Motion detection using field-difference measurements

• US 6,104,755 A
• Filed: 09/12/1997
• Issued: 08/15/2000
• Est. Priority Date: 09/13/1996
• Status: Expired due to Term

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Ohara, Kazuhiro
• Primary Examiner(s): Kelley, Chris S.
• Assistant Examiner(s): PHILIPPE, GIMS S
• Application Number:
  US08/927,990
• Time in Patent Office: 1,068 Days
• Field of Search: 348/416, 348/452, 348/451, 348/448, 348/606, 348/607, 348/620, 348/459, 348/415, 348/446, 348/445, 348/450, 348/625, 348/627, 375/240
• US Class Current: 375/240
• CPC Class Codes: H04N 19/105 Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
• CPC Class Codes: H04N 19/112 according to a given display mode, e.g. for interlaced or progressive display mode
• CPC Class Codes: H04N 19/137 Motion inside a coding unit, e.g. average field, frame or block difference
• CPC Class Codes: H04N 5/145 Movement estimation for video coding H04N19/51
• CPC Class Codes: H04N 7/012 Conversion between an interlaced and a progressive signal for graphics images G09G2310/0229

Abstract

A method of measuring the motion in video image data for a pixel which uses both field-difference and frame-difference motion values to generate a motion value having increased accuracy. Image data (806) from the same pixel in a prior row of the same field (906) is compared to image data (808) from the same pixel in the prior row of the prior frame (908), and the absolute value of the difference is compared to the absolute value of the difference in image data (802) from the same pixel in a following row of the same field (902) and image data (804) from the same pixel in the following line of the prior frame (904). The minimum of these two values is the minimum frame-difference motion value which is input into a logical mixer. Also input into the logical mixer is the minimum field-difference motion value which may be determined by comparing data (802, 806) from the same pixel of an adjacent line of the same field (902, 906) with image data (810) from the same pixel of the same line of the prior field. The average of image data (810) from the same pixel of the same line of the prior field and image data (812, 814) from the same pixel of two rows prior or two rows after of the prior field (912, 914) may be used instead of image data (810) from the same pixel of the same line of the prior field alone, to increase the accuracy of the measurement.

Claims

1. A method of converting interlaced video data to progressive scanned data said method comprising:

receiving a series of fields of said interlaced video data;

calculating a field-difference motion value for a missing pixel by;

determining a prior-field average value equal to the average of same-pixel same-row prior-field image data and same-pixel two-rows-adjacent prior-field image data of said received interlaced video data; and

determining the absolute value of the difference between same-pixel adjacent-row same-field image data and said prior-field average value data of said received interlaced video data; and

selecting a proscan algorithm based on said field-difference motion value.

2. A method of converting interlaced video data to progressive scanned data, said method comprising:

receiving a series of fields of said interlaced video data;

calculating a field-difference motion value for a missing pixel of said received interlaced video data by;

calculating a first field-difference motion value for a missing pixel;

calculating a second field-difference motion value for a pixel one row prior to said missing pixel;

calculating a third field-difference motion value for a pixel one row after said missing pixel; and

selecting a maximum of said first, second, and third field difference motion values as said field-difference motion value for said missing pixel; and

selecting a proscan algorithm based on said field-difference motion value.

3. A method of converting interlaced video data to progressive scanned data, said method comprising:

receiving a series of fields of said interlaced video data,

calculating a field-difference motion value for a missing pixel in said received interlaced video data by;

determining a first prior-field average value equal to the average of same-pixel same-row prior-field image data and same-pixel prior-field two-rows-prior image data of said received interlaced video data; and

determining a first field-difference motion value equal to the absolute value of the difference between same-pixel prior-row same-field image data of said received interlaced video data and said first prior-field average value data;

determining a second prior-field average value equal to the average of same-pixel same-row prior-field image data and same-pixel prior-field two-rows-latter image data of said received interlaced video data; and

determining second field-difference motion value equal to the absolute value of the difference between same-pixel prior-row same-field image data of said received interlaced video data and said second prior-field average value data; and

selecting a proscan algorithm based on said field-difference motion value.

