Image Quality Metrics

You can evaluate and compare the image quality of previews and other single-slice datasets with a number of different metrics, such as Sharpness, Shannon Entropy, SNR, and FactorQ, in the Image Quality Metrics dialog.

Right-click the data that you want to include in your evaluation and then choose Open Image Quality Metrics in the pop-up menu to open the dialog. You can then choose the objects to include in the comparison, apply a mask to limit computations, and select the metrics for comparison. You can also export the results in the comma-separated values file (*.csv extension) format.

Image Quality Metrics dialog

Available metrics

	Description
Sharpness	Is a measure of how accurately a sample is represented and is related to the edge contrast of an image. Note This metric is calculated using the function numpy.gradient (https://numpy.org/doc/stable/reference/generated/numpy.gradient.html).
Shannon Entropy	Is a measure of the uncertainty in predicting voxel values. Decreasing values would mean less uncertainty and better image quality. Note This metric is calculated using the function skimage.measure.shannon_entropy (https://scikit-image.org/docs/0.13.x/api/skimage.measure.html#skimage.measure.shannon_entropy).
SNR	Is the signal-to-noise ratio, which is defined as the ratio of average and standard deviation of a signal or measurement. The higher the ratio, the better the image quality. Note This definition is only useful for variables that are always non-negative.
FactorQ	Is a measure of the degree of separation of two material classes in the analyzed image and is calculated on the base of a grey value histogram. Note FactorQ is only meaningful for datasets with two classes.