高分辨(医学或遥感图像)处理 开源方案简介

语言: CN / TW / HK


环境准备

conda create --name xxx
conda activate xxx
conda install ipykernel


安装 labelimg

https://github.com/tzutalin/labelImg#labelimg

Windows + Anaconda
Download and install Anaconda (Python 3+)

Open the Anaconda Prompt and go to the labelImg directory

conda install pyqt=5
conda install -c anaconda lxml
pyrcc5 -o libs/resources.py resources.qrc
python labelImg.py
python labelImg.py [IMAGE_PATH] [PRE-DEFINED CLASS FILE]


安装 openslide

https://openslide.org/api/python/
https://github.com/openslide/openslide-python

https://www.cnblogs.com/dyc99/p/13754660.html


deepzoom

https://www.microsoft.com/silverlight/deep-zoom/


ASAP

https://github.com/computationalpathologygroup/ASAP

Compilation

To compile the code yourself, some prerequesites are required. First, we use CMake (version >= 3.5) as our build system and Microsoft Visual Studio or GCC as the compiler. The software depends on numerous third-party libraries:

  • Boost (http://www.boost.org/)
  • OpenCV (http://www.opencv.org/)
  • Qt (http://www.qt.io/)
  • libtiff (http://www.libtiff.org/)
  • libjpeg (http://libjpeg.sourceforge.net/)
  • JasPer (http://www.ece.uvic.ca/~frodo/jasper/)
  • DCMTK (http://dicom.offis.de/dcmtk.php.en)
  • SWIG (http://www.swig.org/) (only for Python wrapping of the IO library)
  • OpenSlide (http://openslide.org/)
  • PugiXML (http://pugixml.org/)
  • zlib (http://www.zlib.net/)
  • unittest++ (https://github.com/unittest-cpp/unittest-cpp)

To help developers compile this software themselves we provide the necesarry binaries (Visual Studio 2013, 64-bit) for all third party libraries on Windows except Boost, OpenCV and Qt (due to size constraints). If you want to provide the packages yourself, there are no are no strict version requirements, except for libtiff (4.0.1 and higher), Boost (1.55 or higher), Qt (5.1 or higher) and OpenCV (3.1). On Linux all packages except OpenCV 3.1 can be installed through the package manager on Ubuntu-derived systems (tested on Ubuntu and Kubuntu 16.04 LTS). OpenCV 3.1 can easily be compile yourself.

To compile the source code yourself, first make sure all third-party libraries are installed. If you download the Boost-binaries for Windows (http://sourceforge.net/projects/boost/files/boost-binaries/), you need to rename the folder containing the .lib and .dll files to lib (otherwise the CMake-modules provided by CMake will not be able to find the libraries).

Subsequently, fire up CMake, point it to a source and build directory and hit Configure. Select your compiler of preference and hit ok. This will start the iterative process of CMake trying to find a third party dependency and you specifiying its location. The first one to provide will be Boost. To allow CMake to find Boost add a BOOST_ROOT variable pointing to for example C:/libs/boost_1_57_0. Then press Configure again and CMake will ask for the next library. These should be pretty straightforward to fill in (e.g. TIFF_LIBRRARY should point to tiff.lib, TIFF_INCLUDE_DIRECTORY to |folder to libtiff|\include. If more steps are unclear, please open a ticket on the Github issue-tracker.

During configuration you will notice that several parts of ASAP can be built seperately (e.g. the viewer). To build this part, simply check the component and hit Configure again. The ‘Package on install’-option will allow you to build a binary setup-package like the one provided on the Github-release page. On Windows this requires NSIS to be installed.

After all the dependencies are resolved, hit Generate and CMake will create a Visual Studio Solution or makefile file which can be used to compile the source code.

这个玩意调试比较费劲

要装qt OpenCV boost

环境搭建

boost

https://www.boost.org/

qt

  1. 注册账号
  2. 下载在线安装包

本文同步分享在 博客“shiter”(CSDN)。
如有侵权,请联系 [email protected] 删除。
本文参与“OSC源创计划”,欢迎正在阅读的你也加入,一起分享。

分享到: