第五章： pcDuino上面安裝和測試SimpleCV以及OpenCV

xiaofenglib 2013-08-29

展開全文

pcDuino有計算能力來進行計算機視覺，這就讓很多在Arduino 上不可能的項目可以在pcDuino上實現(xiàn)。我們可以把計算機視覺這一非常強大的技術(shù)用在我們的項目中。在這章，我們來看怎么安裝和使用SimpleCV和OpenCV。

SimpleCV的安裝和例子

在我們開始動手之前，我們要確保手頭有一個 UVC 兼容的USB攝像頭。
下面為安裝SimpleCV的步驟：

$sudo apt-get install ipython python-opencv python-scipy python-numpy python-setuptools python-pip
$sudo pip install https://github.com/ingenuitas/SimpleCV/zipball/master
$sudo apt-get install python-pygame
$sudo apt-get install python-imaging

復(fù)制代碼

在終端里面輸入 “$simpleCV" 就可以啟動simpleCV. simpleCV是一個交互的命令解釋器。我們可以輸入命令來執(zhí)行：

ubuntu@ubuntu:~$ simplecv
+-----------------------------------------------------------+
SimpleCV 1.3.0 [interactive shell] - http://
+-----------------------------------------------------------+
Commands:
"exit()" or press "Ctrl+ D" to exit the shell
"clear" to clear the shell screen
"tutorial" to begin the SimpleCV interactive tutorial
"example" gives a list of examples you can run
"forums" will launch a web browser for the help forums
"walkthrough" will launch a web browser with a walkthrough
Usage:
dot complete works to show library
for example: Image().save("/tmp/test.jpg") will dot complete
just by touching TAB after typing Image().
Documentation:
help(Image), ?Image, Image?, or Image()? all do the same
"docs" will launch webbrowser showing documentation
SimpleCV:1> cam=Camera()
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
SimpleCV:2> img=cam.getImage()
SimpleCV:3> img.show()
SimpleCV:5:

復(fù)制代碼

以下為命令執(zhí)行的結(jié)果：

OpenCV的安裝和例子

OpenCV 是個開源的計算機視覺軟件包。在這里我們詳細的介紹如何安裝OpenCV Python版本到pcDuino上面。并且給出了兩個例子，一個例子用來通過USB 攝像頭來獲取圖形，另外一個例子通過OpenCV來做人臉識別。

安裝步驟：

$ sudo apt-get -y install build-essential cmake cmake-qt-gui pkg-config libpng12-0 libpng12-dev libpng++-dev libpng3 libpnglite-dev zlib1g-dbg zlib1g zlib1g-dev pngtools libtiff4-dev libtiff4 libtiffxx0c2 libtiff-tools
$sudo apt-get -y install libjpeg8 libjpeg8-dev libjpeg8-dbg libjpeg-progs ffmpeg libavcodec-dev libavcodec53 libavformat53 libavformat-dev libgstreamer0.10-0-dbg libgstreamer0.10-0 libgstreamer0.10-dev libxine1-ffmpeg libxine-dev libxine1-bin libunicap2 libunicap2-dev libdc1394-22-dev libdc1394-22 libdc1394-utils swig libv4l-0 libv4l-dev python-numpy libpython2.6 python2.6-dev libgtk2.0-dev pkg-config
$sudo apt-get install libopencv-dev python-opencv
$sudo apt-get install python-dev
$sudo ln -s /usr/lib/arm-linux-gnueabihf/libjpeg.so /usr/lib
$sudo ln -s /usr/lib/arm-linux-gnueabihf/libfreetype.so /usr/lib
$sudo ln -s /usr/lib/arm-linux-gnueabihf/libz.so /usr/lib
$sudo easy_install PIL
$sudo pip install -v PIL

復(fù)制代碼

例子1：用OpenCV通過USB攝像頭來抓圖像：

#!/Users/brent/.virtualenvs/lumber/bin/python
import cv
cv.NamedWindow("w1", cv.CV_WINDOW_AUTOSIZE)
camera_index = 0
capture = cv.CaptureFromCAM(camera_index)
gx = gy = 1
grayscale = blur = canny = False
def repeat():
global capture #declare as globals since we are assigning to them now
global camera_index
global gx, gy, grayscale, canny, blur
frame = cv.QueryFrame(capture)
# import pdb; pdb.set_trace()
if grayscale:
gray = cv.CreateImage(cv.GetSize(frame), frame.depth, 1)
cv.CvtColor(frame, gray, cv.CV_RGB2GRAY)
frame = gray
if blur:
g = cv.CreateImage(cv.GetSize(frame), cv.IPL_DEPTH_8U, frame.channels)
cv.Smooth(frame, g, cv.CV_GAUSSIAN, gx, gy)
frame = g
if grayscale and canny:
c = cv.CreateImage(cv.GetSize(frame), frame.depth, frame.channels)
cv.Canny(frame, c, 10, 100, 3)
frame = c
cv.ShowImage("w1", frame)
c = cv.WaitKey(10)
if c==ord('='): #in "n" key is pressed while the popup window is in focus
gx += 2
gy += 2
elif c == ord('-'):
gx = max(1, gx-2)
gy = max(1, gy-2)
elif c == ord('x'):
gx += 2
elif c == ord('X'):
gx = max(1, gx-2)
elif c == ord('q'):
exit(0)
elif c == ord('b'):
blur = not blur
elif c == ord('g'):
grayscale = not grayscale
elif c == ord('c'):
canny = not canny
while True:
repeat()

復(fù)制代碼

例子二：人臉識別

輸入圖像：

下載上面的圖像，保存為 ”opencv_in.jpg"。

輸出圖像：

#!/usr/bin/env python
#coding=utf-8
import os
from PIL import Image, ImageDraw
import cv
def detect_object(image):
grayscale = cv.CreateImage((image.width, image.height), 8, 1)
cv.CvtColor(image, grayscale, cv.CV_BGR2GRAY)
cascade = cv.Load("/usr/share/opencv/haarcascades/haarcascade_frontalface_alt_tree.xml")
rect = cv.HaarDetectObjects(grayscale, cascade, cv.CreateMemStorage(), 1.1, 2,
cv.CV_HAAR_DO_CANNY_PRUNING, (20,20))
result = []
for r in rect:
result.append((r[0][0], r[0][1], r[0][0]+r[0][2], r[0][1]+r[0][3]))
return result
def process(infile):
image = cv.LoadImage(infile);
if image:
faces = detect_object(image)
im = Image.open(infile)
path = os.path.abspath(infile)
save_path = os.path.splitext(path)[0]+"_face"
try:
os.mkdir(save_path)
except:
pass
if faces:
draw = ImageDraw.Draw(im)
count = 0
for f in faces:
count += 1
draw.rectangle(f, outline=(255, 0, 0))
a = im.crop(f)
file_name = os.path.join(save_path,str(count)+".jpg")
# print file_name
a.save(file_name)
drow_save_path = os.path.join(save_path,"out.jpg")
im.save(drow_save_path, "JPEG", quality=80)
else:
print "Error: cannot detect faces on %s" % infile
if __name__ == "__main__":
process("./opencv_in.jpg")