|
From: | Hartmut |
Subject: | [Octave-patch-tracker] [patch #9235] image package: improve filtering functions (stdfilt, entropyfilt) and add tests (also medfilt2, ordfilt2, rangefilt) |
Date: | Mon, 5 Nov 2018 17:08:35 -0500 (EST) |
User-agent: | Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:63.0) Gecko/20100101 Firefox/63.0 |
Follow-up Comment #2, patch #9235 (project octave): Here is a new patch file (V2) with * Matlab compatible and passing tests for STDFILT, now avoiding truncated floating point numbers * Matlab compatible and passing tests for RANGEFILT, also without trunctated floats now * Matlab compatible but mostly failing tests for ENTROPYFILT. These tests still have truncated floats in them, because I don't have a clue yet how Matlab gets those observed results. In my opinion those additional tests could now be pushed, and this issue then be closed. Should we open a new bug report for the mostly wrong results of entropyfilt? This bug might also concern entropy.m because I couldn't reproduce the Matlab results using this function either. (file #45363) _______________________________________________________ Additional Item Attachment: File name: new_tests_for_filtering_functions_V2.patch Size:4 KB _______________________________________________________ Reply to this item at: <https://savannah.gnu.org/patch/?9235> _______________________________________________ Message sent via Savannah https://savannah.gnu.org/
[Prev in Thread] | Current Thread | [Next in Thread] |