使用 PIL 打开图片时无法通过 cv2 (opencv 4.4) 加载图片答案

【问题标题】：Failing to load a picture through cv2 (opencv 4.4) while opening it with PIL works使用 PIL 打开图片时无法通过 cv2 (opencv 4.4) 加载图片
【发布时间】：2021-02-04 20:04:48
【问题描述】：

我有一堆图片，我可以使用各种方法很好地加载它们，例如 Image.open。但是说到opencv，我什么也得不到：

from PIL import Image
import os
import cv2

 for fi in os.listdir():
            image = cv2.imread(fi)
            print(image)

            image = Image.open(fi)
            print(image)
            print(np.array(image.getdata()))

None
<PIL.PngImagePlugin.PngImageFile image mode=RGB size=1920x1080 at 0x8AF2A30>
[[31 63 84]
 [31 63 84]
 [31 63 84]
 ...
 [16 16 16]
 [16 16 16]
 [16 16 16]]

无论我为 imread 的第二个参数给出什么值，它都保持不变。给出明确的路径也不会带来任何改变。

我的文件是 1920*1080 RGB png 文件，我使用的是 python 3.7.3 和 opencv 4.4.0.44。

感谢您的帮助

编辑 1：

General configuration for OpenCV 4.4.0 =====================================
  Version control:               4.4.0

  Platform:
    Timestamp:                   2020-09-23T00:00:53Z
    Host:                        Windows 6.3.9600 AMD64
    CMake:                       3.18.2
    CMake generator:             Visual Studio 14 2015
    CMake build tool:            C:/Program Files (x86)/MSBuild/14.0/bin/MSBuild.exe
    MSVC:                        1900

  CPU/HW features:
    Baseline:                    SSE SSE2
      requested:                 SSE2
    Dispatched code generation:  SSE4_1 SSE4_2 FP16 AVX
      requested:                 SSE4_1 SSE4_2 AVX FP16
      SSE4_1 (15 files):         + SSE3 SSSE3 SSE4_1
      SSE4_2 (1 files):          + SSE3 SSSE3 SSE4_1 POPCNT SSE4_2
      FP16 (0 files):            + SSE3 SSSE3 SSE4_1 POPCNT SSE4_2 FP16 AVX
      AVX (4 files):             + SSE3 SSSE3 SSE4_1 POPCNT SSE4_2 AVX

  C/C++:
    Built as dynamic libs?:      NO
    C++ standard:                11
    C++ Compiler:                C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC/bin/cl.exe  (ver 19.0.24241.7)
    C++ flags (Release):         /DWIN32 /D_WINDOWS /W4 /GR  /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi  /fp:precise  /arch:SSE /arch:SSE2 /EHa /wd4127 /wd4251 /wd4324 /wd4275 /wd4512 /wd4589 /MP  /MT /O2 /Ob2 /DNDEBUG 
    C++ flags (Debug):           /DWIN32 /D_WINDOWS /W4 /GR  /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi  /fp:precise  /arch:SSE /arch:SSE2 /EHa /wd4127 /wd4251 /wd4324 /wd4275 /wd4512 /wd4589 /MP  /MTd /Zi /Ob0 /Od /RTC1 
    C Compiler:                  C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC/bin/cl.exe
    C flags (Release):           /DWIN32 /D_WINDOWS /W3  /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi  /fp:precise  /arch:SSE /arch:SSE2 /MP   /MT /O2 /Ob2 /DNDEBUG 
    C flags (Debug):             /DWIN32 /D_WINDOWS /W3  /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi  /fp:precise  /arch:SSE /arch:SSE2 /MP /MTd /Zi /Ob0 /Od /RTC1 
    Linker flags (Release):      /machine:X86  /NODEFAULTLIB:atlthunk.lib /INCREMENTAL:NO  /NODEFAULTLIB:libcmtd.lib /NODEFAULTLIB:libcpmtd.lib /NODEFAULTLIB:msvcrtd.lib
    Linker flags (Debug):        /machine:X86  /NODEFAULTLIB:atlthunk.lib /debug /INCREMENTAL  /NODEFAULTLIB:libcmt.lib /NODEFAULTLIB:libcpmt.lib /NODEFAULTLIB:msvcrt.lib
    ccache:                      NO
    Precompiled headers:         YES
    Extra dependencies:          ade wsock32 comctl32 gdi32 ole32 setupapi ws2_32
    3rdparty dependencies:       ittnotify libprotobuf zlib libjpeg-turbo libwebp libpng libtiff libjasper IlmImf quirc ippiw ippicv

