BERATools 0.2.0__py3-none-any.whl → 0.2.2__py3-none-any.whl

beratools/tools/ln_relative_metrics.py

@@ -1,615 +1,615 @@
-import sys
-import os.path
-import geopandas
-import warnings
-
-import time
-import pandas
-import numpy
-import shapely
-import math
-
-from multiprocessing.pool import Pool
-from numpy.lib.stride_tricks import as_strided
-
-from beratools.tools.common import *
-
-# to suppress panadas UserWarning: SettingWithCopyWarning:
-# A value is trying to be set on a copy of a slice from a DataFrame.
-# Try using .loc[row_indexer,col_indexer] = value instead
-warnings.simplefilter(action='ignore', category=UserWarning)
-
-USE_MULTI_PROCESSING = True
-
-
-class OperationCancelledException(Exception):
-    pass
-
-
-# by  Dan Patterson
-
-def _check(a, r_c, subok=False):
-    """Performs the array checks necessary for stride and block.
-    : in_array   - Array or list.
-    : r_c - tuple/list/array of rows x cols.
-    : subok - from numpy 1.12 added, keep for now
-    :Returns:
-    :------
-    :Attempts will be made to ...
-    :  produce a shape at least (1*c).  For a scalar, the
-    :  minimum shape will be (1*r) for 1D array or (1*c) for 2D
-    :  array if r<c.  Be aware
-    """
-    if isinstance(r_c, (int, float)):
-        r_c = (1, int(r_c))
-    r, c = r_c
-    if a.ndim == 1:
-        a = numpy.atleast_2d(a)
-    r, c = r_c = (min(r, a.shape[0]), min(c, a.shape[1]))
-    a = numpy.array(a, copy=False, subok=subok)
-    return a, r, c, tuple(r_c)
-
-
-def _pad(in_array, kernel):
-    """Pad a sliding array to allow for stats"""
-    pad_x = int(kernel.shape[0] / 2)
-    pad_y = int(kernel.shape[0] / 2)
-    result = numpy.pad(in_array, pad_width=(pad_x, pad_y), mode="constant", constant_values=(numpy.NaN, numpy.NaN))
-
-    return result
-
-
-def stride(a, r_c):
-    """Provide a 2D sliding/moving view of an array.
-    :  There is no edge correction for outputs.
-    :
-    :Requires:
-    :--------
-    : _check(a, r_c) ... Runs the checks on the inputs.
-    : a - array or list, usually a 2D array.  Assumes rows is >=1,
-    :     it is corrected as is the number of columns.
-    : r_c - tuple/list/array of rows x cols.  Attempts  to
-    :     produce a shape at least (1*c).  For a scalar, the
-    :     minimum shape will be (1*r) for 1D array or 2D
-    :     array if r<c.  Be aware
-    """
-
-    a, r, c, r_c = _check(a, r_c)
-    shape = (a.shape[0] - r + 1, a.shape[1] - c + 1) + r_c
-    strides = a.strides * 2
-    a_s = (as_strided(a, shape=shape, strides=strides)).squeeze()
-    return a_s
-
-
-def cal_sar(a_s, cell_x, cell_y, diag):
-    # Jenness, J. S. 2004.  Calculating landscape surface area from digital elevation models.
-    # Wildlife Society Bulletin. 32(3):829-839
-    # For SAR
-    # kernel index				cell values (example)
-    # 0,0	0,1	 0,2		4.326477	9.00671	   10.430054
-    # 1,0	1,1	 1,2		7.472778	7.408875	4.323486
-    # 2,0	2,1	 2,2		8.534485	8.106201	7.350098
-    #
-    # Direction
-    # 8	  1	   2
-    # 7	Center 3
-    # 6   5	   4
-    #
-    center = a_s[1, 1]
-    # Pythagorean Theorem
-
-    if not numpy.isnan(center):
-        # Center to 8 Directions
-        dir1 = math.sqrt(abs(center - a_s[0, 1]) ** 2 + (cell_y ** 2))
-        dir2 = math.sqrt(abs(center - a_s[0, 2]) ** 2 + diag ** 2)
-        dir3 = math.sqrt(abs(center - a_s[1, 2]) ** 2 + cell_x ** 2)
-        dir4 = math.sqrt(abs(center - a_s[2, 2]) ** 2 + diag ** 2)
-        dir5 = math.sqrt(abs(center - a_s[2, 1]) ** 2 + cell_y ** 2)
-        dir6 = math.sqrt(abs(center - a_s[2, 0]) ** 2 + diag ** 2)
-        dir7 = math.sqrt(abs(center - a_s[1, 0]) ** 2 + cell_x ** 2)
-        dir8 = math.sqrt(abs(center - a_s[0, 0]) ** 2 + diag ** 2)
-        # 8 Outer sides
-        dir1_2 = math.sqrt(abs(a_s[0, 1] - a_s[0, 2]) ** 2 + cell_x ** 2)
-        dir2_3 = math.sqrt(abs(a_s[0, 2] - a_s[1, 2]) ** 2 + cell_y ** 2)
-        dir3_4 = math.sqrt(abs(a_s[1, 2] - a_s[2, 2]) ** 2 + cell_y ** 2)
-        dir4_5 = math.sqrt(abs(a_s[2, 2] - a_s[2, 1]) ** 2 + cell_x ** 2)
-        dir5_6 = math.sqrt(abs(a_s[2, 1] - a_s[2, 0]) ** 2 + cell_x ** 2)
-        dir6_7 = math.sqrt(abs(a_s[2, 0] - a_s[1, 0]) ** 2 + cell_y ** 2)
-        dir7_8 = math.sqrt(abs(a_s[1, 0] - a_s[0, 0]) ** 2 + cell_y ** 2)
-        dir8_1 = math.sqrt(abs(a_s[0, 0] - a_s[0, 1]) ** 2 + cell_x ** 2)
-
-        # Heron of Alexandria and Archimedes (see also Abramowitz and Stegun [1972, p. 79]):
-        p1 = (dir1 + dir2 + dir1_2) / 2
-        area1 = math.sqrt(p1 * (p1 - dir1) * (p1 - dir2) * (p1 - dir1_2))
-        p2 = (dir2 + dir3 + dir2_3) / 2
-        area2 = math.sqrt(p2 * (p2 - dir2) * (p2 - dir3) * (p2 - dir2_3))
-        p3 = (dir3 + dir4 + dir3_4) / 2
-        area3 = math.sqrt(p3 * (p3 - dir3) * (p3 - dir4) * (p3 - dir3_4))
-        p4 = (dir4 + dir5 + dir4_5) / 2
-        area4 = math.sqrt(p4 * (p4 - dir4) * (p4 - dir5) * (p4 - dir4_5))
-        p5 = (dir5 + dir6 + dir5_6) / 2
-        area5 = math.sqrt(p5 * (p5 - dir5) * (p5 - dir6) * (p5 - dir5_6))
-        p6 = (dir6 + dir7 + dir6_7) / 2
-        area6 = math.sqrt(p6 * (p6 - dir6) * (p6 - dir7) * (p6 - dir6_7))
-        p7 = (dir7 + dir8 + dir7_8) / 2
-        area7 = math.sqrt(p7 * (p7 - dir7) * (p7 - dir8) * (p7 - dir7_8))
-        p8 = (dir8 + dir1 + dir8_1) / 2
-        area8 = math.sqrt(p8 * (p8 - dir8) * (p8 - dir1) * (p8 - dir8_1))
-        areas = (list([area1, area2, area3, area4, area5, area6, area7, area8]))
-        surface_area = 0
-        for area in areas:
-            if not math.isnan(area):
-                surface_area = surface_area + area
-        return surface_area
-    else:
-        surface_area = math.nan
-        return surface_area
-
-
-def cal_tri(a_s):
-    # For TRI
-    # refer https://livingatlas-dcdev.opendata.arcgis.com/content/28360713391948af9303c0aeabb45afd/about
