mongodb python image 图像存储读取

最近做一些数据库调研的工作，目标是实现影像更快的入库、出库、查询，并实现并行访问等操作。

将结果总结成一个mongoImg类，也算是小结吧。

 1 '''
 2 Created on 2013-8-6
 3 class mongoInsert
 4 @author: tree
 5 '''
 6 __metaclass__ = type
 7 
 8 import os
 9 from pymongo.database import Database
10 import time
11 import gridfs
12 
13 class mongoImg(object):
14     """mongoInsert is a class for inserting document
15     
16     
17     """
18     def __init__(self, database, dir):
19         """Create a new instance of :class:mongoInsert
20         :Parameters:
21           - `database`: database to use
22           - `dir` : directory of document 
23           """
24         if not isinstance(database, Database):
25             raise TypeError("database must be an instance of Database")
26         if len(dir) < 1:
27             raise TypeError("dir must be an string of directory")
28         
29 #         self.__con = Connection()
30         self.__imgdb = database
31         self.__imgfs = gridfs.GridFS (self.__imgdb)
32         self.__dir = dir
33         self.__filelist=[]
34 
35     #save filepath in list.txt
36     def __dirwalk(self,topdown=True):
37         """traverse the documents of self.__dir and save in self.__filelist
38         """
39         sum=0
40         self.__filelist.clear()
41         
42         for root,dirs,files in os.walk(self.__dir,topdown):
43             for name in files:
44                 sum+=1
45                 temp=os.path.join(root,name)
46                 self.__filelist.append(temp)
47         print(sum)
48 
49     #insert image 
50     def insert(self):
51         """insert images in mongodb
52         """
53         self.__dirwalk()
54 
55         tStart = time.time()        
56         for fi in self.__filelist:       
57             with open (fi,'rb') as myimage:
58                 data=myimage.read()                  
59                 self.__imgfs.put(data, content_type = "jpg", filename =fi)
60     
61         tEnd =time.time ()
62         print ("It cost %f sec" % (tEnd - tStart))
63         
64     #get image by filename
65     def getbyname(self,filename,savepath):
66         """get img from mongdb by filename
67         """
68         if len(savepath) < 1:
69             raise TypeError("dir must be an string of directory")
70         dataout=self.__imgfs.get_version(filename)
71         try:
72             imgout=open(savepath,'wb')
73             data=dataout.read()
74             imgout.write(data)
75         finally:
76             imgout.close()
77

使用示例：也可以将数据库连接写在类内部

1 from pymongo import Connection
2 import mongoImg
3 
4 filedir=r'D:image'
5 con = Connection()
6 db = con.imgdb
7 imgmongo=mongoImg.mongoImg(db,filedir)
8 imgmongo.insert()

感觉mongodb存储影像切片还是蛮快的，1w多个图片，大约100-200秒左右。

tip：

gridfs.GridFS.put 函数

put(data, **kwargs)
Put data in GridFS as a new file.

Equivalent to doing:

try:
    f = new_file(**kwargs)
    f.write(data)
    finally
f.close()

在存储读取图像时，犯了低级错误，将open得到的file实例当做数据存储，读取的时候怎么也读不出数据。。。囧

另外以字节流形式读取图像数据比较适合。

pipe = open('/dev/input/js0','rb')

如果以str形式存储的话，可能会出现UnicodeDecodeError错误，貌似是因为图像数据有些超出了python默认编码的存储区间。

ps：初学python 数据库操作也忘得差不多欢迎大家批评和指正~