最近在处理Caltech_Pedestrian数据集，需要导入

from scipy.misc import imresize, imreadim = imread(im_file)
im = imresize(im, (desired_sz[0], int(np.round(target_ds * im.shape[1]))), 'bicubic')

出现ImportError: cannot import name 'imresize'错误

其次百度发现可以进行如下替换修改：

from PIL import Image
import numpy as np# im = imresize(im, (h, int(w * aspect_ratio)), interp='bicubic')
im = np.array(Image.fromarray(im).resize((h, int(w * aspect_ratio))))

但是发现以上方法默认采样方式并不是bicubic，可以改为如下形式：

from PIL import Image
import numpy as np# im = imresize(im, (h, int(w * aspect_ratio)), interp='bicubic')
im = np.array(Image.fromarray(im).resize((h, int(w * aspect_ratio)),resample=PIL.Image.BICUBIC))

因为中间转了几个弯，还是不够保证最终测试集处理结果和其他人是一样的，因此决定下载python3.6以及安装scipy=1.0.0。

虽然scipy=1.0.0这个版本的源码里面有imresize这个函数，但还是无法导入以及使用（如果有大佬成功使用过，望分享），因此最终决定直接把包含这几个函数的文件内容直接复制（文件之间的独立性较好），放在需要使用imresize的文件最前面。

