document/TranslateProject
2
0
Fork 0
mirror of https://github.com/LCTT/TranslateProject.git synced 2024-12-23 21:20:42 +08:00
Code Issues Packages Projects Releases Wiki Activity

translated

Browse Source
This commit is contained in:
geekpi 2021-08-16 08:49:37 +08:00
parent 107217e9bc
commit ff07c2563b
2 changed files with 249 additions and 249 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

249
sources/tech/20210802 Use OpenCV on Fedora Linux - part 1.md
View File

@ -1,249 +0,0 @@
[#]: subject: (Use OpenCV on Fedora Linux ‒ part 1)
[#]: via: (https://fedoramagazine.org/use-opencv-on-fedora-linux-part-1/)
[#]: author: (Onuralp SEZER https://fedoramagazine.org/author/thunderbirdtr/)
[#]: collector: (lujun9972)
[#]: translator: (geekpi)
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
Use OpenCV on Fedora Linux ‒ part 1
======
![][1]
Cover image excerpted from Starry Night by [Vincent van Gogh][2], Public domain, via Wikimedia Commons
The technology world changes daily and the demands for computer vision, artificial intelligence, and machine learning are increasing. The technology that allows computers and mobile phones to see their surroundings is called [computer vision][3]. Work on re-creating a human eye started in the 50s. Since then, computer vision technology has come a long way. Computer vision has already made its way to our mobile phones via different applications. This article will introduce [OpenCV][4] on Fedora Linux.
### **What is OpenCV?**
> OpenCV (Open Source Computer Vision Library) is an open-source computer vision and machine learning software library. OpenCV was built to provide a common infrastructure for computer vision applications and to accelerate the use of machine perception in the commercial products. It has more than 2500 optimized algorithms, which includes a comprehensive set of both classic and state-of-the-art computer vision and machine learning algorithms. These algorithms can be used to detect and recognize faces, identify objects, classify human actions in videos and establish markers to overlay it with augmented reality and much more.
>
> [opencv.org – about][5]
### Install OpenCV on Fedora Linux
To get started with OpenCV, install it from the Fedora Linux repositories.
```
$ sudo dnf install opencv opencv-contrib opencv-doc python3-opencv python3-matplotlib python3-numpy
```
**Note:** On Fedora Silverblue or CoreOs, Python 3.9 is part of the core commit. Layer OpenCV and required tools with: _rpm-ostree install opencv opencv-doc python3-opencv python3-matplotlib python3-numpy_.
Next, enter the following commands in a terminal to verify that OpenCV is installed (user input shown in bold).
```
$ python
Python 3.9.6 (default, Jul 16 2021, 00:00:00)
[GCC 11.1.1 20210531 (Red Hat 11.1.1-3)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv2 as cv
>>> print( cv.__version__ )
4.5.2
>>> exit()
```
The current OpenCV version should be displayed when you enter the _print_ command as shown above. This indicates that OpenCV and the Python-OpenCV libraries have been installed successfully.
Additionally, if you want to take notes and write code with Jupyter Notebook and learn more about data science tools, check out the earlier Fedora Magazine article: [_Jupyter and Data Science in Fedora_][6].
### Get started with OpenCV
After installation is complete, load a sample image using Python and the OpenCV libraries (press the **S** key to save a copy of the image in _png_ format and finish the program):
```
$ cp /usr/share/opencv4/samples/data/starry_night.jpg .
$ python starry_night.py
```
Contents of _starry_night.py_:
```
import cv2 as cv
import sys
img = cv.imread(cv.samples.findFile("starry_night.jpg"))
if img is None:
sys.exit("Could not read the image.")
cv.imshow("Display window", img)
k = cv.waitKey(0)
if k == ord("s"):
cv.imwrite("starry_night.png", img)
```
![][7]
Gray-scale the image by adding the parameter **0** to the _cv.imread_ function as shown below.
```
img = cv.imread(cv.samples.findFile("starry_night.jpg"),0)
```
![][8]
These are some alternative values that can be used for the second parameter of the _cv.imread_ function.
* **cv2.IMREAD_GRAYSCALE** or **0:** Load the image in grayscale mode.
* **cv2.IMREAD_COLOR** or **1:** Load the image in color mode. Any transparency in the image will be removed. This is the default.
* **cv2.IMREAD_UNCHANGED** or **-1:** Load the image unaltered; including alpha channel.
#### Display image attributes using OpenCV
Image attributes include the number of rows, columns, and channels; the type of image data; the number of pixels; etc. Suppose you wanted to access the image’s shape and its datatype. This is how you would do it:
```
import cv2 as cv
img = cv.imread(cv.samples.findFile("starry_night.jpg"))
print("Image size is", img.shape)
print("Data type of image is", img.dtype)
Image size is (600, 752, 3)
Data type of image is uint8
print(f"Image 2D numpy array \n {img}")
Image 2D numpy array
[[[0 0 0]
[0 0 0]
[0 0 0]
...