-    # for example: TRI with your elevation data.The results may be interpreted as follows:
-    # a level terrain surface
-    # a nearly level surface
-    # a slightly rugged surface
-    # an intermediately rugged surface
-    # a moderately rugged surface
-    # a highly rugged surface
-    # an extremely rugged surface
-    if not numpy.isnan(a_s[1, 1]):
-        result = math.sqrt(abs((numpy.nanmax(a_s)) ** 2 - (numpy.nanmin(a_s)) ** 2))
-    else:
-        result = math.nan
-    return result
-
-
-def cal_index(in_ndarray, cell_x, cell_y, type):
-    kernel = numpy.arange(3 ** 2)
-    kernel.shape = (3, 3)
-    kernel.fill(1)
-    padded_array = _pad(in_ndarray, kernel)
-    a_s = stride(padded_array, kernel.shape)
-    rows, cols = a_s.shape[0], a_s.shape[1]
-    result = numpy.arange(rows * cols)
-    result.shape = (rows, cols)
-    result = result.astype('float64')
-    result.fill(numpy.nan)
-    diag = math.sqrt(cell_x ** 2 + cell_y ** 2)
-    plannar_area = (cell_y * cell_x) / 2  # area of one cell
-
-    if type == 'SAR':
-        for y in range(rows):
-            for x in range(cols):
-                result[y, x] = cal_sar(a_s[y, x], cell_x, cell_y, diag)
-        total_surface_area = numpy.nansum(result)
-        with numpy.errstate(divide='ignore', invalid='ignore'):
-            result_ratio = numpy.true_divide(result, plannar_area)
-
-        return result_ratio, total_surface_area
-    elif type == 'TRI':
-        for y in range(rows):
-            for x in range(cols):
-                result[y, x] = cal_tri(a_s[y, x])
-        return result
-
-
-def forest_metrics(callback, in_line, out_line, raster_type, in_raster, proc_segments, buffer_ln_dist,
-                   cl_metrics_gap, forest_buffer_dist, processes, verbose, worklnbuffer_dfLR=None):
-    # file_path,in_file_name=os.path.split(in_line)
-    output_path, output_file = os.path.split(out_line)
-    output_filename, file_ext = os.path.splitext((output_file))
-    if not os.path.exists(output_path):
-        print("No output file path found, pls check.")
-        exit()
-    else:
-        if file_ext.lower() != ".shp":
-            print("Output file type should be shapefile, pls check.")
-            exit()
-
-    try:
-        line_seg = geopandas.GeoDataFrame.from_file(in_line)
-    except:
-        print(sys.exc_info())
-        exit()
-
-    # check coordinate systems between line and raster features
-    with rasterio.open(in_raster) as in_image:
-        if not in_image.crs.to_epsg() in [2956]:
-            print("Line and raster spatial references are not same, please check.")
-            exit()
-
-    # # Check the ** column in data. If it is not, new column will be created
-    # if not '**' in line_seg.columns.array:
-    #     print("Cannot find {} column in input line data.\n '{}' column will be create".format("**","**"))
-    #     line_seg['**'] = numpy.nan
-
-    # Check the OLnFID column in data. If it is not, column will be created
-    if not 'OLnFID' in line_seg.columns.array:
-        print(
-            "Cannot find {} column in input line data.\n '{}' column will be create".format('OLnFID', 'OLnFID'))
-        line_seg['OLnFID'] = line_seg.index
-
-    if proc_segments == True:
-        line_seg = split_into_Equal_Nth_segments(line_seg)
-
-    else:
-        pass
-
-    # copy original line input to another Geodataframe
-    workln_dfC = geopandas.GeoDataFrame.copy((line_seg))
-    workln_dfL = geopandas.GeoDataFrame.copy((line_seg))
-    workln_dfR = geopandas.GeoDataFrame.copy((line_seg))
-
-    # copy parallel lines for both side of the input lines
-    print("Creating area for CL....")
-    workln_dfC['geometry'] = workln_dfC['geometry'].simplify(tolerance=0.05, preserve_topology=True)
-    worklnbuffer_dfC = geopandas.GeoDataFrame.copy((workln_dfC))
-    worklnbuffer_dfC['geometry'] = shapely.buffer(workln_dfC['geometry'], distance=float(buffer_ln_dist),
-                                                  cap_style=2, join_style=2, single_sided=False)
-
-    print("Creating offset area for surrounding forest....")
-    workln_dfL, workln_dfR = multiprocessing_copyparallel_lineLR(workln_dfL, workln_dfR, processes,
-                                                                 left_dis=float(buffer_ln_dist + cl_metrics_gap),
-                                                                 right_dist=-float(buffer_ln_dist + cl_metrics_gap))
-    workln_dfR = workln_dfR.sort_values(by=['OLnFID'])
-    workln_dfL = workln_dfL.sort_values(by=['OLnFID'])
-    workln_dfL = workln_dfL.reset_index(drop=True)
-    workln_dfR = workln_dfR.reset_index(drop=True)
-
-    print('%{}'.format(30))
-
-    worklnbuffer_dfL = geopandas.GeoDataFrame.copy((workln_dfL))
-    worklnbuffer_dfR = geopandas.GeoDataFrame.copy((workln_dfR))
-
-    # buffer the parallel line in one side (extend the area into forest)
-
-    worklnbuffer_dfL['geometry'] = shapely.buffer(workln_dfL['geometry'], distance=float(forest_buffer_dist),
-                                                  cap_style=2, join_style=2, single_sided=True)
-    worklnbuffer_dfR['geometry'] = shapely.buffer(workln_dfR['geometry'], distance=-float(forest_buffer_dist),
-                                                  cap_style=2, join_style=2, single_sided=True)
-    print("Creating offset area for surrounding forest....Done")
-    print('%{}'.format(50))
-    # create a New column for surrounding forest statistics:
-    # 1) Height Percentile (add more in the future)
-    # worklnbuffer_dfL['**_L'] = numpy.nan
-    # worklnbuffer_dfR['**_R'] = numpy.nan
-    # worklnbuffer_dfL = worklnbuffer_dfL.reset_index(drop=True)
-    # worklnbuffer_dfR=worklnbuffer_dfR.reset_index(drop=True)
-    # line_seg['L_**'] = numpy.nan
-    # line_seg['R_**'] = numpy.nan
-    print('%{}'.format(80))
-
-    print("Calculating CL metrics ..")
