Demura Compensation Algorithm Based on Strong and Weak Filtering

ABSTRACT AMOLED displays are widely adopted in a broad range of electronic products due to their superior physical properties. However, instabilities in the manufacturing process often result in luminance nonuniformities, commonly known as Mura. As an effective approach to enhancing display performance, Demura compensation algorithms are typically integrated into the driver IC to eliminate such defects. In this work, a boundary‐aware strong/weak filtering framework operating in the luminance error domain is proposed for Demura compensation. The proposed method performs multistep processing on the acquired luminance error data, evaluates local luminance discontinuities according to Mura characteristics and the target correction luminance, and selectively applies strong or weak filtering to optimize the spatial distribution of luminance errors while preserving intrinsic luminance structures. Experimental results demonstrate that the proposed method significantly improves luminance uniformity: For the green channel at a gray level of 32, the uniformity of brightness (UOB) reduces from 0.0818 to 0.0516, corresponding to a 37% improvement; at gray levels of 64 and 192, the UOB decreases by 35% (from 0.0556 to 0.0359) and 24% (from 0.0275 to 0.0210), respectively. Visual comparisons before and after Demura compensation further confirm that the proposed strong/weak filtering–based Demura algorithm effectively suppresses luminance nonuniformity.