• 颜色空间变换(RGB-HSV)


    #!/usr/bin/env python
    #******************************************************************************
    #  $Id$
    # 
    #  Project:  GDAL Python Interface
    #  Purpose:  Script to merge greyscale as intensity into an RGB(A) image, for
    #            instance to apply hillshading to a dem colour relief.
    #  Author:   Frank Warmerdam, warmerdam@pobox.com
    #            Trent Hare (USGS)
    # 
    #******************************************************************************
    #  Copyright (c) 2009, Frank Warmerdam
    #  Copyright (c) 2010, Even Rouault <even dot rouault at mines-paris dot org>
    # 
    #  Permission is hereby granted, free of charge, to any person obtaining a
    #  copy of this software and associated documentation files (the "Software"),
    #  to deal in the Software without restriction, including without limitation
    #  the rights to use, copy, modify, merge, publish, distribute, sublicense,
    #  and/or sell copies of the Software, and to permit persons to whom the
    #  Software is furnished to do so, subject to the following conditions:
    # 
    #  The above copyright notice and this permission notice shall be included
    #  in all copies or substantial portions of the Software.
    # 
    #  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
    #  OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
    #  FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
    #  THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
    #  LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
    #  FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
    #  DEALINGS IN THE SOFTWARE.
    #******************************************************************************
    
    from osgeo import gdal
    import numpy
    import sys
    
    # =============================================================================
    # rgb_to_hsv()
    #
    # rgb comes in as [r,g,b] with values in the range [0,255].  The returned
    # hsv values will be with hue and saturation in the range [0,1] and value
    # in the range [0,255]
    #
    def rgb_to_hsv( r,g,b ):
    
        maxc = numpy.maximum(r,numpy.maximum(g,b))
        minc = numpy.minimum(r,numpy.minimum(g,b))
    
        v = maxc
    
        minc_eq_maxc = numpy.equal(minc,maxc)
    
        # compute the difference, but reset zeros to ones to avoid divide by zeros later.
        ones = numpy.ones((r.shape[0],r.shape[1]))
        maxc_minus_minc = numpy.choose( minc_eq_maxc, (maxc-minc,ones) )
    
        s = (maxc-minc) / numpy.maximum(ones,maxc)
        rc = (maxc-r) / maxc_minus_minc
        gc = (maxc-g) / maxc_minus_minc
        bc = (maxc-b) / maxc_minus_minc
    
        maxc_is_r = numpy.equal(maxc,r)
        maxc_is_g = numpy.equal(maxc,g)
        maxc_is_b = numpy.equal(maxc,b)
    
        h = numpy.zeros((r.shape[0],r.shape[1]))
        h = numpy.choose( maxc_is_b, (h,4.0+gc-rc) )
        h = numpy.choose( maxc_is_g, (h,2.0+rc-bc) )
        h = numpy.choose( maxc_is_r, (h,bc-gc) )
    
        h = numpy.mod(h/6.0,1.0)
    
        hsv = numpy.asarray([h,s,v])
        
        return hsv
    
    # =============================================================================
    # hsv_to_rgb()
    #
    # hsv comes in as [h,s,v] with hue and saturation in the range [0,1],
    # but value in the range [0,255].
    
    def hsv_to_rgb( hsv ):
    
        h = hsv[0]
        s = hsv[1]
        v = hsv[2]
    
        #if s == 0.0: return v, v, v
        i = (h*6.0).astype(int)
        f = (h*6.0) - i
        p = v*(1.0 - s)
        q = v*(1.0 - s*f)
        t = v*(1.0 - s*(1.0-f))
    
        r = i.choose( v, q, p, p, t, v )
        g = i.choose( t, v, v, q, p, p )
        b = i.choose( p, p, t, v, v, q )
    
        rgb = numpy.asarray([r,g,b]).astype(numpy.uint8)
        
        return rgb
    
    # =============================================================================
    # Usage()
    
    def Usage():
        print("""Usage: hsv_merge.py [-q] [-of format] src_color src_greyscale dst_color
    
    where src_color is a RGB or RGBA dataset,
          src_greyscale is a greyscale dataset (e.g. the result of gdaldem hillshade)
          dst_color will be a RGB or RGBA dataset using the greyscale as the
          intensity for the color dataset.
    """)
        sys.exit(1)
        
    # =============================================================================
    #     Mainline
    # =============================================================================
    
    argv = gdal.GeneralCmdLineProcessor( sys.argv )
    if argv is None:
        sys.exit( 0 )
    
    format = 'GTiff'
    src_color_filename = None
    src_greyscale_filename = None
    dst_color_filename = None
    quiet = False
    
