David's Wiki

OpenCV: Difference between revisions

← Older edit
 
(16 intermediate revisions by the same user not shown)
Line 10: Line 10:
pip install opencv-contrib-python
pip install opencv-contrib-python
</pre>
</pre>
* Missing SURF and SIFT.
** The [https://patents.google.com/patent/US6711293B1/en patent for SIFT] expired on March 6, 2020 so it might be added to opencv-contrib-python in a future update.
** The [https://patents.google.com/patent/US20090238460A1/en patent for SURF] should expire around 2029.
* You can use <code>opencv-python</code> if you do not need extra contrib modules.


=== Non-free Algorithms ===
=== Non-free Algorithms ===


You need to compile [https://github.com/opencv/opencv OpenCV] from source alongside [https://github.com/opencv/opencv_contrib OpenCV-contrib].
You need to compile [https://github.com/opencv/opencv OpenCV] from source alongside [https://github.com/opencv/opencv_contrib OpenCV-contrib].<br>
The easiest way is to use [https://github.com/opencv/opencv-python#manual-builds the automated script] to compile opencv-contrib-python.<br>
{{hidden | Compile opencv-contrib-python |
{{hidden | Compile opencv-contrib-python |
You can build a wheel to install which includes NONFREE modules as follows:
You can build a wheel to install which includes NONFREE modules as follows:
Line 21: Line 26:
sudo apt install -y libavcodec-dev libavformat-dev libswscale-dev
sudo apt install -y libavcodec-dev libavformat-dev libswscale-dev
sudo apt install -y libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev
sudo apt install -y libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev
sudo apt install -y libpng-dev libjpeg-dev libopenexr-dev libtiff-dev libwebp-dev libjasper-dev
sudo apt install -y libpng-dev libjpeg-dev libopenexr-dev libtiff-dev libwebp-dev
sudo add-apt-repository "deb http://security.ubuntu.com/ubuntu xenial-security main"
sudo add-apt-repository "deb http://security.ubuntu.com/ubuntu xenial-security main"
sudo apt update
sudo apt update
Line 27: Line 32:
# sudo apt install -y tesseract-ocr* liblept5 leptonica-progs libleptonica-dev
# sudo apt install -y tesseract-ocr* liblept5 leptonica-progs libleptonica-dev
</syntaxhighlight>
</syntaxhighlight>
* Setup a build environment
* Setup a build environment.
** Note that whatever version of qt you use to build, you will also need in the environment you install the wheel in. Ideally, build it in the same environment you want to use it in.
<syntaxhighlight lang="bash">
<syntaxhighlight lang="bash">
conda create -n opencvbuild python=3.7
conda create -n opencvbuild python=3.7
Line 74: Line 80:
import cv2
import cv2


# cv2.IMREAD_ANYCOLOR = 4
# cv2.IMREAD_ANYCOLOR
# cv2.IMREAD_ANYDEPTH = 2
# cv2.IMREAD_ANYDEPTH
# cv2.IMREAD_COLOR = 1
# cv2.IMREAD_COLOR
# cv2.IMREAD_GRAYSCALE = 0
# cv2.IMREAD_GRAYSCALE
# cv2.IMREAD_UNCHANGED # G or BGR or BGRA


# Use cv2.IMREAD_GRAYSCALE to read in grayscale
# Use cv2.IMREAD_GRAYSCALE to read in grayscale
Line 91: Line 98:
===Resizing an Image===
===Resizing an Image===
[https://docs.opencv.org/master/da/d6e/tutorial_py_geometric_transformations.html Reference]
[https://docs.opencv.org/master/da/d6e/tutorial_py_geometric_transformations.html Reference]
<syntaxhighlight lang="bash">
<syntaxhighlight lang="python">
import numpy as np
# Resize to resolution
import cv2
new_img = cv2.resize(img, (500,200), interpolation=v2.INTER_CUBIC)
# Resize by factor
cv2.resize(img, (0, 0), fx=1/4, fy=1/4, interpolation=cv2.INTER_AREA)
</syntaxhighlight>
 
;Interpolation options
* INTER_NEAREST
* INTER_LINEAR
* INTER_CUBIC
* INTER_AREA
* INTER_LANCZOS4
 
For downscaling, use `INTER_AREA` to avoid aliasing. However, `INTER_NEAREST` will give the the optimal speed.<br>
For upscaling, use `INTER_CUBIC` for best results or `INTER_LINEAR` for best performance.


img = cv2.imread('messi5.jpg')
===Face Detection===
new_img = cv2.resize(img, (500,200), interpolation = cv.INTER_CUBIC)
[https://towardsdatascience.com/face-detection-in-2-minutes-using-opencv-python-90f89d7c0f81?gi=2fadb2dbc99d face detection in 2 minutes]
# Download [https://raw.githubusercontent.com/kipr/opencv/master/data/haarcascades/haarcascade_frontalface_default.xml haarcascade_frontalface_default.xml]
<syntaxhighlight lang="python">
face_cascade = cv2.CascadeClassifier(
            'haarcascade_frontalface_default.xml')
image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(image_gray, 1.1, 4)
</syntaxhighlight>
</syntaxhighlight>


Line 108: Line 134:


frame_num = 0
frame_num = 0
while video_capture.isOpened() and frame_num < 30:
while video_capture.isOpened():
     ret, frame = video_capture.read()
     ret, frame = video_capture.read()
     if ret:
     if ret:
Line 128: Line 154:
total_frames = int(video_capture.get(cv2.CAP_PROP_FRAME_COUNT))
total_frames = int(video_capture.get(cv2.CAP_PROP_FRAME_COUNT))
frame_pos = int(video_capture.get(cv2.CAP_PROP_POS_FRAMES))
frame_pos = int(video_capture.get(cv2.CAP_PROP_POS_FRAMES))
fps = video_capture.get(cv2.CAP_PROP_FPS)
</syntaxhighlight>
</syntaxhighlight>


Line 149: Line 176:
output_video.release()
output_video.release()
</syntaxhighlight>
</syntaxhighlight>
==Algorithms==
===Stereo Rectify===
This is a summary of the [https://docs.opencv.org/4.x/d9/d0c/group__calib3d.html#ga617b1685d4059c6040827800e72ad2b6 <code>cv2.stereoRectify</code>] function which produces a rectification rotation and projection matrix for calibrated cameras (i.e. cameras with known intrinsics and extrinsics).
[https://github.com/opencv/opencv/blob/4.x/modules/calib3d/src/calibration.cpp#LL2561C6-L2561C21 source code]
;Algorithm
# Rotation calculation
# Projection calculation
# Bounding box calculation
# Disparity-to-depth matrix calculation


==Resources==
==Resources==


*[https://docs.opencv.org/master/d6/d00/tutorial_py_root.html OpenCV Python Tutorial]
*[https://docs.opencv.org/master/d6/d00/tutorial_py_root.html OpenCV Python Tutorial]
Retrieved from "https://wiki.davidl.me/view/OpenCV"