-    worklnbuffer_dfC, new_col_c = multiprocessing_metrics(worklnbuffer_dfC, in_raster, raster_type, processes,
-                                                          side='center')
-
-    worklnbuffer_dfC = worklnbuffer_dfC.sort_values(by=['OLnFID'])
-    worklnbuffer_dfC = worklnbuffer_dfC.reset_index(drop=True)
-    print("Calculating surrounding forest metrics....")
-    # calculate the Height percentile for each parallel area using CHM
-    worklnbuffer_dfL, new_col_l = multiprocessing_metrics(worklnbuffer_dfL, in_raster, raster_type, processes,
-                                                          side='left')
-
-    worklnbuffer_dfL = worklnbuffer_dfL.sort_values(by=['OLnFID'])
-    worklnbuffer_dfL = worklnbuffer_dfL.reset_index(drop=True)
-
-    worklnbuffer_dfR, new_col_r = multiprocessing_metrics(worklnbuffer_dfR, in_raster, raster_type, processes,
-                                                          side='right')
-
-    worklnbuffer_dfR = worklnbuffer_dfR.sort_values(by=['OLnFID'])
-    worklnbuffer_dfR = worklnbuffer_dfR.reset_index(drop=True)
-
-    print('%{}'.format(90))
-
-    all_new_col = numpy.append(numpy.array(new_col_c), numpy.array(new_col_l))
-    all_new_col = numpy.append(all_new_col, numpy.array(new_col_r))
-
-    for index in (line_seg.index):
-        if raster_type == 'DEM':
-            for col in all_new_col:
-                if "C_" in col:
-                    line_seg.loc[index, col] = worklnbuffer_dfC.loc[index, col]
-                elif "L_" in col:
-                    line_seg.loc[index, col] = worklnbuffer_dfL.loc[index, col]
-                elif "R_" in col:
-                    line_seg.loc[index, col] = worklnbuffer_dfR.loc[index, col]
-    print("Calculating forest metrics....Done")
-
-    print("Saving forest metrics output.....")
-    geopandas.GeoDataFrame.to_file(line_seg, out_line)
-    print("Saving forest metrics output.....Done")
-    del line_seg, worklnbuffer_dfL, worklnbuffer_dfR, workln_dfL, workln_dfR
-
-    print('%{}'.format(100))
-
-
-# TODO: inspect duplicates, split_line_npart in line_footprint_functions.py
-def split_line_nPart(line):
-    from shapely.ops import split, snap
-    # Work out n parts for each line (divided by 10m)
-    n = math.ceil(line.length / 10)
-    if n > 1:
-        # divided line into n-1 equal parts;
-        distances = numpy.linspace(0, line.length, n)
-        points = [line.interpolate(distance) for distance in distances]
-
-        split_points = shapely.multipoints(points)
-        # mline=cut_line_at_points(line,points)
-        mline = split(line, split_points)
-        # mline=split_line_fc(line)
-    else:
-        mline = line
-    return mline
-
-
-def split_into_Equal_Nth_segments(df):
-    odf = df
-    crs = odf.crs
-    if not 'OLnSEG' in odf.columns.array:
-        df['OLnSEG'] = numpy.nan
-    df = odf.assign(geometry=odf.apply(lambda x: split_line_nPart(x.geometry), axis=1))
-    df = df.explode()
-
-    df['OLnSEG'] = df.groupby('OLnFID').cumcount()
-    gdf = geopandas.GeoDataFrame(df, geometry=df.geometry, crs=crs)
-    gdf = gdf.sort_values(by=['OLnFID', 'OLnSEG'])
-    gdf = gdf.reset_index(drop=True)
-    return gdf
-
-
-def split_line_fc(line):
-    return list(map(shapely.LineString, zip(line.coords[:-1], line.coords[1:])))
-
-
-def split_into_every_segments(df):
-    odf = df
-    crs = odf.crs
-    if not 'OLnSEG' in odf.columns.array:
-        df['OLnSEG'] = numpy.nan
-    else:
-        pass
-    df = odf.assign(geometry=odf.apply(lambda x: split_line_fc(x.geometry), axis=1))
-    df = df.explode()
-
-    df['OLnSEG'] = df.groupby('OLnFID').cumcount()
-    gdf = geopandas.GeoDataFrame(df, geometry=df.geometry, crs=crs)
-    gdf = gdf.sort_values(by=['OLnFID', 'OLnSEG'])
-    gdf = gdf.reset_index(drop=True)
-    return gdf
-
-
-def multiprocessing_copyparallel_lineLR(dfL, dfR, processes, left_dis, right_dist):
-    try:
-
-        line_arg = []
-        total_steps = len(dfL)
-
-        for item in dfL.index:
-            item_list = [dfL, dfR, left_dis, right_dist, item]
-            line_arg.append(item_list)
-
-        featuresL = []
-        featuresR = []
-        chunksize = math.ceil(total_steps / processes)
-        with Pool(processes=int(processes)) as pool:
-            step = 0
-            # execute tasks in order, process results out of order
-            for resultL, resultR in pool.imap_unordered(copyparallel_lineLR, line_arg, chunksize=chunksize):
-                if BT_DEBUGGING:
-                    print('Got result: {}{}'.format(resultL, resultR), flush=True)
-                featuresL.append(resultL)
-                featuresR.append(resultR)
-                step += 1
-                print('%{}'.format(step / total_steps * 100))
-            return geopandas.GeoDataFrame(pandas.concat(featuresL)), geopandas.GeoDataFrame(pandas.concat(featuresR))
-    except OperationCancelledException:
-        print("Operation cancelled")
-
-
-def multiprocessing_metrics(df, in_raster, raster_type, processes, side):
-    try:
-        line_arg = []
-        total_steps = len(df)
-        if side == 'left':
-            PerCol = 'L'
-        elif side == 'right':
-            PerCol = 'R'
-        else:
-            PerCol = 'C'
-
-        for item in df.index:
-            item_list = [df.iloc[[item]], in_raster, item, PerCol, raster_type]
-            line_arg.append(item_list)
-        features = []
-        chunksize = math.ceil(total_steps / processes)
-
-        if USE_MULTI_PROCESSING:
-            with Pool(processes=int(processes)) as pool:
-
-                step = 0
-                # execute tasks in order, process results out of order
-                try:
-                    for result in pool.imap_unordered(cal_metrics, line_arg, chunksize=chunksize):
-                        if BT_DEBUGGING:
-                            print('Got result: {}'.format(result), flush=True)
-                        features.append(result)
-                        step += 1
-                        print('%{}'.format(step / total_steps * 100))
-                except Exception:
-                    print(Exception)
-                    raise
-                del line_arg
-                df = geopandas.GeoDataFrame(pandas.concat(features))
-                new_col = []
-                for col in df.columns.array:
-                    if "C_" in col:
-                        new_col.append(col)
-                    elif "L_" in col:
-                        new_col.append(col)
-                    elif "R_" in col:
-                        new_col.append(col)
-
-                return df, new_col
-        else:
-            for row in line_arg:
