Parameter optimization

In this chapter we will apply some strategies to optimize parameters of image processing workflows. In general, it is important to have high-quality ground truth data, such as manually segmented images. Furthermore, a well-engineered fitness function (sometimes also called loss-function) is necessary. For parameter optimization we will use scipy’s optimize package and the Napari plugin napari-workflow-optimizer.