'''
Code for downloading and processing Caltech Pedestrian Dataset - test part (P. Dollar et al. 2009, http://www.vision.caltech.edu/Image_Datasets/CaltechPedestrians/).
Based on code related to PredNet - Lotter et al. 2016 (https://arxiv.org/abs/1605.08104 https://github.com/coxlab/prednet).
Method of resizing was specified (bicubic).
'''
"""
A collection of image utilities using the Python Imaging Library (PIL).Note that PIL is not a dependency of SciPy and this module is not
available on systems that don't have PIL installed."""
# from __future__ import division, print_function, absolute_import# Functions which need the PILimport numpy
import tempfilefrom numpy import (amin, amax, ravel, asarray, arange, ones, newaxis,transpose, iscomplexobj, uint8, issubdtype, array)try:from PIL import Image, ImageFilter
except ImportError:import Imageimport ImageFilterif not hasattr(Image, 'frombytes'):Image.frombytes = Image.fromstring__all__ = ['fromimage', 'toimage', 'imsave', 'imread', 'bytescale','imrotate', 'imresize', 'imshow', 'imfilter']@numpy.deprecate(message="`bytescale` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.")
def bytescale(data, cmin=None, cmax=None, high=255, low=0):"""Byte scales an array (image).Byte scaling means converting the input image to uint8 dtype and scalingthe range to ``(low, high)`` (default 0-255).If the input image already has dtype uint8, no scaling is done.This function is only available if Python Imaging Library (PIL) is installed.Parameters----------data : ndarrayPIL image data array.cmin : scalar, optionalBias scaling of small values. Default is ``data.min()``.cmax : scalar, optionalBias scaling of large values. Default is ``data.max()``.high : scalar, optionalScale max value to `high`.  Default is 255.low : scalar, optionalScale min value to `low`.  Default is 0.Returns-------img_array : uint8 ndarrayThe byte-scaled array.Examples-------->>> from scipy.misc import bytescale>>> img = np.array([[ 91.06794177,   3.39058326,  84.4221549 ],...                 [ 73.88003259,  80.91433048,   4.88878881],...                 [ 51.53875334,  34.45808177,  27.5873488 ]])>>> bytescale(img)array([[255,   0, 236],[205, 225,   4],[140,  90,  70]], dtype=uint8)>>> bytescale(img, high=200, low=100)array([[200, 100, 192],[180, 188, 102],[155, 135, 128]], dtype=uint8)>>> bytescale(img, cmin=0, cmax=255)array([[91,  3, 84],[74, 81,  5],[52, 34, 28]], dtype=uint8)"""if data.dtype == uint8:return dataif high > 255:raise ValueError("`high` should be less than or equal to 255.")if low < 0:raise ValueError("`low` should be greater than or equal to 0.")if high < low:raise ValueError("`high` should be greater than or equal to `low`.")if cmin is None:cmin = data.min()if cmax is None:cmax = data.max()cscale = cmax - cminif cscale < 0:raise ValueError("`cmax` should be larger than `cmin`.")elif cscale == 0:cscale = 1scale = float(high - low) / cscalebytedata = (data - cmin) * scale + lowreturn (bytedata.clip(low, high) + 0.5).astype(uint8)@numpy.deprecate(message="`imread` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.\n""Use ``imageio.imread`` instead.")
def imread(name, flatten=False, mode=None):"""Read an image from a file as an array.This function is only available if Python Imaging Library (PIL) is installed.Parameters----------name : str or file objectThe file name or file object to be read.flatten : bool, optionalIf True, flattens the color layers into a single gray-scale layer.mode : str, optionalMode to convert image to, e.g. ``'RGB'``.  See the Notes for moredetails.Returns-------imread : ndarrayThe array obtained by reading the image.Notes-----`imread` uses the Python Imaging Library (PIL) to read an image.The following notes are from the PIL documentation.`mode` can be one of the following strings:* 'L' (8-bit pixels, black and white)* 'P' (8-bit pixels, mapped to any other mode using a color palette)* 'RGB' (3x8-bit pixels, true color)* 'RGBA' (4x8-bit pixels, true color with transparency mask)* 'CMYK' (4x8-bit pixels, color separation)* 'YCbCr' (3x8-bit pixels, color video format)* 'I' (32-bit signed integer pixels)* 'F' (32-bit floating point pixels)PIL also provides limited support for a few special modes, including'LA' ('L' with alpha), 'RGBX' (true color with padding) and 'RGBa'(true color with premultiplied alpha).When translating a color image to black and white (mode 'L', 'I' or'F'), the library uses the ITU-R 601-2 luma transform::L = R * 299/1000 + G * 587/1000 + B * 114/1000When `flatten` is True, the image is converted using mode 'F'.When `mode` is not None and `flatten` is True, the image is firstconverted according to `mode`, and the result is then flattened usingmode 'F'."""im = Image.open(name)return fromimage(im, flatten=flatten, mode=mode)@numpy.deprecate(message="`imsave` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.\n""Use ``imageio.imwrite`` instead.")
def imsave(name, arr, format=None):"""Save an array as an image.This function is only available if Python Imaging Library (PIL) is installed... warning::This function uses `bytescale` under the hood to rescale images to usethe full (0, 255) range if ``mode`` is one of ``None, 'L', 'P', 'l'``.It will also cast data for 2-D images to ``uint32`` for ``mode=None``(which is the default).Parameters----------name : str or file objectOutput file name or file object.arr : ndarray, MxN or MxNx3 or MxNx4Array containing image values.  If the shape is ``MxN``, the arrayrepresents a grey-level image.  Shape ``MxNx3`` stores the red, greenand blue bands along the last dimension.  An alpha layer may beincluded, specified as the last colour band of an ``MxNx4`` array.format : strImage format. If omitted, the format to use is determined from thefile name extension. If a file object was used instead of a file name,this parameter should always be used.Examples--------Construct an array of gradient intensity values and save to file:>>> from scipy.misc import imsave>>> x = np.zeros((255, 255))>>> x = np.zeros((255, 255), dtype=np.uint8)>>> x[:] = np.arange(255)>>> imsave('gradient.png', x)Construct an array with three colour bands (R, G, B) and store to file:>>> rgb = np.zeros((255, 255, 3), dtype=np.uint8)>>> rgb[..., 0] = np.arange(255)>>> rgb[..., 1] = 55>>> rgb[..., 2] = 1 - np.arange(255)>>> imsave('rgb_gradient.png', rgb)"""im = toimage(arr, channel_axis=2)if format is None:im.save(name)else:im.save(name, format)return@numpy.deprecate(message="`fromimage` is deprecated in SciPy 1.0.0. ""and will be removed in 1.2.0.\n""Use ``np.asarray(im)`` instead.")
