The study of the Extragalactic Background Light ( EBL ) is crucial to understand many astrophysical problems ( as the formation of first stars , the evolution of galaxies and the role of dust emission ) .
At present , one of the most powerful ways to put constraints on EBL is represented by the study of the photon-photon absorption on gamma-ray spectra of TeV blazars .
Adopting this method , we found that , if the only contribution to the optical and Near Infrared ( NIR ) background is given by galaxies , the spectrum of the blazar H 1426+428 can not be fitted .
To reproduce the observational data of H 1426+428 a Near Infrared excess with respect to galaxy counts is required , with amplitude consistent with both the Matsumoto et al .
( 2000 ) data with Kelsall ’ s model of zodiacal light ( ZL ) subtraction and the DIRBE data with Wright ’ s model of ZL subtraction .
The derived constraints on the optical EBL are weaker , because the experimental errors on blazar data are still bigger than the differences among various optical EBL models .
In the mid-infrared the SPITZER measurement of \nu { } I _ { \nu { } } =2.7 nW m ^ { -2 } sr ^ { -1 } at 24 \mu { } m , provides the best fit of the blazar spectrum .