Spectrum reconstruction method based on human vision screening optical filter

The invention discloses a spectrum reconstruction method for screening optical filters based on a human vision system and application, and the method comprises the following steps: weighting original optical filters by using an LMS cone response function, and selecting three optical filters which are most matched with the LMS cone response function of the human vision system as initial optical filters by using a weighted area screening method; after the initial optical filters are determined, the rest optical filters and the three initial optical filters are subjected to exhaustive combination one by one; substituting the generated optical filter group into a spectrum reconstruction model to obtain spectrum recovery and chromaticity errors of each group of optical filters; and sorting according to the self-defined recovery error set, selecting an optimal optical filter group under the L weighting, the M weighting and the S weighting, and selecting an optimal group through ranking of self-defined error scores.