An image processing apparatus includes a setting unit, a generating unit, and a converting unit. The setting unit sets a gloss level of each of plural areas into which an image is divided. The image is formed by forming plural toner images having different colors using plural first toners and at least one second toner having at least one same color as the first toners and by fixing the plural toner images onto a recording medium so as to be superimposed on one another. The generating unit generates multi-level image data for each of the first toners and the at least one second toner in accordance with the gloss levels set by the setting unit. The converting unit converts multi-level image data of toners having the same color into binary image data. The converting unit also converts multi-level image data of toners having different colors into binary image data.