作者: Chunhua Chen , Jeffrey A. Bloom
DOI: 10.1007/978-3-642-15702-8_11
关键词: Block (data storage) 、 Pixel 、 Video processing 、 Block size 、 Artificial intelligence 、 Measure (mathematics) 、 Discrete Fourier transform 、 Absolute difference 、 Robustness (computer science) 、 Computer vision 、 Mathematics
摘要: Some image and video processing algorithms can have the unintended consequence of introducing blocking artifacts into processed imagery. Measuring blockiness plays an important role in many applications. This paper presents a reference-free measurement method. For given image, absolute difference between horizontally adjacent pixels is computed, normalized, averaged along each column. A one-dimensional discrete Fourier transform thereafter employed vertical measure derived. horizontal computed similarly. Finally, for formulated by pooling those two directional measures. The proposed method accurately assess without any priori knowledge block origin size; therefore it blind measure. Experimental results show effectiveness robustness also justified.