The concept of homomorphic filtration

Question

I am trying to understand the concept of homomorphic filtering, I have read several online pages on this topic, such as http://en.wikipedia.org/wiki/Homomorphic_filtering However, I have noticed the scarce online resources that discuss this topic. Therefore, I have two questions.

Firstly, I wonder how it can be decided that an image can be improved using homomorphic filtering? In other words, what is the type of image that requires homomorphic filtering?

Secondly, is homomorphic filtering always used to attenuate illumination (low frequency) and increase reflection coefficient (high frequency)? Or can it also be used for the opposite (weakening reflection and illumination illumination)?

Answer 1

After further research, I found the answer,

1- Images that can be enhanced by homomorphic filtering are images in which the illumination is not evenly distributed, resulting in objects in the image appearing in dark color. In general, these images are dark, so their details are hidden.

2- According to this source, Homomorphic filtering is the process of filtering by the frequency domain, which compresses the brightness (from the lighting condition), while increasing the contrast (from the reflective properties of the object).

Answer 2

Given the fact that homomorphic filtering can be used to separate low-frequency backlight and high-frequency reflection from the image, its obvious use is to detect changes. When changes are detected, when we compare two images, we look for changes in the high-frequency reflection component. If we filter out part of the low-frequency backlight, this helps.

After homomorphic filtering, the bright input image on the left gives the value on the right.

After homomorphic filtering, the dark input image on the left gives the one on the right.

Here is the matlab link

Here are a few other stackoverflow links to questions about homomorphic filtering , more