You are on page 1of 13

IMAGE ENHANCEMENT TECHNIQUES FOR BACKGROUND SUBTRACTION IN INDOOR SEQUENCE WITH ILLUMINATION CHANGES: SURVEY

GUIDE: SATHISH Assistant Professor

BY JAGANATH S.T Mtech, CSE, SBMJCE

BACKGROUND SUBTRACTION: Background Subtraction is the process of separating the foreground object from the background in a sequence of video frames. It is used in many video applications, such as Video Surveillance Traffic monitoring gesture recognition Video compression

Image Enhancement:

Methods:

Spatial domain methods Frequency domain methods

Image enhancement techniques:


Adaptive Histogram Equalization Adaptive Contrast Enhancement Homomorphic Filtering.

Adaptive Histogram Equalization: AHE attempts to improve the contrast of an input image by stretching the peaks of the histogram and compressing the troughs.

Adaptive Contrast Enhancement: It is enhancement method used to increase the contrast of an image. Generally it applies the method to increase the contrast enhancement by making light color lighter and dark color darker at the same time. Fast and adaptive method for Adaptive Contrast Enhancement. Computationally less intensive and Marked improvement in contrast.

Homomorphic Filtering:

This method is used to normalize the brightness and increase the contrast of an image. It is a filter which controls both high frequency and low frequency components . Homomorphic filtering aims at handling large of image intensity. An image f(x, y) can be expressed as the product of its illumination i(x, y) and reflectance r(x, y) components. f(x, y)= i(x, y) r(x, y)

Illuminated image results :

Method/ Images
11.jpg Adaptive histogram equalization Adaptive contrast enhancement Homomorphic Filtering 12.jpg

PSNR Value
13.jpg 14.jpg 15.jpg

9.6107

12.5567

11.9415

13.5292

12.0630

18.7958

15.5132

19.3325

15.0023

19.8824

25.1621

27.8889

27.6503

27.4099

21.1460

PSNR : It measure the effects of image enhancement algorithms on image quality. To compute PSNR:

Block computes the PSNR:

Conclusion : All 3 algorithms are tested on some natural images. The result analysis says that homomorphic filtering is best suitable algorithm for Background Subtraction in indoor sequence with illumination variations.

You might also like