  OpenCV modules:
    To be built:                 calib3d core dnn features2d flann gapi highgui imgcodecs imgproc ml objdetect photo python3 stitching video videoio
    Disabled:                    world
    Disabled by dependency:      -
    Unavailable:                 java js python2 ts
    Applications:                -
    Documentation:               NO
    Non-free algorithms:         NO

  Windows RT support:            NO

  GUI: 
    Win32 UI:                    YES
    VTK support:                 NO

  Media I/O: 
    ZLib:                        build (ver 1.2.11)
    JPEG:                        build-libjpeg-turbo (ver 2.0.5-62)
    WEBP:                        build (ver encoder: 0x020f)
    PNG:                         build (ver 1.6.37)
    TIFF:                        build (ver 42 - 4.0.10)
    JPEG 2000:                   build Jasper (ver 1.900.1)
    OpenEXR:                     build (ver 2.3.0)
    HDR:                         YES
    SUNRASTER:                   YES
    PXM:                         YES
    PFM:                         YES

  Video I/O:
    DC1394:                      NO
    FFMPEG:                      YES (prebuilt binaries)
      avcodec:                   YES (58.54.100)
      avformat:                  YES (58.29.100)
      avutil:                    YES (56.31.100)
      swscale:                   YES (5.5.100)
      avresample:                YES (4.0.0)
    GStreamer:                   NO
    DirectShow:                  YES
    Media Foundation:            YES
      DXVA:                      NO

  Parallel framework:            Concurrency

  Trace:                         YES (with Intel ITT)

  Other third-party libraries:
    Intel IPP:                   2020.0.0 Gold [2020.0.0]
           at:                   C:/Users/appveyor/AppData/Local/Temp/1/pip-req-build-4q0kz99f/_skbuild/win32-3.7/cmake-build/3rdparty/ippicv/ippicv_win/icv
    Intel IPP IW:                sources (2020.0.0)
              at:                C:/Users/appveyor/AppData/Local/Temp/1/pip-req-build-4q0kz99f/_skbuild/win32-3.7/cmake-build/3rdparty/ippicv/ippicv_win/iw
    Lapack:                      NO
    Eigen:                       NO
    Custom HAL:                  NO
    Protobuf:                    build (3.5.1)

  OpenCL:                        YES (NVD3D11)
    Include path:                C:/Users/appveyor/AppData/Local/Temp/1/pip-req-build-4q0kz99f/opencv/3rdparty/include/opencl/1.2
    Link libraries:              Dynamic load

  Python 3:
    Interpreter:                 C:/Python37/python.exe (ver 3.7.5)
    Libraries:                   C:/Python37/libs/python37.lib (ver 3.7.5)
    numpy:                       C:/Users/appveyor/AppData/Local/Temp/1/pip-build-env-5cllkua4/overlay/Lib/site-packages/numpy/core/include (ver 1.14.5)
    install path:                python

  Python (for build):            C:/Python27/python.exe

  Java:                          
    ant:                         NO
    JNI:                         C:/Program Files/Java/jdk1.8.0/include C:/Program Files/Java/jdk1.8.0/include/win32 C:/Program Files/Java/jdk1.8.0/include
    Java wrappers:               NO
    Java tests:                  NO

  Install to:                    C:/Users/appveyor/AppData/Local/Temp/1/pip-req-build-4q0kz99f/_skbuild/win32-3.7/cmake-install

