awin/Obsidian-Main
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

vault backup: 2025-02-10 17:27:57

Browse Source
This commit is contained in:
Awin Huang
2025-02-10 17:27:57 +08:00
parent 60a1c02c6a
commit 5752038f86
159 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
21.01. Programming/OpenCV/OpenCV API.md Normal file
View File

@@ -0,0 +1,38 @@
## cv2.erode()
### 用途
1. Erosion 影像侵蝕對於移除影像中的小白雜點很有幫助，可用來去噪，例如影像中的小雜點，雜訊。
2. 細化影像，消除毛刺。
### Define
```python
cv.erode(src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) ->dst
```
### Arguments
`src`: 二值化的影像
`kernel`: 捲積 kernel
`iterations`: 迭代次數（預設為1）
## cv2.dilate()
### 用途
1. Dilation 影像膨脹通常是配合著影像侵蝕 Erosion 使用，先使用侵蝕的方式使影像中的線條變窄，同時也去除雜訊，之後再透過 Dilation 將影像膨脹回來。
2. 用來連接兩個很靠近但分開的物體。
### Define
```python
cv.dilate(src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) ->dst
```
### Arguments
`src`: 二值化的影像
`kernel`: 捲積 kernel
`iterations`: 迭代次數（預設為1）
## cv2.GaussianBlur()
高斯濾波與平均濾波 Averaging 類似，平均濾波 Averaging 的 kernel 裡的每個 pixel 權重都是1，而高斯濾波給予每個 pixel 不同權重，中心 pixel 的權重最高，越往邊角權重就越低，相較於平均濾波 Averaging 這樣可以讓圖片失真較少，高斯濾波通常去除雜訊也有不錯的效果。
## cv2.canny()
```python
cv.Canny( image, threshold1, threshold2[, edges[, apertureSize[, L2gradient]]] ) -> edges
```
Canny函数的使用很简单，指定最大（`threshold2`）和最小 threshold（`threshold1`），其中 `apertureSize` 預設為3。

389
21.01. Programming/OpenCV/OpenCV.md Normal file
View File

@@ -0,0 +1,389 @@
## Build code
### 前置作業
1. 要有NVIDIA GPU
2. 安裝NVIDIA GPU driver
3. 安裝[CUDA](https://developer.nvidia.com/cuda-downloads)
4. 安裝[cuDNN](https://developer.nvidia.com/rdp/cudnn-download)
5. 安裝[CMake](https://cmake.org/download/)
### CMake && Visual Studio
1. 下載OpenCV：[opencv](https://github.com/opencv/opencv)
2. 解壓縮後放到`D:\temp\build_opencv\opencv-4.5.3\source`
3. 建立`D:\temp\build_opencv\opencv-4.5.3\build`
4. 下載OpenCV contrib：[opencv_contrib](https://github.com/opencv/opencv_contrib)
5. 解壓縮後放到`D:\temp\build_opencv\opencv_contrib`
6. 打開cmake-gui
7. 如圖設定
![[Pasted image 20210818115317.png]]
8. 用`Add Entry`先加入以下define
![[Pasted image 20210818115507.png]]
如果是檔名，type就選FILEPATH，如果是目錄，type就選PATH
- PYTHON3_EXECUTABLE=C:/python39/python.exe
- PYTHON3_INCLUDE_DIR=C:/python39/include
- PYTHON3_LIBRARY=C:/python39/libs/python39.lib
- PYTHON3_NUMPY_INCLUDE_DIRS=C:/python39/Lib/site-packages/numpy/core/include
- PYTHON3_PACKAGES_PATH=C:/python39/Lib/site-packages
如果要build win32 + Python 3.6.3 x86的話，改為以下設定：
- PYTHON3_EXECUTABLE=C:/Python363/python.exe
- PYTHON3_INCLUDE_DIR=C:/Python363/include
- PYTHON3_LIBRARY=C:/Python363/libs/python36.lib
- PYTHON3_NUMPY_INCLUDE_DIRS=C:/Python363/Lib/site-packages/numpy/core/include
- PYTHON3_PACKAGES_PATH=C:/Python363/Lib/site-packages
9. 按Configure，會跳出一個視窗，platform選x64
![[Pasted image 20210818115809.png]]
10. 這些要打勾
- BUILD_opencv_world
- BUILD_opencv_python3
- OPENCV_DNN_CUDA
- OPENCV_PYTHON3_VERSION
- OPENCV_FORCE_PYTHON_LIBS
- OPENCV_ENABLE_NONFREE
- ENABLE_FAST_MATH
- WITH_CUDA
- WITH_OPENMP
- OPENCV_EXTRA_MODULES_PATH=D:/temp/build_opencv/opencv_contrib/modules
11. 再按一次Configure，必須沒有錯誤的跑完，像是
```
General configuration for OpenCV 4.5.3 =====================================
Version control: unknown
Extra modules:
Location (extra): D:/temp/build_opencv/opencv_contrib/modules
