• Journal of Broadcast Engineering
• /
• v.17 no.3
• /
• pp.437-446
• /
• 2012

A face detection algorithms using two-dimensional (2-D) intensity or color images have been studied for decades. Recently, with the development of low-cost range sensor, three-dimensional (3-D) information (i.e., depth image that represents the distance between a camera and objects) can be easily used to reliably extract facial features. Most people have a similar pattern of 3-D facial structure. This paper proposes a face detection method using intensity and depth images. At first, adaboost algorithm using intensity image classifies face and nonface candidate regions. Each candidate region is divided into $5{\times}5$ blocks and depth values are averaged in each block. Then, $5{\times}5$ block rank pattern is constructed by sorting block averages of depth values. Finally, candidate regions are classified as face and nonface regions by matching the constructed depth map based block rank patterns and a template pattern that is generated from training data set. For template matching, the $5{\times}5$ template block rank pattern is prior constructed by averaging block ranks using training data set. The proposed algorithm is tested on real images obtained by Kinect range sensor. Experimental results show that the proposed algorithm effectively eliminates most false positives with true positives well preserved.

• Journal of Broadcast Engineering
• /
• v.5 no.1
• /
• pp.113-122
• /
• 2000

It is sometimes necessary to remove the captions and recover original images from video images already broadcast, When the number of images requiring such recovery is small, manual processing is possible, but as the number grows it would be very difficult to do it manually. Therefore, a method for recovering original image for the caption areas in needed. Traditional research on image restoration has focused on restoring blurred images to sharp images using frequency filtering or video coding for transferring video images. This paper proposes a method for automatically recovering original image using BMA(Block Matching Algorithm). We extract information on caption regions and scene change that is used as a prior-knowledge for recovering original image. From the result of caption information detection, we know the start and end frames of captions in video and the character areas in the caption regions. The direction for the recovery is decided using information on the scene change and caption region(the start and end frame for captions). According to the direction, we recover the original image by performing block matching for character components in extracted caption region. Experimental results show that the case of stationary images with little camera or object motion is well recovered. We see that the case of images with motion in complex background is also recovered.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.5
• /
• pp.56-65
• /
• 2009

Stereoscopic images consist of the left image and the right image. Thus, stereoscopic images have much amounts of data than single image. Then an efficient image compression technique is needed, the DPCM-based predicted coding compression technique is used in most video coding standards. Motion and disparity estimation are needed to realize the predicted coding compression technique. Their performing algorithm is block matching algorithm used in most video coding standards. Full search algorithm is a base algorithm of block matching algorithm which finds an optimal block to compare the base block with every other block in the search area. This algorithm presents the best efficiency for finding optimal blocks, but it has very large computational loads. In this paper, we have proposed fast disparity estimation algorithm using motion and disparity vector information of the prior frame in stereo image coding. We can realize fast disparity vector estimation in order to reduce search area by taking advantage of global disparity vector and to decrease computational loads by limiting search points using motion vectors and disparity vectors of prior frame. Experimental results show that the proposed algorithm has better performance in the simple image sequence than complex image sequence. We conclude that the fast disparity vector estimation is possible in simple image sequences by reducing computational complexities.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.05a
• /
• pp.389-398
• /
• 2007

Case-based reasoning (CBR) is one of the most popular prediction techniques for medical diagnosis because it is easy to apply, has no possibility of overfitting, and provides a good explanation for the output. However, it has a critical limitation - its prediction performance is generally lower than other artificial intelligence techniques like artificial neural networks (ANNs). In order to obtain accurate results from CBR, effective retrieval and matching of useful prior cases for the problem is essential, but it is still a controversial issue to design a good matching and retrieval mechanism for CBR systems. In this study, we propose a novel approach to enhance the prediction performance of CBR. Our suggestion is the simultaneous optimization of feature weights, instance selection, and the number of neighbors that combine using genetic algorithms (GAs). Our model improves the prediction performance in three ways - (1) measuring similarity between cases more accurately by considering relative importance of each feature, (2) eliminating redundant or erroneous reference cases, and (3) combining several similar cases represent significant patterns. To validate the usefulness of our model, this study applied it to a real-world case for evaluating cytological features derived directly from a digital scan of breast fine needle aspirate (FNA) slides. Experimental results showed that the prediction accuracy of conventional CBR may be improved significantly by using our model. We also found that our proposed model outperformed all the other optimized models for CBR using GA.

• Journal of KIISE:Software and Applications
• /
• v.37 no.1
• /
• pp.80-84
• /
• 2010

Tracking hands is an essential technique for hand gesture recognition which is an efficient way in Human Computer Interaction (HCI). Recently, many researchers have focused on hands tracking using a 3D hand model and showed robust tracking results compared to using 2D hand models. In this paper, we propose a novel 3D hand tracking method based on a coupled particle filter. This provides robust and fast tracking results by estimating each part of global hand poses and local finger motions separately and then utilizing the estimated results as a prior for each other. Furthermore, in order to improve the robustness, we apply a multi-cue based method by integrating a color-based area matching method and an edge-based distance matching method. In our experiments, the proposed method showed robust tracking results for complex hand motions in a cluttered background.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.6
• /
• pp.287-295
• /
• 2007

This paper proposes a novel data editing techniques with genetic algorithm (GA) in case-based reasoning (CBR) for the prediction of Korea Stock Price Index (KOSPI). CBR has been widely used in various areas because of its convenience and strength in compelax problem solving. Nonetheless, compared to other machine teaming techniques, CBR has been criticized because of its low prediction accuracy. Generally, in order to obtain successful results from CBR, effective retrieval of useful prior cases for the given problem is essential. However. designing a good matching and retrieval mechanism for CBR system is still a controversial research issue. In this paper, the GA optimizes simultaneously feature weights and a selection task for relevant instances for achieving good matching and retrieval in a CBR system. This study applies the proposed model to stock market analysis. Experimental results show that the GA approach is a promising method for data editing in CBR.