4. The method of claim 3, further comprising the step of:

selecting a minimum of said first and second field-difference motion values as said field-difference motion value.

5. A method of converting interlaced video data to progressive scanned video data, said method comprising:

calculating a field-difference motion value for a missing pixel;

calculating a frame-difference motion value for a missing pixel; and

selecting a proscan algorithm for creating data for said missing pixel both on said frame-difference and said field-difference motion values;

comparing the frame-difference motion value to a first threshold;

said selecting a proscan algorithm step comprising the step of using the frame-difference motion value to select an algorithm for creating data for said missing pixel when the frame-difference motion value is less than the first threshold and using a maximum of the field-difference motion value and the frame difference motion value to select an algorithm for creating data for said missing pixel when the frame-difference motion value is greater than the first threshold.

6. A method of converting interlaced video data to progressive scanned data, said method comprising:

calculating a field-difference motion value for a missing pixel;

calculating a frame-difference motion value for a missing pixel; and

selecting a proscan algorithm for creating data for said missing pixel both on said frame-difference and said field-difference motion values;

comparing said frame-difference motion value to a first threshold;

comparing said frame-difference motion value to a second threshold;

said selecting a proscan algorithm based on the field-difference motion value step further comprising;

using said frame-difference motion value to select an algorithm for creating data for the missing pixel when the frame-difference motion value is less than the first threshold;

using a maximum of said field-difference motion value and said frame-difference motion value to select an algorithm for creating data for said missing pixel when said frame-difference motion value is greater than the second threshold;

using a weighted average of said frame-difference and field-difference motion values to select an algorithm for creating data for said missing pixel when said frame-difference motion value is less than the first threshold and greater than said second threshold.

7. A method of determining a motion value for a pixel location in a video signal, said method comprising:

determining a first frame-difference motion value by comparing same-pixel prior-row image data from a current frame and a prior frame;

determining a second frame-difference motion value by comparing same-pixel next-row image data from said current frame and said prior frame; and

setting said motion value equal to a minimum of said first frame-difference motion value and said second frame-difference motion value.

8. A display system comprising:

a video processor for receiving an interlaced video signal, calculating a field-difference motion value using said received interlaced video signal, and converting said interlaced video signal to a progressive-scan video signal using an algorithm selected by said calculated field-difference motion value said video processor performing said conversion based on a pixel motion value for said interlaced video signal, said pixel motion value calculated by determining a first frame-difference motion value by comparing same-pixel prior-row image data from a current frame and a prior frame, determining a second frame-difference motion value by comparing same-pixel next-row image data from said current frame and said prior frame, selecting a minimum frame-difference motion value from said first frame-difference motion value and said second frame-difference motion value, determining a first prior-field average value by averaging same-pixel same-row image data image data from the prior field and same-pixel two-rows-before from the prior field, determining a second prior-field average value by averaging same-pixel same-row image data image data from the prior field and same-pixel two-rows-after from the prior field, determining a first field-difference motion value by comparing same-pixel prior-row image data from said current frame with said first prior-field average value, determining a second field-difference motion value by comparing same-pixel next-row image data from said current frame with said second prior-field average value, selecting a minimum field-difference motion value from said first field-difference motion value and said second field-difference motion value, selecting a maximum motion value equal to the greater of said frame-difference motion value and said field-difference motion value, providing a first threshold motion value, providing a second threshold motion value, and computing said motion value for said pixel location equal to said minimum frame-difference motion value when said minimum frame-difference motion value is less than said first threshold motion value, said pixel motion value for said pixel location equal to said maximum motion value when said minimum frame-difference motion value is greater than said second threshold motion value; and

a display for receiving said progressive-scan video signal from said video processor and for displaying said progressive scan video signal.

9. The display system of claim 8, wherein said first threshold motion value equals said second threshold motion value.