Method and apparatus for zooming images on a video display

• US 5,657,047 A
• Filed: 02/10/1995
• Issued: 08/12/1997
• Est. Priority Date: 01/12/1995
• Status: Expired due to Term

• Current Assignee: NVIDIA Corporation
• Original Assignee: AccelGraphics, Inc. (Elevate Entertainment, Inc.)
• Inventors: Tarolli, Gary M.
• Primary Examiner(s): Hjerpe, Richard
• Assistant Examiner(s): Mengistu, Amare
• Application Number:
  US08/386,742
• Time in Patent Office: 914 Days
• Field of Search: 395/102, 395/139, 395/134, 345/127, 345/130, 345/185, 398/451
• US Class Current: 345/671
• CPC Class Codes: G06T 3/40 Scaling of whole images or parts thereof, e.g. expanding or contracting
• CPC Class Codes: G09G 5/393 Arrangements for updating the contents of the bit-mapped memory

Abstract

An apparatus and method for enlarging a source image on a computer display is described. A scanline buffer stores pixels in a first row of the source image. Coordination circuitry receives from a processor and stores pixels in a second row of the source image. Blend circuitry generates output pixels, such that each of the output pixels corresponds to a weighted average of a plurality of input pixels. The plurality of input pixels comprises two pixels stored in the scanline buffer and two pixels stored in the coordination circuitry. Control circuitry determines weight values and provides the weight values to the blend circuitry for calculating the weighted average. The control circuitry also determines, based on the weight values, which pixels stored in the scanline buffer and in the coordination circuitry are used by the blend circuitry as the input pixels. In addition, the control circuitry replaces the pixels in the scanline buffer with corresponding pixels from the coordination circuitry after the last time the pixels in the scanline buffer are used as the input pixels.

Claims

1. An apparatus for enlarging a source image for display on a display device, the apparatus comprising:

a scanline buffer for storing pixel values of a first row of the source image;

coordination circuitry for receiving and storing pixel values of a second row of the source image;

blend circuitry coupled to the scanline buffer and the coordination circuitry, the blend circuitry for generating output pixel values, each of the output pixel values corresponding to a weighted average of a plurality of input pixel values, the plurality of input pixel values comprising a plurality of pixel values stored in the scanline buffer and a plurality of pixel values stored in the coordination circuitry; and

control circuitry coupled to the scanline buffer, the coordination circuitry, and the blend circuitry, the control circuitry for determining weight values for computing the weighted average.

2. The apparatus of claim 1, wherein the weight values comprise:

a value Xfrac representing a horizontal component of the weighted average; and

a value Yfrac representing a vertical component of the weighted average, wherein Xfrac is incremented by a predetermined value Xfrac_-- incr once for each time an output pixel value is generated by the blend circuitry and Yfrac is incremented by a predetermined value Yfrac_-- incr once for each time the blend circuitry generates a complete row of output pixel values,

wherein Xfrac_-- incr corresponds to a factor by which the source image is to be enlarged along a horizontal axis, and wherein Yfrac_-- incr corresponds to a factor by which the source image is to be enlarged along a vertical axis.

3. The apparatus of claim 2, wherein Xfrac is incremented by a predetermined value Xfrac_-- incr once for each time an output pixel value is generated by the blend circuitry and Yfrac is incremented by a predetermined value Yfrac_-- incr once for each time the blend circuitry generates a complete row of output pixel values.

4. The apparatus of claim 3, wherein Xfrac_-- incr corresponds to a factor by which the source image is to be enlarged along a horizontal axis.

5. The apparatus of claim 3, wherein Yfrac_-- incr corresponds to a factor by which the source image is to be enlarged along a vertical axis.

6. The apparatus of claim 3, wherein the input pixel values comprise two input pixel values from the scanline buffer and two input pixel values from the coordination circuitry, and wherein the control circuitry replaces a first of the two input pixel values from the scanline buffer with a first of the two input pixel values from the coordination circuitry when Xfrac exceeds a first predetermined value and Yfrac exceeds a second predetermined value.

7. The apparatus of claim 2, wherein the weighted average is computed according to the formula:

space="preserve" listing-type="equation">T1=A+Yfrac(C-A)

space="preserve" listing-type="equation">T2=B+Yfrac(D-B)

space="preserve" listing-type="equation">OUTPUT=T1+Xfrac (T2-T1)

wherein

A represents a first of the input pixel values from the stored scanline;

B represents a second of the input pixel values from the stored scanline;

C represents a first of the input pixel values from the coordination circuitry;

D represents a second of the input pixel values from the stored scanline; and

OUTPUT represents the output pixel value.

8. The apparatus of claim 7, wherein a pixel represented by A is horizontally adjacent to a pixel represented by B, a pixel represented by C is horizontally adjacent to a pixel represented by D, and the second row of the source image is vertically adjacent to the first row of the source image.

