OpenCV使用CUDA處理圖像的教程與實(shí)戰(zhàn)

重磅干貨，第一時(shí)間送達(dá)

import?cv2?as?cv

gpu_frame?=?cv.cuda_GpuMat()

screenshot?=?cv.imread('media/drip.png')

gpu_frame.upload(screenshot)

gpu_frame.download()

• 在單張圖像上使用

• 在多張圖像上使用

• 對(duì)多張圖像使用Dask進(jìn)行并行延時(shí)處理

第1步：上傳

import?cv2?as?cv

gpu_frame?=?cv.cuda_GpuMat()

screenshot?=?cv.imread('media/drop.png')

gpu_frame.upload(screenshot)

第2步：處理圖像

screenshot?=?cv.cuda.cvtColor(gpu_frame,?cv.COLOR_RGB2BGR)

screenshot?=?cv.cuda.resize(screenshot,?(400,?400))

第3步：下載

screenshot.download()

import?cv2?as?cv

img_files?=?['bear.png',?'drip.png',?'tldr.png',?'frog.png']

#?創(chuàng)建幀來保存圖片(cv2.cuda_GpuMat)
gpu_frame?=?cv.cuda_GpuMat()

for?i?in?range(len(img_files)):
????#?加載圖像(CPU)
????screenshot?=?cv.imread(f"media/{img_files[i]}")

????#?上傳到GPU
????gpu_frame.upload(screenshot)
????
????#?轉(zhuǎn)換顏色到opencv?(numpy)?ndarray→cv2.cuda_GpuMat
????screenshot?=?cv.cuda.cvtColor(gpu_frame,?cv.COLOR_RGB2BGR)
????
????#?反向閾值@?100
????screenshot?=?cv.cuda.threshold(screenshot,?105,?255,?cv.THRESH_BINARY_INV)
????????
????#?調(diào)整圖像
????screenshot?=?cv.cuda.resize(screenshot[1],?(200,?200))

????#?從GPU下載圖像(cv2)?cuda_GpuMat→numpy.ndarray
????screenshot?=?screenshot.download()

import?cv2?as?cv
import?dask.delayed


@dask.delayed
def?preprocess(files):
????#?復(fù)制圖像文件
????i_files?=?files.copy()
????
????#?創(chuàng)建GPU幀來保存圖像
????gpu_frame?=?cv.cuda_GpuMat()
????
????for?i?in?range(len(i_files)):
????????#?加載圖像(CPU)
????????screenshot?=?cv.imread(f'media/{i_files[i]}')

????????#?上傳到GPU
????????gpu_frame.upload(screenshot)

????????#?轉(zhuǎn)換顏色到opencv?(numpy)?ndarray→cv2.cuda_GpuMat
????????screenshot?=?cv.cuda.cvtColor(gpu_frame,?cv.COLOR_RGB2BGR)
????????screenshot?=?cv.cuda.cvtColor(screenshot,?cv.COLOR_BGR2GRAY)

????????#?反向閾值@?100
????????screenshot?=?cv.cuda.threshold(screenshot,?125,?255,?cv.THRESH_BINARY)

????????#?調(diào)整圖像
????????screenshot?=?cv.cuda.resize(screenshot[1],?(200,?200))
????????
????????#?從GPU下載圖像?(cv2.cuda_GpuMat?->?numpy.ndarray)
????????screenshot?=?screenshot.download()

????????#?用新圖像
????????i_files[i]?=?screenshot
????
????#?輸出預(yù)處理圖像
????return?i_files

from?dask?import?compute

img_files?=?['bear.png',?'drip.png',?'tldr.png',?'frog.png']
img_files_2?=?['apple.png',?'eye.png',?'window.png',?'blinds.png']

#?設(shè)置延遲
set_a?=?dask.delayed(preprocess)(img_files)
set_b?=?dask.delayed(preprocess)(img_files_2)

#?開始計(jì)算
out_a,?out_b?=?compute(*[set_a,?set_b])

進(jìn)階閱讀：

• https://medium.com/dropout-analytics/opencv-cuda-for-videos-f3dcf346e398
• https://medium.com/dropout-analytics/pycuda-on-jetson-nano-7990decab299
• https://medium.com/dropout-analytics/beginners-guide-to-knn-with-cuml-ddca099f9e9d

參考引用

1. Koriukina, Valeriia. “Getting Started with OpenCV CUDA Module.” Learn OpenCV, Learnopencv.com, 15 Sept. 2020, learnopencv.com/getting-started-opencv-cuda-modul.
2. McWhorter, Paul. “AI on the Jetson Nano LESSON 10: Installing OpenCV for Python 3.” Paul McWhorter — YouTube, Youtube.com/User/Mcwhorpj, 2 Nov. 2019, youtu.be/3QYayL5y2hk.
3. Pulli, Kari; Baksheev, Anatoly; Kornyakov, Kirill; Eruhimov, Victor. “Realtime Computer Vision with OpenCV.” Realtime Computer Vision with OpenCV — ACM Queue, Association for Computing Machinery, 22 Apr. 2012, queue.acm.org/detail.cfm?id=2206309.