尝试使用 findContours() 但不断收到 cpp:197 错误(-210: 不支持的格式或格式组合)
我在其他文件中使用了相同的格式,效果很好。不知道为什么它在这里不起作用。
完整错误:
Traceback (most recent call last):
File "C:/Users/FreddyMac/PycharmProjects/TestProj/ballTrackingAbsDiff.py", line 33, in <module>
cnts = cv2.findContours(th1, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
cv2.error: OpenCV(4.4.0) C:\Users\appveyor\AppData\Local\Temp\1\pip-req-build-cff9bdsm\opencv\modules\imgproc\src\contours.cpp:197: error: (-210:Unsupported format or combination of formats) [Start]FindContours supports only CV_8UC1 images when mode != CV_RETR_FLOODFILL otherwise supports CV_32SC1 images only in function 'cvStartFindContours_Impl'
Run Code Online (Sandbox Code Playgroud)
我检查了图像的类型,它是正确的“uint8”类型。
请参阅下面的代码。
import cv2
import imutils
vs = cv2.VideoCapture('ballsFlying.MP4')
while True:
# read frame1, resize and convert to grayscale
ret, frame1 = vs.read()
if frame1 is None:
break
frame1 = imutils.resize(frame1, width=600)
gray1 = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)
# read frame2, resize and convert to grayscale
ret2, frame2 = vs.read()
if frame2 is None:
break
frame2 = imutils.resize(frame2, width=600)
gray2 = cv2.cvtColor(frame2, cv2.COLOR_BGR2GRAY)
# compute the difference between frames
dist = cv2.absdiff(frame1, frame2)
# blur image
blurred = cv2.GaussianBlur(dist, (9, 9), 0)
# global thresholding
ret3, th1 = cv2.threshold(blurred, 85, 255, cv2.THRESH_BINARY)
print(th1.dtype)
cnts = cv2.findContours(th1, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
# other way to find contours = same error
# hierarchy, contours = cv2.findContours(th1, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
cv2.imshow('dist', frame1)
cv2.imshow('thresh', th1)
cv2.imshow('blurred', blurred)
# show the frame to our screen
key = cv2.waitKey(100) & 0xFF
# if the 'q' key is pressed, stop the loop
if key == ord("q"):
break
# otherwise, release the camera
vs.release()
# close all windows
cv2.destroyAllWindows()
Run Code Online (Sandbox Code Playgroud)
好吧,错误给出了答案:
当 mode != CV_RETR_FLOODFILL 时,FindContours 仅支持 CV_8UC1 图像,否则仅在函数中支持 CV_32SC1 图像
由于您没有使用CV_RETR_FLOODFILL,因此您的图像应该是CV_32SC1单通道图像。findContours适用于单通道图像。
使用灰度图像,问题就可以解决。
dist = cv2.absdiff(gray1, gray2)
Run Code Online (Sandbox Code Playgroud)
结果:
th
模糊