[0 0 0]
[0 0 0]
[0 0 0]]
[[0 0 0]
[0 0 0]
[0 0 0]
...
```
* **img.shape:** return a tuple of the number of rows, columns, and channels (if it is a color image)
* **img.dtype:** return the datatype of the image
Next display image with Matplotlib:
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),0)
plt.imshow(img)
plt.show()
```
![][9]
#### What happened?
The image was read in as a gray-scale image, however it won’t necessarily display in gray-scale when using Matplotlib’s _imshow_ fucntion. This is because the _imshow_ function uses a different color map by default. To specify that a gray-scale color map should be used, set the second parameter of the _imshow_ function to _cmap=’gray’_ as shown below.
```
plt.imshow(img,cmap='gray')
```
![][10]
This problem is also going to happen when opening a picture in color mode because Matplotlib expects the image in RGB (red, green, blue) format whereas OpenCV stores images in BGR (blue, green, red) format. For correct display, you need to reverse the channels of the BGR image.
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),cv.IMREAD_COLOR)
fig, (ax1, ax2) = plt.subplots(1,2)
ax1.imshow(img)
ax1.set_title('BGR Colormap')
ax2.imshow(img[:,:,::-1])
ax2.set_title('Reversed BGR Colormap(RGB)')
plt.show()
```
![][11]
#### Splitting and merging color channels
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),cv.IMREAD_COLOR)
b,g,r = cv.split(img)
fig,ax = plt.subplots(2,2)
ax[0,0].imshow(r,cmap='gray')
ax[0,0].set_title("Red Channel");
ax[0,1].imshow(g,cmap='gray')
ax[0,1].set_title("Green Channel");
ax[1,0].imshow(b,cmap='gray')
ax[1,0].set_title("Blue Channel");
# Merge the individual channels into a BGR image
imgMerged = cv.merge((b,g,r))
# Show the merged output
ax[1,1].imshow(imgMerged[:,:,::-1])
ax[1,1].set_title("Merged Output");
plt.show()
```
![][12]
* **cv2.split:** Divide a multi-channel array into several single-channel arrays.
* **cv2.merge:** Merge several arrays to make a single multi-channel array. All the input matrices must have the same size.
**Note:** Images with more white have a higher density of color. Contrarily, images with more black have a lower density of color. In the above example the red color has the lowest density.
#### Converting to different color spaces
The _cv2.cvtColor_ function converts an input image from one color space to another. When transforming between the RGB and BGR color spaces, the order of the channels should be specified explicitly (_RGB2BGR_ or _BGR2RGB_). **Note that the default color format in OpenCV is often referred to as RGB but it is actually BGR (the bytes are reversed).** So the first byte in a standard (24-bit) color image will be an 8-bit blue component, the second byte will be green, and the third byte will be red. The fourth, fifth, and sixth bytes would then be the second pixel (blue, then green, then red), and so on.
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),cv.IMREAD_COLOR)
img_rgb = cv.cvtColor(img, cv.COLOR_BGR2RGB)
plt.imshow(img_rgb)
plt.show()
```
![][13]
### Further information
More details on OpenCV are available in the [online documentation][14].
Thank you.
