19614cc8d7
euler number now behaves like MATLAB implementation
2012-05-22 23:28:13 +02:00
4eb3028a64
fixed typo and improved example in doc string
2012-05-22 23:28:13 +02:00
1b2bed7501
renamed regionprops source files for better separation between function name and source file
2012-05-22 23:28:12 +02:00
e993124329
most functionality of regionprops now pure python except for moments
2012-05-22 23:28:12 +02:00
0027b91e84
added test: labelled image must be of integer type
2012-05-22 23:28:12 +02:00
6b0c0e1948
changed reference notes in doc string
2012-05-22 23:28:12 +02:00
658dbec381
removed old description of moments in doc string
2012-05-22 23:28:12 +02:00
e845dc7746
moments and central moments no longer use separate functions
2012-05-22 23:28:12 +02:00
cb673ba516
extended and improved description of properties in doc string
2012-05-22 23:28:12 +02:00
6905603b98
adapted to PEP8 whitespace conventions
2012-05-22 23:28:12 +02:00
8c1acfb7f5
added regionprops function
2012-05-22 23:28:12 +02:00
7e51e519b1
Fix whitespace
2012-05-22 12:21:58 -07:00
85670d60b6
Update docstring
2012-05-22 12:13:51 -07:00
8e67dd901f
Fix spelling
2012-05-22 11:49:17 -07:00
754f0e5d39
Docstring additions
2012-05-22 11:39:38 -07:00
64da74c91a
Fix grammar
2012-05-22 11:12:47 -07:00
4d999c313f
feature.hog(): use y,x coords consistently, now works on non-square images
2012-05-22 17:35:11 +10:00
fc42d86894
Adjust rescale_intensity function to new dtype conversion rules
2012-05-22 00:28:32 -07:00
e4b49239c4
Adjust tests for new type conversion rules
2012-05-22 00:25:42 -07:00
3ff1ebf526
Rewrite convert function
2012-05-22 00:21:32 -07:00
8f43ee7739
BUG: Catch all exceptions, including NotImplementedError, when checking for number of CPUs in multi-processing.
2012-05-21 19:20:20 -07:00
34505fef46
BUG: Fix FreeImage Py3k compatibility [patch by Almar Klein].
2012-05-21 18:24:32 -07:00
4e87a1dfce
Add new 64-bit metadata types to METADATA_DATATYPE (new in FreeImage 3.15.3)
2012-05-19 13:23:32 -07:00
048c6c06eb
BUG: Fix stackcopy for grayscale image.
2012-05-15 15:56:26 -04:00
fc676d6ae0
BUG: Correctly import _feature in Python 3.
2012-05-09 15:42:04 -07:00
7326b1949f
BUG: fix test images (float images between 0 and 1).
2012-05-09 18:35:38 -04:00
d19d45850f
BUG: Fix internal import in template matching.
2012-05-09 12:53:29 -07:00
06bd3ebd6f
Add doctest example to match_template.
2012-05-08 21:32:09 -04:00
a2f91585ef
Add test for pad_input = True.
2012-05-08 21:32:08 -04:00
193d5abd3c
Rename pad_output parameter to pad_input.
2012-05-08 21:32:08 -04:00
383ca6220a
Set match_template to default to no padding and fix test.
2012-05-08 21:32:08 -04:00
ad99285bc0
Prevent match_template from returning NaNs.
2012-05-08 21:32:08 -04:00
0c1e3541b3
Check that image is larger than template.
2012-05-08 21:32:08 -04:00
b2006da346
Change flipud/fliplr to array indexing.
2012-05-08 21:32:08 -04:00
43ed4b9fb7
Pad input image instead of output.
2012-05-08 21:32:08 -04:00
5d79537566
Fix convolution input to get correlation results.
2012-05-08 21:32:07 -04:00
7caf85a5c1
Renaming for clarity.
2012-05-08 21:32:07 -04:00
f21f032bfe
Add normalization test and remove unnecessary clipping.
2012-05-08 21:32:07 -04:00
5682d27eb0
Rewrite normalization algorithm.
...
This is a major revision that removes the `method` parameter of `match_template` and uses a new normalization method.
Note that the example result is different with this new normalization.
2012-05-08 21:32:07 -04:00
753e999a7a
Minor cleanup
...
* Rename some variables.
* Delete some old comments.
* Move some variable initializations to the top of the function.
2012-05-08 21:32:07 -04:00
883dd24cdb
Fix bug when indexing into summed-area table.
2012-05-08 21:32:07 -04:00
2e8fcef89b
Simply equation in docstring of match_docstring.
2012-05-08 21:32:06 -04:00
cbdea0d36e
Add pad_output argmument to match_template.
2012-05-08 21:32:06 -04:00
3c7ae849b9
Remove unnecessary truncation at boundary.
2012-05-08 21:32:06 -04:00
545bdab985
Refactor template matching.
...
* Change Cython function to take names of correlation method instead of numbers representing the methods.
* Use alternate formula for `template_norm` of 'norm-corr' method, but note that both formulas need to be checked for correctness.
* Add note that `match_template` output has a different shape than the input image. This needs to be fixed before merging.
* Change 'Sigma' to 'Sum' in docstring to avoid confusion with standard deviation.
* Other minor changes for readability.
2012-05-08 21:32:06 -04:00
8c816109c9
Remove OpenCV version of match_template
2012-05-08 21:32:06 -04:00
425a4ea707
Use feature.peak_local_max instead of custom peak detection.
2012-05-08 21:32:06 -04:00
9d3072d3f8
Fix type errors in match_template by casting inputs to float32.
2012-05-08 21:32:06 -04:00
2a00e24b12
Change _template.match_template variables to float.
...
Template and image in test are converted to float32 before passing to `match_template`. This change is temporary: `match_template` should convert these variables internally.
2012-05-08 21:32:06 -04:00
e8461e22dd
Move template matching to feature subpackage
2012-05-08 21:32:05 -04:00
Johannes Schönberger