gvSIG bugs #3096

Problems using Unsupervised Classification

Added by Mario Carrera over 9 years ago. Updated almost 8 years ago.

Status:New% Done:

0%

Priority:NormalSpent time:-
Assignee:-
Category:Geoprocess
Target version:-
Severity:Minor Add-on version:
gvSIG version:2.1.0 Add-on build:
gvSIG build:2259 Add-on resolve version:
Operative System: Add-on resolve build:
Keywords: Proyecto:
Has patch: Hito:
Add-on name:Unknown

Description

There have been these problems using the Unsupervised Classification on gvSIG 2.1 on a Landsat image:

- The results are moved from the original location, they are not in the correct place.
- The final area is not complete (it contains NoData values at right side and at the bottom of the image).
- The top of the image hasn't been rasterized.

It worked good in gvSIG 1.12.

I attach the log file.

gvSIG-clasificacion-no-supervisada.log (365 KB) Mario Carrera, 01/07/2015 01:24 PM

Model_2.jpg - Input image (83 KB) Mario Carrera, 01/08/2015 01:16 PM

Model_3.jpg - Output image (96.6 KB) Mario Carrera, 01/08/2015 01:16 PM

L7_before_UC.jpg (290 KB) Antonio Falciano, 01/10/2015 12:05 PM

L7_after_UC.jpg (267 KB) Antonio Falciano, 01/10/2015 12:05 PM

1228
1229
1232
1233

Related issues

Related to Application: gvSIG desktop - gvSIG bugs #3097: Wrong output region if "Use extent from view" Closed 01/08/2015
Related to Application: gvSIG desktop - gvSIG bugs #1987: Rasterize result shifted Won't fix 08/07/2013

History

#1 Updated by Álvaro Anguix over 9 years ago

  • Category changed from Raster to Geoprocess

#2 Updated by Antonio Falciano over 9 years ago

I can reproduce the issue: the output raster is shifted of one pixel in SW direction and its spatial extent is not complete (see also #3097).

#3 Updated by Antonio Falciano over 9 years ago

See also #1987.

#4 Updated by Álvaro Anguix over 9 years ago

  • Related to gvSIG bugs #3097: Wrong output region if "Use extent from view" added

#5 Updated by Álvaro Anguix over 9 years ago

#6 Updated by Mario Carrera over 9 years ago

I attach the input (Model_2.jpg) and output (Model_3.jpg) images to see the results.

#7 Updated by Antonio Falciano over 9 years ago

Hi Mario,
maybe the shift is too wide because of #2343. Have you tried the same task without maximizing the view? What's happened in this case?

#8 Updated by Mario Carrera over 9 years ago

Hi Antonio,
I'm going to ask the user who tested it and I tell you about the results.

#9 Updated by Antonio Falciano over 9 years ago

Hi Mario,
I've tried to replicate the issue(s) under very similar conditions ;) with the view maximized (see screenshots in attachment), "Use the view extent" in the Analysis region and the same cell size of the input layers. It seems that it doesn't depend by #2343. There's a less consistent shift in the NS direction in my case and the output extent is cut symmetrically along the X axis. A further issue that affects the "Unsupervised classification" is #3098, because the second output is a table.

#10 Updated by Álvaro Anguix about 9 years ago

  • Target version set to 2.2.0-2311-rc2
  • Assignee set to Francisco Díaz Carsí

#11 Updated by Álvaro Anguix about 9 years ago

  • Target version deleted (2.2.0-2311-rc2)

#12 Updated by Joaquín del Cerro Murciano almost 8 years ago

  • Assignee deleted (Francisco Díaz Carsí)

Also available in: Atom PDF