CN100533485C - Image describing method - Google Patents
Image describing method Download PDFInfo
- Publication number
- CN100533485C CN100533485C CNB2007100009843A CN200710000984A CN100533485C CN 100533485 C CN100533485 C CN 100533485C CN B2007100009843 A CNB2007100009843 A CN B2007100009843A CN 200710000984 A CN200710000984 A CN 200710000984A CN 100533485 C CN100533485 C CN 100533485C
- Authority
- CN
- China
- Prior art keywords
- block
- row
- present
- image
- property value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
A method for describing image includes setting the most left block of highest block column to be an initial block and setting reference pixel of said initial block to be an initial pixel, calculating block interval amount at X direction between the most left block and said initial block on current block column as well as block interval amount at Y direction, utilizing a formula to calculate out image attribute value of reference pixel at the most left block and block interval amount between current block and the most left block then utilizing a formula to calculate out image attribute value of reference pixel at current block.
Description
Technical field
The invention relates to a kind of image treatment method, and be particularly to a kind of plotting method of 3-dimensional image object.
Background technology
Computer graphics (Computer Graphics) can be applied in a lot of aspects, comprises aspects such as simulation, CAD, CAI, and the function of various application need just can obtain through the collocation of computer hardware and software.In computer drawing technology, describing (rendering) is meant the three dimensional object in the database, convert the program of the two-dimensional projection on the view plane to, it mainly comprises sets up straight line (Line), circular (Circle), oval (Ellipse), polygon (Polygon) or the like, and processes and displays polygon and do not show polygon and painted (shading) program or the like.
About straight line and polygonal foundation, it mainly is to utilize plane formula (PlaneEquation), digital differential analyser (Digital Differential Analyzer, DDA) algorithm and Improvement type DDA algorithm, wherein Improvement type DDA algorithm comprises that again the segment formula cuts digital differential analyser (Tile Based DDA) algorithm and sweep trace formula digital differential analyser (Scanline Based DDA) algorithm.
Please refer to Fig. 1, is the mode of describing that cuts the digital differential analyser algorithm for the segment formula, promptly is that picture is cut into many blocks (Tile).As shown in Figure 1, this triangle is cut into 19 4 * 4 block (Tile, respectively with T1～T19 name), then with top left pixel (for example, the A of each block
1, A
2, A
3... A
19) be starting point, describe 19 blocks in this triangle in regular turn.For instance, if will describe pixel B
1The image attribute, then can utilize formula T
A(tx, ty)=T
A(tx-1, ty)+4 * dA/dx obtains block T6 top left corner pixel A earlier
6Property value, promptly T6 is that T5 adds 4 times differential value (wherein fix a number 4 obtain all be made up of 4 * 4 pixels because of each block), then, again by formula f
A(x, y)=(T
A(tx, ty)+n * dA/dy)+m * dA/dx is by A
6Pixel derive to calculate and obtain desire image pixel B
1, wherein tx and ty represent block A
6The segment coordinate, m=x mod 4, n=ymod 4, A represents image attribute (depth of field, color, material, coordinate or the like).Promptly must carry out describing of block one by one according to the direction of arrow that is illustrated among Fig. 1.
Please refer to Fig. 2, being the mode of describing for sweep trace formula digital differential analyser (Scanline BasedDDA) algorithm, is that this triangle image is cut into a plurality of pixels that are positioned at same row for describing the benchmark processing of drawing, as shown in Figure 2, this triangle image is cut into 18 row, with R
1Classify as and describe original bench mark, from top to bottom and from a left side and the right image of describing each row pixel in regular turn.For instance, if will describe block B
2Image, then can utilize formula f
A(x, y)=f
A(x, y-1)+dA/dy is first from R
7Be painted into R
8The most left pixel, again by formula f
A(x, y)=f
A(x-1, y)+dA/dx describes to desire image pixel B one by one from the most left pixel
1, wherein A represents image attribute (depth of field, color, material, coordinate or the like).Promptly must carry out describing of pixel column one by one according to the direction of arrow that is illustrated among Fig. 2.
