hello all, i need your help..what will i do. for automatic done adjustment of brightness, object detect, deblurred image.

if i have one image, which is blurred, which has faces, which have low brightness, so these 3 all done automatic when image is selected. how automatic it detecet that their is face, so extrace face in new image, automatic adjust the brightness in image. if image is blurred then automatic deblurr the image. what will be code for this. please help me to solve out with this problem. how deep learning and machine learning will apply for automatic. and what will be code for deep learning and machine learing

[KALYAN ACHARJYA]

That means you are not sure about input images. These all parameters measurement based on human perception.

I read somewhere, you cannot measure the brightness in numeric data, you can say high or low or mid? What I have undestand is-

If brightness is less than certain threshold,

then only adjust the brightness,

else

no operation....

end

Are you looking for this one?

[Anshika Jain]

yes sir, i want this. but i have one doubt the brightness minimum value is what and the maximum value will be what. or for blurr image what will do to clear. like given below is input image.

Answer 1

[Image Analyst]

Use fft2() and sum up the energy in high spatial frequencies. Or use stdfilt() and get the mean intensity of the local standard deviation image. If they're low, then there are few details and the image may be blurred.

[KALYAN ACHARJYA]

Sir About Brightness? Can it measurable?

[Image Analyst]

Yes. What is your definition of brightness? The mean gray level in RGB color space? The mean L value in the LAB color space version or the mean V in the HSV color space version?