作者: Jorge Caviedes , Yingwei Chen , Deepak S. Turaga
DOI:
关键词: Objective quality 、 Quantization (signal processing) 、 Discrete cosine transform 、 Trellis quantization 、 Computer science 、 Computer vision 、 Source data 、 No reference 、 Peak signal-to-noise ratio 、 Rate–distortion optimization 、 Artificial intelligence
摘要: The present invention relates to a method and system for estimating the quality of encoded video data without gaining access source data. is configured decode compressed using an MPEG/H.263 decoder produce decompressed decoded subject discrete cosine transform (DCT) set DCT coefficients determined be intra-coded. Meanwhile, quantization matrix including quantizier step size each block are extracted. Following this, statistical properties extracted from video. Then, average error both AC DC information regarding properties. Finally, this estimated used in computing peak signal noise ratio (PSNR).