编辑 2


File: ... (1273217 bytes)
  chunk IHDR at offset 0x0000c, length 13
    1920 x 1080 image, 24-bit RGB, non-interlaced
  chunk pHYs at offset 0x00025, length 9: 11811x11811 pixels/meter (300 dpi)
  chunk IDAT at offset 0x0003a, length 65536
    zlib: deflated, 32K window, default compression
  chunk IDAT at offset 0x10046, length 65536
  chunk IDAT at offset 0x20052, length 65536
  chunk IDAT at offset 0x3005e, length 65536
  chunk IDAT at offset 0x4006a, length 65536
  chunk IDAT at offset 0x50076, length 65536
  chunk IDAT at offset 0x60082, length 65536
  chunk IDAT at offset 0x7008e, length 65536
  chunk IDAT at offset 0x8009a, length 65536
  chunk IDAT at offset 0x900a6, length 65536
  chunk IDAT at offset 0xa00b2, length 65536
  chunk IDAT at offset 0xb00be, length 65536
  chunk IDAT at offset 0xc00ca, length 65536
  chunk IDAT at offset 0xd00d6, length 65536
  chunk IDAT at offset 0xe00e2, length 65536
  chunk IDAT at offset 0xf00ee, length 65536
  chunk IDAT at offset 0x1000fa, length 65536
  chunk IDAT at offset 0x110106, length 65536
  chunk IDAT at offset 0x120112, length 65536
  chunk IDAT at offset 0x13011e, length 27727
  chunk IEND at offset 0x136d79, length 0
No errors detected in ... (23 chunks, 79.5% compression).

编辑 3

Image:
  Filename: ...png
  Format: PNG (Portable Network Graphics)
  Mime type: image/png
  Class: DirectClass
  Geometry: 1920x1080+0+0
  Resolution: 118.11x118.11
  Print size: 16.256x9.14402
  Units: PixelsPerCentimeter
  Colorspace: sRGB
  Type: TrueColor
  Base type: Undefined
  Endianness: Undefined
  Depth: 8-bit
  Channel depth:
    Red: 8-bit
    Green: 8-bit
    Blue: 8-bit
  Channel statistics:
    Pixels: 2073600
    Red:
      min: 0  (0)
      max: 255 (1)
      mean: 88.4175 (0.346735)
      standard deviation: 48.6837 (0.190916)
      kurtosis: 2.88645
      skewness: 1.60217
      entropy: 0.667925
    Green:
      min: 0  (0)
      max: 255 (1)
      mean: 78.7912 (0.308985)
      standard deviation: 52.3865 (0.205437)
      kurtosis: 2.33529
      skewness: 1.54681
      entropy: 0.672343
    Blue:
      min: 0  (0)
      max: 255 (1)
      mean: 134.848 (0.528816)
      standard deviation: 81.9142 (0.321232)
      kurtosis: -1.36843
      skewness: -0.507794
      entropy: 0.668643
  Image statistics:
    Overall:
      min: 0  (0)
      max: 255 (1)
      mean: 100.686 (0.394846)
      standard deviation: 60.9948 (0.239195)
      kurtosis: -0.758096
      skewness: 0.627416
      entropy: 0.669637
  Rendering intent: Perceptual
  Gamma: 0.454545
  Chromaticity:
    red primary: (0.64,0.33)
    green primary: (0.3,0.6)
    blue primary: (0.15,0.06)
    white point: (0.3127,0.329)
  Matte color: grey74
  Background color: white
  Border color: srgb(223,223,223)
  Transparent color: none
  Interlace: None
  Intensity: Undefined
  Compose: Over
  Page geometry: 1920x1080+0+0
  Dispose: Undefined
  Iterations: 0
  Compression: Zip
  Orientation: Undefined
  Convex hull: 0,0 1919,0 1919,1039 1915,1079 104,1079 17,1066 16,1065 0,1039 0,0
  Minimum bounding box: 1919,0 1919,1079 0,1079 0,0
  Properties:
    date:create: 2020-10-21T19:06:24+00:00
    date:modify: 2020-10-21T19:22:08+00:00
    minimum-bounding-box:_p: 0,0
    minimum-bounding-box:_q: 1919,0
    minimum-bounding-box:_v: 1915,1079
    minimum-bounding-box:angle: 0
    minimum-bounding-box:area: 2.0706e+06
    minimum-bounding-box:height: 1919
    minimum-bounding-box:unrotate: -0
    minimum-bounding-box:width: 1079
    png:IHDR.bit-depth-orig: 8
    png:IHDR.bit_depth: 8
    png:IHDR.color-type-orig: 2
    png:IHDR.color_type: 2 (Truecolor)
    png:IHDR.interlace_method: 0 (Not interlaced)
    png:IHDR.width,height: 1920, 1080
    png:pHYs: x_res=11811, y_res=11811, units=1
    png:sRGB: intent=0 (Perceptual Intent)
    signature: bcf111b6c823a95bd31bbec26225847dffa81c1a07cbf95cf07396f5404ffa5c
  Artifacts:
    verbose: true
  Tainted: False
  Filesize: 1.21423MiB
  Number pixels: 2073600
  Pixels per second: 28.8118MP
  User time: 0.063u
  Elapsed time: 0:01.071
  Version: ImageMagick 7.0.10-34 Q16 x64 2020-10-08 http://www.imagemagick.org

