SegmentMeanShift function requires Output Raster Dataset?

JosephFogarty · ‎07-11-2019

Hi all,

I am fairly new to using ArcGIS Pro, but I am slightly experienced in Python in general. I'm not sure if this is the exact place to post this question, but I am stuck. I have successfully used ArcPy to clip a raster from a shapefile, and I now want to perform the SegmentMeanShift() function on said clipped raster.

This process works in ArcGIS Pro, and I would like to have a script to automate this process for many different rasters. Below is the snippet of my code that gives me issues (most of it was copied from the documentation here)

# script to clip and segment a raster

# import libraries
import arcpy
import matplotlib.pyplot as plt
import numpy as np
import os
from arcpy.sa import *

#%% Clip raster
in_ras = r"data\s20180403.tif"
clip_ras = in_ras[:5] + r"clip_ras\\" + in_ras[5:-4] + r"_clip.tif"
feature_template = r"shapefiles\square_domain_0723.shp"

arcpy.Clip_management(
        in_ras, "#", clip_ras, feature_template, "0", "ClippingGeometry")

#%% Segment raster
#check for spatial analyst license
arcpy.CheckOutExtension("Spatial")

# segment mean shift the clipped raster                 
spec_detail = "10"
spat_detail = "10"
min_seg_size = "20"
band_indexes = "1"
seg_clip_ras = r"data\\seg_clip_ras\\" + "s20180403" + "_clip_seg.tif"

seg_ras = SegmentMeanShift(in_ras, spec_detail,
                           spat_detail, min_seg_size, band_indexes)
seg_ras.save(seg_clip_ras)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

The output I receive from running this chunk of the script is:

Traceback (most recent call last):

  File "<ipython-input-33-232592bd359c>", line 11, in <module>
    spat_detail, min_seg_size, band_indexes)

  File "D:\ArcGISPro\Resources\ArcPy\arcpy\sa\Functions.py", line 7768, in SegmentMeanShift
    band_indexes)

  File "D:\ArcGISPro\Resources\ArcPy\arcpy\sa\Utils.py", line 53, in swapper
    result = wrapper(*args, **kwargs)

  File "D:\ArcGISPro\Resources\ArcPy\arcpy\sa\Functions.py", line 7761, in Wrapper
    band_indexes)

  File "D:\ArcGISPro\Resources\ArcPy\arcpy\geoprocessing\_base.py", line 506, in <lambda>
    return lambda *args: val(*gp_fixargs(args, True))

ExecuteError: Failed to execute. Parameters are not valid.
ERROR 000735: Output Raster Dataset: Value is required
Failed to execute (SegmentMeanShift).‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

I am unsure what is happening here, when I try to add another argument, the error says that SegmentMeanShift() only takes 5 arguments, not 6, which makes sense and is what I thought originally. So why is a value required for Output Raster Dataset?

Basically, when this script is finished, I would like a raster saved in my directory that is clipped and segmented from the first raster. I have looked over the documentation and other questions and I cannot seem to find this specific issue.

Thanks for help in advance, I have been stuck on this for awhile now.

DanPatterson_Retired · ‎07-17-2019

seg_clip_ras = r"data\\seg_clip_ras\\" + "s20180403" + "_clip_seg.tif"

your path... what does 'data' refer to? or are you just anonymizing your actual path? (ie you have no drive specification, if you are working locally)

JosephFogarty · ‎07-17-2019

Hi Dan,

Thanks for the reply. 'data' simple refers to a folder named data. I have created an environment after I imported the libraries in the code above, something along the lines of:

arcpy.env.workspace = "D:\my\project\path"‍‍

but I have omitted this line for anonymity. Yes, I am working locally on my computer. I think the path works fine, because the program is able to clip the raster, but not segment it.