Get you quickly to producing simple algorithms, without getting bogged down by more information than you need

Give you a broad, working mental model of image analysis algorithms, to facilitate further learning of more detailed information

Experiment, come up with your own solutions, learn by trial-and-error.

Basic thresholding

Recognizing when noise reduction would be helpful

Segmentation cleanup

Simple measurements

Planning desired selections

Comparing types of noise reduction

Considerations when choosing how to clean a segmentation

Noticing variation in your image

Dealing with variation in your image

•During pre-processing

•During segmentation

Measurement interpretation: trusting numbers in isolation

Recognizing features in an image that can be targeted with filters

Compensating for filter side-effects

Check your measurements

Proper image illumination

Being aware of filter side-effects

Combining multiple segmentations

Using segmentations as masks for subsequent processing

Identify features that might confound your segmentation

Ways to handle these features, depending on how they interfere

•Grayscale pre-processing vs. segmentation mathematics

2-stage segmentation: What defines a feature vs. what differentiates a feature from others

Identify errors in your segmentation and whether those errors need to be addressed

Masks and Regions of Interest

Centers vs edges

Be OK with having errors in your first pass at segmentation.

Understand the strengths of different filters and segmentation methods, so you can apply them where appropriate

•Also, being willing to experiment

How to double-check your work

Memory management

How morphological operations will affect different parts of your segmentation

Continued two-stage segmentation / define vs. distinguish

•Returning to your original segmentation to reduce side-effects

Segmentation mathematics to combine segmentations and achieve optimal results

Measuring one segmentation relative to another

Understand needed segmentation accuracy

Consider potential variation between your images

Deciding the best way to measure a feature