3.3. Scikit-image：图像处理.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "> **作者**: Emmanuelle Gouillart\n",
    "\n",
    "[scikit-image](http://scikit-image.org/)是专注于图像处理的Python包，并且使用原生的Numpy数组作为图像对象。本章描述如何在不同图像处理任务上使用`scikit-image`，并且保留了其他科学Python模块比如Numpy和Scipy的链接。\n",
    "\n",
    "---\n",
    "**也可以看一下：**对于基本图像处理，比如图像剪切或者简单过滤，大量简单操作可以用Numpy和SciPy来实现。看一下[使用Numpy和Scipy图像操作和处理部分](http://www.scipy-lectures.org/advanced/image_processing/index.html#basic-image)。\n",
    "\n",
    "注意，在阅读本章之前你应该熟悉前面章节的内容，比如基础操作，比如面具和标签作为先决条件。\n",
    "\n",
    "---\n",
    "\n",
    "---\n",
    "**章节内容**\n",
    "- **介绍和观点**\n",
    "    - `scikit-image` 和 `SciPy`生态系统\n",
    "    - `scikit-image`能发现什么\n",
    "- **输入/输出, 数据类型和 颜色空间**\n",
    "    - 数据类型\n",
    "    - 颜色空间\n",
    "- **图像预处理/增强**\n",
    "    - 本地过滤器\n",
    "    - 非-本地过滤器\n",
    "    - 数学形态学\n",
    "- **图像细分**\n",
    "    - 二元细分: 前景 + 背景\n",
    "    - 基于标记的方法\n",
    "- **测量区域的属性**\n",
    "- **数据可视化和交互**\n",
    "\n",
    "## 3.3.1 介绍和观点\n",
    "\n",
    "图像是NumPy的数组`np.ndarray`\n",
    "\n",
    "|图像:\t|np.ndarray|\n",
    "|--|--|\n",
    "|像素:|array values: a[2, 3]|\n",
    "|渠道:|array dimensions|\n",
    "|图像编码:|dtype (np.uint8, np.uint16, np.float)|\n",
    "|过滤器:|functions (numpy, skimage, scipy)|"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x105717610>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPYAAAD7CAYAAABZjGkWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x103bdbd10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "check = np.zeros((9, 9))\n",
    "check[::2, 1::2] = 1\n",
    "check[1::2, ::2] = 1\n",
    "import matplotlib.pyplot as plt\n",
    "plt.imshow(check, cmap='gray', interpolation='nearest') "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.3.1.1 scikit-image 和 SciPy 生态系统\n",
    "\n",
    "最新版的`scikit-image`包含在大多数的科学Python发行版中，比如，Anaconda或Enthought Canopy。它也包含在 Ubuntu/Debian。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import skimage\n",
    "from skimage import data  # 大多数函数在子包中"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "大多数`scikit-image`函数用NumPy ndarrays作为参数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('uint8')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "camera = data.camera()\n",
    "camera.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(512, 512)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "camera.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skimage import restoration\n",
    "filtered_camera = restoration.denoise_bilateral(camera)\n",
    "type(filtered_camera)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "其他Python包也可以用于图像处理，并且使用Numpy数组:\n",
    "- [scipy.ndimage](http://docs.scipy.org/doc/scipy/reference/ndimage.html#module-scipy.ndimage) : 对于 nd-arrays。基础过滤、数学形态学和区域属性\n",
    "- [Mahotas](http://luispedro.org/software/mahotas)\n",
    "同时，强大的图形处理库有Python封装:\n",
    "- [OpenCV](https://opencv-python-tutroals.readthedocs.org/en/latest/) (计算机视觉)\n",
    "- [ITK](http://www.itk.org/itkindex.html) (3D图像和注册)\n",
    "- 其他\n",
    "(但是，他们没有那么Pythonic也没有Numpy友好，在一定范围)。\n",
    "\n",
    "### 3.3.1.2 scikit-image能发现什么\n",
    "\n",
    "- 网站: http://scikit-image.org/\n",
    "- 例子库 (就像在 [matplotlib](http://matplotlib.org/gallery.html) 或 [scikit-learn](http://scikit-learn.org/)): http://scikit-image.org/docs/stable/auto_examples/ 不同类的函数，从基本的使用函数到高级最新算法。\n",
    "\n",
    "- 过滤器: 函数将图像转化为其他图像。\n",
    "    - NumPy组件\n",
    "    - 通用过滤器算法\n",
    "- 数据简化函数: 计算图像直方图、局部极值位置、角。\n",
    "- 其他动作: I/O, 可视化，等。\n",
    "\n",
    "## 3.3.2 输入/输出, 数据类型和颜色空间\n",
    "\n",
    "I/O: [skimage.io](http://scikit-image.org/docs/stable/api/skimage.io.html#module-skimage.io)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from skimage import io"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "读取文件: [skimage.io.imread()](http://scikit-image.org/docs/stable/api/skimage.io.html#skimage.io.imread)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import os\n",
    "filename = os.path.join(skimage.data_dir, 'camera.png')\n",
    "camera = io.imread(filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](http://www.scipy-lectures.org/_images/plot_camera_1.png)\n",
    "\n",
    "支持所有被Python Imaging Library（或者`imread` `plugin`关键词提供的任何I/O插件）的数据格式。\n",
    "也支持URL图片路径:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "logo = io.imread('http://scikit-image.org/_static/img/logo.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "存储文件:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "io.imsave('local_logo.png', logo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(`imsave`也用外部插件比如PIL)\n",
    "\n",
    "### 3.3.2.1 数据类型\n",
    "\n",
    "![](http://www.scipy-lectures.org/_images/plot_camera_uint_1.png)\n",
    "\n",
    "图像ndarrays可以用整数（有符号或无符号）或浮点来代表。\n",
    "\n",
    "小心整数类型的溢出"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('uint8')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "camera = data.camera()\n",
    "camera.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "camera_multiply = 3 * camera"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可用不同的整型大小: 8-, 16- 或 32-字节, 有符号或无符号。\n",
    "\n",
    "---\n",
    "一个重要的 (如果有疑问的话) `skimage` **惯例**: 图像浮点支持在[-1, 1] (与所以浮点图像相对)\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(255, 1.0)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skimage import img_as_float\n",
    "camera_float = img_as_float(camera)\n",
    "camera.max(), camera_float.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "一些图像处理程序需要应用在浮点数组上，因此，输出的数组可能类型和数据范围都与输入数组不同"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5915023652179584"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "try:\n",
    "    from skimage import filters\n",
    "except ImportError:\n",
    "    from skimage import filter as filters\n",
    "camera_sobel = filters.sobel(camera)\n",
    "camera_sobel.max() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "在上面的例子中，我们使用`scikit-image`的子模块`filters`，在0.11到0.10版本间，`filter`被重命名为`filters`，为了避免与Python内置的`filter`冲突。\n",
    "\n",
    "---\n",
    "\n",
    "在[skimage](http://scikit-image.org/docs/stable/api/skimage.html#module-skimage)提供了下列skimage实用的函数来转化dtype和data range: `util.img_as_float`、 `util.img_as_ubyte`等。\n",
    "\n",
    "看一下[用户手册](http://scikit-image.org/docs/stable/user_guide/data_types.html)来了解细节。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "An important (if questionable) skimage convention: float images are supposed to lie in [-1, 1] (in order to have comparable contrast for all float images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "3.3.1. Introduction and concepts\n",
    "\n",
    "Images are NumPy’s arrays np.ndarray"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}