A reference to Digital image processing, may most people think of MATLAB, but MATLAB has its own shortcomings:
1. Not open source, the price is expensive;
2. Large software capacity. Generally more than 3GB, the high version even reached 5GB or more.
3. Easy to do research, not easily converted into software.
Therefore, I tend to learn python to do image processing work. To use Python, the first thing to do is to install Python, which is typically above version 2.7, whether it's in Windows or Linux, and the installation work is very simple.
To use Python for all kinds of development, we have to install the corresponding libraries or. Just like the toolbox we installed in MATLAB (Toolbox). Install these libraries, typically using the PIP.
Using Python for digital image processing, you also have to install the pillow package. Although Python has a PIL (Python image Library), but this library has now stopped updating, so using pillow, it was developed by PIL.
Pip Install Pillow
Because I installed the anaconda, so save this step. With regard to anaconda, a topic will be written in a later section.
Use the PIP list to view the existing installation packages in Anaconda and find that there are already pillow.
(Don't repeat this condition later)
Opening and displaying of a picture
From PIL import Image
img=image.open (' e:/a.png ')
img.show ()
Although the use of pillow, but it is from the PiL fork, so still want to import from the PiL. Use the open () function to open the picture and use the show () function to display the picture.
The above method is sometimes inconvenient, so we can also use the other way, let the program to draw the picture.
From PIL import Image
import matplotlib.pyplot at Plt
img=image.open (' e:/a.png ')
plt.figure ("a")
Plt.imshow (IMG)
plt.show ()
Although this approach is a bit more complicated, it is recommended that you use a matplotlib library to draw pictures for display. Matplotlib is a professional drawing library, equivalent to the plot in MATLAB, you can set up a number of figure, set figure title, can even use subplot in a figure display more than one picture.
Matplotlib can be installed directly
Pip Install Matplotlib
Figure the default is axis, we can turn it off if there is no need.
Plt.axis (' off ')
After you open a picture, you can use some properties to view picture information, such as:
Print img.size #图片尺寸
print img.mode #图片模式
print Img.format #图片格式
Display results as:
(690, 1226)
Rgb
Jpeg
Preservation of two pictures
Img.save (' e:/b.jpg ')
On one line of code, this line of code not only saves pictures, but also converts formatting. This time, the original PNG format is saved in JPG format.