Combustion is often used in the industry to produce required energy. Due to the increasing need of minimizing energy loss or pollution of combustors, keeping good performance of combustion control is highly desired. To achieve tight combustion control is not straightforward, because most of the flames encountered are turbulent flames. A novel method is proposed in this study to improve the control performance of product quality by applying the digital flame color images to control loops. It will lead to a substantial reduction in oxygen quality variability. With the minimal oxygen quality variability, the requirement of the excess combustion air in a furnace is minimized and it will have less loss of heat. Based on the novel structure, the designable performance bound and the corresponding optimal parameters of the controller structure computed from the closed loop operating data are proposed. The performance bound can assess the performance of the current given controller. To demonstrate the advantages of the proposed method, data from a real combustion system are presented to delve into the matter.