• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.11
• /
• pp.5459-5473
• /
• 2017

Recently, a tracking algorithm called the spatial-temporal context model has been proposed to locate a target by using the contextual information around the target. This model has achieved excellent results when the target undergoes slight occlusion and appearance changes. However, the target location in the current frame is based on the location in the previous frame, which will lead to failure in the presence of fast motion because of the lack of a prediction mechanism. In addition, the spatial context model is updated frame by frame, which will undoubtedly result in drift once the target is occluded continuously. This paper proposes two improvements to solve the above two problems: First, four possible positions of the target in the current frame are predicted based on the displacement between the previous two frames, and then, we calculate four confidence maps at these four positions; the target position is located at the position that corresponds to the maximum value. Second, we propose a target reliability criterion and design an adaptive threshold to regulate the updating speed of the model. Specifically, we stop updating the model when the reliability is lower than the threshold. Experimental results show that the proposed algorithm achieves better tracking results than traditional STC and other algorithms.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.9-16
• /
• 2008

In this paper, we present an algorithm for synthesizing intermediate view image from a stereoscopic pair of images. An image of multiview is need for people in order to easily recognize 3D image. However, if many cameras are use for that, not only does system get more complicated but also transmission rating cause a big trouble. Hence, stereo images are photograph and issue on the sending side and algorithm to generate several intermediate view image is able to be use on the receiving side. The proposed method is based on disparity space image. First of all, disparity space image that is depicted by the gap of pixel followed by disparity of stereo image is generated. Disparity map is made by utilizing disparity space image for searching for optimal disparity path then eventual intermediate view image is generated after occlusion region which does not match is processed. Experimental results illustrate the performance of the proposed technique and we obtained a high quality image of more than 30 dB PSNR.

• Journal of IKEEE
• /
• v.26 no.4
• /
• pp.686-692
• /
• 2022

In this paper, the matching error rates of modified area-based, energy-based algorithms, and learning-based structures were compared for stereo image matching. Census transform (CT) based on region and life propagation (BP) algorithm based on energy were selected, respectively.Existing algorithms have been improved and implemented in an embedded processor environment so that they can be used for stereo image matching in mobile systems. Even in the case of the learning base to be compared, a neural network structure that utilizes small-scale parameters was adopted. To compare the error rates of the three matching methods, Middlebury's Tsukuba was selected as a test image and subdivided into non-occlusion, discontinuous, and disparity error rates for accurate comparison. As a result of the experiment, the error rate of modified CT matching improved by about 11% when compared with the existing algorithm. BP matching was about 87% better than conventional CT in the error rate. Compared to the learning base using neural networks, BP matching was about 31% superior.

• Tuberculosis and Respiratory Diseases
• /
• v.45 no.1
• /
• pp.140-152
• /
• 1998

Background: Prone position improves oxygenation in patients with ARDS probably by reducing shunt Reduction of shunt in prone position is thought to be effected by lowering of the critical opening pressure (COP) of the dorsal lung because the pleural pressure becomes less positive in prone position compared to supine position. It can then be assumed that prone position would bring about greater improvement in oxygenation when PEEP applied in supine position is just beneath COP than when PEEP is above COP. Hemodynamically, prone position is expected to attenuate the lifting of cardiac fossa induced by PEEP. Based on these backgrounds, we investigated whether the effect of prone position on oxygenation differs in magnitude according to the level of PEEP applied in supine position, and whether impaired cardiac output in supine position by PEEP can be restored in prone position. Methods: In seven mongrel dogs, $PaO_2/F_1O_2$(P/F) was measured in supine position and at prone position 30 min. Cardiac output (CO), stroke volume (SV), pulse rate (PR), and pulmonary artery occlusion pressure (PAOP) were measured in supine position, at prone position 5 min, and at prone position 30 min. After ARDS was established with warmed saline lavage(P/F ratio $134{\pm}72$ mm Hg), inflection point was measured by constant flow method($6.6{\pm}1.4cm$ $H_2O$), and the above variables were measured in supine and prone positions under the application of Low PEEP($5.0{\pm}1.2cm$ $H_2O$), and Optimal PEEP($9.0{\pm}1.2cm$ $H_2O$)(2 cm $H_2O$ below and above the inflection point, respectively) consecutively. Results : P/F ratio in supine position was $195{\pm}112$ mm Hg at Low PEEP and $466{\pm}63$ mm Hg at Optimal PEEP(p=0.003). Net increase of P/F ratio at prone position 30 min, however, was far greater at Low PEEP($205{\pm}90$ mm Hg) than at Optimal PEEP($33{\pm}33$ mm Hg)(p=0.009). Compared to CO in supine position at Optimal PEEP($2.4{\pm}0.5$ L/min), CO in prone improved to $3.4{\pm}0.6$ L/min at prone position 5 min (p=0.0180) and $3.6{\pm}0.7$ L/min at prone position 30 min (p=0.0180). Improvement in CO was attributable to the increase in SV: $14{\pm}2$ ml in supine position, $20{\pm}2$ ml at prone position 5 min (p=0.0180), and $21{\pm}2$ ml at prone position 30 min (p=0.0180), but not to change in PR or PAOP. When the dogs were turned to supine position again, MAP ($92{\pm}23$ mm Hg, p=0.009), CO ($2.4{\pm}0.5$ L/min, p=0.0277) and SV ($14{\pm}1$ ml, p=0.0277) were all decreased compared to prone position 30 min. Conclusion: Prone position in a dog with saline-lavaged acute lung injury appeared to augment the effect of relatively low PEEP on oxygenation, and also attenuate the adverse hemodynamic effect of relatively high PEEP. These findings suggest that a PEEP lower than Optimal PEEP can be adopted in prone position to achieve the goal of alveolar recruitment in ARDS avoiding the hemodynamic complications of a higher PEEP at the same time.