A simple gray-edge automatic white balance method with FPGA implementation

Automatic white balance is one of the most important functions in digital cameras. In addition to effectively correct color bias, the automatic white balance technology should also be fit for resource-constrained hardware and meet its real-time requirements. Based on gray-edge hypothesis, this paper proposes a simple automatic white balance method using image horizontal down-sampling with the averaging filter and horizontal first-order difference on Bayer image, and discusses its algorithm flow on FPGA. The test results show that the proposed method can correct image color bias powerfully. Moreover, the analysis of resource usage on FPGA indicates that the method consumes less hardware resource and achieves high real-time capability, and its parameter selection is unrelated with resource consumption.