Method and systems for selecting test stimuli for use in evaluating performance of video watermarking methods

A method of selecting test stimuli for use in evaluating performance of a watermarking algorithm, the method comprising the steps of:

• calculating an energy distribution rate (EDR) for each of a plurality of test stimulus candidates;
• classifying the plurality of test stimulus candidates into M levels of the test stimuli, based on the calculated EDR values,
• where M is an integer greater than or equal to 2;
• and
• selecting at least one test stimulus from each of the M levels of the test stimuli, wherein said step of calculating the energy distribution rate (EDR) for each of the plurality of test stimulus candidates includes the steps of:
• applying DCT transformation to each 2D image(s) of each of the test stimulus candidates;
• reordering AC coefficients of the DCT transformed image(s) to generate 1D image data;
• calculating and normalizing energy distribution (E) of the AC coefficients from the generated 1D image data using the following Equation:
• E(s)=T(s)2, sε[0,N]
• where s is the position in the 1-D distribution, N is the number of AC coefficients, and T is the reordered AC coefficients and E is the energy distribution;
• and
• calculating the EDR based on the normalized energy distribution (E) using the following Equation:
• EDR
• =
• 1
• N
• limit
• ⁢
• ∑
• i
• =
• 1
• N
• limit
• ⁢
• ·
• i
• N
• ·
• log
• ⁡
• (
• E
• ⁡
• (
• i
• )
• )
• where Nlimit is the number of AC coefficients except 30% of high frequency.

A method of selecting test stimuli for use in evaluating performance of a video watermarking algorithm, the method comprising:

• calculating a motion rate (MR) for each of a plurality of test stimulus candidates;
• classifying the plurality of test stimulus candidates into N levels of the test stimuli, based on the calculated MR values,
• where N is an integer greater than or equal to 2;
• and
• selecting at least one test stimulus from each of the N levels of the test stimuli, wherein the MR is calculated using the following Equation:
• M
• =
• 1
• NF
• ⁢
• ∑
• i
• =
• 1
• NF
• ⁢
• ∑
• j
• =
• 0
• TMV
• i
• ⁢
• 
• v
• i
• ,
• j
• 
• where NF is the number of frames of stimulus and TMVi is the number of motion vectors in the ith frame and vi,j is the jth motion vector in the ith frame.

A method of selecting test stimuli for use in evaluating performance of a watermarking algorithm, the method comprising the steps of:

• calculating an image complexity rate (ICR) for each of a plurality of test stimulus candidates and classifying them into a plurality of L levels based on the calculated ICR values,
• where L in an integer greater than 2 and
• wherein each of the plurality of L levels has a predetermined ICR range,
• wherein each of the plurality of L levels has a predetermined ICR range, and
• wherein the ICR is calculated using the following Equation:
• U
• =
• -
• 1
• *
• -
• 1
• w
• ×
• h
• ⁢
• ∑
• x
• =
• 1
• w
• ⁢
• ∑
• y
• =
• 1
• h
• ⁢
• [
• f
• ⁡
• (
• x
• ,
• y
• )
• -
• f
• _
• ⁡
• (
• x
• ,
• y
• )
• ]
• 2
• where f(x,y) is the intensity-level at pixel (x, y) and f(x,y) is the average gray-level in a 3×3 window centered at (x, y) and w and h are the width and height of the image;
• and
• selecting at least one test stimulus from each of the L levels of the test stimuli;
• calculating an energy distribution rate (EDR) for each of the plurality of test stimulus candidates, and classifying them into M (where M is an integer more than or equal to 2) levels based on the calculated EDR values;
• calculating a motion rate (MR) for each of the plurality of test stimulus candidates and classifying them into N (where N is an integer more than or equal to 2) levels of the calculated MR values;
• and
• selecting at least one test stimulus from each of the L levels, each of the M levels, each of the N levels of the test stimuli, respectively.

A system for selecting test stimuli for use in evaluating performance of a video watermarking algorithm, the system comprising:

• first classification unit for calculating an image complexity rate (ICR) for each of a plurality of test stimulus candidates and classifying them into a plurality of L levels based on the calculated ICR values,
• wherein L is an integer greater than 2, and
• wherein each of the plurality of L levels has a predetermined ICR range, and
• wherein the ICR is calculated using the following Equation:
• U
• =
• -
• 1
• *
• 1
• w
• ×
• h
• ⁢
• ∑
• x
• =
• 1
• w
• ⁢
• ∑
• y
• =
• 1
• h
• ⁢
• [
• f
• ⁡
• (
• x
• ,
• y
• )
• -
• f
• _
• ⁡
• (
• x
• ,
• y
• )
• ]
• 2
• where f(x, y) is the intensity-level at pixel (x, y) and f(x, y) is the average gray-level in a 3×3 window centered at (x, y) and w and h are the width and height of the image;
• second classification unit for calculating an energy distribution rate (EDR) for each of the plurality of test stimulus candidates, and classifying them into M levels based on the calculated EDR values,
• where M is an integer greater than or equal to 2;
• third classification unit for calculating a motion rate (MR) for each of the plurality of test stimulus candidates and classifying them into N levels based on the calculated MR values,
• where N is an integer greater than or equal to 2;
• and
• selection unit for selecting at least one test stimulus from each of the L levels, each of the M levels, each of the N levels of the test stimuli, respectively.