chore: Described edge detection and median subtract

docs/commands/command_edge_detection.md

@@ -0,0 +1,29 @@
+
+(command-edge-detection)=
+# Edge detection
+
+:::{sidebar} Edge detection
+
+Edge detection is the process of finding edges in an image and converting them to a gradient representation.
+Sharper changes in the intensity values lead to a higher gradient.
+
+```{figure} images/edge_detection.drawio.svg
+:class: full-image
+```
+
+:::
+
+```{figure} images/edge_detection_screenshot.png
+:class: tiny-image
+```
+
+Edge detection is a function used to highlight the edges within an image.
+An edge is defined as a change in grey level.
+Larger rates of change in the values lead to a larger output value in the algorithm.
+
+ImageC provides {guilabel}`Sobel` and {guilabel}`Canny` edge detection.
+
+
+The main advantages of the Sobel operator are that it is simple and time efficient but produces rough edges.
+On the other hand, the Canny technique produces smoother edges due to the implementation of non-maxima suppression and thresholding. 
+The disadvantage of the Canny algorithm is that it is more complex and less time efficient than Sobel.

docs/commands/command_image_math.md

@@ -0,0 +1,18 @@
+
+(command-image-math)=
+# Image math
+
+:::{sidebar} Edge detection
+
+Edge detection is the process of finding edges in an image and converting them to a gradient representation.
+Sharper changes in the intensity values lead to a higher gradient.
+
+```{figure} images/edge_detection.drawio.svg
+:class: full-image
+```
+
+:::
+
+```{figure} images/image_math_screenshot.png
+:class: tiny-image
+```

docs/commands/command_median_subtract.md

@@ -0,0 +1,33 @@
+
+(command-median-subtract)=
+# Median subtract
+
+:::{sidebar} Rank filter
+
+Rank-order filters belong to the class of non-linear filters in digital image processing. 
+These are filters that cannot be described by a convolution.
+
+With ranking filters, the gray values of the pixels in a defined area around a pixel are collected, sorted by size and ranked. 
+A gray value is then selected from this sorted list to replace the gray value of the current pixel.
+
+The choice of position determines the type of ranking filter. With ascending sorting, you get the:
+
+- Minimum filter, for the minimum gray value, first position in the list
+- Median filter, for the gray value in the middle of the list
+- Maximum filter, for the maximum gray value, last position in the list.
+
+:::
+
+```{figure} images/median_subtract_screenshot.png
+:class: tiny-image
+```
+
+Median subtraction is a sort of noise filtering which can be used to reduce background noise similar to the [rolling ball]() algorithm.
+Behind the scenes a rank filter is used to calculate the median of the image intensity, which is subtracted afterwards from the original image.
+
+The kernel size defines the window size used for the rank filter to calculate the median.
+Bigger kernels reduces more details from the image than smaller ones.
+
+:::{hint}
+Bigger filter kernels removes more noise and details from the image.
+:::

docs/commands/index.md

@@ -19,7 +19,8 @@ command_color_filter
 command_rolling_ball
 command_blur
 command_intensity
-
+command_median_subtract
+command_edge_detection
 :::
 
 (commands-object-segmentation)=