Abstract
This paper proposes a tone mapping operation (TMO) with a low memory cost and a low computational cost. A TMO generates a low dynamic range (LDR) image from a high dynamic range (HDR) image by compressing its dynamic range. Due to an HDR image is generally expressed in floating-point data, a TMO is also executed with floating-point arithmetic. The proposed method treats a floating-point number as two 8-bit integer numbers which correspond to a exponent part and a mantissa part, and applies tone mapping to these integer numbers separately. Moreover, the method conducts calculations in the tone mapping with fixed-point arithmetic. As a result, the method reduces a memory cost and a computational cost. The experimental and evaluation results show that the method has high accuracy of a TMO, though reducing memory and computational cost.
