Efficient predictor of subjective video quality rating measures

• US 6,690,839 B1
• Filed: 01/17/2000
• Issued: 02/10/2004
• Est. Priority Date: 01/17/2000
• Status: Expired due to Term

• Current Assignee: Project Giants, LLC
• Original Assignee: Tektronix Incorporated (Fortive Corp.)
• Inventors: Ferguson, Kevin M.
• Primary Examiner(s): MARIAM, DANIEL G
• Application Number:
  US09/483,834
• Time in Patent Office: 1,485 Days
• Field of Search: 382/112, 382/149, 382/162, 382/167, 382/190, 382/209, 382/210, 382/212, 382/213, 382/218, 382/219, 382/220, 382/232, 382/238, 382/239, 382/251, 382/260, 382/269, 382/274, 382/275, 348/189, 348/192, 348/188, 375/240.03, 375/240.21, 375/240.26, 714/712
• US Class Current: 382/274
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: H04N 17/00 Diagnosis, testing or measuring for television systems or their details

Abstract

An efficient predictor for subjective video quality rating measures processes a reference video and an impaired video corresponding to the reference video in respective processing channels. Each channel applies a contrast sensitivity function from a human vision system model to the respective videos and a contrast discrimination function for self-masking to the output of the contrast sensitivity function. The outputs from the channels are input to a differencing circuit to obtain absolute difference values. The absolute difference values are modified by a global masking factor derived from the output from the reference video channel, and the resulting values are converted into units conforming to a standard quality rating metric.

Claims

1. An apparatus for efficient prediction of subjective video quality rating measures comprising:

a pair of processing channels, a first channel for processing a reference video and a second channel for processing an impaired video corresponding to the reference video, the reference and impaired videos being processed directly in the respective channels according to a human vision system model without pyramid decomposition to produce a processed reference video and a processed impaired video; and

means for obtaining an absolute difference between the processed reference video and the processed impaired video as a quality rating measure.

2. The apparatus as recited in claim 1 further comprising means for applying a global masking factor as a function of the processed reference video to the absolute difference, the output of the applying means being the quality rating measure.

3. The apparatus as recited in claims 1 or 2 further comprising means for converting the quality rating measure to units of a standard quality rating measure.

4. The apparatus as recited in claim 1 wherein each channel comprises:

a two-dimensional filter for obtaining a spatial frequency response for an input video, the two-dimensional filter emulating the human vision system model; and

a contrast discrimination stage having the spatial frequency response as an input to account for the effects of self-masking, the output from the contrast discrimination stage being a processed video for input to the obtaining means.