def fromimage(im, flatten=False, mode=None):"""Return a copy of a PIL image as a numpy array.This function is only available if Python Imaging Library (PIL) is installed.Parameters----------im : PIL imageInput image.flatten : boolIf true, convert the output to grey-scale.mode : str, optionalMode to convert image to, e.g. ``'RGB'``.  See the Notes of the`imread` docstring for more details.Returns-------fromimage : ndarrayThe different colour bands/channels are stored in thethird dimension, such that a grey-image is MxN, anRGB-image MxNx3 and an RGBA-image MxNx4."""if not Image.isImageType(im):raise TypeError("Input is not a PIL image.")if mode is not None:if mode != im.mode:im = im.convert(mode)elif im.mode == 'P':# Mode 'P' means there is an indexed "palette".  If we leave the mode# as 'P', then when we do `a = array(im)` below, `a` will be a 2-D# containing the indices into the palette, and not a 3-D array# containing the RGB or RGBA values.if 'transparency' in im.info:im = im.convert('RGBA')else:im = im.convert('RGB')if flatten:im = im.convert('F')elif im.mode == '1':# Workaround for crash in PIL. When im is 1-bit, the call array(im)# can cause a seg. fault, or generate garbage. See# https://github.com/scipy/scipy/issues/2138 and# https://github.com/python-pillow/Pillow/issues/350.## This converts im from a 1-bit image to an 8-bit image.im = im.convert('L')a = array(im)return a_errstr = "Mode is unknown or incompatible with input array shape."@numpy.deprecate(message="`toimage` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.\n""Use Pillow's ``Image.fromarray`` directly instead.")
def toimage(arr, high=255, low=0, cmin=None, cmax=None, pal=None,mode=None, channel_axis=None):"""Takes a numpy array and returns a PIL image.This function is only available if Python Imaging Library (PIL) is installed.The mode of the PIL image depends on the array shape and the `pal` and`mode` keywords.For 2-D arrays, if `pal` is a valid (N,3) byte-array giving the RGB values(from 0 to 255) then ``mode='P'``, otherwise ``mode='L'``, unless modeis given as 'F' or 'I' in which case a float and/or integer array is made... warning::This function uses `bytescale` under the hood to rescale images to usethe full (0, 255) range if ``mode`` is one of ``None, 'L', 'P', 'l'``.It will also cast data for 2-D images to ``uint32`` for ``mode=None``(which is the default).Notes-----For 3-D arrays, the `channel_axis` argument tells which dimension of thearray holds the channel data.For 3-D arrays if one of the dimensions is 3, the mode is 'RGB'by default or 'YCbCr' if selected.The numpy array must be either 2 dimensional or 3 dimensional."""data = asarray(arr)if iscomplexobj(data):raise ValueError("Cannot convert a complex-valued array.")shape = list(data.shape)valid = len(shape) == 2 or ((len(shape) == 3) and((3 in shape) or (4 in shape)))if not valid:raise ValueError("'arr' does not have a suitable array shape for ""any mode.")if len(shape) == 2:shape = (shape[1], shape[0])  # columns show up firstif mode == 'F':data32 = data.astype(numpy.float32)image = Image.frombytes(mode, shape, data32.tostring())return imageif mode in [None, 'L', 'P']:bytedata = bytescale(data, high=high, low=low,cmin=cmin, cmax=cmax)image = Image.frombytes('L', shape, bytedata.tostring())if pal is not None:image.putpalette(asarray(pal, dtype=uint8).tostring())# Becomes a mode='P' automagically.elif mode == 'P':  # default gray-scalepal = (arange(0, 256, 1, dtype=uint8)[:, newaxis] *ones((3,), dtype=uint8)[newaxis, :])image.putpalette(asarray(pal, dtype=uint8).tostring())return imageif mode == '1':  # high input gives threshold for 1bytedata = (data > high)image = Image.frombytes('1', shape, bytedata.tostring())return imageif cmin is None:cmin = amin(ravel(data))if cmax is None:cmax = amax(ravel(data))data = (data*1.0 - cmin)*(high - low)/(cmax - cmin) + lowif mode == 'I':data32 = data.astype(numpy.uint32)image = Image.frombytes(mode, shape, data32.tostring())else:raise ValueError(_errstr)return image# if here then 3-d array with a 3 or a 4 in the shape length.# Check for 3 in datacube shape --- 'RGB' or 'YCbCr'if channel_axis is None:if (3 in shape):ca = numpy.flatnonzero(asarray(shape) == 3)[0]else:ca = numpy.flatnonzero(asarray(shape) == 4)if len(ca):ca = ca[0]else:raise ValueError("Could not find channel dimension.")else:ca = channel_axisnumch = shape[ca]if numch not in [3, 4]:raise ValueError("Channel axis dimension is not valid.")bytedata = bytescale(data, high=high, low=low, cmin=cmin, cmax=cmax)if ca == 2:strdata = bytedata.tostring()shape = (shape[1], shape[0])elif ca == 1:strdata = transpose(bytedata, (0, 2, 1)).tostring()shape = (shape[2], shape[0])elif ca == 0:strdata = transpose(bytedata, (1, 2, 0)).tostring()shape = (shape[2], shape[1])if mode is None:if numch == 3:mode = 'RGB'else:mode = 'RGBA'if mode not in ['RGB', 'RGBA', 'YCbCr', 'CMYK']:raise ValueError(_errstr)if mode in ['RGB', 'YCbCr']:if numch != 3:raise ValueError("Invalid array shape for mode.")if mode in ['RGBA', 'CMYK']:if numch != 4:raise ValueError("Invalid array shape for mode.")# Here we know data and mode is correctimage = Image.frombytes(mode, shape, strdata)return image@numpy.deprecate(message="`imrotate` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.\n""Use ``skimage.transform.rotate`` instead.")