Yet, no matter be with the segment formula for describe benchmark or with the sweep trace formula for describing benchmark, it is when carrying out object drawing processing, the error that produced can be drawn order and accumulation is always gone down in proper order because be, be that the drawing processing that the error that produces can influence the back is handled in the front drawing, cause describe the gained image may be very big with expection drop originally, cause error span excessive because of the accumulation of error.For instance, as shown in Figure 1, if desire to describe pixel B
1, then must be in regular turn with A
1, A
2, A
3, A
4With A
5The image attribute of pixel is obtained according to above-mentioned formula earlier can describe A
6Be that later pixel B is carried out on the basis with it
1Describe pixel A wherein
2Be reference pixel A
1Describe to form pixel A
3Be reference pixel A
2Describe to form, by that analogy to A
6And obtain all and need get through above-mentioned differential formulas owing to the property value of each pixel, therefore, describing pixel A
1, A
2, A
3, A
4With A
5When the time produces error because of differential respectively and wherein have the error span of a certain pixel when especially big, then desire to describe block B
1The time, because of accumulating A
1, A
2, A
3With A
4Deng the error of property value, so block B
1Possibly can't the output expected effect.Again, as shown in Figure 2, if desire to describe pixel B
2, then must be earlier with R
1, R
2..., R
7Can describe after describing to finish Deng pixel column, wherein R
2Pixel column is reference columns R
1Pixel column is described to form, R
3Pixel column is with reference to R
2Pixel column is described to form, by that analogy to R
7Pixel column.And obtain all and need get through above-mentioned differential formulas owing to the property value of each row pixel, therefore, describing R
1, R
2..., R
7When wrong respectively difference produces when the row pixel, then desire to describe pixel B
2The time, because of accumulating R
1, R
2..., R
7Deng the pixel column error and wherein have the error span of a certain pixel column when especially big, so block B
2Possibly can't the output expected effect.
Please refer to Fig. 3 A to Fig. 3 D, wherein, Fig. 3 A is the actual straight line L that describes of expression
1With L
2, if see over, then on screen, should show shown in Fig. 3 B in theory by the direction of arrow that indicates depth, promptly can't see by straight line L
1With L
2The part of covering respectively (being represented by dotted lines).But because the influence of previously described cumulative errors, cause to occur the situation shown in Fig. 3 C and Fig. 3 D.
Lift the consequence that a routine specification error accumulation is caused again.As previously mentioned, when carrying out image describing, a pair of liking depicted complete image according to last reference pixel, therefore, when the amplitude of makeing mistakes when a certain reference pixel is excessive, the block or the pixel column at follow-up reference pixel and this reference pixel place will be accumulated and influence to its mistake, to cause this object unusual situation takes place.Shown in Fig. 4 A, it is that triangle object O is described in expression
1With triangle object O
2The time, the synoptic diagram of the relative position of two triangle object.When normally describing triangle, triangle O
2Be to have part to cover triangle O
1On, shown in Fig. 4 B.And, then may make the triangle object display abnormality of depicting when if as indicated earlier cumulative error takes place, and shown in Fig. 4 C, triangle O
1The lower right corner sub-fraction is arranged not by triangle O
2Complete covering.
In addition, above-mentioned known way is if be all with software and realize what the mode of algorithm realized, if desire to depict fast a plane, though hardware device that can high-order is reached, its hardware cost is higher relatively.Therefore, be badly in need of a kind of generation of image describing method, with improve existing DDA algorithm with fast and more accurate inerrably depict desired image.
Summary of the invention
Based on above-mentioned purpose, the present invention has disclosed a kind of image describing method, be to be used to describe an imaged object, this imaged object comprises a plurality of block row, and each block row comprises at least one block, and the size of wherein said block is m * m, m is a positive integer, and each block respectively comprises a reference pixel, and this reference pixel is the top left pixel of each block, and this image describing method comprises: the highest region piece row (RTOP), minimum block row and the present block that define an imaged object are listed as; The most left block of setting this highest region piece row (RTOP) is an original block (TINIT), and the reference pixel of setting this original block is an initial pixel; Between the most left block that calculates these present block row and this original block at the block amount of space (Δ x) of a directions X and at the block amount of space (Δ y) of a Y direction; Utilize formula A
LEFT=A
INIT+ dA/dx * n * Δ x+dA/dy * n * Δ y is with the image property value (A of the reference pixel of the most left block that directly calculates these present block row
LEFT), wherein, A is the image attribute, A
INITBe the image property value of this initial pixel, and the n value is relevant with the block size; Calculate the block amount of space (Δ z) between the most left block of the present block row of a present block and this, wherein this present block is positioned at same block with the most left block that should present block be listed as and is listed as; And utilize formula A
CURRENT=A
LEFT+ dA/dx * n * Δ z is with the image property value (A of the reference pixel that directly calculates this present block
CURRENT), Δ z value is the block amount of space between the most left block of the present block row of this present block and this, and the n value is relevant with the block size.