9. The apparatus of claim 1, wherein the control circuitry is further for determining a plurality of weight values and for determining, based on the weight values, which pixel values stored, in the scanline buffer and in the coordination circuitry are used by the blend circuitry as the input pixel values.

10. The apparatus of claim 9, wherein the control circuitry is further for replacing the pixel values stored in the scanline buffer with corresponding pixel values stored in the coordination circuitry after the last time the pixel values stored in the scanline buffer are used as input pixel values.

11. An apparatus for enlarging a source image on a computer display, comprising:

a scanline buffer for storing a scanline, the scanline comprising pixel values for each pixel in a first row of the source image, the source image having a plurality of rows of pixels, each of the rows having P pixels, where P is an integer;

a memory device coupled to a processor for receiving from the processor pixel values of pixels in a second row of the source image;

a data word register coupled to the memory device for receiving from the memory device and storing data words, each data word corresponding to at least one of the pixels of the second row of the source image;

a C-D register coupled to the data word register for receiving and storing pixel values of the Nth pixel and the (N+1)th pixel of the second row of the source image, where N is an integer;

blending circuitry coupled to the C-D register and the scanline buffer for generating output pixel values of an output image, each of the output pixel values representing a weighted average of a plurality of input pixel values, the plurality of input pixel values comprising

the Nth and (N+1)th pixel values of the scanline, and

the Nth and (N+1)th pixel values of the second row of the source image stored in the coordination circuitry, and

control circuitry coupled to the scanline buffer, the blending circuitry, the C-D register, and the data word register, the control circuitry storing values Xfrac and Yfrac to be used by the blending circuitry in computing the weighted average, the control circuitry incrementing Xfrac by a predetermined value Xfrac_-- incr each time the blending circuitry generates an output pixel value, such that N is incremented by one each time Xfrac exceeds a first predetermined value and N is reset to one if N exceeds P, the control circuitry resetting Xfrac to a second predetermined value if Xfrac exceeds the first predetermined value, the control circuitry causing the pixel value of the Nth pixel of the scanline to be replaced with the pixel value of the Nth pixel of the second row of the source image stored in the coordination circuit after the blending circuitry generates the last output intensity level for which the Nth pixel value of the scanline is used as one of the input pixel values, the control circuitry incrementing Yfrac by a predetermined value Yfrac_-- incr once each time the blending circuitry generates a complete row of output pixel values, such that the second row of the source image is replaced by the next vertically adjacent row of the source image each time Yfrac exceeds a fourth predetermined value, the control circuitry resetting Yfrac to a third predetermined value if Yfrac exceeds the fourth predetermined values.

12. The apparatus of claim 11, wherein N is initially one.

13. The apparatus of claim 11, wherein the weighted average is computed by the blending circuitry according to the formula:

space="preserve" listing-type="equation">T1=A+Yfrac(C-A)

space="preserve" listing-type="equation">T2=B+Yfrac(D-B)

space="preserve" listing-type="equation">OUTPUT=T1+Xfrac (T2-T1)

wherein

A represents the Nth pixel value of the scanline;

B represents the (N+1)th pixel value of the scanline;

C represents the Nth pixel value of the Mth row of the source image; and

D represents the (N+1)th pixel value of the Mth row of the source image; and

OUTPUT represents the output pixel value.

14. The apparatus of claim 11, wherein the control circuitry comprises:

an Xfrac register for storing Xfrac, Xfrac representing a horizontal component of a weighted intensity of one of the input pixel values;

a Yfrac register for storing Yfrac, Yfrac representing a vertical component of the weighted intensity of one of the input pixel values; and

an Xfrac_-- incr register for storing Xfrac_-- incr, Xfrac_-- incr corresponding to a factor by which the source image is to be enlarged horizontally.

15. The apparatus of claim 14, wherein the control circuitry further comprises:

a control register for storing a STORE bit and a DRAW bit, the DRAW bit for enabling the blending circuitry to generate the output pixel value, the STORE bit for enabling the Nth pixel value of the scanline to be replaced with the Nth pixel value of the second row of the source image stored in the coordination circuit; and

a command circuit for causing Xfrac to be incremented, for causing the Nth pixel value of the scanline to be replaced with the the Nth pixel value of the second row of the source image stored in the coordination circuit when the STORE bit has a fifth predetermined value, and for enabling the blending circuitry to generate the output intensity level when the DRAW bit has a sixth predetermined value.