    # Parse command line arguments.
    i = 1
    while i < len(argv):
        arg = argv[i]
    
        if arg == '-of':
            i = i + 1
            format = argv[i]
    
        elif arg == '-q' or arg == '-quiet':
            quiet = True
    
        elif src_color_filename is None:
            src_color_filename = argv[i]
    
        elif src_greyscale_filename is None:
            src_greyscale_filename = argv[i]
    
        elif dst_color_filename is None:
            dst_color_filename = argv[i]
        else:
            Usage()
    
        i = i + 1
    
    if dst_color_filename is None:
        Usage()
    
    datatype = gdal.GDT_Byte
    
    hilldataset = gdal.Open( src_greyscale_filename, gdal.GA_ReadOnly )
    colordataset = gdal.Open( src_color_filename, gdal.GA_ReadOnly )
    
    #check for 3 or 4 bands in the color file
    if (colordataset.RasterCount != 3 and colordataset.RasterCount != 4):
        print('Source image does not appear to have three or four bands as required.')
        sys.exit(1)
    
    #define output format, name, size, type and set projection
    out_driver = gdal.GetDriverByName(format)
    outdataset = out_driver.Create(dst_color_filename, colordataset.RasterXSize, 
                       colordataset.RasterYSize, colordataset.RasterCount, datatype)
    outdataset.SetProjection(hilldataset.GetProjection())
    outdataset.SetGeoTransform(hilldataset.GetGeoTransform())
    
    #assign RGB and hillshade bands
    rBand = colordataset.GetRasterBand(1)
    gBand = colordataset.GetRasterBand(2)
    bBand = colordataset.GetRasterBand(3)
    if colordataset.RasterCount == 4:
        aBand = colordataset.GetRasterBand(4)
    else:
        aBand = None
    
    hillband = hilldataset.GetRasterBand(1)
    hillbandnodatavalue = hillband.GetNoDataValue()
    
    #check for same file size
    if ((rBand.YSize != hillband.YSize) or (rBand.XSize != hillband.XSize)):
        print('Color and hilshade must be the same size in pixels.')
        sys.exit(1)
    
    #loop over lines to apply hillshade
    for i in range(hillband.YSize):
        #load RGB and Hillshade arrays
        rScanline = rBand.ReadAsArray(0, i, hillband.XSize, 1, hillband.XSize, 1)
        gScanline = gBand.ReadAsArray(0, i, hillband.XSize, 1, hillband.XSize, 1)
        bScanline = bBand.ReadAsArray(0, i, hillband.XSize, 1, hillband.XSize, 1)
        hillScanline = hillband.ReadAsArray(0, i, hillband.XSize, 1, hillband.XSize, 1)
    
        #convert to HSV
        hsv = rgb_to_hsv( rScanline, gScanline, bScanline )
        
        # if there's nodata on the hillband, use the v value from the color
        # dataset instead of the hillshade value.
        if hillbandnodatavalue is not None:
            equal_to_nodata = numpy.equal(hillScanline, hillbandnodatavalue)
            v = numpy.choose(equal_to_nodata,(hillScanline,hsv[2]))
        else:
            v = hillScanline
    
        #replace v with hillshade
        hsv_adjusted = numpy.asarray( [hsv[0], hsv[1], v] )
        
        #convert back to RGB
        dst_color = hsv_to_rgb( hsv_adjusted )
    
        #write out new RGB bands to output one band at a time
        outband = outdataset.GetRasterBand(1)
        outband.WriteArray(dst_color[0], 0, i)
        outband = outdataset.GetRasterBand(2)
        outband.WriteArray(dst_color[1], 0, i)
        outband = outdataset.GetRasterBand(3)
        outband.WriteArray(dst_color[2], 0, i)
        if aBand is not None:
            aScanline = aBand.ReadAsArray(0, i, hillband.XSize, 1, hillband.XSize, 1)
            outband = outdataset.GetRasterBand(4)
            outband.WriteArray(aScanline, 0, i)
        
        #update progress line
        if not quiet:
            gdal.TermProgress_nocb( (float(i+1) / hillband.YSize) )
  • 相关阅读:
    python中的编码与解码
    python中的迭代器
    python中高阶函数与装饰器(3)
    python中高阶函数与装饰器(2)
    python中高阶函数与装饰器
    python中几种常用的数据类型
    python 启航
    Python [习题] 字典排序
    SQL-TSQL
    SQL基础语句
  • 原文地址:https://www.cnblogs.com/gispathfinder/p/5790488.html
Copyright © 2020-2023  润新知