The present invention has more disclosed a kind of image describing method, be to be used to describe an imaged object, this imaged object comprises a plurality of block row, each block row comprises at least one block, the size of wherein said block is m * m, and m is a positive integer, and each block respectively comprises a reference pixel, this reference pixel is the top left pixel of each block, and this image describing method comprises: a highest region piece row R who defines this imaged object
TOP, minimum block row and a present block be listed as; Set these highest region piece row (R
TOP) the most left block be an original block (T
INIT), and the reference pixel of setting this original block is an initial pixel; Calculate between these present block row and this original block block amount of space (Δ y) in a Y direction; Utilize formula A
YTILE=A
INIT+ dA/dx * n * Δ y is directly to calculate the image property value (A that lists the reference pixel of vertical common factor with this initial pixel at this present block
YTILE), wherein, A is the image attribute, A
INITBe the image property value of this initial pixel, and the n value is relevant with the block size; Calculate between a present block and this original block block amount of space (Δ x) at a directions X, wherein this present block is positioned at same block row with reference pixel with the vertical common factor of this initial pixel; And utilize formula A
CURRENT=A
YTILE+ dA/dx * n * Δ x is with the image property value (A of the reference pixel that directly calculates this present block
CURRENT), wherein the n value is relevant with the block size.
Image describing method of the present invention has been improved existing DDA algorithm, depict fast and more accurately desire image.
Description of drawings
Fig. 1 shows to utilize Tile Based DDA algorithm to describe leg-of-mutton synoptic diagram.
Fig. 2 shows to utilize Scanline Based DDA algorithm to describe leg-of-mutton synoptic diagram.
Fig. 3 A to Fig. 3 D shows actual depiction, theoretical displayed map and the excessive synoptic diagram of accumulation amplitude that makes a mistake when utilizing traditional image describing method to describe line segment.
Fig. 4 A to Fig. 4 C figure shows actual depiction, theoretical displayed map and the excessive synoptic diagram of accumulation amplitude that makes a mistake when utilizing traditional image describing method to describe triangle.
Fig. 5 shows the image describing method synoptic diagram that utilizes one embodiment of the invention.
Fig. 6 is the image describing method process flow diagram that shows according to Fig. 5.
Fig. 7 is the image describing method synoptic diagram that shows another embodiment of the present invention.
Fig. 8 shows to utilize the image describing method process flow diagram of the present invention according to Fig. 7.
Embodiment
For purpose of the present invention, feature and advantage can be become apparent, preferred embodiment cited below particularly, and cooperate appended pictorial image 5 to Fig. 8 to do a detailed explanation.Instructions of the present invention provides different embodiment that the technical characterictic of the different embodiments of the present invention is described.Wherein, the configuration of each element among the embodiment is the usefulness for explanation, is not in order to restriction the present invention, and the part of reference numerals repeats among the embodiment, is for the purpose of simplifying the description, is not the relevance that means between the different embodiment.
The embodiment of the invention has disclosed a kind of image describing method, and it is improved existing DDA algorithm and is desired image to depict fast and more accurately.
Explanation is in advance, and the image attribute (A) of the following stated comprises the depth of field, color, material, coordinate or the like, and it is about the required correction of image data of using when carrying out image describing, but is not given unnecessary details in the present invention.
In order to overcome the repeatedly shortcoming of previous differential calculation error of traditional image describing method accumulation, the image describing method of the embodiment of the invention can reduce the wrong amplitude of differential calculation accumulation, to reduce the probability that abnormal show takes place, also or even abnormal show takes place, can slow down the degree of its abnormal show.
Please merge with reference to figure 5 and Fig. 6, be that the image describing method that shows one embodiment of the invention is respectively described leg-of-mutton synoptic diagram and process flow diagram.Wherein, the image describing method that the present invention proposes, be to be used to describe an imaged object, this imaged object mainly is that a kind of triangle kenel is main in following disclosed embodiment, but it is not to describe on other kenel objects as square, circle etc. in order to limit the present invention.Each imaged object all comprises a plurality of block row, and each block row comprises at least one block, and each block respectively comprises a reference pixel, this reference pixel promptly is positioned at the top left pixel of each block, its meaning is except as the reference data about rest of pixels image property value in the same block, also as in all the other blocks about the reference data of reference pixel image property value.With Fig. 5, this triangle imaged object promptly includes 6 blocks row, and wherein first block row comprise a block and (include reference pixel C
1), second block row comprise two blocks and (include reference pixel C respectively
2, C
3), the 3rd block and the 4th block row respectively comprise four blocks and (include reference pixel C respectively
4～C
7With C
8～C
11), the 5th block row comprise six blocks and (include reference pixel C respectively
12, C
13, C
14, C
15, C
16, C
17), SECTOR-SEVEN piece row comprise two blocks and (include reference pixel C respectively
18, C
19).