def imrotate(arr, angle, interp='bilinear'):"""Rotate an image counter-clockwise by angle degrees.This function is only available if Python Imaging Library (PIL) is installed... warning::This function uses `bytescale` under the hood to rescale images to usethe full (0, 255) range if ``mode`` is one of ``None, 'L', 'P', 'l'``.It will also cast data for 2-D images to ``uint32`` for ``mode=None``(which is the default).Parameters----------arr : ndarrayInput array of image to be rotated.angle : floatThe angle of rotation.interp : str, optionalInterpolation- 'nearest' :  for nearest neighbor- 'bilinear' : for bilinear- 'lanczos' : for lanczos- 'cubic' : for bicubic- 'bicubic' : for bicubicReturns-------imrotate : ndarrayThe rotated array of image."""arr = asarray(arr)func = {'nearest': 0, 'lanczos': 1, 'bilinear': 2, 'bicubic': 3, 'cubic': 3}im = toimage(arr)im = im.rotate(angle, resample=func[interp])return fromimage(im)@numpy.deprecate(message="`imshow` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.\n""Use ``matplotlib.pyplot.imshow`` instead.")
def imshow(arr):"""Simple showing of an image through an external viewer.This function is only available if Python Imaging Library (PIL) is installed.Uses the image viewer specified by the environment variableSCIPY_PIL_IMAGE_VIEWER, or if that is not defined then `see`,to view a temporary file generated from array data... warning::This function uses `bytescale` under the hood to rescale images to usethe full (0, 255) range if ``mode`` is one of ``None, 'L', 'P', 'l'``.It will also cast data for 2-D images to ``uint32`` for ``mode=None``(which is the default).Parameters----------arr : ndarrayArray of image data to show.Returns-------NoneExamples-------->>> a = np.tile(np.arange(255), (255,1))>>> from scipy import misc>>> misc.imshow(a)"""im = toimage(arr)fnum, fname = tempfile.mkstemp('.png')try:im.save(fname)except Exception:raise RuntimeError("Error saving temporary image data.")import osos.close(fnum)cmd = os.environ.get('SCIPY_PIL_IMAGE_VIEWER', 'see')status = os.system("%s %s" % (cmd, fname))os.unlink(fname)if status != 0:raise RuntimeError('Could not execute image viewer.')@numpy.deprecate(message="`imresize` is deprecated in SciPy 1.0.0, ""and will be removed in 1.3.0.\n""Use Pillow instead: ``numpy.array(Image.fromarray(arr).resize())``.")
def imresize(arr, size, interp='bilinear', mode=None):"""Resize an image.This function is only available if Python Imaging Library (PIL) is installed... warning::This function uses `bytescale` under the hood to rescale images to usethe full (0, 255) range if ``mode`` is one of ``None, 'L', 'P', 'l'``.It will also cast data for 2-D images to ``uint32`` for ``mode=None``(which is the default).Parameters----------arr : ndarrayThe array of image to be resized.size : int, float or tuple* int   - Percentage of current size.* float - Fraction of current size.* tuple - Size of the output image (height, width).interp : str, optionalInterpolation to use for re-sizing ('nearest', 'lanczos', 'bilinear','bicubic' or 'cubic').mode : str, optionalThe PIL image mode ('P', 'L', etc.) to convert `arr` before resizing.If ``mode=None`` (the default), 2-D images will be treated like``mode='L'``, i.e. casting to long integer.  For 3-D and 4-D arrays,`mode` will be set to ``'RGB'`` and ``'RGBA'`` respectively.Returns-------imresize : ndarrayThe resized array of image.See Also--------toimage : Implicitly used to convert `arr` according to `mode`.scipy.ndimage.zoom : More generic implementation that does not use PIL."""im = toimage(arr, mode=mode)ts = type(size)if issubdtype(ts, numpy.signedinteger):percent = size / 100.0size = tuple((array(im.size)*percent).astype(int))elif issubdtype(type(size), numpy.floating):size = tuple((array(im.size)*size).astype(int))else:size = (size[1], size[0])func = {'nearest': 0, 'lanczos': 1, 'bilinear': 2, 'bicubic': 3, 'cubic': 3}imnew = im.resize(size, resample=func[interp])return fromimage(imnew)@numpy.deprecate(message="`imfilter` is deprecated in SciPy 1.0.0, ""and will be removed in 1.2.0.\n""Use Pillow filtering functionality directly.")
def imfilter(arr, ftype):"""Simple filtering of an image.This function is only available if Python Imaging Library (PIL) is installed... warning::This function uses `bytescale` under the hood to rescale images to usethe full (0, 255) range if ``mode`` is one of ``None, 'L', 'P', 'l'``.It will also cast data for 2-D images to ``uint32`` for ``mode=None``(which is the default).Parameters----------arr : ndarrayThe array of Image in which the filter is to be applied.ftype : strThe filter that has to be applied. Legal values are:'blur', 'contour', 'detail', 'edge_enhance', 'edge_enhance_more','emboss', 'find_edges', 'smooth', 'smooth_more', 'sharpen'.Returns-------imfilter : ndarrayThe array with filter applied.Raises------ValueError*Unknown filter type.*  If the filter you are tryingto apply is unsupported."""_tdict = {'blur': ImageFilter.BLUR,'contour': ImageFilter.CONTOUR,'detail': ImageFilter.DETAIL,'edge_enhance': ImageFilter.EDGE_ENHANCE,'edge_enhance_more': ImageFilter.EDGE_ENHANCE_MORE,'emboss': ImageFilter.EMBOSS,'find_edges': ImageFilter.FIND_EDGES,'smooth': ImageFilter.SMOOTH,'smooth_more': ImageFilter.SMOOTH_MORE,'sharpen': ImageFilter.SHARPEN}im = toimage(arr)if ftype not in _tdict:raise ValueError("Unknown filter type.")return fromimage(im.filter(_tdict[ftype]))#下面加上自己需要的文件，也可以放在另一个文件，导入该函数