Version control (extra): 4.5.3-6-g907efb96
Platform:
Timestamp: 2021-08-18T03:40:06Z
Host: Windows 10.0.19043 AMD64
CMake: 3.21.1
CMake generator: Visual Studio 16 2019
CMake build tool: C:/Program Files (x86)/Microsoft Visual Studio/2019/Professional/MSBuild/Current/Bin/MSBuild.exe
MSVC: 1929
Configuration: Debug Release
CPU/HW features:
Baseline: SSE SSE2 SSE3
requested: SSE3
Dispatched code generation: SSE4_1 SSE4_2 FP16 AVX AVX2 AVX512_SKX
requested: SSE4_1 SSE4_2 AVX FP16 AVX2 AVX512_SKX
SSE4_1 (17 files): + SSSE3 SSE4_1
SSE4_2 (2 files): + SSSE3 SSE4_1 POPCNT SSE4_2
FP16 (1 files): + SSSE3 SSE4_1 POPCNT SSE4_2 FP16 AVX
AVX (5 files): + SSSE3 SSE4_1 POPCNT SSE4_2 AVX
AVX2 (31 files): + SSSE3 SSE4_1 POPCNT SSE4_2 FP16 FMA3 AVX AVX2
AVX512_SKX (7 files): + SSSE3 SSE4_1 POPCNT SSE4_2 FP16 FMA3 AVX AVX2 AVX_512F AVX512_COMMON AVX512_SKX
C/C++:
Built as dynamic libs?: YES
C++ standard: 11
C++ Compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Professional/VC/Tools/MSVC/14.29.30037/bin/Hostx64/x64/cl.exe (ver 19.29.30040.0)
C++ flags (Release): /DWIN32 /D_WINDOWS /W4 /GR /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /fp:fast /EHa /wd4127 /wd4251 /wd4324 /wd4275 /wd4512 /wd4589 /MP /MD /O2 /Ob2 /DNDEBUG
C++ flags (Debug): /DWIN32 /D_WINDOWS /W4 /GR /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /fp:fast /EHa /wd4127 /wd4251 /wd4324 /wd4275 /wd4512 /wd4589 /MP /MDd /Zi /Ob0 /Od /RTC1
C Compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Professional/VC/Tools/MSVC/14.29.30037/bin/Hostx64/x64/cl.exe
C flags (Release): /DWIN32 /D_WINDOWS /W3 /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /fp:fast /MP /MD /O2 /Ob2 /DNDEBUG
C flags (Debug): /DWIN32 /D_WINDOWS /W3 /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /fp:fast /MP /MDd /Zi /Ob0 /Od /RTC1
Linker flags (Release): /machine:x64 /INCREMENTAL:NO
Linker flags (Debug): /machine:x64 /debug /INCREMENTAL
ccache: NO
Precompiled headers: NO
Extra dependencies: cudart_static.lib nppc.lib nppial.lib nppicc.lib nppidei.lib nppif.lib nppig.lib nppim.lib nppist.lib nppisu.lib nppitc.lib npps.lib cublas.lib cudnn.lib cufft.lib -LIBPATH:C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.4/lib/x64
3rdparty dependencies:
OpenCV modules:
To be built: aruco barcode bgsegm bioinspired calib3d ccalib core cudaarithm cudabgsegm cudacodec cudafeatures2d cudafilters cudaimgproc cudalegacy cudaobjdetect cudaoptflow cudastereo cudawarping cudev datasets dnn dnn_objdetect dnn_superres dpm face features2d flann fuzzy gapi hfs highgui img_hash imgcodecs imgproc intensity_transform line_descriptor mcc ml objdetect optflow phase_unwrapping photo plot python3 quality rapid reg rgbd saliency shape stereo stitching structured_light superres surface_matching text tracking ts video videoio videostab wechat_qrcode world xfeatures2d ximgproc xobjdetect xphoto
Disabled: -
Disabled by dependency: -
Unavailable: alphamat cvv freetype hdf java julia matlab ovis python2 python2 sfm viz
Applications: tests perf_tests apps
Documentation: NO
Non-free algorithms: NO
Windows RT support: NO
GUI:
Win32 UI: YES
VTK support: NO
Media I/O:
ZLib: build (ver 1.2.11)
JPEG: build-libjpeg-turbo (ver 2.1.0-62)
WEBP: build (ver encoder: 0x020f)