-                features.append(cal_metrics(row))
-            df = geopandas.GeoDataFrame(pandas.concat(features))
-            new_col = []
-            for col in df.columns.array:
-                if "C_" in col:
-                    new_col.append(col)
-                elif "L_" in col:
-                    new_col.append(col)
-                elif "R_" in col:
-                    new_col.append(col)
-
-            return df, new_col
-
-    except OperationCancelledException:
-        print("Operation cancelled")
-
-
-def cal_metrics(line_arg):
-    try:
-        df = line_arg[0]
-        raster = line_arg[1]
-        row_index = line_arg[2]
-        PerCol = line_arg[3]
-        raster_type = line_arg[4]
-        line_buffer = df.loc[row_index, 'geometry']
-    except:
-        print("Assigning variable on index:{} Error: ".format(line_arg) + sys.exc_info())
-        exit()
-    try:
-        with rasterio.open(raster) as image:
-
-            clipped_raster, out_transform = rasterio.mask.mask(image, [line_buffer], crop=True,
-                                                               nodata=BT_NODATA, filled=True)
-            clipped_raster = numpy.squeeze(clipped_raster, axis=0)
-            cell_x, cell_y = image.res
-            cell_area = cell_x * cell_y
-            # mask all -9999 (nodata) value cells
-            masked_raster = numpy.ma.masked_where(clipped_raster == BT_NODATA, clipped_raster)
-            filled_raster = numpy.ma.filled(masked_raster, numpy.nan)
-
-            # Calculate the metrics
-            if raster_type == "DEM":
-                # Surface area ratio (SAR)
-
-                SAR, total_surface_area = cal_index(filled_raster, cell_x, cell_y, 'SAR')
-                SAR_mean = numpy.nanmean(SAR)
-                SAR_percentile90 = numpy.nanpercentile(SAR, 90, method='hazen')
-                SAR_median = numpy.nanmedian(SAR)
-                SAR_std = numpy.nanstd(SAR)
-                SAR_max = numpy.nanmax(SAR)
-                SAR_min = numpy.nanmin(SAR)
-
-                # General Statistics
-                total_planar_area = numpy.ma.count(masked_raster) * cell_area
-                cell_volume = masked_raster * cell_area
-                total_volume = numpy.ma.sum(cell_volume)
-                mean = numpy.nanmean(filled_raster)
-                percentile90 = numpy.nanpercentile(filled_raster, 90, method='hazen')
-                median = numpy.nanmedian(filled_raster)
-                std = numpy.nanstd(filled_raster)
-                max = numpy.nanmax(filled_raster)
-                min = numpy.nanmin(filled_raster)
-
-                # Terrain Ruggedness Index (TRI)
-                # TRI = cal_index(filled_raster, cell_x, cell_y, 'TRI')
-                # TRI_mean = numpy.nanmean(TRI)
-                # TRI_percentile90 = numpy.nanpercentile(TRI, 90, method='hazen')
-                # TRI_median = numpy.nanmedian(TRI)
-                # TRI_std = numpy.nanstd(TRI)
-                # TRI_max = numpy.nanmax(TRI)
-                # TRI_min = numpy.nanmin(TRI)
-
-            del clipped_raster, out_transform
-
-    # return the generated value
-    except:
-        print(sys.exc_info())
-    try:
-
-        # Writing General statisic
-        df.at[row_index, PerCol + '_PlArea'] = total_planar_area
-        df.at[row_index, PerCol + '_mean'] = mean
-        # df.at[row_index, PerCol + '_P90'] = percentile90
-        # df.at[row_index, PerCol + '_median'] = median
-        df.at[row_index, PerCol + '_StdDev'] = std
-        df.at[row_index, PerCol + '_max'] = max
-        df.at[row_index, PerCol + '_min'] = min
-        # df.at[row_index, PerCol + '_Vol'] = total_volume
-
-        # Writing SAR statisic
-        df.at[row_index, PerCol + '_SurArea'] = total_surface_area
-        df.at[row_index, PerCol + '_SARmean'] = SAR_mean
-        # df.at[row_index, PerCol + '_SARP90'] = SAR_percentile90
-        # df.at[row_index, PerCol + '_SARmed'] = SAR_median
-        df.at[row_index, PerCol + '_SARStd'] = SAR_std
-        df.at[row_index, PerCol + '_SARmax'] = SAR_max
-        df.at[row_index, PerCol + '_SARmin'] = SAR_min
-
-        # highest=df[PerCol + '_max'].max()
-        # lowest = df[PerCol + '_min'].min()
-
-        # Writing TRI statisic
-        # df.at[row_index, PerCol + '_TRImean'] = TRI_mean
-        # # df.at[row_index, PerCol + '_TRIP90'] = TRI_percentile90
-        # # df.at[row_index, PerCol + '_TRImed'] = TRI_median
-        # df.at[row_index, PerCol + '_TRIStd'] = TRI_std
-        # df.at[row_index, PerCol + '_TRImax'] = TRI_max
-        # df.at[row_index, PerCol + '_TRImin'] = TRI_min
-
-        return df
-    except:
-        print("Error writing forest metrics into table: " + sys.exc_info())
-
-
-def copyparallel_lineLR(line_arg):
-    dfL = line_arg[0]
-    dfR = line_arg[1]
-    # Simplify input center lines
-    lineL = dfL.loc[line_arg[4], 'geometry'].simplify(tolerance=0.05, preserve_topology=True)
-    lineR = dfL.loc[line_arg[4], 'geometry'].simplify(tolerance=0.05, preserve_topology=True)
-    offset_distL = float(line_arg[2])
-    offset_distR = float(line_arg[3])
-
-    # Older alternative method to the offset_curve() method,
-    # but uses resolution instead of quad_segs and a side keyword (‘left’ or ‘right’) instead
-    # of sign of the distance. This method is kept for backwards compatibility for now,
-    # but it is recommended to use offset_curve() instead.
-    # (ref: https://shapely.readthedocs.io/en/stable/manual.html#object.offset_curve)
-
-    # parallel_lineL = shapely.offset_curve(geometry=lineL, distance=offset_distL,
-    #                                       join_style=shapely.BufferJoinStyle.mitre)
-    # parallel_lineR = shapely.offset_curve(geometry=lineR, distance=offset_distR,
-    #                                       join_style=shapely.BufferJoinStyle.mitre)
-
-    parallel_lineL = lineL.parallel_offset(distance=offset_distL, side='left',
-                                           join_style=shapely.BufferJoinStyle.mitre)
-    parallel_lineR = lineR.parallel_offset(distance=-offset_distR, side='right',
-                                           join_style=shapely.BufferJoinStyle.mitre)
-
-    if not parallel_lineL.is_empty:
-        dfL.loc[line_arg[4], 'geometry'] = parallel_lineL
-    if not parallel_lineR.is_empty:
-        dfR.loc[line_arg[4], 'geometry'] = parallel_lineR
-    return dfL.iloc[[line_arg[4]]], dfR.iloc[[line_arg[4]]]
-
-
-if __name__ == '__main__':
-    start_time = time.time()
-    print('Starting forest metrics calculation processing\n @ {}'.format(
-        time.strftime("%a, %d %b %Y %H:%M:%S", time.localtime())))
-
-    print("Checking input parameters....")