【问题讨论】：

同意@MarcSances。问题是，os.listdir() 只提供文件名，但 cv2 需要图像的绝对路径（或者至少由于此类问题，强烈建议始终使用绝对路径）。
也许您可以通过 Dropbox 或 Google Drive 或类似工具分享 "unhappy" PNG 文件。
也许你也可以分享cv2.getBuildInformation() 的输出
甚至可能是一个实际的错误...呃搜索我在 SO 中发现了这个问题，如果不重复，则非常相关：stackoverflow.com/questions/17949268/…
您的 PNG 文件可能存在问题，PIL 比 OpenCV 准备接受的更宽容。我建议通过pngcheck 运行您的PNG，并打开所有详细选项。也可以通过 ImageMagick 使用 magick identify -verbose YOURIMAGE.PNG 并检查输出是否存在问题。

标签： python opencv

【解决方案1】：

它不能解决 imread 问题，但我已经找到了一种方法来满足我的目的：

    image = Image.open(fi)
    im_arr = np.array(image.getdata(), dtype=np.uint16).reshape(1080, 1920, 3)
    im_arr = np.vectorize(lambda x: x/255)(im_arr)
    cv2.imshow("image", im_arr)
    cv2.waitKey()

不过我会进一步调查这个问题。

编辑：

尽管读写过程消耗了时间和空间，这似乎也有效并且更令人满意：

    image = Image.open(fi)
    im_arr = np.array(image.getdata()).reshape(1080, 1920, 3)
    cv2.imwrite(os.path.join(path_data, "image_data_clean", rep, fi), im_arr)
    image = cv2.imread(os.path.join(path_data, "image_data_clean", rep, fi))

在这种情况下，我确实不需要类型声明和映射（但仍然需要重塑）

编辑 2：

想探索文件可能已损坏的可能性，我从源头重新解压缩了它们。我不小心把它们放在了同一个地方，所以我失去了以前的那些，从而失去了比较的可能性。但是，当我在新文件上运行时：

    test = cv2.imread(os.path.join(path_data, "image_data", rep, fi))
    print(test)
    im_arr = np.array(Image.open(os.path.join(path_data, "image_data", rep, fi)))
    print(im_arr.shape)
    print(im_arr.dtype)
    cv2.imshow("image", im_arr)
    cv2.waitKey()
>>>
None
(1080, 1920, 3)
uint8

然后我现在得到了 shape 和 dtype 的预期答案，并且 imshow 可以完美地工作而无需重塑/类型声明！我真的很确定今天早上不是这样，所以一定是文件有问题——或者我的大脑——而不是 PIL

但是，直接尝试使用 imread 仍然失败

【讨论】：

应该够了im_arr = np.array(Image.open(fi))
如果我不添加类型参数，我会得到：``` 错误：(-215:Assertion failed) src_depth != CV_16F && src_depth != CV_32S in function 'convertToShow' ```如果我不重塑，我会得到一个 (2073600, 3) 数组，该数组在技术上不会引发异常，但也不会导致预期结果。另一方面，您是对的，在显示图像时实际上不需要 x/255 映射。我只需要它来进行进一步的 cv 处理，例如 cv2.threshold() 非常感谢各位的帮助！
这里有问题。请在这里运行我上面两行评论中的代码，然后打印im_arr.shape 和im_arr.dtype PIL 知道图像的形状和类型，所以你不应该告诉 Numpy 或 OpenCV。
发生了一些奇怪的事情，为了更清楚，我将在帖子中进行编辑。