PNG: build (ver 1.6.37)
TIFF: build (ver 42 - 4.2.0)
JPEG 2000: build (ver 2.4.0)
OpenEXR: build (ver 2.3.0)
HDR: YES
SUNRASTER: YES
PXM: YES
PFM: YES
Video I/O:
DC1394: NO
FFMPEG: YES (prebuilt binaries)
avcodec: YES (58.134.100)
avformat: YES (58.76.100)
avutil: YES (56.70.100)
swscale: YES (5.9.100)
avresample: YES (4.0.0)
GStreamer: NO
DirectShow: YES
Media Foundation: YES
DXVA: YES
Parallel framework: Concurrency
Trace: YES (with Intel ITT)
Other third-party libraries:
Intel IPP: 2020.0.0 Gold [2020.0.0]
at: D:/temp/build_opencv/opencv-4.5.3/build/3rdparty/ippicv/ippicv_win/icv
Intel IPP IW: sources (2020.0.0)
at: D:/temp/build_opencv/opencv-4.5.3/build/3rdparty/ippicv/ippicv_win/iw
Lapack: NO
Eigen: NO
Custom HAL: NO
Protobuf: build (3.5.1)
NVIDIA CUDA: YES (ver 11.4, CUFFT CUBLAS)
NVIDIA GPU arch: 35 37 50 52 60 61 70 75 80 86
NVIDIA PTX archs:
cuDNN: YES (ver 8.2.2)
OpenCL: YES (NVD3D11)
Include path: D:/temp/build_opencv/opencv-4.5.3/source/3rdparty/include/opencl/1.2
Link libraries: Dynamic load
Python 3:
Interpreter: C:/python39/python.exe (ver 3.9.6)
Libraries: C:/python39/libs/python39.lib (ver 3.9.6)
numpy: C:/python39/Lib/site-packages/numpy/core/include (ver 1.19.5)
install path: C:/python39/Lib/site-packages/cv2/python-3.9
Python (for build): C:/python39/python.exe
Java:
ant: NO
JNI: NO
Java wrappers: NO
Java tests: NO
Install to: D:/temp/build_opencv/opencv-4.5.3/build/install
-----------------------------------------------------------------
Configuring done
```
12. 按下Generate按鈕就會生sln檔案。
### CMake by command
用command line的話就可以將多個平台一次編起來，不用一直改GUI。
寫了一個`build.bat`可以編譯不同版本與平台：
```bat
rem build.bat
echo on
rem build.bat
rem win32 ARCH=%1, win32/x64
rem 15 VS_CODE=%2, 15/16
rem 2017 VS_VERSION=%3, 2017/2019
rem DEBUG BUILD_TYPE=%4, DEBUG/RELEASE
rem "C:/Python363/python.exe" PYTHON3_EXE=%5
rem "C:/Python363/include" PYTHON3_INCLUDE=%6
rem "C:/Python363/libs/python36.lib" PYTHON3_LIB=%7
rem "C:/Python363/Lib/site-packages/numpy/core/include" PYTHON3_NP_INCLUDE=%8
rem "C:/Python363/Lib/site-packages" PYTHON3_PACKAGES=%9
rem 3.6.3 PY_VERSION=%10
rem 1 CLEAN_BUILD=%11, 1: Delete build folder, 0: Do nothing
set DO_CONFIG=1
set DO_BUILD=1
set ARCH=%1
set VS_CODE=%2
set VS_VERSION=%3
set BUILD_TYPE=%4
set PYTHON3_EXE=%5
set PYTHON3_INCLUDE=%6
set PYTHON3_LIB=%7
set PYTHON3_NP_INCLUDE=%8
set PYTHON3_PACKAGES=%9
set GENERATOR="Visual Studio %VS_CODE% %VS_VERSION%"
shift
set PY_VERSION=%9
set CV_VERSION=4.5.3
set CV_SOURCE="opencv-%CV_VERSION%\source"
set CV_BUILD=build\vs%VS_VERSION%.%ARCH%.%BUILD_TYPE%
set CV_EXTRA_MODULES="opencv_contrib\modules"
shift
set CLEAN_BUILD=%9
if %CLEAN_BUILD% == 1 (
rmdir /s /q %CV_BUILD%
)
if %DO_CONFIG% == 1 (
mkdir %CV_BUILD%
"C:\Program Files\CMake\bin\cmake.exe" ^
-B%CV_BUILD% ^
-H%CV_SOURCE% ^
-G%GENERATOR% ^
-A%ARCH% ^
-DCMAKE_BUILD_TYPE=%BUILD_TYPE% ^
-DOPENCV_EXTRA_MODULES_PATH=%CV_EXTRA_MODULES% ^
-DCMAKE_INSTALL_PREFIX=%CV_BUILD% ^
-DINSTALL_PYTHON_EXAMPLES=OFF ^
-DINSTALL_C_EXAMPLES=OFF ^
-DBUILD_opencv_python3=ON ^
-DPYTHON3_EXECUTABLE=%PYTHON3_EXE% ^
-DPYTHON3_INCLUDE_DIR=%PYTHON3_INCLUDE% ^
-DPYTHON3_LIBRARY=%PYTHON3_LIB% ^
-DPYTHON3_NUMPY_INCLUDE_DIRS=%PYTHON3_NP_INCLUDE% ^
-DPYTHON3_PACKAGES_PATH=%PYTHON3_PACKAGES% ^