Image describing method of the present invention includes step S101, defines the highest region piece row (R of this triangle object in advance
TOP, i.e. C
1The block position at place), a minimum block row (R
BOTTOM, i.e. C
18With C
19The block position at place) and present block row (R
CURRENT, be assumed to be C
4～C
7The block position at place), this in advance explanation be because the process of describing is to begin to describe from highest region piece row, therefore in program is write, can be about to these present block row earlier to be defined as these highest region piece row (R
CURRENT=R
TOP), even the also therefore selected at present target of describing is C
4～C
7The block row at place, but the present block row of original state still are these highest region piece row.Then, define present block row (R at step S102
CURRENT) the most left block (T
LEFT), the rightest block (T
RIGHT) and present block (T
CURRENT), owing to be in original state, and highest region piece row (R
TOP) block is only arranged, so C
1The block at place is present block row (R
CURRENT) the most left block (T
LEFT), the rightest block (T
RIGHT) and present block (T
CURRENT).
In step S103, judge present block row whether to be that the highest region piece is listed as, (R when judging this present block and classify these highest region piece row as
CURRENT=R
TOP), promptly carry out the initialization procedure of step S104, be about to these highest region piece row (R
TOP) the most left block be set at an original block (T
INIT,, be C therefore because the highest region piece of present embodiment row only comprise a block
1The block at place), and with this original block (T
INIT) the top left corner pixel image property value (A that is set at initial pixel and calculates this initial pixel
INIT).(R when judging these present block row and be not these highest region piece row
CURRENT≠ R
TOP) or behind the initialization procedure that carries out step S104, promptly carry out the calculating of step S105, comprise between the most left block that calculates these present block row and this original block at the block amount of space (Δ x) of a directions X and at the block amount of space (Δ y) of a Y direction, for the purpose of simplifying the description, be about to the process of describing at this and skip to the selected present block row of describing from these highest region piece row, therefore hypothesis makes that via a plurality of flow process circulations the present block among the step S112 is listed as (R
CURRENT) proceeded to C
4～C
7During the block row at place, be back to step S102, define present block row (R
CURRENT) the most left block (T
LEFT, i.e. C
4The block at place), block is listed as (R at present
CURRENT) the rightest block (T
RIGHT, i.e. C
7The block at place), with as a computing interval of in program is write, setting, in addition, also can define present block row (R earlier
CURRENT) the most left block (T
LEFT) be the block (T that will describe at present
CURRENT=T
LEFT, i.e. C
4The block at place is that the most left block also is a present block).
To describe B
3During the present block at place, the most left block that need depict these present block row earlier (is C
4The block at place), also therefore must in step S105, calculate C
4The block at place and C
1The block at place is at the block amount of space (Δ x) of directions X and block amount of space (Δ y) in the Y direction, by among the figure as can be known, block amount of space (Δ x) is all 2 with block amount of space (Δ y).
Then can in step S106, directly utilize formula A
LEFT=A
INIT+ dA/dx * n * Δ x+dA/dy * n * Δ y calculates the reference pixel C of left block
4Image property value (A
LEFT), wherein, A
INITBe initial pixel C
1The image property value, and the n value is (for example, when the size of a block be 4 * 4, i.e. n=4, present embodiment promptly as a n value, but be not with this limit) relevant with block size.When in step S106, learning reference pixel C as calculated
4Image property value (A
LEFT) time, promptly in step S107, depict be positioned at this most left block remove C
4The image property value of outer rest of pixels.It should be noted that according to step S106, the present invention promptly need not as known Fig. 1 when describing the most left block of present block row, except original block (A
1The block at place) outside, still needs and calculate reference pixel A earlier
3, A
2, A
5The image property value, just obtained A
4The image property value, the present invention can obtain the initial pixel C of original block
1Image property value (A
INIT) after, directly utilize step S105 and step S106 great-jump-forward to obtain C
4Image property value (A
LEFT), therefore can not accumulate reference pixel C
3, C
2, C
5Image property value error.
Then, not during right block of these present block row when judge this present block at step S108, then specifying the next one block that moves to right at step S109 is this present block, for the purpose of simplifying the description, supposes to move to right to B in step S109
3The present block at place then in order to describe present block, must calculate C earlier in step S110
6The present block and the C at place
4Block amount of space between the most left block at place (Δ z, Δ z=2 in this embodiment) utilizes formula A
CURRENT=A
LEFT+ dA/dx * n * Δ z directly calculates the reference pixel C of this block that moves to right
6Image property value (A
CURRENT), wherein this present block is to be positioned at same block row with this most left block, Δ z value is the block amount of space between this present block and this most left block, and the n value is and block size relevant (wherein the definition of n value and above-mentioned steps S106 is identical), when proceeding to step S107 when continuing, can be according to reference pixel C
6Describe benchmark depicting the image property value of the rest of pixels that is positioned at this present block, and promptly can be painted into the pixel B 3 desiring to describe this moment.It should be noted that according to step S110, the present invention promptly need not as known Fig. 1 when describing the present block of present block row, except the most left block (A
4The block at place) outside, still needs and calculate reference pixel A earlier
5The image property value, just obtained A
6The image property value, the present invention can obtain the reference pixel C of left block
4Image property value (A
LEFT) after, directly utilize step S108, S109 and step S110 great-jump-forward to obtain C
6Image property value (A
CURRENT), therefore can not be accumulated to reference pixel C
5Image property value error.
