diff --git a/episodes/07-thresholding.md b/episodes/07-thresholding.md
index 43b79e84..bd3b0341 100644
--- a/episodes/07-thresholding.md
+++ b/episodes/07-thresholding.md
@@ -26,7 +26,8 @@ exercises: 50
 In this episode, we will learn how to use scikit-image functions to apply
 thresholding to an image.
 Thresholding is a type of *image segmentation*,
-where we change the pixels of an image to make the image easier to analyze.
+which is the partitioning of an image into disjoint regions, or *segments*.
+The different segments can then be analyzed separately.
 In thresholding, we convert an image from colour or grayscale into
 a *binary image*, i.e., one that is simply black and white.
 Most frequently,