本文整理汇总了Python中scipy.ndimage.filters.percentile_filter方法的典型用法代码示例。如果您正苦于以下问题：Python filters.percentile_filter方法的具体用法？Python filters.percentile_filter怎么用？Python filters.percentile_filter使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类scipy.ndimage.filters的用法示例。
在下文中一共展示了filters.percentile_filter方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: estimate_skew_angle
# 需要导入模块: from scipy.ndimage import filters [as 别名]
# 或者: from scipy.ndimage.filters import percentile_filter [as 别名]
def estimate_skew_angle(raw):
"""
估计图像文字角度
"""
def resize_im(im, scale, max_scale=None):
f=float(scale)/min(im.shape[0], im.shape[1])
if max_scale!=None and f*max(im.shape[0], im.shape[1])>max_scale:
f=float(max_scale)/max(im.shape[0], im.shape[1])
return cv2.resize(im, (0, 0), fx=f, fy=f)
raw = resize_im(raw, scale=600, max_scale=900)
image = raw-amin(raw)
image = image/amax(image)
m = interpolation.zoom(image,0.5)
m = filters.percentile_filter(m,80,size=(20,2))
m = filters.percentile_filter(m,80,size=(2,20))
m = interpolation.zoom(m,1.0/0.5)
w,h = min(image.shape[1],m.shape[1]),min(image.shape[0],m.shape[0])
flat = np.clip(image[:h,:w]-m[:h,:w]+1,0,1)
d0,d1 = flat.shape
o0,o1 = int(0.1*d0),int(0.1*d1)
flat = amax(flat)-flat
flat -= amin(flat)
est = flat[o0:d0-o0,o1:d1-o1]
angles = range(-15,15)
estimates = []
for a in angles:
roest =interpolation.rotate(est,a,order=0,mode='constant')
v = np.mean(roest,axis=1)
v = np.var(v)
estimates.append((v,a))
_,a = max(estimates)
return a