Hanyieh,

- I recommend you the paper "Cross-Comparison of Vegetation Indices Derived from Landsat-7 Enhanced Thematic Mapper Plus (ETM+) and Landsat-8 operational Land Imager (OLI) Sensors " by Li, Jiang and Feng (Remote Sensing, Vol 6, 2014, p. 310-329, at www.mdpi.com/journal/remotesensing).

- In addition, you may look to a quite similar RG question addressed on Dec 5, 2013 by Vassilis Trigkas together with the comments.

Thanks dear Olena, I did not know that.

Hanyieh,

- I recommend you the paper "Cross-Comparison of Vegetation Indices Derived from Landsat-7 Enhanced Thematic Mapper Plus (ETM+) and Landsat-8 operational Land Imager (OLI) Sensors " by Li, Jiang and Feng (Remote Sensing, Vol 6, 2014, p. 310-329, at www.mdpi.com/journal/remotesensing).

- In addition, you may look to a quite similar RG question addressed on Dec 5, 2013 by Vassilis Trigkas together with the comments.

You may try/compare a different approach to change detection. Have a look at my papers on forest cover change in Italy and Bolivia.

Thanks Haniyeh for asking this useful question and thanks others like Olena, Ruxandra and Hein for replying to this question. There is another index which is developed by my supervisor based on Pattern Decomposition Method (PDM). This index is considered as sensor independent as compared to other contemporary indices. It also incorporates Spectral Response Function (SRF) for any sensor. The paper is under process. As soon as it is accepted, I will post its idl and matlab code for the services of our scientific community.

Some good suggestions have been given by Olena, plus what you have done was a good start.  One possible issue that could explain some of what you are seeing with Landsat 8 vs 5 and 7 is that the radiometric resolution (DN dynamic range) is 11-bits vs 8-bits.  Make sure that the images are not only radiometrically corrected, but also transformed into the same dynamic range scale (e.g., if the forested vegetation have a spectral surface reflectance of say 6% that this is represented by the same DN value in the corrected form of ALL the images).

Pat

Hi pat, do you have a real example of DN correction between different images? thank you a lot in advance

Giovanni,

It has been awhile since I did some multi-temporal image calibration and normalization so I do not have actual images and DN values.  I have included a list of papers and a presentation / slides that are under my contributions section here on Research Gate that may be of interest to you and related to this topic.  I am relatively new to research gate and not sure if I did the links correctly; if not, you should be able to find them under my contributions list.  Look them over and let me know if you have any specific questions.

1.    Presentation - Radiometric Calibration and Atmospheric Corrections

2.    Image-Based Atmospheric Corrections - Revisited and Improved

3.    Radiometric Calibration of Landsat Thematic Mapper Multispectral Images

4.    An improved dark-object subtraction technique for atmospheric scattering correction of multispectral data

5.    Automatic detection of vegetation changes in the southwestern United States using remotely sensed images

6.    Comparison of the Spectral Information Content of Landsat Thematic Mapper and SPOT for Three Different Sites in the Phoenix, Arizona Region

Pat 

 Thanks a lot Pat, I'll see your pubblications very soon ;-). Thanks again