So far, the present invention desires the pixel B described describing
3(be the pixel B of known Fig. 1
1) time, can not accumulate reference pixel C
3, C
2With twice C
5Image property value error, to reduce the amplitude and the number of times of error accumulation.
As the present block (T that in step S108, judges in describing
CURRENT) when being the rightest block in the same block row, then continue to judge at step S111 whether present block row are minimum block row, if, then finish above-mentioned image describing flow process, if not, then specify next block to classify present block row as and get back to step S102 and continue to describe next block row at step S112.
Please merge with reference to figure 7 and Fig. 8, be that the image describing method that shows another embodiment of the present invention is respectively described leg-of-mutton synoptic diagram and process flow diagram.Wherein, employed imaged object in the image describing method that the present invention proposes to disclose explanation in the foregoing description, is promptly given unnecessary details no longer in detail at this, and wherein is denoted as D
1～D
19Promptly be all the reference pixel of each block.
Image describing method of the present invention includes step S201, defines the highest region piece row (R of this triangle object in advance
TOP, i.e. D
1The block at place), a minimum block row (R
BOTTOM, i.e. D
18With D
19The block at place) and present block row (R
CURRENT, be assumed to be D
4～D
7The block at place), wherein because the process of describing is to begin to describe from highest region piece row, therefore in program is write, can be about to these present block row earlier to be defined as these highest region piece row (R
CURRENT=R
TOP), even the therefore selected at present target of describing is D
4～D
7The block row at place, but the present block row of original state still are these highest region piece row.
Then, define present block row (R at step S202
CURRENT) the most left block (T
LEFT), the rightest block (T
RIGHT) and present block (T
CURRENT), owing to be in original state, and highest region piece row (R
TOP) block is only arranged, so D
1The block at place is present block row (R
CURRENT) the most left block (T
LEFT), the rightest block (T
RIGHT) and present block (T
CURRENT).
Then in step S203, judge present block row whether to be that the highest region piece is listed as, (R when judging this present block and classify these highest region piece row as
CURRENT=R
TOP), promptly carry out the initialization procedure of step S204, be about to these highest region piece row (R
TOP) the most left block be set at an original block (T
INIT,, be D therefore because the highest region piece of present embodiment row only comprise a block
1The block at place), and with this original block (T
INIT) the top left corner pixel image property value (A that is set at initial pixel and calculates this initial pixel
INIT).(R when judging these present block row and be not these highest region piece row
CURRENT≠ R
TOP) or behind the initialization procedure that carries out step S204, representing this moment has not been original state, promptly carry out the calculating of step S205, comprise the block amount of space (Δ y) that calculates between these present block row and this original block in a Y direction, for the purpose of simplifying the description, be about to the process of describing at this and skip to selected present block row, therefore will describe B from these highest region piece row
3During the present block at place, need depict the image property value (A of these present block row and the reference pixel of the vertical common factor of this initial pixel earlier
YTILE), also therefore must in step S205, calculate D
4～D
7Block row and D
1The block at place is at the block amount of space (Δ y) of Y direction, by among the figure as can be known, block amount of space (Δ y) is 2, utilizes formula A again
YTILE=A
INIT+ dA/dy * n * Δ y is directly to calculate the image property value (A of this vertical reference pixel that occurs simultaneously
YTILE), wherein, A
INITBe the image property value of this initial pixel, and the n value is and block size relevant (this n value defined has illustrated in the foregoing description, does not repeat them here).
Then, calculating this present block in step S206 (is D
6The place block) with this original block (D
1The place block) between at the block amount of space (Δ x) of a directions X, this present block reference pixel (A that is vertical common factor wherein with this
YTILE) be positioned at same block row, in this example, this present block is except the reference pixel (A of vertical common factor with this
YTILE) be arranged in same block row, and be arranged in same block, so its block amount of space (Δ x) equals 0.Then utilize formula A
CURRENT=A
YTILE+ dA/dx * n * Δ x is with the image property value (A of the reference pixel that directly calculates a present block
CURRENT), wherein the n value is relevant with the block size, also in above-mentioned explanation it.