--------------------------------------------------------------------------------
via: https://fedoramagazine.org/use-opencv-on-fedora-linux-part-1/
作者：[Onuralp SEZER][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://fedoramagazine.org/author/thunderbirdtr/
[b]: https://github.com/lujun9972
[1]: https://fedoramagazine.org/wp-content/uploads/2021/08/starry-night-1-816x345.jpg
[2]: https://commons.wikimedia.org/wiki/File:Van_Gogh_-_Starry_Night_-_Google_Art_Project.jpg
[3]: https://en.wikipedia.org/wiki/Computer_vision
[4]: https://en.wikipedia.org/wiki/OpenCV
[5]: https://opencv.org/about/
[6]: https://fedoramagazine.org/jupyter-and-data-science-in-fedora/
[7]: https://fedoramagazine.org/wp-content/uploads/2021/06/image.png
[8]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-1.png
[9]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-2.png
[10]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-3.png
[11]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-4.png
[12]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-5.png
[13]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-7.png
[14]: https://docs.opencv.org/4.5.2/index.html

249
translated/tech/20210802 Use OpenCV on Fedora Linux - part 1.md Normal file
View File

@ -0,0 +1,249 @@
[#]: subject: (Use OpenCV on Fedora Linux ‒ part 1)
[#]: via: (https://fedoramagazine.org/use-opencv-on-fedora-linux-part-1/)
[#]: author: (Onuralp SEZER https://fedoramagazine.org/author/thunderbirdtr/)
[#]: collector: (lujun9972)
[#]: translator: (geekpi)
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
在 Fedora Linux 上使用 OpenCV ‒ 第一部分
======
![][1]
封面图片选自[文森特-凡高][2]的《星空》，公共领域，通过维基共享资源发布
技术世界每天都在变化，对计算机视觉、人工智能和机器学习的需求也在增加。让计算机和手机能够看到周围环境的技术被称为[计算机视觉][3]。重新创造人眼的工作始于 50 年代。从那时起，计算机视觉技术有了长足的发展。计算机视觉已经通过不同的应用进入了我们的手机。这篇文章将介绍 Fedora Linux 上的[OpenCV][4]。
### **什么是 OpenCV?**
> OpenCV （开源计算机视觉库）是一个开源的计算机视觉和机器学习软件库。OpenCV 的建立是为了给计算机视觉应用提供一个通用的基础设施，并加速机器感知在商业产品中的应用。它有超过 2500 种优化算法，其中包括一套全面的经典和最先进的计算机视觉和机器学习算法。这些算法可用于检测和识别人脸，识别物体，对视频中的人类行为进行分类，并建立标记，将其与增强现实叠加等等。
>
> [opencv.org – about][5]
### 在 Fedora Linux 上安装 OpenCV
要开始使用 OpenCV，请从 Fedora Linux 仓库中安装它。
```
$ sudo dnf install opencv opencv-contrib opencv-doc python3-opencv python3-matplotlib python3-numpy
```
**注意：**在 Fedora Silverblue 或 CoreOs 上，Python 3.9 是核心提交的一部分。用以下方法安装 OpenCV 和所需工具：_rpm-ostree install opencv opencv-doc python3-opencv python3-matplotlib python3-numpy_。
接下来，在终端输入以下命令，以验证 OpenCV 是否已经安装（用户输入的内容以粗体显示）。
```
$ python
Python 3.9.6 (default, Jul 16 2021, 00:00:00)
[GCC 11.1.1 20210531 (Red Hat 11.1.1-3)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv2 as cv
>>> print( cv.__version__ )
4.5.2
>>> exit()
```
当你输入 _print_ 命令时，应该显示当前的 OpenCV 版本，如上图所示。这表明 OpenCV 和 Python-OpenCV 库已经成功安装。
此外，如果你想用 Jupyter Notebook 做笔记和写代码，并了解更多关于数据科学工具的信息，请查看早期的 Fedora Magazine 文章：[_Fedora 中的 Jupyter 和数据科学_][6]。
### 开始使用 OpenCV
安装完成后，使用 Python 和 OpenCV 库加载一个样本图像（按 **S** 键以 _png_ 格式保存图像的副本并完成程序）：
```
$ cp /usr/share/opencv4/samples/data/starry_night.jpg .
$ python starry_night.py
```
_starry_night.py_ 的内容：
```
import cv2 as cv
import sys
img = cv.imread(cv.samples.findFile("starry_night.jpg"))
if img is None:
sys.exit("Could not read the image.")
cv.imshow("Display window", img)
k = cv.waitKey(0)
if k == ord("s"):
cv.imwrite("starry_night.png", img)
```
![][7]
通过在 _cv.imread_ 函数中添加参数 **0**，对图像进行灰度处理，如下所示。
```
img = cv.imread(cv.samples.findFile("starry_night.jpg"),0)
```
![][8]
这些是一些可以用于 _cv.imread_ 函数的第二个参数的替代值。
* **cv2.IMREAD_GRAYSCALE** 或 **0：** 以灰度模式加载图像。
* **cv2.IMREAD_COLOR** 或 **1：** 以彩色模式载入图像。图像中的任何透明度将被移除。这是默认的。
* **cv2.IMREAD_UNCHANGED** 或 **-1：**载入未经修改的图像。包括 alpha 通道。
#### 使用 OpenCV 显示图像属性
图像属性包括行、列和通道的数量、图像数据的类型、像素的数量等等。假设你想访问图像的形状和它的数据类型。你可以这样做：
```
import cv2 as cv
img = cv.imread(cv.samples.findFile("starry_night.jpg"))
print("Image size is", img.shape)
print("Data type of image is", img.dtype)
Image size is (600, 752, 3)
Data type of image is uint8
print(f"Image 2D numpy array \n {img}")
Image 2D numpy array
[[[0 0 0]
[0 0 0]
[0 0 0]
...
[0 0 0]
[0 0 0]
[0 0 0]]
[[0 0 0]
[0 0 0]
[0 0 0]
...