-    in_args, in_verbose = check_arguments()
-
-    verbose = True if in_args.verbose == 'True' else False
-    forest_metrics(print, **in_args.input, processes=int(in_args.processes), verbose=verbose)
-
-    print('%{}'.format(100))
-    print('Starting forest metrics calculation processing is done @ {}\n(or in {} second)'.format(
-        time.strftime("%a, %d %b %Y %H:%M:%S", time.localtime()), round(time.time() - start_time, 5)))
+import sys
+import os.path
+import geopandas
+import warnings
+
+import time
+import pandas
+import numpy
+import shapely
+import math
+
+from multiprocessing.pool import Pool
+from numpy.lib.stride_tricks import as_strided
+
+from beratools.tools.common import *
+
+# to suppress panadas UserWarning: SettingWithCopyWarning:
+# A value is trying to be set on a copy of a slice from a DataFrame.
+# Try using .loc[row_indexer,col_indexer] = value instead
+warnings.simplefilter(action='ignore', category=UserWarning)
+
+USE_MULTI_PROCESSING = True
+
+
+class OperationCancelledException(Exception):
+    pass
+
+
+# by  Dan Patterson
+
+def _check(a, r_c, subok=False):
+    """Performs the array checks necessary for stride and block.
+    : in_array   - Array or list.
+    : r_c - tuple/list/array of rows x cols.
+    : subok - from numpy 1.12 added, keep for now
+    :Returns:
+    :------
+    :Attempts will be made to ...
+    :  produce a shape at least (1*c).  For a scalar, the
+    :  minimum shape will be (1*r) for 1D array or (1*c) for 2D
+    :  array if r<c.  Be aware
+    """
+    if isinstance(r_c, (int, float)):
+        r_c = (1, int(r_c))
+    r, c = r_c
+    if a.ndim == 1:
+        a = numpy.atleast_2d(a)
+    r, c = r_c = (min(r, a.shape[0]), min(c, a.shape[1]))
+    a = numpy.array(a, copy=False, subok=subok)
+    return a, r, c, tuple(r_c)
+
+
+def _pad(in_array, kernel):
+    """Pad a sliding array to allow for stats"""
+    pad_x = int(kernel.shape[0] / 2)
+    pad_y = int(kernel.shape[0] / 2)
+    result = numpy.pad(in_array, pad_width=(pad_x, pad_y), mode="constant", constant_values=(numpy.NaN, numpy.NaN))
+
+    return result
+
+
+def stride(a, r_c):
+    """Provide a 2D sliding/moving view of an array.
+    :  There is no edge correction for outputs.
+    :
+    :Requires:
+    :--------
+    : _check(a, r_c) ... Runs the checks on the inputs.
+    : a - array or list, usually a 2D array.  Assumes rows is >=1,
+    :     it is corrected as is the number of columns.
+    : r_c - tuple/list/array of rows x cols.  Attempts  to
+    :     produce a shape at least (1*c).  For a scalar, the
+    :     minimum shape will be (1*r) for 1D array or 2D
+    :     array if r<c.  Be aware
+    """
+
+    a, r, c, r_c = _check(a, r_c)
+    shape = (a.shape[0] - r + 1, a.shape[1] - c + 1) + r_c
+    strides = a.strides * 2
+    a_s = (as_strided(a, shape=shape, strides=strides)).squeeze()
+    return a_s
+
+
+def cal_sar(a_s, cell_x, cell_y, diag):
+    # Jenness, J. S. 2004.  Calculating landscape surface area from digital elevation models.
+    # Wildlife Society Bulletin. 32(3):829-839
+    # For SAR
+    # kernel index				cell values (example)
+    # 0,0	0,1	 0,2		4.326477	9.00671	   10.430054
+    # 1,0	1,1	 1,2		7.472778	7.408875	4.323486
+    # 2,0	2,1	 2,2		8.534485	8.106201	7.350098
+    #
+    # Direction
+    # 8	  1	   2
+    # 7	Center 3
+    # 6   5	   4
+    #
+    center = a_s[1, 1]
+    # Pythagorean Theorem
+
+    if not numpy.isnan(center):
+        # Center to 8 Directions
+        dir1 = math.sqrt(abs(center - a_s[0, 1]) ** 2 + (cell_y ** 2))
+        dir2 = math.sqrt(abs(center - a_s[0, 2]) ** 2 + diag ** 2)
+        dir3 = math.sqrt(abs(center - a_s[1, 2]) ** 2 + cell_x ** 2)
+        dir4 = math.sqrt(abs(center - a_s[2, 2]) ** 2 + diag ** 2)
+        dir5 = math.sqrt(abs(center - a_s[2, 1]) ** 2 + cell_y ** 2)
+        dir6 = math.sqrt(abs(center - a_s[2, 0]) ** 2 + diag ** 2)
+        dir7 = math.sqrt(abs(center - a_s[1, 0]) ** 2 + cell_x ** 2)
+        dir8 = math.sqrt(abs(center - a_s[0, 0]) ** 2 + diag ** 2)
+        # 8 Outer sides
+        dir1_2 = math.sqrt(abs(a_s[0, 1] - a_s[0, 2]) ** 2 + cell_x ** 2)
+        dir2_3 = math.sqrt(abs(a_s[0, 2] - a_s[1, 2]) ** 2 + cell_y ** 2)
+        dir3_4 = math.sqrt(abs(a_s[1, 2] - a_s[2, 2]) ** 2 + cell_y ** 2)
+        dir4_5 = math.sqrt(abs(a_s[2, 2] - a_s[2, 1]) ** 2 + cell_x ** 2)
+        dir5_6 = math.sqrt(abs(a_s[2, 1] - a_s[2, 0]) ** 2 + cell_x ** 2)
+        dir6_7 = math.sqrt(abs(a_s[2, 0] - a_s[1, 0]) ** 2 + cell_y ** 2)
+        dir7_8 = math.sqrt(abs(a_s[1, 0] - a_s[0, 0]) ** 2 + cell_y ** 2)
+        dir8_1 = math.sqrt(abs(a_s[0, 0] - a_s[0, 1]) ** 2 + cell_x ** 2)
+
+        # Heron of Alexandria and Archimedes (see also Abramowitz and Stegun [1972, p. 79]):
+        p1 = (dir1 + dir2 + dir1_2) / 2
+        area1 = math.sqrt(p1 * (p1 - dir1) * (p1 - dir2) * (p1 - dir1_2))
+        p2 = (dir2 + dir3 + dir2_3) / 2
+        area2 = math.sqrt(p2 * (p2 - dir2) * (p2 - dir3) * (p2 - dir2_3))
+        p3 = (dir3 + dir4 + dir3_4) / 2
+        area3 = math.sqrt(p3 * (p3 - dir3) * (p3 - dir4) * (p3 - dir3_4))
+        p4 = (dir4 + dir5 + dir4_5) / 2
+        area4 = math.sqrt(p4 * (p4 - dir4) * (p4 - dir5) * (p4 - dir4_5))
+        p5 = (dir5 + dir6 + dir5_6) / 2
+        area5 = math.sqrt(p5 * (p5 - dir5) * (p5 - dir6) * (p5 - dir5_6))
+        p6 = (dir6 + dir7 + dir6_7) / 2
+        area6 = math.sqrt(p6 * (p6 - dir6) * (p6 - dir7) * (p6 - dir6_7))
+        p7 = (dir7 + dir8 + dir7_8) / 2
+        area7 = math.sqrt(p7 * (p7 - dir7) * (p7 - dir8) * (p7 - dir7_8))
+        p8 = (dir8 + dir1 + dir8_1) / 2
+        area8 = math.sqrt(p8 * (p8 - dir8) * (p8 - dir1) * (p8 - dir8_1))
+        areas = (list([area1, area2, area3, area4, area5, area6, area7, area8]))
+        surface_area = 0
+        for area in areas:
+            if not math.isnan(area):
+                surface_area = surface_area + area
+        return surface_area
+    else:
+        surface_area = math.nan
+        return surface_area
+
+
+def cal_tri(a_s):
+    # For TRI
+    # refer https://livingatlas-dcdev.opendata.arcgis.com/content/28360713391948af9303c0aeabb45afd/about
+    # for example: TRI with your elevation data.The results may be interpreted as follows:
+    # a level terrain surface
+    # a nearly level surface
+    # a slightly rugged surface
+    # an intermediately rugged surface
+    # a moderately rugged surface
+    # a highly rugged surface
+    # an extremely rugged surface
+    if not numpy.isnan(a_s[1, 1]):
+        result = math.sqrt(abs((numpy.nanmax(a_s)) ** 2 - (numpy.nanmin(a_s)) ** 2))
+    else:
+        result = math.nan
+    return result
+
+
+def cal_index(in_ndarray, cell_x, cell_y, type):
+    kernel = numpy.arange(3 ** 2)
+    kernel.shape = (3, 3)
+    kernel.fill(1)
+    padded_array = _pad(in_ndarray, kernel)
+    a_s = stride(padded_array, kernel.shape)
+    rows, cols = a_s.shape[0], a_s.shape[1]
+    result = numpy.arange(rows * cols)
+    result.shape = (rows, cols)
+    result = result.astype('float64')
+    result.fill(numpy.nan)
+    diag = math.sqrt(cell_x ** 2 + cell_y ** 2)
+    plannar_area = (cell_y * cell_x) / 2  # area of one cell
+
+    if type == 'SAR':
+        for y in range(rows):
+            for x in range(cols):
+                result[y, x] = cal_sar(a_s[y, x], cell_x, cell_y, diag)
+        total_surface_area = numpy.nansum(result)
+        with numpy.errstate(divide='ignore', invalid='ignore'):
+            result_ratio = numpy.true_divide(result, plannar_area)
+
+        return result_ratio, total_surface_area
+    elif type == 'TRI':
+        for y in range(rows):
+            for x in range(cols):
+                result[y, x] = cal_tri(a_s[y, x])
+        return result
+
+
+def forest_metrics(callback, in_line, out_line, raster_type, in_raster, proc_segments, buffer_ln_dist,
+                   cl_metrics_gap, forest_buffer_dist, processes, verbose, worklnbuffer_dfLR=None):
+    # file_path,in_file_name=os.path.split(in_line)
+    output_path, output_file = os.path.split(out_line)
+    output_filename, file_ext = os.path.splitext((output_file))
+    if not os.path.exists(output_path):
+        print("No output file path found, pls check.")
+        exit()
+    else:
+        if file_ext.lower() != ".shp":
+            print("Output file type should be shapefile, pls check.")
+            exit()
+
+    try:
+        line_seg = geopandas.GeoDataFrame.from_file(in_line)
+    except:
+        print(sys.exc_info())
+        exit()
+
+    # check coordinate systems between line and raster features
+    with rasterio.open(in_raster) as in_image:
+        if not in_image.crs.to_epsg() in [2956]:
+            print("Line and raster spatial references are not same, please check.")
+            exit()
+
+    # # Check the ** column in data. If it is not, new column will be created
+    # if not '**' in line_seg.columns.array:
+    #     print("Cannot find {} column in input line data.\n '{}' column will be create".format("**","**"))
+    #     line_seg['**'] = numpy.nan
+
+    # Check the OLnFID column in data. If it is not, column will be created
+    if not 'OLnFID' in line_seg.columns.array:
+        print(
+            "Cannot find {} column in input line data.\n '{}' column will be create".format('OLnFID', 'OLnFID'))
+        line_seg['OLnFID'] = line_seg.index
+
+    if proc_segments == True:
+        line_seg = split_into_Equal_Nth_segments(line_seg)
+
+    else:
+        pass
+
+    # copy original line input to another Geodataframe
+    workln_dfC = geopandas.GeoDataFrame.copy((line_seg))
+    workln_dfL = geopandas.GeoDataFrame.copy((line_seg))
+    workln_dfR = geopandas.GeoDataFrame.copy((line_seg))
+
+    # copy parallel lines for both side of the input lines
+    print("Creating area for CL....")
+    workln_dfC['geometry'] = workln_dfC['geometry'].simplify(tolerance=0.05, preserve_topology=True)
+    worklnbuffer_dfC = geopandas.GeoDataFrame.copy((workln_dfC))
+    worklnbuffer_dfC['geometry'] = shapely.buffer(workln_dfC['geometry'], distance=float(buffer_ln_dist),
+                                                  cap_style=2, join_style=2, single_sided=False)
+
+    print("Creating offset area for surrounding forest....")
+    workln_dfL, workln_dfR = multiprocessing_copyparallel_lineLR(workln_dfL, workln_dfR, processes,
+                                                                 left_dis=float(buffer_ln_dist + cl_metrics_gap),
+                                                                 right_dist=-float(buffer_ln_dist + cl_metrics_gap))
+    workln_dfR = workln_dfR.sort_values(by=['OLnFID'])
+    workln_dfL = workln_dfL.sort_values(by=['OLnFID'])
+    workln_dfL = workln_dfL.reset_index(drop=True)
+    workln_dfR = workln_dfR.reset_index(drop=True)
+
+    print('%{}'.format(30))
+
+    worklnbuffer_dfL = geopandas.GeoDataFrame.copy((workln_dfL))
+    worklnbuffer_dfR = geopandas.GeoDataFrame.copy((workln_dfR))
+
+    # buffer the parallel line in one side (extend the area into forest)
+
+    worklnbuffer_dfL['geometry'] = shapely.buffer(workln_dfL['geometry'], distance=float(forest_buffer_dist),
+                                                  cap_style=2, join_style=2, single_sided=True)
+    worklnbuffer_dfR['geometry'] = shapely.buffer(workln_dfR['geometry'], distance=-float(forest_buffer_dist),
+                                                  cap_style=2, join_style=2, single_sided=True)
+    print("Creating offset area for surrounding forest....Done")
+    print('%{}'.format(50))
+    # create a New column for surrounding forest statistics:
+    # 1) Height Percentile (add more in the future)
+    # worklnbuffer_dfL['**_L'] = numpy.nan
+    # worklnbuffer_dfR['**_R'] = numpy.nan
+    # worklnbuffer_dfL = worklnbuffer_dfL.reset_index(drop=True)
+    # worklnbuffer_dfR=worklnbuffer_dfR.reset_index(drop=True)
+    # line_seg['L_**'] = numpy.nan
+    # line_seg['R_**'] = numpy.nan
+    print('%{}'.format(80))
+
+    print("Calculating CL metrics ..")
+    worklnbuffer_dfC, new_col_c = multiprocessing_metrics(worklnbuffer_dfC, in_raster, raster_type, processes,
+                                                          side='center')
+
+    worklnbuffer_dfC = worklnbuffer_dfC.sort_values(by=['OLnFID'])
+    worklnbuffer_dfC = worklnbuffer_dfC.reset_index(drop=True)
+    print("Calculating surrounding forest metrics....")
