247-250
• /
• 2003

This paper describes the extraction method for brain region using cellular automata from the brain MR image. In the first removing the background from the brain MR image, and then extracting the brain region by applying the cellular automata rule obtained from histogram analysis information. The results on some experimental results showed that the PSNR is 42.11(dB) on image quality and also the correlation factor is estimated 98.46%. The result of this study can be used as the auto-diagnostics system.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2515-2517
• /
• 2002

The process of character recognition goes through 5 steps; image acquisition, character region extraction, preprocessing, character region segmentation, character recognition. Therefore the final recognition rate of character recognition is directly affected by the performance of each step. This paper is a leading research for object recognition using image processing algorithm which is one of the field of study in computer vision. And this paper will suggest an algorithm to extract the portion of number chain, which is part of the research embodying a system to perceive the data of manufacture and the name of the producer on the wrapping of groceries. In addition, this can extract the number chain comparatively accurate without using many complex algorithm by diving and extracting the moving number region at the same time.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.408-410
• /
• 2014

In this paper a license plate area extraction of labeling the vehicle images is proposed. Studies on license plate recognition systems have largely been conducted and there is a tendency of increasing license plate recognition rates. In this paper a license plate region is extracted from an image labeling for the region of interest and research on technology for labeling sample image using the Otsu algorithm to binary.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.157-165
• /
• 2021

Recently, as the spatial resolution of satellite and aerial images has improved, various studies using remotely sensed data with high spatial resolution have been conducted. In particular, since the building extraction is essential for creating digital thematic maps, high accuracy of building extraction result is required. In this manuscript, building extraction models were generated using SegNet, U-Net, FC-DenseNet, and HRNetV2, which are representative semantic segmentation models in deep learning techniques, and then the evaluation of building extraction results was performed. Training dataset for building extraction were generated by using aerial orthophotos including various buildings, and evaluation was conducted in three areas. First, the model performance was evaluated through the region adjacent to the training dataset. In addition, the applicability of the model was evaluated through the region different from the training dataset. As a result, the f1-score of HRNetV2 represented the best values in terms of model performance and applicability. Through this study, the possibility of creating and modifying the building layer in the digital map was confirmed.

• 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.