In this paper, we have a tendency to propose a generalized leveling model for image improvement. supported our analysis on the relationships between image bar chart and distinction improvement white reconciliation, we have a tendency to initial establish a generalized leveling model desegregation distinction improvement and white reconciliation into a unified framework of protrusive programming of image bar chart. We have a tendency to show that several image improvement tasks may be accomplished by the projected model exploitation totally different configurations of parameters. With 2 shaping properties of bar chart remodel, specifically distinction gain and nonlinearity, the model parameters for various improvement applications may be optimized. We have a tendency to then derive associate degree best image improvement formula that in theory achieves the most effective joint distinction improvement and white reconciliation result with trading-off between distinction improvement and tonal distortion.