-DOPENCV_PYTHON3_VERSION=%PY_VERSION% ^
-DOPENCV_FORCE_PYTHON_LIBS=ON ^
-DBUILD_opencv_world=ON ^
-DOPENCV_ENABLE_NONFREE=ON ^
-DENABLE_FAST_MATH=ON ^
-DWITH_OPENMP=ON ^
-DWITH_OPENGL=ON
)
if %DO_BUILD% == 1 (
"C:\Program Files\CMake\bin\cmake.exe" --build %CV_BUILD% --target INSTALL --config %BUILD_TYPE%
)
```
這樣以後就可以用參數的方法來設定，譬如要使用vs2019來編譯x64 release的話，就可以這樣下：
```bat
rem build_vs2019.x64.RELEASE.bat
echo off
rem ----------------------------------------------------------------------------------------------------
rem
rem x64, RELEASE, 2019, CLEAN
rem
set ARCH=x64
set VS_CODE=16
set VS_VERSION=2019
set BUILD_TYPE=RELEASE
set PYTHON3_EXE="C:/Python39/python.exe"
set PYTHON3_INCLUDE="C:/Python39/include"
set PYTHON3_LIB="C:/Python39/libs/python39.lib"
set PYTHON3_NP_INCLUDE="C:/Python39/Lib/site-packages/numpy/core/include"
set PYTHON3_PACKAGES="C:/Python39/Lib/site-packages"
set PY_VERSION=3.9
set CLEAN_BUILD=1
echo "============================================================"
echo "Build %ARCH%.%VS_VERSION%(%VS_CODE%).%BUILD_TYPE%, Python=%PY_VERSION%, CLEAN_BUILD=%CLEAN_BUILD%"
echo ""
call build.bat ^
%ARCH% %VS_CODE% %VS_VERSION% %BUILD_TYPE% %PYTHON3_EXE% ^
%PYTHON3_INCLUDE% %PYTHON3_LIB% %PYTHON3_NP_INCLUDE% %PYTHON3_PACKAGES% ^
%PY_VERSION% %CLEAN_BUILD%
rem ----------------------------------------------------------------------------------------------------
```
### 參考
- [Build OpenCV GPU Version On Windows 10](https://medium.com/chung-yi/build-opencv-gpu-version-on-windows-10-c37a33437525)
- [Cannot install openCV 3.1.0 with python3. CMAKE not including or linking python correctly](https://stackoverflow.com/questions/42638342/cannot-install-opencv-3-1-0-with-python3-cmake-not-including-or-linking-python)
- [在Windows上编译带CUDA(GPU)的OpenCV](https://shaogui.life/2021/03/08/%E5%9C%A8Windows%E4%B8%8A%E7%BC%96%E8%AF%91%E5%B8%A6CUDA(GPU)%E7%9A%84opencv/)
- [How to use OpenCV DNN Module with Nvidia GPU on Windows](https://learnopencv.com/how-to-use-opencv-dnn-module-with-nvidia-gpu-on-windows/)
- [Accelerate OpenCV 4.5.0 on Windows – build with CUDA and python bindings - James Bowley](https://jamesbowley.co.uk/accelerate-opencv-4-5-0-on-windows-build-with-cuda-and-python-bindings/#python_bindings)
## 影像處理
- sift
```python
def adjust_gamma(image, gamma=1.0):
invGamma = 1.0 / gamma
table = np.array([((i / 255.0) ** invGamma) * 255
for i in np.arange(0, 256)]).astype("uint8")
return cv2.LUT(image, table)
def sift_compare(path_a, path_b):
'''
Use SIFT features to measure image similarity
@args:
{str} path_a: the path to an image file
{str} path_b: the path to an image file
@returns:
TODO
'''
# initialize the sift feature detector
orb = cv2.ORB_create()
# get the images
img_a = cv2.imread(path_a)
img_b = cv2.imread(path_b)
img_a = adjust_gamma(img_a, 0.1)
img_b = adjust_gamma(img_b, 0.1)
# find the keypoints and descriptors with SIFT
kp_a, desc_a = orb.detectAndCompute(img_a, None)
kp_b, desc_b = orb.detectAndCompute(img_b, None)
# print(f'len kp_a = {len(kp_a)}, len desc_a = {len(desc_a)}')
# print(f'type kp_b = {type(kp_b)}, type desc_b = {type(desc_b)}')
# print(f'len kp_b = {len(kp_b)}, len desc_b = {len(desc_b) if desc_b else 0}')
if desc_a is None or desc_b is None:
