Then you can estimate Point Spread Function of an Image.
Do you just have the blurred image, or do you also have sharp version of Image? If you only have blurred Image then it is difficult to find PSF because you don’t have enough information.
But Blind Deconvolution method can be used if Initial estimate value of Point Spread Function is known. deconvblind(I,psfi) returns  both the deblurred image and a restored PSF. Or Bayesian Blind deconvolution method can also be used.

Hey Aufa, Our webinars are completely free and you can register from this page itself by clicking on “Click here to reserve your spot” and filling your details. Happy learning!

Thank you for the response. May I know where and when is your webinar? How about the cost? Does it free?

Depending on the type on noise present in the medical image a suitable filter must be used. Wiener filter is among the popularly used filters, if you want to increase the complexity of the filter you can try adaptive median filter, wavelet transforms, adaptive Gaussian filter or homomorphic wiener filter according to the requirements. As for enhancement I would suggest contrast enhancement instead of histogram equalization if it’s a medical image like a MIR scan. Join our webinar to get a clearer view on the algorithms mentioned.