python 扫描枪_树莓派做条码扫描枪

首先安装必要的包python-zbar，

Shell

sudo apt-get install python-zbar

1

sudoapt-getinstallpython-zbar

然后安装SimpleCV:(参考这里的)

Monkey

sudo apt-get install ipython python-opencv python-scipy python-numpy python-setuptools python-pip

1

sudoapt-getinstallipythonpython-opencvpython-scipypython-numpypython-setuptoolspython-pip

Shell

sudo pip install https://github.com/sightmachine/SimpleCV/zipball/master

1

sudopipinstallhttps://github.com/sightmachine/SimpleCV/zipball/master

还有一个包svgwrite：(不安装的话，运行会出错：No module named svgwrite)，同时一起安装了一个pyparsing

Shell

sudo pip install svgwrite

1

sudopipinstallsvgwrite

还要安装一个lsof,(不安装的话，运行cam = Camera()会报错：/bin/sh: 1: lsof: not found)，参考：https://www.raspberrypi.org/forums/viewtopic.php?t=56103&p=460672

Shell

sudo aptitude install lsof

1

sudoaptitudeinstalllsof

上程序：

Python

import time

t0 = time.time()

from SimpleCV import *

def readFromCamera(cam):

ok = False

n=0

max_try = 50

while(ok == False and n

img = cam.getImage()

barcodes = img.findBarcode()

if type(barcodes) == FeatureSet and len(barcodes) >=1:

ok = True

img.save("1.jpg")

return barcodes[0].data, n

else:

ok = False

n = n +1

return "",n

if __name__ == "__main__":

t1 = time.time()

print t1-t0, "[firom SimpleCV import * time used]"

cam = Camera()

t2 = time.time()

print t2-t1, "[cam = Camera() time used]"

print readFromCamera(cam)

t3 = time.time()

print t3-t2, "[get image and findBarcode time used]"

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

importtime

t0=time.time()

fromSimpleCVimport*

defreadFromCamera(cam):

ok=False

n=0

max_try=50

while(ok==Falseandn

img=cam.getImage()

barcodes=img.findBarcode()

iftype(barcodes)==FeatureSetandlen(barcodes)>=1:

ok=True

img.save("1.jpg")

returnbarcodes[0].data,n

else:

ok=False

n=n+1

return"",n

if__name__=="__main__":

t1=time.time()

printt1-t0,"[firom SimpleCV import * time used]"

cam=Camera()

t2=time.time()

printt2-t1,"[cam = Camera() time used]"

printreadFromCamera(cam)

t3=time.time()

printt3-t2,"[get image and findBarcode time used]"

运行结果(手动对焦摄头)：

Shell

pi@raspberrypi:~/jlink/jlink/readSN $ python sn.py

2.31966090202 [firom SimpleCV import * time used]

0.699511051178 [cam = Camera() time used]

('640522710850', 0)

0.160804033279 [get image and findBarcode time used]

1

2

3

4

5

pi@raspberrypi:~/jlink/jlink/readSN$pythonsn.py

2.31966090202[firomSimpleCVimport*timeused]

0.699511051178[cam=Camera()timeused]

('640522710850',0)

0.160804033279[getimageandfindBarcodetimeused]

运行结果(自动对焦摄头)：

Shell

pi@raspberrypi:~/jlink/jlink/readSN $ python sn.py

2.35182714462 [firom SimpleCV import * time used]

1.07666397095 [cam = Camera() time used]

('640522710850', 13)

2.31718182564 [get image and findBarcode time used]

1

2

3

4

5

pi@raspberrypi:~/jlink/jlink/readSN$pythonsn.py

2.35182714462[firomSimpleCVimport*timeused]

1.07666397095[cam=Camera()timeused]

('640522710850',13)

2.31718182564[getimageandfindBarcodetimeused]

可以看到

导入SimpleCV包用掉了2.3秒，

获取手动对焦摄像头用掉0.7秒，获取自动对焦摄像头用掉1秒，

获取手动对焦图像及分析条码加起来只用了0.16秒，获取自动对焦图像及分析条码加起来用了2.3秒(循环了解14次)，

所以在正常循环使用时，可以先import包，获取到摄像头，然后当需要读sn时就比较快了。

结果输出里有显示循环了几次才正确读到条码，这个0表示只用一次就读到了。这款手动对焦摄像头其实很普通，是好多年前电脑CPU还是奔腾4时代买来玩的。对好焦后一次过读取到SN。 相反一款罗技带自动对焦功能的摄像头反而要用多次才能读到，时间上就慢了一些。

反复试验焦点可以在5-14cm之间

这是焦点设在12cm时获取到的图像，条码长度占图片宽度约1/3.

参考：

使用usb webcam获取图片：

打赏

微信扫一扫，打赏作者吧～

Post Views:

5,173