• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.5
• /
• pp.1690-1710
• /
• 2014

In this paper, we propose the rapid implementation of a 3-dimensional (3D) facial reconstruction from a single frontal face image and introduce a design for its application on a mobile device. The proposed system can effectively reconstruct human faces in 3D using an approach robust to lighting conditions, and a fast method based on a Canonical Correlation Analysis (CCA) algorithm to estimate the depth. The reconstruction system is built by first creating 3D facial mapping from a personal identity vector of a face image. This mapping is then applied to real-world images captured with a built-in camera on a mobile device to form the corresponding 3D depth information. Finally, the facial texture from the face image is extracted and added to the reconstruction results. Experiments with an Android phone show that the implementation of this system as an Android application performs well. The advantage of the proposed method is an easy 3D reconstruction of almost all facial images captured in the real world with a fast computation. This has been clearly demonstrated in the Android application, which requires only a short time to reconstruct the 3D depth map.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.51-55
• /
• 2020

In this paper, an improved FAM is proposed by adopting similarity learning in the existing FAM (Fuzzy Associative Memory) used in image restoration. Image restoration refers to the recovery of the latent clean image from its noise-corrupted version. In serious application like face recognition, this process should be noise-tolerant, robust, fast, and scalable. The existing FAM is a simple single layered neural network that can be applied to this domain with its robust fuzzy control but has low capacity problem in real world applications. That similarity measure is implied to the connection strength of the FAM structure to minimize the root mean square error between the recovered and the original image. The efficacy of the proposed algorithm is verified with significant low error magnitude from random noise in our experiment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.1109-1111
• /
• 2017

본 논문은 다양한 센서들과 디바이스들이 실시간으로 정보를 주고 산업 IoT 환경에서 실시간 분산 제어 역할을 하는 마스터 노드(Master Node)인 CPS시스템이 시스템장애 또는 내, 외부적인 요인으로 인해 정상적인 수행을 하지 못할 때 일시적으로 마스터 노드의 기능을 수행할 수 있는 싱크노드(Sink Node)를 선정하여 대리 마스터 노드를 결정하는 방안을 제시한다. 산업 IoT환경에서 마스터 노드의 역할로써 중요한 파라미터들을 선정한다. 이 후 상황에 따라 파라미터들의 가중치를 변경하여 우선순위 알고리즘을 통해 지속적으로 싱크 노드들의 우선순위에 대한 정보를 마스터 노드와 싱크노드들 간 공유하게 된다. 마스터 노드의 결함이 발생 시 우선순위가 높은 싱크노드가 마스터 노드의 역할을 대신 수행하여 예기치 못한 상황에도 데이터를 유지할 수 있도록 한다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2003.05a
• /
• pp.311-315
• /
• 2003

Recently, there is asymmetrical transmission in Internet data stream. The asymmetrical transmission has much more downstream than upstream. Owing to this point, it needs to reduce the acknowledgement as element of the obsrtuction in downstream. In this paper, according to simulation's result, we know that Sack has good performance than New Reno in bottleneck environment Comparing two protocols in packet loss rate, NewReno is lower than Sack. And also comparing two protocols in throughput of ack packet, not only NewReno processes ack packet more quickly than Sack, but also NewReno processes more ack packet than Sack protocol during ten seconds in simulation. As a result, NewReno is batter than Sack in throughput of asymmetrical link.

• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.1
• /
• pp.81-87
• /
• 2016

It is important to analyze the exact damage information for fast recovery when natural disasters cause damage on river-side facilities such as dams, bridges, embankments etc. In this study, we shows the method to effectively damage analysis plan using UAV(Unmanned aerial vehicle) images and accuracy assessment of it. The UAV images are captured on area near the river-side facilities and the core methodology for damage analysis are image matching and change detection algorithm. The result(point cloud) from image matching is to construct 3-dimensional data using by 2-dimensional images, it extracts damage areas by comparing the height values on same area with reference data. The results are tested absolute locational precision compared by post-processed aerial LiDAR data named reference data. The assessment analysis test shows our matching results 10-20 centimeter level precision if external orientation parameters are very accurate. This study shows suggested method is very useful for damage analysis in a large size structure like river-side facilities. But the complexity building can't apply this method, it need to the other method for damage analysis.