Revision history [back]

you can either store the numpy array as is (wrapped in a buffer object):

import cv2
import sqlite3

im = cv2.imread("pattern.png")
db = sqlite3.connect("my.db")
cur = db.cursor()
cur.execute("create table images(id string, img blob)")
db.commit()
cur.execute("insert into images values(?,?)",("pattern",buffer(im)))
db.commit()

or encode it as an image ( like on disk, only in memory / db ) :

_,enc = cv2.imencode(".png",im)
cur.execute("insert into images values(?,?)",("pattern",buffer(enc)))
db.commit()

you can either store the numpy array as is (wrapped in a buffer object):

import cv2
import sqlite3

im = cv2.imread("pattern.png")
db = sqlite3.connect("my.db")
cur = db.cursor()
cur.execute("create table images(id string, img blob)")
db.commit()
cur.execute("insert into images values(?,?)",("pattern",buffer(im)))
db.commit()

or encode it as an image ( like on disk, only in memory / db ) :

_,enc = cv2.imencode(".png",im)
cur.execute("insert into images values(?,?)",("pattern",buffer(enc)))
db.commit()

you can either store after loading the numpy array as is (wrapped in a buffer object):image and setting up the db,

import cv2
import sqlite3

im = cv2.imread("pattern.png")
db = sqlite3.connect("my.db")
cur = db.cursor()
cur.execute("create table images(id string, img blob)")
db.commit()

you can either store the numpy array as is (wrapped in a buffer object):

cur.execute("insert into images values(?,?)",("pattern",buffer(im)))
db.commit()

or encode it as an image ( like on disk, only in memory / db ) :

_,enc = cv2.imencode(".png",im)
cur.execute("insert into images values(?,?)",("pattern",buffer(enc)))
db.commit()

use the first, if you need it for further opencv processing, the second, if you want to e.g serve it to the outside world