This paper aims at providing aperture corrections for emission lines in a sample of spiral galaxies from the Calar Alto Legacy Integral Field Area Survey (CALIFA) database. In particular, we explore the behavior of the log([OIII]5007/Hbeta)/([NII]6583/Halpha) (O3N2) and log[NII]6583/Halpha (N2) flux ratios since they are closely connected to different empirical calibrations of the oxygen abundances in star forming galaxies. We compute median growth curves of Halpha, Halpha/Hbeta, O3N2 and N2 up to 2.5R_50 and 1.5 disk R_eff. The growth curves simulate the effect of observing galaxies through apertures of varying radii. The median growth curve of the Halpha/Hbeta ratio monotonically decreases from the center towards larger radii, showing for small apertures a maximum value of ~10% larger than the integrated one. The median growth curve of N2 shows a similar behavior, decreasing from the center towards larger radii. No strong dependence is seen with the inclination, morphological type and stellar mass for these growth curves. Finally, the median growth curve of O3N2 increases monotonically with radius. However, at small radii it shows systematically higher values for galaxies of earlier morphological types and for high stellar mass galaxies. Applying our aperture corrections to a sample of galaxies from the SDSS survey at 0.02<=z<=0.3 shows that the average difference between fiber-based and aperture corrected oxygen abundances, for different galaxy stellar mass and redshift ranges, reaches typically to ~11%, depending on the abundance calibration used. This average difference is found to be systematically biased, though still within the typical uncertainties of oxygen abundances derived from empirical calibrations. Caution must be exercised when using observations of galaxies for small radii (e.g. below 0.5R_eff) given the high dispersion shown around the median growth curves.