把以上源码放在需要使用imresize、imread前面即可。又由于scipy==1.0.0源码里面,需要使用PIL，因此需要安装该库：

pip install Pillow

接下来就可以直接正常使用这两个函数了，和老版本一模一样。

高效准确处理scipy.misc 中imresize、imread导入错误问题相关推荐

Python扩展库scipy.misc中图像转换成pillow图像
众所周知,在数字图像处理领域中有很多基准测试图像,这些图像用来作为科研人员PK自己的算法时的参考,给大家提供一个公平的样本,针对同一个问题进行处理时,可以用这些基准图像做实验,比较常见的应该就是len ...
unity中Standard Assets导入错误的解决方法
点击错误,由VS打开SimpleActivatorMenu.cs 把里面的GUIText改为UnityEngine.UI.Text,然后保存就可以了
imresize导入错误ImportError: cannot import name 'imresize'
python代码中导入imresize: from scipy.misc import imresize 出现如下错误,无法导入: ImportError: cannot import name 'i ...
python报错AttributeError module ‘scipy.misc‘ has no attribute ‘imresize‘和 ‘imread‘
python报错AttributeError: module 'scipy.misc' has no attribute 'imresize'和 'imread' 报错原因:scipy版本过高解决方 ...
解决cs321n 的 python教程中 name 'imread' ‘imsave’ 'imresize' is not defined 的问题
将from scipy.misc import imread 改成 from scipy.misc.pilutil import imread, imsave,imresize 但这在后边的作业的ju ...
scipy.misc.imresize的替换方案
错误提示当scipy版本>=1.3.0时,导入scipy.misc.imresize会出现如下错误 AttributeError: module 'scipy.misc' has no att ...
python 导入库：from scipy.misc import imread出错问题
这是因为scipy库中的imread要被弃用也就是不能用了,这时需要改成这样from imageio import imread
cannot import name 'imresize' from 'scipy.misc'
问题:执行以下python命令,出现这样的错误 cannot import name 'imresize' from 'scipy.misc' from scipy.misc import i ...
scipy.misc.imresize
此函数的作用是: 对图像进行成比例放大或缩小,且归一化到0-255区间 https://blog.csdn.net/u013066730/article/details/59528939 >&g ...
图片操作scipy.ndimage.imread和scipy.misc.imresize
读图片str or file object--ndarray scipy.ndimage.imread(*args, **kwds) imread is deprecated! imread is d ...

高效准确处理scipy.misc 中imresize、imread导入错误问题

出现ImportError: cannot import name 'imresize'错误

高效准确处理scipy.misc 中imresize、imread导入错误问题相关推荐

最新文章

热门文章