Calculation of indicators PSNR, SSIM and RMSE is different from other papers.

[xwt11]

Hello, I use the code of this project, and the calculated PSNR and SSIM are smaller than the results of other papers. The value of RMSE is large.

After denormalize_ function, the pixel value range of the image changes to [-1024,3072], but trunc function is used to directly change the pixel value greater than 240 to 240, and the pixel value less than -160 to -160.
Why the change? Why is the pixel value negative?

def denormalize_(self, image):
    image = image * (self.norm_range_max - self.norm_range_min) + self.norm_range_min
    return image


def trunc(self, mat):
    mat[mat <= self.trunc_min] = self.trunc_min
    mat[mat >= self.trunc_max] = self.trunc_max
    return mat

[Nioolek]

我猜测，由于测试的患者L506的CT图是腹部图像，而腹部图像在CT中成像HU值都比较接近，一般都会使用-160~240这样的窗宽窗位，所以这里就按照这样来设置了

[eveW1nd]

My PSNR and SSIM are also much smaller than the results of other papers. Even the RMSE between original noised CT and the clean one is larger than yours in papers. I m sure I used same dataset and test set.Have you already done with this issue?

[eveW1nd]

My PSNR and SSIM are also much smaller than the results of other papers. Even the RMSE between original noised CT and the clean one is larger than yours in papers. I m sure I used same dataset and test set.Have you already done with this issue?

I've figured this out after my testing. It is reasonable to adjust the range of image truncation to obtain different PSNR and SSIM values, which is reasonable because truncation will lead to different degrees of image information loss. On this basis, it can be explained that the same test set gets different RMSE.