autothresholdr 1.3.6

BUG FIXES

Cope with ijtiff not working on 32-bit windows.

autothresholdr 1.3.5

BUG FIXES

• Insist on bug-fixed ijtiff v2.0.2.

autothresholdr 1.3.4

BUG FIXES

• Insist on bug-fixed filesstrings v3.1.5.

autothresholdr 1.3.3

BUG FIXES

• Insist on updated, less buggy filesstrings and ijtiff.

autothresholdr 1.3.2

BUG FIXES

• Remove superfluous stringr from Imports.

autothresholdr 1.3.1

BUG FIXES

• Require necessary version of glue.
• Require debugged filesstrings 3.1.1.

autothresholdr 1.3.0

NEW FEATURES

• Better vignettes.
• A pkgdown website.

MINOR IMPROVEMENTS

• Better error messages.

BUG FIXES

• Insist on latest versions of ijtiff and filesstrings (which have important bug fixes).

autothresholdr 1.2.1

BUG FIXES

• For new versions of filesstrings and ijtiff.
• Update to checkmate broke the package so this is the fix.

autothresholdr 1.2.0

MINOR IMPROVEMENTS

• The package is now lighter by depending on filesstrings instead of RSAGA.

autothresholdr 1.1.2

BUG FIXES

• Ensure using latest version of ijtiff. Old version is buggy.

autothresholdr 1.1.1

BUG FIXES

• Fix some poor C++ typing.

autothresholdr 1.1.0

NEW FEATURES

• The package now has it’s own S3 class system:
  • Class th for thresholds.
  • Class threshed_arr for thresholded arrays.
  • Class stack_threshed_arr for stack-thresholded arrays.
  • Class arr_mask for a mask of an array.

MINOR IMPROVEMENTS

• The package no longer depends on EBImage package from Bioconductor. EBImage is a great package but CRAN packages are easier to install if they depend on CRAN packages only. Now ijtiff is used instead.

DEPRECATED

• can_be_integer() is gone. Use checkmate::check_integerish() instead.
• Pillar statistics are no longer exported. Should you be looking for them, they are exported in the detrendr package.

autothresholdr 1.0.0

NEW FEATURES

• The rJava (and hence java) dependency has been removed. All functions which used to be in java are now done in Rcpp.

MINOR IMPROVEMENTS

• The fail arguments of auto_thresh_apply_mask(), mean_stack_thresh() and med_stack_thresh() now work like the na argument of nandb::WriteIntImage().
• auto_thresh_mask() and auto_thresh_apply_mask() now have the abbreviations mask() and apply_mask().
• The functions now have an ignore_na argument for NA handling.

BUG FIXES

• It is now the case that for a threshold x, values greater than or equal to x are deemed to “pass” the thresholding and values less than x are deemed to “fail” the thresholding. Before what was happening was that values greater than x were deemed to have passed and values less than or equal to x were deemed to have failed.

autothresholdr 0.6.0

NEW FEATURES

• There’s a new thresholding method "Huang2" which is very similar to (but not the same as and hence should not be assumed to give the same results as) "Huang" and is much faster when applied to 16-bit images.
• mean_stack_thresh() can now handle thresholds between 0 and 1.

MINOR IMPROVEMENTS

• Use ignore_black and ignore_white instead of ignore.black and ignore.white to comply with tidyverse style guide.
• Add ignore_black and ignore_white options to mean_stack_thresh() and med_stack_thresh().

BUG FIXES

• Fix issues for Mean and Otsu methods concerning 16-bit images.

autothresholdr 0.5.0

MINOR IMPROVEMENTS

• Renamed all exported functions to be in snake_case.
• The skip.consts option in the stack_thresh functions is gone. Now these functions error if you pass them a constant array.

autothresholdr 0.4.0

MINOR IMPROVEMENTS

• Add MeanStackThresh() and MedStackThresh().

autothresholdr 0.3.0

MINOR IMPROVEMENTS

• Add option to manually set threshold.
• Improve OSX installation instructions in README.md.

autothresholdr 0.2.0

• The first CRAN-worthy version.