In the last years, the use of multispectral cameras have been increasing due to their ability to analyze different characteristics of materials. Thus, multiples methods to develop cameras with multispectral filters arrays have been studied. These cameras must always have the option to combine the multispectral information to create a regular color image. In this work, we present a method to estimate the regular colors from a novel prototype of a multispectral camera that been developed using nanowires sensors. Real information from nine frequencies are analyzed using classification and regression methods.