openCV

OpenCV的全稱是Open Source Computer Vision Library，是一個跨平台的電腦視覺庫。OpenCV是由英特爾公司發起並參與開發，以BSD授權條款授權發行，可以在商業和研究領域中免費使用。OpenCV可用於開發即時的圖像處理、電腦視覺以及圖型識別程式。

Raspberry Pi安裝OpenCV
sudo apt-get install libopencv-dev
on Ubuntu: sudo apt-get install python3-opencv

A.安裝 OpenCV 相關軟體。步驟 1:先安裝好作業系統。
可參考安裝Linux 在Raspberry Pi4。
步驟2:更新當前安裝的軟體。
$ sudo apt-get update
$ sudo apt-get upgrate
步驟3:安裝 OpenCV 編譯所需的軟體。
$ sudo apt-get install cmake build-essential pkg-config git
步驟4:安裝常用圖像和視頻格式的支援軟體。
$ sudo apt-get install libjpeg-dev libtiff-dev libjasper-dev libpng-dev libwebp-dev libopenexr-dev
$ sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libxvidcore-dev libx264-dev libdc1394–22-dev libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev
步驟5:安裝 OpenCV 界面所需的軟體。
$ sudo apt-get install libgtk-3-dev libqtgui4 libqtwebkit4 libqt4-test python3-pyqt5
步驟6:安裝提高 OpenCV 運行速度的軟體。
$ sudo apt-get install libatlas-base-dev liblapacke-dev gfortran
步驟7:安裝 HDF5軟體。
$ sudo apt-get install libhdf5-dev libhdf5–103
步驟8:安裝 Python。
$ sudo apt-get install python3-dev python3-pip python3-numpy

OpenCV函數

imread("file")

# coding=big5
# Python program to use openCV functions
import cv2
src = cv2.imread("square-1.png")  #read-in an image file
gray = cv2.imread("square-1.png",cv2.IMREAD_GRAYSCALE)  #convert it to a gray scale
print(type(src))
print(len(src))
print(src.shape)
(x,y,z)=src.shape
print((x,y,z))
for i in range(0,x,4):
    if(src[i][50][0] > 233 ): cc='White'
    elif(src[i][50][0] > 163 and src[i][50][2] < 220 ): cc='BLUE'
    else: cc='Black'
    print(i,50,src[i][50][0],src[i][50][1],src[i][50][2], cc)


160
(160, 202, 3)
(160, 202, 3)
0 50 255 255 255 White
4 50 255 255 255 White
8 50 255 255 255 White
12 50 255 255 255 White
16 50 255 255 255 White
20 50 255 255 255 White
24 50 255 255 255 White
28 50 255 255 255 White
32 50 0 0 0 Black
36 50 0 0 0 Black
40 50 232 162 0 BLUE
44 50 232 162 0 BLUE
48 50 232 162 0 BLUE
52 50 232 162 0 BLUE
56 50 232 162 0 BLUE
60 50 232 162 0 BLUE
64 50 232 162 0 BLUE
68 50 232 162 0 BLUE
72 50 232 162 0 BLUE
76 50 232 162 0 BLUE
80 50 232 162 0 BLUE
84 50 232 162 0 BLUE
88 50 232 162 0 BLUE
92 50 232 162 0 BLUE
96 50 232 162 0 BLUE
100 50 232 162 0 BLUE
104 50 232 162 0 BLUE
108 50 232 162 0 BLUE
112 50 232 162 0 BLUE
116 50 232 162 0 BLUE
120 50 0 0 0 Black
124 50 0 0 0 Black
128 50 255 255 255 White
132 50 255 255 255 White
136 50 255 255 255 White
140 50 255 255 255 White
144 50 255 255 255 White
148 50 255 255 255 White
152 50 255 255 255 White
156 50 255 255 255 White

RaspberryOpenCV 範例程式

# Python program to use openCV functions
import cv2
src = cv2.imread("star.jpg")  #read-in image
# Using cv2.COLOR_BGR2GRAY  convert to gray scale
image = cv2.cvtColor(src, cv2.COLOR_BGR2GRAY )
cv2.imwrite("star_gray.jpg", image)  #write image to a new file
# cv2.rectangle() method  draw a rectange
start_point = (5, 5)
end_point = (220, 220)
color = (255, 0, 0) # Blue color in BGR
thickness = 4  # Line thickness of 4 px
image = cv2.rectangle(src, start_point, end_point, color, thickness)
image = cv2.rectangle(image, (100,100),(400,400), (0,0,255), thickness)
cv2.imwrite("star_rect.jpg", image)
#---------------------------------------------------------
# put 2 images in parallel(hstack) and serial(vstack)
import numpy as np
a=cv2.imread("star.jpg")
b=cv2.imread("star_rect.jpg")
c=np.hstack((a,b))
d=np.vstack((a,b))
cv2.imwrite("star_hstack.jpg", c)
cv2.imwrite("star_vstack.jpg", d)

openCV

Raspberry Pi安裝OpenCV

OpenCV函數

RaspberryOpenCV 範例程式