Abstract
The Gaussian-random-sphere model is employed for morphological characterization of nonspherical, irregular particles using an inverse light scattering technique. The synthetic measurement data consist of reduced scattering spectra caused by an aggregate of irregular particles randomly oriented in turbid media and are generated using the discrete dipole approximation. The proposed method simultaneously retrieves the concentration and shape parameters of particles using the data collected at multiple wavelengths. The performance of the inverse algorithm is tested using noise-corrupted data, in which up to 50% noise may be added to the observed scattering spectra.