Table 2 Summary of testing the method over a synthetic dataset.

From: Tomographic reconstruction from planar thermal imaging using convolutional neural network

Metrics

Mean ± std. dev.

Median (25–75 percentile)

NMSE

0.036 ± 0.049

0.021

(0.013–0.035)

CC

0.981 ± 0.026

0.989

(0.981–0.993)

PSNR

43.135 ± 3.386

43.239

(40.856–45.556)

SSIM

0.748 ± 0.114

0.790

(0.697–0.826)

HSLE

1.398 ± 1.146

0.967

(0.689–1.592)