When in step S206, learning reference pixel D as calculated
6Image property value (A
CURRENT) time, promptly in step S207, depict the D that removes that is positioned at this present block
6The image property value of outer rest of pixels.It should be noted that according to step S206, the present invention promptly need not as known Fig. 1 when describing present block, except original block (A
1The block at place) outside, still needs and calculate reference pixel A earlier
3, A
2, A
5, A
4The image property value, just obtained A
6The image property value, the present invention can obtain the initial pixel D of original block
1Image property value (A
INIT) after, directly utilize step S205 and step S206 great-jump-forward to obtain D
6Image property value (A
CURRENT), therefore can not accumulate reference pixel D
3, D
2, D
5, D
4Image property value error.
Then, not during right block of these present block row when judge this present block at step S208, then specifying the next one block that moves to right at step S209 is this present block, is D
7The block at place is present block, continues to step S206 and continues to describe this block.But as the present block (T that in step S208, judges in describing
CURRENT) when being the rightest block in the same block row, then continue to judge at step S210 whether present block row are minimum block row (R
BOTTOM), if, then finish above-mentioned image describing flow process, if not, then specify next block to classify present block row as and get back to step S202 and continue to describe next block row at step S211.
In the present embodiment, block institute reference at present be not as known, to be last block, also be not the most left block of a last embodiment for same block row, but with reference to same block row but with the direct vertical reference block of initial pixel.
Because the reference pixel of each block is except as the reference data about rest of pixels image property value in the same block, also as in all the other blocks about the reference data of reference pixel image property value, therefore if as known each block describing Shi Jiexu with reference to last block, then can allow the error amount that produces because of differential be accumulate to present block always, but among the one embodiment of the present invention, each block is when describing, each present block is also all only with reference to the most left block in the same block row, and the most left block of each block row is all only with reference to original block, though therefore also having the error that produces because of differential takes place, but can be an error (please refer to the image property value operational formula of step S106 and step S110) qualitatively, and among another embodiment, each block is when describing, each present block is the reference pixel of the vertical common factor with original block of reference all only, and each vertical reference block of occuring simultaneously is also all only with reference to original block, therefore also can be to have only an error (please refer to the image property value operational formula of step S205 and step S206) qualitatively, so can significantly reduce the amplitude that abnormal show takes place.
Though the present invention by the preferred embodiment explanation as above, this preferred embodiment is not in order to limit the present invention.Those skilled in the art without departing from the spirit and scope of the present invention, should have the ability this preferred embodiment is made various changes and replenished, so protection scope of the present invention is as the criterion with the scope of claims.
Being simply described as follows of symbol in the accompanying drawing:
A
1..A
19、、C
1..C
19、D
1..D
19: reference pixel
B
1..B
3: pixel R1..R
18: pixel column
L
1、L
2: straight line
O
1、O
2: triangle object
Step S101～S112: the implementation step of one embodiment of the invention
Step S201～S211: the implementation step of another embodiment of the present invention
Claims (12)
1. image describing method, this image describing method is to be used to describe an imaged object, this imaged object comprises a plurality of block row, each block row comprises at least one block, and the size of wherein said block is m * m, and m is a positive integer, and each block respectively comprises a reference pixel, this reference pixel is the top left pixel of each block, it is characterized in that this image describing method comprises:
Define a highest region piece row R of this imaged object
TOP, minimum block row and a present block be listed as;
Set these highest region piece row R
TOPThe most left block be an original block T
INIT, and set this original block T
INITReference pixel be an initial pixel; Calculate the most left block and this original block T of these present block row
INITBetween at the block amount of space Δ x of a directions X and block amount of space Δ y in a Y direction;
Utilize formula A
LEFT=A
INIT+ dA/dx * n * Δ x+dA/dy * n * Δ y is with the image property value A of the reference pixel of the most left block that directly calculates these present block row
LEFT, wherein, A is the image attribute, A
INITIt is the image property value of this initial pixel;
Calculate the block amount of space Δ z between the most left block of the present block row of a present block and this, wherein this present block is positioned at same block with the most left block that should present block be listed as and is listed as; And
Utilize formula A
CURRENT=A
LEFT+ dA/dx * n * Δ z is with the image property value A of the reference pixel that directly calculates this present block
CURRENT, Δ z value is the block amount of space between the most left block of the present block row of this present block and this, and the n value is relevant with the block size.
2. image describing method according to claim 1 is characterized in that, more comprises:
Image property value A at the reference pixel of the most left block that calculates these present block row
LEFTAfter, depict the image property value of the rest of pixels of the most left block that is positioned at these present block row; And
Image property value A at the reference pixel that calculates this present block
CURRENTAfter, depict the image property value of the rest of pixels that is positioned at this present block.
