Merge pull request #48 from lauracmurphy/patch-1

Fix typo in README description

Author: petebankhead

README.md

@@ -4,6 +4,6 @@
 [![License: CC BY 4.0](https://img.shields.io/badge/License-CC_BY_4.0-blue.svg)](https://creativecommons.org/licenses/by/4.0/)
 [![Twitter](https://img.shields.io/twitter/follow/petebankhead?style=flat)](https://twitter.com/petebankhead)
 
-This book tries explain the main ideas of bioimage analysis in a practical and engaging way.
+This book tries to explain the main ideas of bioimage analysis in a practical and engaging way.
 
 You're currently looking at the ReadMe - for the actual book, see https://bioimagebook.github.io

chapters/1-concepts/3-bit_depths/imagej.md

@@ -242,7 +242,7 @@ However, if we first subtract the smallest of our 16-bit values (i.e. 4000), our
 Make sure that the {guilabel}`Scale When Converting` option is turned on (it should be by default).
 Then using a suitable 8-bit sample image, e.g. {menuselection}`File --> Open Samples --> Boats`, explore the effects of brightness/contrast settings when increasing or decreasing bit-depths.
 
-Can you destroy the image by simply 1) increasing the bit-depth, and the then 2) decreasing the bit-depth to its original value?
+Can you destroy the image by simply 1) increasing the bit-depth, and then 2) decreasing the bit-depth to its original value?
 
 [![launch ImageJ.JS](https://ij.imjoy.io/assets/badge/open-in-imagej-js-badge.svg)](https://ij.imjoy.io/?run=https://gist.github.com/petebankhead/6f9f451fdc0116197501ae504a57d5e7)
 ```

chapters/2-processing/3-thresholding/thresholding.md

@@ -1007,7 +1007,7 @@ The top half of {numref}`fig-thresholds_noisy` reproduces the nuclei from {numre
 Although the nuclei are still clearly visible in the image (A), the two classes of pixels (which were previously easy to separate) have now been merged together in the histogram (B).
 The triangle threshold method, which had performed well before, now gives less attractive results \(C), because the noise has caused the ranges of background and nuclei pixels to overlap.
 
-*However,* if we apply a Gaussian filter to smooth the image, a lot of the the random noise is reduced (see {ref}`chap_filters`).
+*However,* if we apply a Gaussian filter to smooth the image, a lot of the random noise is reduced (see {ref}`chap_filters`).
 This results in a histogram dramatically more similar to that in the original, (almost) noise-free image, and the threshold is again quite successful (F).
 
 ```{code-cell} ipython3

chapters/2-processing/7-multidimensional_processing/multidimensional_processing.md

@@ -420,7 +420,7 @@ Thinning algorithms are often designed to work in 3D, although not usually highe
 The distance and watershed transforms extend readily to 3D, but require a little caution.
 
 One thing to look out for, especially with the distance transform, is whether pixel anisotropy is taken into consideration.
-If not, the the distance transform will not be capable of properly identifying the 'nearest' foreground or background pixel in calibrated units, but rather only in pixel units.
+If not, then the distance transform will not be capable of properly identifying the 'nearest' foreground or background pixel in calibrated units, but rather only in pixel units.
 
 A cumbersome workaround may be to resize the image so that the pixels *are* isotropic, but that may make every other analysis step more complicated and/or require a huge amount more memory to store the image.
 A preferable approach is to try to find a distance transform implementation that incorporates pixel size information into its calculations.