熟悉latex后真心觉得word好费事，一般latex论文都会有模板，只需要替换把原有内容替换一下，就会生成比较好看的文档。
废话不多说，总结一下使用后的体会
1、先说说插图
插图的话，我插入的是".esp"格式。
当然先导包usepackage{graphicx}，
我把图片都放在了fig2文件夹中，fig2的文件夹与latex文档在同一目录下。
（1) 、插入一张图片
egin{figure}[htbp]
centering
includegraphics[height=6.0cm,width=9.5cm]{fig2/Xbee.eps}%fig2文件夹下的xbee.esp图片，
caption{Campus environment detection system}
end{figure}
宽度，高度自己调节。
（2）、 并排插入俩张图片
egin{figure}[htbp]
egin{minipage}[t]{0.4linewidth}
%并排插图时，线宽很重要，自己慢慢试，俩张图就不要超过0.5，三张图不要超过0.33之类的，自己看着办
centering
includegraphics[height=7.5cm,width=2.5cm]{fig2/xitong1.eps}
caption{Fatiguedetection overview}
end{minipage}
hfill%分栏的意思吧
egin{minipage}[t]{0.5linewidth}
centering
includegraphics[height=7.5cm,width=5.5cm]{fig2/tupianchuli1.eps}
caption{The imageprocessing}
end{minipage}
end{figure}
（3）、并排插入三张图片，线宽很重要，要不然插不进去
egin{figure}[htbp]
egin{minipage}[t]{0.2linewidth}
centering
includegraphics[height=7.5cm,width=2.5cm]{fig2/xitong1.eps}
caption{Fatigue detection overview}
end{minipage}
hfill
egin{minipage}[t]{0.2linewidth}
centering
includegraphics[height=7.5cm,width=2.5cm]{fig2/tupianchuli1.eps}
caption{The image processing}
end{minipage}
hfill
egin{minipage}[t]{0.2linewidth}
centering
includegraphics[height=7.5cm,width=2.5cm]{fig2/tupianchuli1.eps}
caption{The image processing}
end{minipage}
end{figure}
（4）、 插入并排子图
导包 usepackage{graphicx}和usepackage{subfigure}
egin{figure}
centering
subfigure[图1]{
label{figa} %% label for first subfigure
includegraphics[width=1.5in]{figs/tupianchuli1.eps}}
hspace{1in}
subfigure[图2]{
label{fig:subfig:b} %% label for secondsubfigure
includegraphics[width=1.5in]{figs/tupianchuli1.eps}}
caption{说明介绍}
label{figb} %% label for entire figure
end{figure}
（5）、 并排三张子图，第一个占一般空间
egin{figure}
centering
subfigure[]{
label{fig:a} %% label for first subfigure
includegraphics[width=2cm]{fig2/Seeed_Stalker3.eps}}
hspace{1in}%使第一个子图占一半空间
subfigure[]{
label{fig:subfig:b} %% label for secondsubfigure
includegraphics[width=2cm]{fig2/temper_humidity_sensor1.eps}}
subfigure[]{
label{fig:subfig:c} %% label for secondsubfigure
includegraphics[width=1.5cm]{fig2/xbee_s1.eps}}
caption{bingpai}
label{figb} %% label for entire figure
end{figure}
5.2 并排三张子图
egin{figure}
centering
subfigure[]{
label{fig:a} %% label for first subfigure
includegraphics[width=2cm]{fig2/Seeed_Stalker3.eps}}
%hspace{1in}%使第一个子图占一半空间
subfigure[]{
label{fig:subfig:b} %% label for secondsubfigure
includegraphics[width=2cm]{fig2/temper_humidity_sensor1.eps}}
subfigure[]{
label{fig:subfig:c} %% label for secondsubfigure
includegraphics[width=1.5cm]{fig2/xbee_s1.eps}}
caption{bingpai}
label{figb} %% label for entire figure
end{figure}
5.3 四张子图，分行
egin{figure}
centering
subfigure[]{
label{fig:a} %% label for first subfigure
includegraphics[width=3cm]{fig2/Seeed_Stalker3.eps}}
%hspace{1in}%使第一个子图占一半空间
vfill%分行命令
subfigure[]{
label{fig:subfig:b} %% label for secondsubfigure
includegraphics[width=2cm]{fig2/temper_humidity_sensor1.eps}}
subfigure[]{
label{fig:subfig:c} %% label for secondsubfigure
includegraphics[width=2cm]{fig2/xbee_s1.eps}}
subfigure[]{
label{fig:a} %% label for first subfigure
includegraphics[width=1.5cm]{fig2/ESP_01.eps}}
caption{bingpai}
label{figb} %% label for entire figure
end{figure}
6.文本文档分俩栏
导包usepackage{multicol}
egin{multicols}{2}
Viola-Jones algorithm [7] [10] is commonlyused for fast appearance-based detection of different kind of objects. Faceclassifier and eyes classifier are trained by using Haar-like features. TheHaar-like features are the input to the classifier and are specified by theirshapes, position within the region of interest, and the scale (Fig3). Toincrease the accuracy of eye detection, a classifier was used to detect botheyes. When judging the state of the eye, the system will read an image from thevideo, using the Viola-Jones algorithm to mark the face area from the originalimage, as in Fig (4-a) shows, using the same method, and then find out the eyesarea from the facial region, as Fig (4-b) shows. The eyes area will be croppedas a region of interest (ROI), as Fig (4-c) show. The subsequent binaryprocessing of the image and the use of the area ratio to determine the eyesstate will based on the ROI.
end{multicols}
6.1文本文档分三栏
egin{multicols}{3}
。。。。。
end{multicols}