3. image describing method according to claim 1 is characterized in that, more comprises:
Described highest region piece row R in this imaged object of definition
TOP, after described minimum block row and the described present block row, classify these highest region piece row R as when judging this present block
TOPThe time, promptly carry out this original block T
INITWith the setting of this initial pixel and the image property value A of this initial pixel
INITCalculating; And
When judging these present block row is not these highest region piece row R
TOPThe time, promptly carry out the calculating of this block amount of space Δ x and this block amount of space Δ y, and the image property value A of the reference pixel of the most left block of these present block row
LEFTCalculating.
4. image describing method according to claim 1 is characterized in that, more comprises:
In utilizing formula A
LEFT=A
INIT+ dA/dx * n * Δ x+dA/dy * n * Δ y is with the image property value A of the reference pixel of the most left block that directly calculates these present block row
LEFTAfterwards, when this present block is not during a right block of these present block row, then specifying the next one block that moves to right is present block;
Utilize formula A
CURRENT=A
LEFT+ dA/dx * n * Δ z is with the image property value A of the reference pixel that directly calculates this present block
CURRENT, and depict the image property value of the rest of pixels that is positioned at this present block;
Wherein, above-mentioned n value is relevant with the block size; The most left block of these present block row is the original block of this present block.
5. image describing method according to claim 1 is characterized in that, more comprises:
In utilizing formula A
LEFT=A
INIT+ dA/dx * n * Δ x+dA/dy * n * Δ y is with the image property value A of the reference pixel of the most left block that directly calculates these present block row
LEFTAfterwards, when this present block is the rightest block for these present block row, and be during for these minimum block row when judging that this present block is listed as, finish the flow process of above-mentioned image describing method; And
When this present block is the rightest block for these present block row, and when judging these present block row and be not these minimum block row, specify next block to classify present block row as, and continue to describe next block and be listed as;
Wherein, the most left block of these present block row is the original block of this present block.
6. according to claim 1 or 4 described image describing methods, it is characterized in that described n value equals m.
7. image describing method, this image describing method is to be used to describe an imaged object, this imaged object comprises a plurality of block row, each block row comprises at least one block, and the size of wherein said block is m * m, and m is a positive integer, and each block respectively comprises a reference pixel, this reference pixel is the top left pixel of each block, it is characterized in that this image describing method comprises:
Define a highest region piece row R of this imaged object
TOP, minimum block row and a present block be listed as;
Set these highest region piece row R
TOPThe most left block be an original block T
INIT, and set this original block T
INITReference pixel be an initial pixel;
Calculate this present block row and this original block T
INITBetween at the block amount of space Δ y of a Y direction;
Utilize formula A
YTILE=A
INIT+ dA/dx * n * Δ y is directly to calculate the image property value A that lists the reference pixel of vertical common factor with this initial pixel at this present block
YTILE, wherein, A is the image attribute, A
INITIt is the image property value of this initial pixel;
Calculate a present block and this original block T
INITBetween at the block amount of space Δ x of a directions X, wherein this present block is positioned at same block row with reference pixel with the vertical common factor of this initial pixel; And
Utilize formula A
CURRENT=A
YTILE+ dA/dx * n * Δ x is with the image property value A of the reference pixel that directly calculates this present block
CURRENT, wherein the n value is relevant with the block size.
8. image describing method according to claim 7 is characterized in that, more comprises: at the image property value A of the reference pixel that calculates this present block
CURRENTAfter, depict the image property value of the rest of pixels that is positioned at this present block.
9. image describing method according to claim 7 is characterized in that, more comprises:
Described highest region piece row R in this imaged object of definition
TOP, after described minimum block row and the described present block row, classify these highest region piece row R as when judging this present block
TOPThe time, promptly carry out this original block T
INITWith the setting of this initial pixel and the image property value A of this initial pixel
INITCalculating; And
When judging these present block row is not these highest region piece row R
TOPThe time, promptly carry out the calculating of this block amount of space Δ x and this block amount of space Δ y, and with the image property value A of the reference pixel of the vertical common factor of this initial pixel
YTILECalculating.
10. image describing method according to claim 7 is characterized in that, more comprises:
In utilizing formula A
CURRENT=A
YTILE+ dA/dx * n * Δ x is with the image property value A of the reference pixel that directly calculates this present block
CURRENTAfterwards, when this present block is not during a right block of these present block row, then specifying the next one block that moves to right is present block;
Utilize formula A once more
CURRENT=A
YTILE+ dA/dx * n * Δ x is with the image property value A of the reference pixel that directly calculates this block that moves to right
CURRENTAnd
Depict the image property value of the rest of pixels that is positioned at this block that moves to right;
Wherein, above-mentioned n value is relevant with the block size.
