• Journal of the Institute of Convergence Signal Processing
• /
• 제14권3호
• /
• pp.162-168
• /
• 2013

This paper examines the implementation of automatic compensation modules for digital camera image when a person is closing his/her eyes. The modules detect the face and eye region and then recognize the eye state. If the image is taken when a person is closing his/her eyes, the function corrects the eye and produces the image by using the most satisfactory image of the eye state among the past frames stored in the buffer. In order to recognize the face and eye precisely, the pre-process of image correction is carried out using SURF algorithm and Homography method. For the detection of face and eye region, Haar-like feature algorithm is used. To decide whether the eye is open or not, similarity comparison method is used along with template matching of the eye region. The modules are tested in various facial environments and confirmed to effectively correct the images containing faces.

• Korean Journal of Veterinary Research
• /
• 제36권1호
• /
• pp.195-207
• /
• 1996

To make the genomic DNA probe of Theileria sergenti, the merozoites were purified from erythrocytes of Korean cattle, The previous studies on the probe of T sergenti had resulted in two probes as KTS1 and KTS3 DNA fragment. Nucleotide sequence of both ends of the KTS1 and KST3 were determined in order to design primers for polymerase chain reaction. A pair of an uper primer(5'-CCTCTTGAAGTCATCCATGT-3'; nucleotide position 48) and a lower primer(5'-CACTGAGCTG GAAAGAGCTA-3'; nucleotide position 156) in pKTS1 were synthesized. The anticipated PCR product was 128bp in length. To examine the sensitivity of the PCR, KTS1 DNA and purified T sergenti DNA were serially diluted by tenfolds with distilled water. The primers were sensitive enough to detect 4ag of the authentic template DNA and 4fg of the purified T sergenti DNA by PCR. Furthermore, when the blood was serially diluted by two-folds with 0.9% saline, the pair could detect up to 0.00029%(about 164 parasites in $10{\mu}l$ of blood) of T sergenti infection in bovine erythrocytes by PCR. In a comparison of microscopic and PCR detection of T sergenti in the same samples from Chonbuk area, 47 and 51 out of 70 sample(67.1%) were positive by the former and by the latter method, respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권9호
• /
• pp.4491-4509
• /
• 2017

The present paper proposes a novel dynamic system for hand gesture recognition. The approach involved is comprised of three main steps: detection, tracking and recognition. First, the gesture contour captured by a 2D-camera is detected by combining the three-frame difference method and skin-color elliptic boundary model. Then, the trajectory of the hand gesture is extracted via a gesture-tracking algorithm based on an occlusion-direction oriented linear extrapolation predictor, where the gesture coordinate in next frame is predicted by the judgment of current occlusion direction. Finally, to overcome the interference of insignificant trajectory segments, the longest common subsequence (LCS) is employed with the aid of velocity information. Besides, to tackle the subgesture problem, i.e., some gestures may also be a part of others, the most probable gesture category is identified through comparison of the relative LCS length of each gesture, i.e., the proportion between the LCS length and the total length of each template, rather than the length of LCS for each gesture. The gesture dataset for system performance test contains digits ranged from 0 to 9, and experimental results demonstrate the robustness and effectiveness of the proposed approach.

• Journal of Broadcast Engineering
• /
• 제27권3호
• /
• pp.273-282
• /
• 2022

MPEG-VCM(Video Coding for Machine) aims to standardize video codec for machines. VCM provides data sets and anchors, which provide reference data for comparison, for several machine vision tasks including object detection, object segmentation, and object tracking. The evaluation template can be used to compare compression and machine vision task performance between anchor data and various proposed video codecs. However, performance comparison is carried out separately for each machine vision task, and information related to performance evaluation of multiple machine vision tasks on a single bitstream is not provided currently. In this paper, we propose a performance evaluation method of a video codec for AI-based multi-tasks. Based on bits per pixel (BPP), which is the measure of a single bitstream size, and mean average precision(mAP), which is the accuracy measure of each task, we define three criteria for multi-task performance evaluation such as arithmetic average, weighted average, and harmonic average, and to calculate the multi-tasks performance results based on the mAP values. In addition, as the dynamic range of mAP may very different from task to task, performance results for multi-tasks are calculated and evaluated based on the normalized mAP in order to prevent a problem that would be happened because of the dynamic range.

• Journal of the Korea Computer Graphics Society
• /
• 제20권2호
• /
• pp.1-11
• /
• 2014

The quality of 2D-to-3D conversion depends on the accuracy of the assigned depth to scene objects. Manual depth painting for given objects is labor intensive as each frame is painted. Specifically, a human is one of the most challenging objects for a high-quality conversion, as a human body is an articulated figure and has many degrees of freedom (DOF). In addition, various styles of clothes, accessories, and hair create a very complex silhouette around the 2D human object. We propose an efficient method to estimate visually pleasing depths of a human at every frame in a monocular video. First, a 3D template model is matched to a person in a monocular video with a small number of specified user correspondences. Our pose estimation with sequential joint angular constraints reproduces a various range of human motions (i.e., spine bending) by allowing the utilization of a fully skinned 3D model with a large number of joints and DOFs. The initial depth of the 2D object in the video is assigned from the matched results, and then propagated toward areas where the depth is missing to produce a complete depth map. For the effective handling of the complex silhouettes and appearances, we introduce a partial depth propagation method based on color segmentation to ensure the detail of the results. We compared the result and depth maps painted by experienced artists. The comparison shows that our method produces viable depth maps of humans in monocular videos efficiently.