+    # calculate the Height percentile for each parallel area using CHM
+    worklnbuffer_dfL, new_col_l = multiprocessing_metrics(worklnbuffer_dfL, in_raster, raster_type, processes,
+                                                          side='left')
+
+    worklnbuffer_dfL = worklnbuffer_dfL.sort_values(by=['OLnFID'])
+    worklnbuffer_dfL = worklnbuffer_dfL.reset_index(drop=True)
+
+    worklnbuffer_dfR, new_col_r = multiprocessing_metrics(worklnbuffer_dfR, in_raster, raster_type, processes,
+                                                          side='right')
+
+    worklnbuffer_dfR = worklnbuffer_dfR.sort_values(by=['OLnFID'])
+    worklnbuffer_dfR = worklnbuffer_dfR.reset_index(drop=True)
+
+    print('%{}'.format(90))
+
+    all_new_col = numpy.append(numpy.array(new_col_c), numpy.array(new_col_l))
+    all_new_col = numpy.append(all_new_col, numpy.array(new_col_r))
+
+    for index in (line_seg.index):
+        if raster_type == 'DEM':
+            for col in all_new_col:
+                if "C_" in col:
+                    line_seg.loc[index, col] = worklnbuffer_dfC.loc[index, col]
+                elif "L_" in col:
+                    line_seg.loc[index, col] = worklnbuffer_dfL.loc[index, col]
+                elif "R_" in col:
+                    line_seg.loc[index, col] = worklnbuffer_dfR.loc[index, col]
+    print("Calculating forest metrics....Done")
+
+    print("Saving forest metrics output.....")
+    geopandas.GeoDataFrame.to_file(line_seg, out_line)
+    print("Saving forest metrics output.....Done")
+    del line_seg, worklnbuffer_dfL, worklnbuffer_dfR, workln_dfL, workln_dfR
+
+    print('%{}'.format(100))
+
+
+# TODO: inspect duplicates, split_line_npart in line_footprint_functions.py
+def split_line_nPart(line):
+    from shapely.ops import split, snap
+    # Work out n parts for each line (divided by 10m)
+    n = math.ceil(line.length / 10)
+    if n > 1:
+        # divided line into n-1 equal parts;
+        distances = numpy.linspace(0, line.length, n)
+        points = [line.interpolate(distance) for distance in distances]
+
+        split_points = shapely.multipoints(points)
+        # mline=cut_line_at_points(line,points)
+        mline = split(line, split_points)
+        # mline=split_line_fc(line)
+    else:
+        mline = line
+    return mline
+
+
+def split_into_Equal_Nth_segments(df):
+    odf = df
+    crs = odf.crs
+    if not 'OLnSEG' in odf.columns.array:
+        df['OLnSEG'] = numpy.nan
+    df = odf.assign(geometry=odf.apply(lambda x: split_line_nPart(x.geometry), axis=1))
+    df = df.explode()
+
+    df['OLnSEG'] = df.groupby('OLnFID').cumcount()
+    gdf = geopandas.GeoDataFrame(df, geometry=df.geometry, crs=crs)
+    gdf = gdf.sort_values(by=['OLnFID', 'OLnSEG'])
+    gdf = gdf.reset_index(drop=True)
+    return gdf
+
+
+def split_line_fc(line):
+    return list(map(shapely.LineString, zip(line.coords[:-1], line.coords[1:])))
+
+
+def split_into_every_segments(df):
+    odf = df
+    crs = odf.crs
+    if not 'OLnSEG' in odf.columns.array:
+        df['OLnSEG'] = numpy.nan
+    else:
+        pass
+    df = odf.assign(geometry=odf.apply(lambda x: split_line_fc(x.geometry), axis=1))
+    df = df.explode()
+
+    df['OLnSEG'] = df.groupby('OLnFID').cumcount()
+    gdf = geopandas.GeoDataFrame(df, geometry=df.geometry, crs=crs)
+    gdf = gdf.sort_values(by=['OLnFID', 'OLnSEG'])
+    gdf = gdf.reset_index(drop=True)
+    return gdf
+
+
+def multiprocessing_copyparallel_lineLR(dfL, dfR, processes, left_dis, right_dist):
+    try:
+
+        line_arg = []
+        total_steps = len(dfL)
+
+        for item in dfL.index:
+            item_list = [dfL, dfR, left_dis, right_dist, item]
+            line_arg.append(item_list)
+
+        featuresL = []
+        featuresR = []
+        chunksize = math.ceil(total_steps / processes)
+        with Pool(processes=int(processes)) as pool:
+            step = 0
+            # execute tasks in order, process results out of order
+            for resultL, resultR in pool.imap_unordered(copyparallel_lineLR, line_arg, chunksize=chunksize):
+                if BT_DEBUGGING:
+                    print('Got result: {}{}'.format(resultL, resultR), flush=True)
+                featuresL.append(resultL)
+                featuresR.append(resultR)
+                step += 1
+                print('%{}'.format(step / total_steps * 100))
+            return geopandas.GeoDataFrame(pandas.concat(featuresL)), geopandas.GeoDataFrame(pandas.concat(featuresR))
+    except OperationCancelledException:
+        print("Operation cancelled")
+
+
+def multiprocessing_metrics(df, in_raster, raster_type, processes, side):
+    try:
+        line_arg = []
+        total_steps = len(df)
+        if side == 'left':
+            PerCol = 'L'
+        elif side == 'right':
+            PerCol = 'R'
+        else:
+            PerCol = 'C'
+
+        for item in df.index:
+            item_list = [df.iloc[[item]], in_raster, item, PerCol, raster_type]
+            line_arg.append(item_list)
+        features = []
+        chunksize = math.ceil(total_steps / processes)
+
+        if USE_MULTI_PROCESSING:
+            with Pool(processes=int(processes)) as pool:
+
+                step = 0
+                # execute tasks in order, process results out of order
+                try:
+                    for result in pool.imap_unordered(cal_metrics, line_arg, chunksize=chunksize):
+                        if BT_DEBUGGING:
+                            print('Got result: {}'.format(result), flush=True)
+                        features.append(result)
+                        step += 1
+                        print('%{}'.format(step / total_steps * 100))
+                except Exception:
+                    print(Exception)
+                    raise
+                del line_arg
+                df = geopandas.GeoDataFrame(pandas.concat(features))
+                new_col = []
+                for col in df.columns.array:
+                    if "C_" in col:
+                        new_col.append(col)
+                    elif "L_" in col:
+                        new_col.append(col)
+                    elif "R_" in col:
+                        new_col.append(col)
+
+                return df, new_col
+        else:
+            for row in line_arg:
+                features.append(cal_metrics(row))
+            df = geopandas.GeoDataFrame(pandas.concat(features))
+            new_col = []
+            for col in df.columns.array:
+                if "C_" in col:
+                    new_col.append(col)
+                elif "L_" in col:
+                    new_col.append(col)
+                elif "R_" in col:
+                    new_col.append(col)
+