# rr.LOG('Error: desc_a = {}, desc_b = {}'.format(type(desc_a), type(desc_b)))
# rr.LOG('Error: Return score 0.')
return 0
# Initialize the bruteforce matcher
bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)
# match.distance is a float between {0:100} - lower means more similar
matches = bf.match(desc_a, desc_b)
similar_regions = [i for i in matches if i.distance < 70]
# print('Length of similar_regions = {}'.format(len(similar_regions)))
# print('Length of matches = {}'.format(len(matches)))
sorted_matches = sorted(matches, key = lambda x:x.distance)
good_matches = sorted_matches[:int(len(sorted_matches) * 0.5)]
similar_regions2 = [i for i in good_matches if i.distance < 70]
# print('Length of good_matches = {}'.format(len(good_matches)))
if len(matches) == 0:
return 0
return len(similar_regions) / len(matches)
```
### Color tempertature
code
```
img1 = cv2.imread(file)
B, G, R = cv2.split(img1)
avgB = cv2.mean(B)[0]
avgG = cv2.mean(G)[0]
avgR = cv2.mean(R)[0]
# X = avgR * -0.14282 + avgG * 1.54924 + avgB * -0.95641
# Y = avgR * -0.32466 + avgG * 1.57837 + avgB * -0.73191
# Z = avgR * -0.68202 + avgG * 0.77073 + avgB * 0.56332
X = 2.789 * avgR + 1.7517 * avgG + 1.1302 * avgB
Y = 1 * avgR + 4.5907 * avgG + 0.0601 * avgB
Z = 0 * avgR + 0.0565 * avgG + 5.5943 * avgB
x = X / (X + Y + Z)
y = Y / (X + Y + Z)
n = (x - 0.3320) / (0.1858 - y)
# n = (0.23881 * avgR + 0.25499 * avgG + -0.58291 * avgB) / (0.11109 * avgR + -0.85406 * avgG + 0.52289 * avgB)
CCT = 449 * pow(n, 3) + 3525 * pow(n, 2) + 6823.3 * n + 5520.33
# CCT = 437 * pow(n, 3) + 3601 * pow(n, 2) + 6831 * n + 5517
# print(f'x = {x}')
# print(f'y = {y}')
# print(f'n = {n}')
print(f'{file}: CCT = {CCT}')
```
- [How to Calculate the Color Temperature / Tint of the Colors in an Image?](https://dsp.stackexchange.com/questions/8949/how-to-calculate-the-color-temperature-tint-of-the-colors-in-an-image)
- [Calculating Color Temperature and Illuminance using the TAOS TCS3414CS Digital Color Sensor_](https://ams.com/documents/20143/80162/TCS34xx_AN000517_1-00.pdf)
- [Java RGB转色温(CCT)](https://www.codeleading.com/article/16404945886/)
- [Calculate color temperature (CCT) from CIE 1931 xy coordinates](https://www.waveformlighting.com/tech/calculate-color-temperature-cct-from-cie-1931-xy-coordinates)
- [OpenCV: Color conversions](https://docs.opencv.org/3.4/de/d25/imgproc_color_conversions.html)
- [sRGB色彩空間 - 維基百科，自由的百科全書](https://zh.wikipedia.org/wiki/SRGB%E8%89%B2%E5%BD%A9%E7%A9%BA%E9%97%B4)
- [Color temperature - Wikipedia](https://en.wikipedia.org/wiki/Color_temperature#Approximation)
- [What color is a blackbody? - some pixel rgb values](http://www.vendian.org/mncharity/dir3/blackbody/)