• The Journal of Korea Robotics Society
• /
• v.12 no.2
• /
• pp.217-227
• /
• 2017

Crosswalk detection is an important part of the Pedestrian Protection System in autonomous vehicles. Different methods of crosswalk detection have been introduced so far using crosswalk edge features, the distance between crosswalk blocks, laser scanning, Hough Transformation, and Fourier Transformation. However, most of these methods failed to detect crosswalks accurately, when they are damaged, faded away or partly occluded. Furthermore, these methods face difficulties when applying on real road environment where there are lot of vehicles. In this paper, we solve this problem by first using a region based binarization technique and x-axis histogram to detect the candidate crosswalk areas. Then, we apply Support Vector Machine (SVM) based classification method to decide whether the candidate areas contain a crosswalk or not. Experiment results prove that our method can detect crosswalks in different environment conditions with higher recognition rate even they are faded away or partly occluded.

• Journal of Korea Multimedia Society
• /
• v.18 no.10
• /
• pp.1216-1224
• /
• 2015

In this paper, a detection algorithm for the removal of needle in oriental clinic is presented. First, in the proposed method, the candidate areas of each needle penetrated are selected by using the RGB color information of needle head, and the false candidates are removed by considering their area size. Next, two main edges of the needle are extracted through using the edges of selected candidate areas and their radon transformation. The final verification of penetrated needle is accomplished by using the morphological analysis of these two edge lines. In the experiments, the detection rate of proposed method reaches to 99% for the 36 images containing 294 needles.

• International Journal of Internet, Broadcasting and Communication
• /
• v.9 no.1
• /
• pp.51-55
• /
• 2017

LINE, an instant message App with a powerful capability of transmitting various forms of data, has been overwhelmed in Asia since launched in 2011. Due to its popularity, LINE was first used in the 2016 Taiwanese presidential election. This research utilized a functional approach of campaign communication discourse and political visual images to analyze how candidates managed and presented themselves by textual and visual information on LINE. Regarding the textual information, results revealed their strategy inclined to reverse gender stereotype because the female candidate emphasized policy over character, while the male emphasized character over policy. Both candidates did not fully employ ten image functions because they utilized mostly the emotional and image building functions. The female candidate message achieved the largest total amount of 'like' and 'share'. This study probes into the App campaigning and improve the feasibility and practicability to share knowledge of political communication by new media.

• The Journal of the Korea Contents Association
• /
• v.5 no.2
• /
• pp.199-205
• /
• 2005

In this paper, we present a face detection algorithm which uses the features of color and Geometry of faces and hairs appeared in images. after candidate area detection using color features, background areas are removed by the deviation of luminance in each of candidate areas. And then, final face area is detected using feature of geometry between face and hair. Performance of the presented algorithm is evaluated by detection rate test. The test result showed high detection rate.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.559-561
• /
• 2003

Human face detection plays an important role in variable applications. A face detection method based on skin-color information and facial feature in color images is proposed in this paper. First, the RGB color space is transformed to YCbCr space and only the skin region is extracted with the skin color information. And then, the candidate where face is likely to exist is selected after labeling processing. Finally, we detect facial features in face candidate. The experimental results show that the method proposed here is effective.

• International journal of advanced smart convergence
• /
• v.8 no.2
• /
• pp.116-125
• /
• 2019

The fire should extinguish as soon as possible because it causes economic loss and loses precious life. In this study, we propose a new atypical fire and smoke detection algorithm using deep learning and color histogram of fire and smoke. First, input frame images obtain from the ONVIF surveillance camera mounted in factory search motion candidate frame by motion detection algorithm and mean square error (MSE). Second deep learning (Faster R-CNN) is used to extract the fire and smoke candidate area of motion frame. Third, we apply a novel algorithm to detect the fire and smoke using color histogram algorithm with local area motion, similarity, and MSE. In this study, we developed a novel fire and smoke detection algorithm applied the local motion and color histogram method. Experimental results show that the surveillance camera with the proposed algorithm showed good fire and smoke detection results with very few false positives.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.2 s.308
• /
• pp.12-18
• /
• 2006

This paper presents a method to extract a good high-resolution image from a low-resolution video in an automatic manner. Since a high-resolution image reconstruction method utilizing several low-resolution input images works better than a conventional interpolation method utilizing single low-resolution input image only if the input images are well registered onto a common high-resolution grid, low-resolution input images should be carefully chosen so that the registration errors can be carefully considered. In this paper, the statistics obtained from the motion-compensated low-resolution images are utilized to evaluate the feasibility of the input image candidates. Maximum motion-compensation error is estimated from the high-resolution image observation model. U the motion-compensation error of the input image candidate is greater than the estimated maximum motion-compensation error, the input image candidate is discarded. The number of good input image candidates and the statistics of the motion-compensation errors are used to choose final input images. The final input images chosen from the input image selection block are given to the following high-resolution image reconstruction block. It is expected that the proposed method is utilized to extract a good high-resolution image efficiently from a low-resolution video without any user intervention.

• Journal of the Institute of Convergence Signal Processing
• /
• v.9 no.1
• /
• pp.18-23
• /
• 2008

In this paper, we propose the input image selection-method to improve the reconstructed high-resolution (HR) image quality. To obtain ideal super-resolution (SR) reconstruction image, all input images are well-registered. However, the registration is not ideal in practice. Due to this reason, the selection of input images with low registration error (RE) is more important than the number of input images in order to obtain good quality of a HR image. The suitability of a candidate input image can be determined by using statistical and restricted registration properties. Therefore, we propose the proper candidate input Low Resolution(LR) image selection-method as a pre-processing for the SR reconstruction in automatic manner. In video sequences, all input images in specified region are allowed to use SR reconstruction as low-resolution input image and/or the reference image. The candidacy of an input LR image is decided by the threshold value and this threshold is calculated by using the maximum motion compensation error (MMCE) of the reference image. If the motion compensation error (MCE) of LR input image is in the range of 0 < MCE < MMCE then this LR input image is selected for SR reconstruction, else then LR input image are neglected. The optimal reference LR (ORLR) image is decided by comparing the number of the selected LR input (SLRI) images with each reference LR input (RLRI) image. Finally, we generate a HR image by using optimal reference LR image and selected LR images and by using the Hardie's interpolation method. This proposed algorithm is expected to improve the quality of SR without any user intervention.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.265-269
• /
• 2007

In this paper, we propose an automatic extraction method of gastric cancer region from electronic endoscopic images. We use the brightness and saturation of HSI in removing noises by illumination and shadows by the crookedness occurring in the endoscopic process. We partition the image into several areas with similar pigments of hemoglobin using IHb. The candidate areas for gastric cancer are defined as the areas that have high hemoglobin pigments and high value in every channel of RGB. Then the morphological characteristics of gastric cancer are used to decide the target region. In experiment, our method is sufficiently accurate in that it correctly identifies most cases (18 out of 20 cases) from real electronic endoscopic images, obtained by expert endoscopists.

• Journal of Korea Multimedia Society
• /
• v.3 no.6
• /
• pp.575-584
• /
• 2000

This paper presents a new image retrieval system with color and morphological informations based on binary sets. Each of them can be obtained from color binary sets and regional segmentation separately. For retrieval processes, the candidate images are decided by comparing color and their image binary sets of the database with query images. Particularly, it is possible that the retrieval of similar-measurements has a weight of color spatial distribution and its objective morphological features. We proposed a new idea for performing simply the complicated similar-measurement of candidated images to improve queried processes. The retrieval method using spatial and morphological features is shown with the effectiveness on the result of implementation on database with 3,000 images.