OpenCV videoccapture到一个图像数组

Question

我试图使用OpenCV的videoccapture函数读取视频，提取帧，然后将它们放入数组中，这样我就有了一个3D图像堆栈。以下是目前为止的内容:

cap = cv2.VideoCapture('/home/jryan/Videos/RMS/20211019_234125-75mm.mkv')
frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
fps = cap.get(cv2.CAP_PROP_FPS)
image_array = []
for frame in cap:
if frame_count <= 1000: # only want the first 1000 frames
continue
capture = cap.read()
image_array += capture

然而，我得到错误'cv2.VideoCapture' object is not iterable。据我所知，没有什么可以从视频捕获中迭代-否则我如何引用单个帧?

Answer 1

您只需在for循环中迭代int变量，而不使用cap，这是不可迭代的。例如:

for i in range(1000): #only the first 1000 frames
ret, frame = cap.read()
#do your stuff