计算机科学
人工智能
图像(数学)
滤波器(信号处理)
计算机视觉
绘图
算法
模式识别(心理学)
计算机图形学(图像)
作者
Changmeng Peng,Maohan Xia,Zhizhong Fu,Jin Xu,Xiaofeng Li
标识
DOI:10.1109/lsp.2021.3099962
摘要
Bit-depth enhancement is of significant research interest, and its main challenge lies in eliminating false contours in low-bit images. Existing algorithms either fail to effectively suppress false contours or rely on a vast number of samples and high-performance graphics cards, resulting in poor practicality. To solve these problems, we propose a bilateral false contour elimination filter-based bit-depth enhancement algorithm: BEF-BDE, which contains three innovative modules: B ilateral false contour E limination F ilter (BEF), iterative false contour elimination, and content-adaptive iteration number selection. Experiments on three datasets show that our method can achieve superior subjective and objective results.
科研通智能强力驱动
Strongly Powered by AbleSci AI