```
* **img.shape：** 返回一个行数、列数和通道数的元组（如果是彩色图像）。
* **img.dtype：** 返回图像的数据类型。
接下来用 Matplotlib 显示图像：
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),0)
plt.imshow(img)
plt.show()
```
![][9]
#### 发生了什么？
该图像是作为灰度图像读入的，但是当使用 Matplotlib 的 _imshow_ 函数时，它不一定会以灰度显示。这是因为 _imshow_ 函数默认使用不同的颜色映射。要指定使用灰度颜色映射，请将 _imshow_ 函数的第二个参数设置为 _cmap='gray'_，如下所示。
```
plt.imshow(img,cmap='gray')
```
![][10]
这个问题在以彩色模式打开图片时也会发生，因为 Matplotlib 期望图片为 RGB（红、绿、蓝）格式，而 OpenCV 则以 BGR（蓝、绿、红）格式存储图片。为了正确显示，你需要将 BGR 图像的通道反转。
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),cv.IMREAD_COLOR)
fig, (ax1, ax2) = plt.subplots(1,2)
ax1.imshow(img)
ax1.set_title('BGR Colormap')
ax2.imshow(img[:,:,::-1])
ax2.set_title('Reversed BGR Colormap(RGB)')
plt.show()
```
![][11]
#### 分割和合并颜色通道
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),cv.IMREAD_COLOR)
b,g,r = cv.split(img)
fig,ax = plt.subplots(2,2)
ax[0,0].imshow(r,cmap='gray')
ax[0,0].set_title("Red Channel");
ax[0,1].imshow(g,cmap='gray')
ax[0,1].set_title("Green Channel");
ax[1,0].imshow(b,cmap='gray')
ax[1,0].set_title("Blue Channel");
# Merge the individual channels into a BGR image
imgMerged = cv.merge((b,g,r))
# Show the merged output
ax[1,1].imshow(imgMerged[:,:,::-1])
ax[1,1].set_title("Merged Output");
plt.show()
```
![][12]
* **cv2.split：**将一个多通道数组分割成几个单通道数组。
* **cv2.merge：** 将几个数组合并成一个多通道数组。所有的输入矩阵必须具有相同的大小。
**注意：**白色较多的图像具有较高的颜色密度。相反，黑色较多的图像，其颜色密度较低。在上面的例子中，红色的密度是最低的。
#### 转换到不同的色彩空间
_cv2.cvtColor_ 函数将一个输入图像从一个颜色空间转换到另一个颜色空间。在 RGB 和 BGR 色彩空间之间转换时，应明确指定通道的顺序（_RGB2BGR_ 或 _BGR2RGB_）。**注意，OpenCV 中的默认颜色格式通常被称为 RGB，但它实际上是 BGR（字节是相反的）。**因此，标准（24 位）彩色图像的第一个字节将是一个 8 位蓝色分量，第二个字节是绿色，第三个字节是红色。然后第四、第五和第六个字节将是第二个像素（蓝色，然后是绿色，然后是红色），以此类推。
```
import cv2 as cv
import matplotlib.pyplot as plt
img = cv.imread(cv.samples.findFile("starry_night.jpg"),cv.IMREAD_COLOR)
img_rgb = cv.cvtColor(img, cv.COLOR_BGR2RGB)
plt.imshow(img_rgb)
plt.show()
```
![][13]
### 更多信息
关于 OpenCV 的更多细节可以在[在线文档][14]中找到。
谢谢。
--------------------------------------------------------------------------------
via: https://fedoramagazine.org/use-opencv-on-fedora-linux-part-1/
作者：[Onuralp SEZER][a]
选题：[lujun9972][b]
译者：[geekpi](https://github.com/geekpi)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://fedoramagazine.org/author/thunderbirdtr/
[b]: https://github.com/lujun9972
[1]: https://fedoramagazine.org/wp-content/uploads/2021/08/starry-night-1-816x345.jpg
[2]: https://commons.wikimedia.org/wiki/File:Van_Gogh_-_Starry_Night_-_Google_Art_Project.jpg
[3]: https://en.wikipedia.org/wiki/Computer_vision
[4]: https://en.wikipedia.org/wiki/OpenCV
[5]: https://opencv.org/about/
[6]: https://fedoramagazine.org/jupyter-and-data-science-in-fedora/
[7]: https://fedoramagazine.org/wp-content/uploads/2021/06/image.png
[8]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-1.png
[9]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-2.png
[10]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-3.png
[11]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-4.png
[12]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-5.png
[13]: https://fedoramagazine.org/wp-content/uploads/2021/06/image-7.png
[14]: https://docs.opencv.org/4.5.2/index.html