+            return df, new_col
+
+    except OperationCancelledException:
+        print("Operation cancelled")
+
+
+def cal_metrics(line_arg):
+    try:
+        df = line_arg[0]
+        raster = line_arg[1]
+        row_index = line_arg[2]
+        PerCol = line_arg[3]
+        raster_type = line_arg[4]
+        line_buffer = df.loc[row_index, 'geometry']
+    except:
+        print("Assigning variable on index:{} Error: ".format(line_arg) + sys.exc_info())
+        exit()
+    try:
+        with rasterio.open(raster) as image:
+
+            clipped_raster, out_transform = rasterio.mask.mask(image, [line_buffer], crop=True,
+                                                               nodata=BT_NODATA, filled=True)
+            clipped_raster = numpy.squeeze(clipped_raster, axis=0)
+            cell_x, cell_y = image.res
+            cell_area = cell_x * cell_y
+            # mask all -9999 (nodata) value cells
+            masked_raster = numpy.ma.masked_where(clipped_raster == BT_NODATA, clipped_raster)
+            filled_raster = numpy.ma.filled(masked_raster, numpy.nan)
+
+            # Calculate the metrics
+            if raster_type == "DEM":
+                # Surface area ratio (SAR)
+
+                SAR, total_surface_area = cal_index(filled_raster, cell_x, cell_y, 'SAR')
+                SAR_mean = numpy.nanmean(SAR)
+                SAR_percentile90 = numpy.nanpercentile(SAR, 90, method='hazen')
+                SAR_median = numpy.nanmedian(SAR)
+                SAR_std = numpy.nanstd(SAR)
+                SAR_max = numpy.nanmax(SAR)
+                SAR_min = numpy.nanmin(SAR)
+
+                # General Statistics
+                total_planar_area = numpy.ma.count(masked_raster) * cell_area
+                cell_volume = masked_raster * cell_area
+                total_volume = numpy.ma.sum(cell_volume)
+                mean = numpy.nanmean(filled_raster)
+                percentile90 = numpy.nanpercentile(filled_raster, 90, method='hazen')
+                median = numpy.nanmedian(filled_raster)
+                std = numpy.nanstd(filled_raster)
+                max = numpy.nanmax(filled_raster)
+                min = numpy.nanmin(filled_raster)
+
+                # Terrain Ruggedness Index (TRI)
+                # TRI = cal_index(filled_raster, cell_x, cell_y, 'TRI')
+                # TRI_mean = numpy.nanmean(TRI)
+                # TRI_percentile90 = numpy.nanpercentile(TRI, 90, method='hazen')
+                # TRI_median = numpy.nanmedian(TRI)
+                # TRI_std = numpy.nanstd(TRI)
+                # TRI_max = numpy.nanmax(TRI)
+                # TRI_min = numpy.nanmin(TRI)
+
+            del clipped_raster, out_transform
+
+    # return the generated value
+    except:
+        print(sys.exc_info())
+    try:
+
+        # Writing General statisic
+        df.at[row_index, PerCol + '_PlArea'] = total_planar_area
+        df.at[row_index, PerCol + '_mean'] = mean
+        # df.at[row_index, PerCol + '_P90'] = percentile90
+        # df.at[row_index, PerCol + '_median'] = median
+        df.at[row_index, PerCol + '_StdDev'] = std
+        df.at[row_index, PerCol + '_max'] = max
+        df.at[row_index, PerCol + '_min'] = min
+        # df.at[row_index, PerCol + '_Vol'] = total_volume
+
+        # Writing SAR statisic
+        df.at[row_index, PerCol + '_SurArea'] = total_surface_area
+        df.at[row_index, PerCol + '_SARmean'] = SAR_mean
+        # df.at[row_index, PerCol + '_SARP90'] = SAR_percentile90
+        # df.at[row_index, PerCol + '_SARmed'] = SAR_median
+        df.at[row_index, PerCol + '_SARStd'] = SAR_std
+        df.at[row_index, PerCol + '_SARmax'] = SAR_max
+        df.at[row_index, PerCol + '_SARmin'] = SAR_min
+
+        # highest=df[PerCol + '_max'].max()
+        # lowest = df[PerCol + '_min'].min()
+
+        # Writing TRI statisic
+        # df.at[row_index, PerCol + '_TRImean'] = TRI_mean
+        # # df.at[row_index, PerCol + '_TRIP90'] = TRI_percentile90
+        # # df.at[row_index, PerCol + '_TRImed'] = TRI_median
+        # df.at[row_index, PerCol + '_TRIStd'] = TRI_std
+        # df.at[row_index, PerCol + '_TRImax'] = TRI_max
+        # df.at[row_index, PerCol + '_TRImin'] = TRI_min
+
+        return df
+    except:
+        print("Error writing forest metrics into table: " + sys.exc_info())
+
+
+def copyparallel_lineLR(line_arg):
+    dfL = line_arg[0]
+    dfR = line_arg[1]
+    # Simplify input center lines
+    lineL = dfL.loc[line_arg[4], 'geometry'].simplify(tolerance=0.05, preserve_topology=True)
+    lineR = dfL.loc[line_arg[4], 'geometry'].simplify(tolerance=0.05, preserve_topology=True)
+    offset_distL = float(line_arg[2])
+    offset_distR = float(line_arg[3])
+
+    # Older alternative method to the offset_curve() method,
+    # but uses resolution instead of quad_segs and a side keyword (‘left’ or ‘right’) instead
+    # of sign of the distance. This method is kept for backwards compatibility for now,
+    # but it is recommended to use offset_curve() instead.
+    # (ref: https://shapely.readthedocs.io/en/stable/manual.html#object.offset_curve)
+
+    # parallel_lineL = shapely.offset_curve(geometry=lineL, distance=offset_distL,
+    #                                       join_style=shapely.BufferJoinStyle.mitre)
+    # parallel_lineR = shapely.offset_curve(geometry=lineR, distance=offset_distR,
+    #                                       join_style=shapely.BufferJoinStyle.mitre)
+
+    parallel_lineL = lineL.parallel_offset(distance=offset_distL, side='left',
+                                           join_style=shapely.BufferJoinStyle.mitre)
+    parallel_lineR = lineR.parallel_offset(distance=-offset_distR, side='right',
+                                           join_style=shapely.BufferJoinStyle.mitre)
+
+    if not parallel_lineL.is_empty:
+        dfL.loc[line_arg[4], 'geometry'] = parallel_lineL
+    if not parallel_lineR.is_empty:
+        dfR.loc[line_arg[4], 'geometry'] = parallel_lineR
+    return dfL.iloc[[line_arg[4]]], dfR.iloc[[line_arg[4]]]
+
+
+if __name__ == '__main__':
+    start_time = time.time()
+    print('Starting forest metrics calculation processing\n @ {}'.format(
+        time.strftime("%a, %d %b %Y %H:%M:%S", time.localtime())))
+
+    print("Checking input parameters....")
+    in_args, in_verbose = check_arguments()
+
+    verbose = True if in_args.verbose == 'True' else False
+    forest_metrics(print, **in_args.input, processes=int(in_args.processes), verbose=verbose)
+
+    print('%{}'.format(100))
+    print('Starting forest metrics calculation processing is done @ {}\n(or in {} second)'.format(
+        time.strftime("%a, %d %b %Y %H:%M:%S", time.localtime()), round(time.time() - start_time, 5)))