Python提取图片的ROI

图像处理经常需要提取图片的ROI，本文使用Python提取图片的ROI。

使用的Module是PIL (Pillow)，一个图像处理库，用到的函数为类 Image 中的 crop 方法。

函数原型为：

Image.crop(box=None)

Returns a rectangular region from this image. The box is a 4-tuple defining the left, upper, right, and lower pixel coordinate.
This is a lazy operation. Changes to the source image may or may not be reflected in the cropped image. To break the connection, call the load() method on the cropped copy.
Parameters:    box – The crop rectangle, as a (left, upper, right, lower)-tuple.
Return type:    Image
Returns:    An Image object.

知道矩形的左上角的坐标和右下角的坐标，即可构造box，例如下面的代码

box = (100, 100, 400, 400)
region = im.crop(box)

知道如何提取除ROI时，上面例子为 region，保存ROI到图像则使用类 Image 的 save 方法

region.save(filename)

给出一个Demo，使用人脸数据库GENKI部分的图像做实验，该数据的数字子集GENKI-SZSL提供人脸区域的坐标和大小。提取代码提供如下

from PIL import Image
import os

src = ‘.‘

imlist = open(src + ‘/GENKI-SZSL_Images.txt‘, ‘r‘).readlines()

rs = [float(line.split()[1]) for line in open(src + ‘/GENKI-SZSL_labels.txt‘, ‘r‘).readlines()]
cs = [float(line.split()[0]) for line in open(src + ‘/GENKI-SZSL_labels.txt‘, ‘r‘).readlines()]
ss = [float(line.split()[2]) for line in open(src + ‘/GENKI-SZSL_labels.txt‘, ‘r‘).readlines()]

for i in range(0, len(rs)):
    path = src + ‘/images/‘ + imlist[i].strip()
    filename = src + ‘/output/‘ + imlist[i].strip()

    try:
        im = Image.open(path)
    except:
        continue

    r = rs[i]
    c = cs[i]
    s = ss[i]

    xLeft   = int(c - s/2)
    yUpper  = int(r - s/2)
    xRight  = int(c + s/2)
    yLower  = int(r + s/2) 

    region = im.crop((xLeft, yUpper, xRight, yLower))
    region.save(filename)

代码打包下载：http://pan.baidu.com/s/1dD4opKP 密码：ygu7

Pillow的项目文档地址： http://pillow.readthedocs.org