From 5050e51e206580c4129443d3fdb9faec146feed5 Mon Sep 17 00:00:00 2001 From: Hobe Date: Sun, 7 Jun 2020 12:14:18 +0000 Subject: [PATCH] Fix some typo. git-svn-id: http://newslabx.csie.ntu.edu.tw/svn/Ginger@71 5747cdd2-2146-426f-b2b0-0570f90b98ed --- trunk/RTCSA_SS/01Introduction.tex | 4 ++-- trunk/RTCSA_SS/02Background.tex | 2 +- trunk/RTCSA_SS/03Design.tex | 8 ++++---- trunk/RTCSA_SS/04Evaluation.tex | 4 ++-- 4 files changed, 9 insertions(+), 9 deletions(-) diff --git a/trunk/RTCSA_SS/01Introduction.tex b/trunk/RTCSA_SS/01Introduction.tex index da4967b..ab2124c 100644 --- a/trunk/RTCSA_SS/01Introduction.tex +++ b/trunk/RTCSA_SS/01Introduction.tex @@ -5,7 +5,7 @@ The turn over frequency while sleeping is an important index to quantify the health of elderly. Many wearable devices can also achieve the same purpose, but many study show that the elderly feel uncomfortable with wearing such devices all days.\textcolor{red}{(source?)}. By the low resolution thermal camera, we -can obtain the daily activities informations, but not reveal too much privacy +can obtain the daily activities information, but not reveal too much privacy like the RGB camera. {\bf Contribution} @@ -16,7 +16,7 @@ Super-resolution techniques, and our method, we can have 50\% recall rate and 83 on turning over detection. -The remaining of this paper is organized as follow. Section~\ref{sec:bk_related} +The remaining of this paper is organized as follows. Section~\ref{sec:bk_related} presents background for developing the methods. Section~\ref{sec:design} presents the system architecture, and the developed mechanisms. Section~\ref{sec:eval} presents the evaluation results of proposed mechanism and Section~\ref{sec:conclusion} summaries our works. diff --git a/trunk/RTCSA_SS/02Background.tex b/trunk/RTCSA_SS/02Background.tex index 908e761..e71958b 100644 --- a/trunk/RTCSA_SS/02Background.tex +++ b/trunk/RTCSA_SS/02Background.tex @@ -26,7 +26,7 @@ patches. %% \end{figure} \subsection{Thermal cameras} -In this work, we use two different resolution thermal camera to play the role +In this work, we use two different resolution thermal cameras to play the role of low-resolution and high-resolution camera. For low-resolution camera, we use Grid-EYE thermal camera. Grid-EYE is a thermal camera that can output $8 \times 8$ pixels thermal data with $2.5^\circ C$ accuracy and $0.25^\circ C$ diff --git a/trunk/RTCSA_SS/03Design.tex b/trunk/RTCSA_SS/03Design.tex index 4277659..04d3299 100644 --- a/trunk/RTCSA_SS/03Design.tex +++ b/trunk/RTCSA_SS/03Design.tex @@ -5,7 +5,7 @@ We designed a thermal-box to collect the data. It has four Grideye sensors on the corners of a 10 cm square and a Lepton 3 at the central. Figure~\ref{fig:method} shows the system of our method. It consists four parts. The first part is to fuse multiple -Grideye image, since the resolution of singel Grideye sensor only has 64 pixels. The +Grideye image, since the resolution of single Grideye sensor only has 64 pixels. The second part, we train the SRCNN model with fused Grideye image as low-resolution and downscaled Lepton 3 image as high-resolution image. The third part, we use the @@ -29,7 +29,7 @@ On the thermal-box, there are four Grideye sensors. At the beginning, we let the thermal-box faces to an empty bed and records the background temperature. All the following frames will subtract this background temperature. After that, we resize four $8 \times 8$ Grideye images to $64 \times 64$ by bilinear -interpolation and than merge them dependence on the distance between thermal-box and +interpolation and then merge them dependence on the distance between thermal-box and bed, width of sensor square and the FOV of Grideye sensor. \begin{enumerate} @@ -51,8 +51,8 @@ distinguish the residual heat on bed and the person as Figure~\ref{fig:residual_ situation will slowly disappear after one or two minutes. To determination the pose, first we use a median filter with a window size of five -to filter out the noise. Than, find the curve hull line of the upper bound and -lower bound of the data. Finally calculate the middle line of upper bound and lower bound. +to filter out the noise. Then, find the curve hull line of the upper bound and +lower bound of the data. Finally, calculate the middle line of upper bound and lower bound. Figure~\ref{fig:trend} shows the data and these lines. We divide every data into 10 second time windows. If the middle line of the time window diff --git a/trunk/RTCSA_SS/04Evaluation.tex b/trunk/RTCSA_SS/04Evaluation.tex index f126101..70654db 100644 --- a/trunk/RTCSA_SS/04Evaluation.tex +++ b/trunk/RTCSA_SS/04Evaluation.tex @@ -27,7 +27,7 @@ For training the pose recognition model, we collect 200 images of lay on back an of lay on right or left side. The result shows that the accuracy of single frame detection can be improved about 5\% by SRCNN. -We let a person lay on bed and change his pose every minutes. The pose is repeating -lay on back, lay on left, lay on back and lay on right. Our method will output the currect pose +We let a person lay on bed and change his pose every minute. The pose is repeating +lay on back, lay on left, lay on back and lay on right. Our method will output the current pose every 10 seconds and check if the pose changed. The accuracy of pose detection is 65\%, and the turning over detection has 50\% recall rate and 83\% precision.