Processed Image Comparator

For processed image data, you can select metrics such as Mean Squared Error (MSE), Normalized Root Mean Square Error (NRMSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM), to evaluate the quality of images generated by denoising deep models or that were processed by filtering or any other operation.

To open the Processed Image Comparator, shown below, select the reference image and processed image data in the Data Properties and Settings panel, right-click, and then choose Open Processed Image Comparator in the pop-up menu. You can then select a mask and the required metrics to evaluate your data.

Processed Image Comparator dialog

Processed Image Comparator

Available metrics
  Description
MSE Mean squared error (MSE)… Represents the second moment of the error, which is the mean value of the square of the error and the expected loss of the square of the signal at each pixel position. The closer MSE is to 0, the less noise in the image.
NRMSE Normalized root mean square error (NRMSE)… Is the normalized square root of the Mean Square Error.
PSNR Peak signal-to-noise ratio (PSNR)… Is the ratio between the maximum power that can be present in a signal and the power of noise affecting the signal. Similar to SNR, the greater the value of PSNR, the higher the quality of the processed image.
SSIM Structural similarity index (SSIM)… Shows how much the processed image matches the reference or ground-truth image. In the calculation of SSIM, three factors are included to quantify the similarity between two images — the similarity of brightness, contrast, and structure. The value of SSIM is between 0 and 1. The closer the value of SSIM is to 1, the greater the similarity between the two images.
The images that you choose to compare must have the same size and shape as the selected ground truth.
How to Evaluate Processed Images

  1. Do one of the following:

    • Select the ground truth and processed image data in the Data Properties and Settings panel, right-click, and then choose Open Processed Image Comparator in the pop-up menu

    • Choose Utilities >Structured Grids Comparator on the menu bar to open the Structured Grids Comparator dialog.

      You can then choose a reference image and the processed data you to evaluate in the drop-down menus.

  2. Select an ROI mask, optional.

  3. Select the required metric(s) in the Metrics drop-down menu.

  4. Click Compare.

    You should note that results are cumulative and can be exported in the comma-separated values (*.csv extension) file format for further analysis or archiving.

  5. Select another dataset to compare, optional.