• Korean Journal of Social Welfare
• /
• v.60 no.4
• /
• pp.55-77
• /
• 2008

After the economic crisis of Korea, the character of Korean Poverty has changed. Most notably, many people are working but poor. Therefore it is important to understand the characteristics of the working poor, especially the unstable work experiences of the working poor since one of the causes of poverty is that. Prior research about the working poor has not fully proven this issue. This study is to examine the job sequences of the working poor. Thus I utilized the KLIPS(Korea Labor and Income Study), and analyzed it by event sequence analysis and optimal matching methods. The job sequences are divided as follows: total years of working in the labor market, the number of gaps and the length of gaps in their careers, and the characteristics of experienced jobs since they have entered the labor market from age 15. As a result, there are no statistically significant in the total years of working in the labor market. And the number of gaps and the length of gaps in their careers, and the characteristics of experienced jobs show that working poor have been experiencing more unstable than non-poor. Thus, almost all of the male working poor has unstable jobs their whole lives, and the female working poor's job sequences show distinct features according to women's life course. These results can give political implications to the anti-poverty policy in Korea.

• Journal of the Korean Society for Precision Engineering
• /
• v.26 no.6
• /
• pp.81-89
• /
• 2009

Dancer systems are typical equipment for attenuation of tension disturbances. Lately, demands for high speed roll-to-roll machines are rising but it is prior to attenuate the tension variation on the web entering into the printing zone to achieve the speed increment. Maintaining a constant tension before the first printing cylinder is the key of high speed, high quality printing. Dancer has been researched in two ways, whether it is controlled or not. The first one is active dancer and the other one is passive dancer. In the active dancer, a position of idle roll of dancer is measured and the roll is moved by external hydraulic cylinder to control tension disturbances. While the passive one composed with spring, damper and idle roll has no external actuator to position the idle roll. The tension disturbance causes movement of dancer roll and the displacement of the roll regulates the tension variation. On the other hand a composite type of dancer is applied for roll-to-roll printing machines. It has same apparatus as passive dancer. The displacement of roll is measured and front(or rear) driven roller is controlled to position the roll. In this paper, it is presented an analysis of pendulum dancer including position feedback PI control and logic for PI gain tuning in roll-to-roll machines. Pole-zero map and root locus with varying system parameters gives a design method for control of the dancer.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_2
• /
• pp.949-959
• /
• 2020

In this paper, as a prior study related to the generation of topographic information using the CAS500-1/2 satellite, we propose a method of extraction DSM for each 1:5000 scaled map in North Korea using KOMPSAT-3A strip images. This technique is designed to set the processing area by receiving shape file, only to generate output for every 1:5000 scaled map. In addition, dense point clouds and the DSM were extracted by applying MDR, a robust stereo image matching technique. Considering that the strip images are input in the units of scenes, we attempted to extract a DSM by processing and merging multiple image pairs in one 1:5000 map area. As a result, it was possible to confirm the generation of an integrated DSM with minimal separation at the junction, and as a result of the accuracy analysis, it was confirmed that the accuracy was within 5m compared to GCP.

• Radiation Oncology Journal
• /
• v.37 no.1
• /
• pp.43-50
• /
• 2019

Purpose: The aim of this retrospective study was to investigate the use of a radiopaque tissue fiducial marker (TFM) in the treatment of prostate cancer patients who undergo post-prostatectomy radiotherapy (PPRT). TFM safety, its role and benefit in quantifying the set-up uncertainties in patients undergoing PPRT image-guided radiotherapy were assessed. Materials and Methods: A total of 45 consecutive PPRT patients underwent transperineal implantation of TFM at the level of vesicourethral anastomosis in the retrovesical tissue prior to intensity-modulated radiotherapy. Prostate bed motion was calculated by measuring the position of the TFM relative to the pelvic bony anatomy on daily cone-beam computed tomography. The stability and visibility of the TFM were assessed in the initial 10 patients. Results: No postoperative complications were recorded. A total of 3,500 images were analysed. The calculated prostate bed motion for bony landmark matching relative to TFM were 2.25 mm in the left-right, 5.89 mm in the superior-inferior, and 6.59 mm in the anterior-posterior directions. A significant 36% reduction in the mean volume of rectum receiving 70 Gy (rV₇₀) was achieved for a uniform planning target volume (PTV) margin of 7 mm compared with the Australian and New Zealand Faculty of Radiation Oncology Genito-Urinary Group recommended PTV margin of 10 mm. Conclusion: The use of TFM was safe and can potentially eliminate set-up errors associated with bony landmark matching, thereby allowing for tighter PTV margins and a consequent favourable reduction in dose delivered to the bladder and rectum, with potential improvements in toxicities.