如何通过 python 中的 np.array 点裁剪图像的所有位置？答案

【问题标题】：How do I crop all locations of image through a np.array of points within python?如何通过 python 中的 np.array 点裁剪图像的所有位置？
【发布时间】：2019-01-31 19:20:11
【问题描述】：

我正在设置一个 for 循环，以从我的 22 的 np.array 中裁剪图像的不同部分。如何正确设置 for 循环并使其快速运行？

我知道我需要一个空数组来放作物。

crop_array = np.empty([])



#test with y1,y2,x1,x2 points 
rect = np.array([[137,166,14,80],[174,204,14,80],[214,244,14,80], 
[252,282,14,80],[290,320,14,80],[328,358,14,80],[366,396,14,80], 
[402,432,14,80],[65,94,161,227],[102,131,199,265],[139,168,234,300], 
[179,208,270,336],[217,246,306,372],[254,283,344,410],[293,322,381,447],
[65,94,235,301],[102,131,273,339],[139,168,308,374],[179,208,344,410], 
[217,246,380,446],[254,283,418,478]])

#this is one point and works
crop0 = small2gray[137:166,14:80].copy()

【问题讨论】：

标签： python-3.x for-loop numpy-ndarray

【解决方案1】：

鉴于为 rect 给出的 np.array。

#y1,y2,x1,x2 points within each list
rect = np.array([[137,166,14,80],[174,204,14,80],[214,244,14,80], 
[252,282,14,80],[290,320,14,80],[328,358,14,80],[366,396,14,80], 
[402,432,14,80],[65,94,161,227],[102,131,199,265],[139,168,234,300], 
[179,208,270,336],[217,246,306,372],[254,283,344,410],[293,322,381,447],
[65,94,235,301],[102,131,273,339],[139,168,308,374],[179,208,344,410], 
[217,246,380,446],[254,283,418,478]])

#For loop to contents of list that are nested
for (a,b,c,d) in rect:
   #Crop ROI Location
   crop_empty = smallgray[a:b,c:d].copy()
   crop = small2gray[a:b,c:d].copy

解释嵌套列表的网站。 https://cognitiveclass.ai/blog/nested-lists-multidimensional-numpy-arrays/

这是一个解释视频。 https://www.youtube.com/watch?v=rUDdguWk-QQ

【讨论】：