Laser-induced incandescence (LII) is a powerful and robust optical method for in situ determination of soot volume fraction and/or soot absorption/emission properties in flames and engine exhaust. The laser-induced signal is interpreted as thermal emission based on the Planck law. Up to now, the evaluation and interpretation of LII signal have been largely based on contributions from isolated primary particles that are assumed much smaller than wavelengths. In the present paper, the morphology, wavelength, and aggregate size-dependent effects of multiple scattering within fractal soot aggregates on their absorption and emission cross sections are taken into account in the evaluation of LII signal by proposing correction terms to the traditional model. The impact of accounting for the correction to soot aggregate emission due to multiple scattering on LII signal and on the two excitation wavelength-induced incandescence method for inferring the soot absorption function, E(m), is discussed. For wavelengths shorter than 532 nm, E(m, lambda)/E(m, 1064 nm) increases more significantly with decreasing wavelength. For wavelengths longer than 532 nm, the wavelength dependence of E(m, lambda)/E(m, 1064 nm) becomes very small and can be neglected. The proposed corrections, along with the soot morphology, are applied to re-analyze the experimental data of Bejaoui et al. (Appl Phys B Lasers Opt, 116:313, 2014) for deriving the relative soot absorption function variation with wavelength at different locations in a rich premixed methane flat flame at atmospheric pressure. The present analysis showed that the soot absorption function varies with the height above the burner exit and can be correlated with the degree of soot maturation.