11. image describing method according to claim 7 is characterized in that, more comprises:
In utilizing formula A
CURRENT=A
YTILE+ dA/dx * n * Δ x is with the image property value A of the reference pixel that directly calculates this present block
CURRENTAfterwards, when this present block is the rightest block for these present block row, and be during for minimum block row when judging that this present block is listed as, finish the flow process of above-mentioned image describing method; And
When this present block is the rightest block for these present block row, and when judging these present block row and be not these minimum block row, specify next block to classify present block row as, and continue to describe next block and be listed as.
12., it is characterized in that described n value equals m according to claim 7 or 10 described image describing methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007100009843A CN100533485C (en) | 2007-01-17 | 2007-01-17 | Image describing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007100009843A CN100533485C (en) | 2007-01-17 | 2007-01-17 | Image describing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101000693A CN101000693A (en) | 2007-07-18 |
| CN100533485C true CN100533485C (en) | 2009-08-26 |
Family
ID=38692651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2007100009843A Active CN100533485C (en) | 2007-01-17 | 2007-01-17 | Image describing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100533485C (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1248032A (en) * | 1998-06-08 | 2000-03-22 | 松下电器产业株式会社 | Graphic obliterator |
| US6377265B1 (en) * | 1999-02-12 | 2002-04-23 | Creative Technology, Ltd. | Digital differential analyzer |
| US6510442B1 (en) * | 2000-01-12 | 2003-01-21 | Microsoft Corporation | System and method for improved digital differential analyzer |
| US6556203B1 (en) * | 2000-06-30 | 2003-04-29 | Intel Corporation | Tile-based digital differential analyzer rasterization |
| CN1491401A (en) * | 2001-12-12 | 2004-04-21 | ���ṫ˾ | Image processing apparatus and method thereof |
-
2007
- 2007-01-17 CN CNB2007100009843A patent/CN100533485C/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1248032A (en) * | 1998-06-08 | 2000-03-22 | 松下电器产业株式会社 | Graphic obliterator |
| US6377265B1 (en) * | 1999-02-12 | 2002-04-23 | Creative Technology, Ltd. | Digital differential analyzer |
| US6510442B1 (en) * | 2000-01-12 | 2003-01-21 | Microsoft Corporation | System and method for improved digital differential analyzer |
| US6556203B1 (en) * | 2000-06-30 | 2003-04-29 | Intel Corporation | Tile-based digital differential analyzer rasterization |
| CN1491401A (en) * | 2001-12-12 | 2004-04-21 | ���ṫ˾ | Image processing apparatus and method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101000693A (en) | 2007-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5538435B2 (en) | Image feature extraction method and system | |
| US20220044474A1 (en) | Method for constructing grid map by using binocular stereo camera | |
| CN109992226A (en) | Image display method and device and spliced display screen | |
| CN103838437A (en) | Touch positioning control method based on projection image | |
| CN112154445A (en) | Method and device for determining lane line in high-precision map | |
| CN104035664A (en) | Display method and device for user interface of application program | |
| CN104183016A (en) | Method for rapidly constructing 2.5-dimensional building model | |
| CN103984720A (en) | Tile map establishment method and device based on OpenGL (open graphics library) | |
| CN103686064A (en) | Picture segment display method and client-side | |
| CN102957895A (en) | Satellite map based global mosaic video monitoring display method | |
| CN105447815A (en) | Method and device for processing image of mobile terminal equipment | |
| CN104346779A (en) | Image correction method and apparatus using creation of feature points | |
| US20130162674A1 (en) | Information processing terminal, information processing method, and program | |
| KR102023087B1 (en) | Method for camera calibration | |
| CN111698491B (en) | Multi-projection image display method and device, electronic equipment and storage medium | |
| US20210201522A1 (en) | System and method of selecting a complementary image from a plurality of images for 3d geometry extraction | |
| JP6636963B2 (en) | Image processing apparatus and image processing method | |
| CN100533485C (en) | Image describing method | |
| CN107608670B (en) | Method and system for self-adaptive display of report UI (user interface) elements | |
| CN111612838A (en) | Image acquisition method and device and electronic equipment | |
| CN113535878B (en) | Line selection method, device, terminal and storage medium based on three-dimensional geological modeling | |
| CN102903097B (en) | Method and device for image perspective correction | |
| US20140348416A1 (en) | Stereo image rectification apparatus and method | |
| CN110083947B (en) | Method for constructing GIS switch three-dimensional solid model based on reverse engineering technology | |
| US6529642B1 (en) | Method for enlarging/reducing digital images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |