• 대한임베디드공학회논문지
• /
• 제10권1호
• /
• pp.41-48
• /
• 2015

To deal with the emergency of the solitary aged people, we have developed a collapse-sensing phone, in which a collapse sensor, a GPS receiving chipset and a CDMA sending chipset are included. The general cellular phone is somewhat expensive communication device using sound and characters, but the collapse-sensing phone is a cheaper and popular version. If the collapse sensor recognizes a certain of collapse of the aged people, CDMA sending chipset will send the location of the phone which is received from satellite by GPS receiving chipset. In this paper, a collapse recognition algorithm which is developed by using much experimental data, will be introduced to explain how to recognize the real collapse from fast sitting or immediate standing after collapse. Once a true collapse is ecognized, the phone-ID and the coordinate will be sent to the server of administrative office via CDMA network. And the position of emergency will be displayed on the GIS with the rescue center.

• 대한임베디드공학회논문지
• /
• 제9권4호
• /
• pp.219-227
• /
• 2014

Natural disasters such as a ground collapse and a landslide have broken out due to the climate change of the Korea and the reckless expansion of cities and roads. The climate changes and the reckless urbanization have made the ground weak. Thus, it is important to keep a close eye on the highly weakened landslide and to prevent its natural disasters. In order to prevent these disasters, this paper presents a system of recognizing the road slide condition by measuring the displacements using laser scanner instrument. The previous system of monitoring the road slide has some problems as inaccurate recognition due to using only images from a camera, or expensive system such as artificial satellites and aircraft systems. To solve this problem, our proposed system uses the 3D range data from the laser scanner for measuring the accurate displacement of the road slide and optical flows from the Lucas-Kanade algorithm for recognizing the road slide in the image.

• 한국산업융합학회 논문집
• /
• 제26권1호
• /
• pp.113-119
• /
• 2023

In this paper, we propose an optimization of a pose estimation deep learning model for recognition of abnormal behavior in unmanned stores using radio frequencies. The radio frequency use millimeter wave in the 30 GHz to 300 GHz band. Due to the short wavelength and strong straightness, it is a frequency with less grayness and less interference due to radio absorption on the object. A millimeter wave radar is used to solve the problem of personal information infringement that may occur in conventional CCTV image-based pose estimation. Deep learning-based pose estimation models generally use convolution neural networks. The convolution neural network is a combination of convolution layers and pooling layers of different types, and there are many cases of convolution filter size, number, and convolution operations, and more cases of combining components. Therefore, it is difficult to find the structure and components of the optimal posture estimation model for input data. Compared with conventional millimeter wave-based posture estimation studies, it is possible to explore the structure and components of the optimal posture estimation model for input data using genetic algorithms, and the performance of optimizing the proposed posture estimation model is excellent. Data are collected for actual unmanned stores, and point cloud data and three-dimensional keypoint information of Kinect Azure are collected using millimeter wave radar for collapse and property damage occurring in unmanned stores. As a result of the experiment, it was confirmed that the error was moored compared to the conventional posture estimation model.