A new perceptual model for video sequence encoding

摘要

A method of encoding monochrome video sequences is presented. Thismethod is applicable to slow-moving scenes with a zoom factor close tounity. It incorporates a perceptual model for the detection and trackingof image blocks perceived to be in apparent motion. The perceptual modelis a threshold model based on contrast. Weber fractions in thespatiotemporal domain are used to set the threshold for the detection ofmotion. The threshold is determined on a scene-adaptive basis. Theinformation content of the scene is derived from the spectraldistribution of image data and the masking characteristics of the humanvisual system. The perceptual model is used in the block-classificationprocess, which in turn determines the coding strategy used for eachblock type. The discrete cosine transform is used for encoding imageblocks which have changed beyond a perceptual threshold. A comparison ofthe perceptually encoded image sequence with that using no perceptualmodel indicates substantially improved quality and performance of theperceptual encoder for compression ratios up to 100:1