16. The apparatus of claim 11, wherein the processor is a central processing unit (CPU) of a computer.

17. A method of enlarging a source image on a video display, comprising the steps of:

a) storing in a storage buffer pixel values of a single row of the source image received from a processor;

b) determining weight values corresponding to factors by which the source image is to be enlarged;

c) providing to a blend circuit a plurality of sets of input pixel values, each of the sets of input pixel values comprising two pixel values stored in the storage buffer and two pixel values from a second row of the source image received from the processor;

d) determining based on the weight values which of the pixel values in the storage buffer and in the second row of the source image are to be included in each of the sets of input pixel values;

e) using the blend circuit to generate a plurality of output pixel values, each of the output pixel values corresponding to a weighted average of one of the sets of input pixel values; and

f) replacing predetermined pixel values in the storage buffer with corresponding pixel values from the second row of the source image, each of the predetermined pixel values being replaced after the blend circuit has generated the last output pixel value for which the predetermined pixel value was used as an input pixel value wherein the predetermined pixel values are determined based on the weight values;

g) repeating steps b) through f) such that the pixel value of each pixel of the source image is used by the blend circuit in at least one of the sets input pixel values for generating at least one of the output pixel values.

18. The method of claim 17, wherein the weight values comprise:

a value Xfrac corresponding to a horizontal component of the weighted average; and

a value Yfrac corresponding to a vertical component of the weighted average.

19. The method of claim 17, wherein Xfrac is incremented by a predetermined value Xfrac_-- incr once for each time an output pixel value is generated by the blend circuit and Yfrac is incremented by a predetermined value Yfrac_-- incr once for each time the blend circuit generates a complete row of output pixel values.

20. The method of claim 19, wherein Xfrac_-- incr corresponds to a factor by which the source image is to be enlarged horizontally, and wherein Yfrac_-- incr corresponds to a factor by which the source image is to be enlarged vertically.

21. The method of claim 17, wherein the weighted average is computed according to the formula:

space="preserve" listing-type="equation">T1=A+Yfrac(C-A)

space="preserve" listing-type="equation">T2=B+Yfrac(D-B)

space="preserve" listing-type="equation">OUTPUT=T1+Xfrac (T2-T1)

wherein

A represents a first of the two pixel values stored in the storage buffer;

B represents a second of the two pixel values stored in the storage buffer;

C represents a first of the two pixel values from the second row of the source image;

D represents a second of the two pixel values from the second row of the source image; and

OUTPUT represents the output pixel value.

22. The method of claim 21, wherein a pixel represented by A is horizontally adjacent to a pixel represented by B in the source image, a pixel represented by C is horizontally adjacent to a pixel represented by D in the source image, and the second row of the source image is vertically adjacent to the first row of the source image.

23. The method of claim 17, wherein the first of the two input pixel values from the storage buffer is replaced by the first of the two input pixel values from the second row of the source image when Xfrac exceeds a first predetermined value and Yfrac exceeds a second predetermined value.

24. The method of claim 17, further comprising the step of providing each of the output pixel values generated by the blend circuit to a frame buffer.

25. The method of claim 17, wherein the processor is a central processing unit (CPU) of a computer.

26. An apparatus for enlarging a source image, comprising:

a buffer for storing pixel values of a single row of the source image;

a first circuit for receiving from a processor and storing pixel values of a second row of the source image;

a second circuit coupled to the buffer and the first circuit, the second circuit for generating output pixel values, each of the output pixel values corresponding to a weighted average of a plurality of input pixel values; and

a third circuit coupled to the buffer, the first circuit, and the second circuit, the third circuit for determining the weighted average.

27. In a computer system including a processor and a display device, a method of enlarging a source image for display by the display device, the method comprising the steps of:

a) receiving pixel values of a single row of the source image from the processor;

b) storing the pixel values of the single row of the source image in a storage buffer;

c) receiving pixel values of a second row of the source image from the processor;

d) determining weight values corresponding to factors by which the source image is to be enlarged;

e) determining, based on the weight values, which of the pixel values in the storage buffer and in the second row of the source image are to be used as input pixel values for generating each of a plurality of output pixel values; and

f) generating the plurality of output pixel values, each of the output pixel values corresponding to a weighted average of a set of the input pixel values.

28. The method of claim 27, further comprising the step of:

(g) replacing predetermined pixel values in the storage buffer with corresponding pixel values from the second row of the source image, each of the predetermined pixel values being replaced after generating the last output pixel value for which the predetermined pixel value was used as an input pixel value.

29. The method of claim 28, further comprising the step of:

h) repeating steps a) through g) such that each pixel value of the source image is used in at least one set of input pixel values for generating at least one of the output pixel values.

30. An apparatus for enlarging a source image for display on a display device, the apparatus comprising:

buffer means for storing pixel values of only a first row of the source image;

coordination means for receiving from a processor and storing pixel values of a second row of the source image;

blend means for generating a plurality of output pixel values, each of the output pixel values representing a weighted average of a plurality of input pixel values, the plurality of input pixel values comprising a plurality of the pixel values stored in the buffer means and a plurality of the pixel values stored in the coordination means; and

control means for providing weight values to the blend means corresponding to the weighted average and for determining which pixels stored in the buffer means and in the coordination means are used by the blend means as the input pixel values.

31. The apparatus of claim 30, wherein the control means is further for replacing the pixel values stored in the buffer means with corresponding pixel values from the coordination means after the last time the pixel values in the buffer means are used as input pixel values.