JP2011039664A - Line segment drawing processing method and device, and pickup image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a line segment drawing processing method for shortening the processing time of a line segment whose inclination is especially small. <P>SOLUTION: The inclination of a line segment is checked, and the end point pixel of a scanning direction line segment is retrieved by binary search to a line segment having an inclination whose scale is equal to or less than a prescribed value. In one embodiment, when the x directional difference and y directional difference of the start point/end point of the line segment are respectively defined as dx and dy, the inclination dy/dx of the line segment is calculated, and whether dy/dx≤α is satisfied is determined, and when the above formula is satisfied, the binary search is applied. In the binary search, the position of a pixel under current consideration is moved to an x direction by Δx<SB>p</SB>, and an error e<SB>p</SB>=Δx<SB>p</SB>*dy/dx in the moved pixel is calculated, and when the error e<SB>p</SB>is 1≤e<SB>p</SB>≤2, the moved pixel is defined as the terminal pixel of the scanning direction line segment, and when the error e<SB>p</SB>is 0<e<SB>p</SB><1 or e<SB>p</SB>≥2, Δx<SB>p</SB>←1/2*Δx<SB>p</SB>is applied, and the pixel is further moved to a direction where x coordinates become large or small, and this is repeated until 1≤e<SB>p</SB><2 is satisfied. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a line segment drawing processing method and apparatus suitable for displaying an arbitrary line segment superimposed on a captured image, and a captured image processing apparatus to which the line segment drawing processing method and apparatus are applied.

  As one of the functions of an imaging apparatus such as a camera, a function that supports a user operation or the like by superimposing and displaying an arbitrary line segment on a captured image is now common. For example, in an in-vehicle camera, a vehicle width line is displayed by superimposing it on an image obtained by photographing the rear of the vehicle.

  Conventionally, there is a Bresenham method as one of algorithms for obtaining each point (pixel) on a line segment connected by two arbitrary points on a digital coordinate plane. This Bresenham method is known as an algorithm with a wide application range because it does not require multiplication / division and can draw a line segment only by addition / subtraction.

  In the Bresenham method, a process for obtaining a line segment in the scanning direction of each row constituting the line segment is performed. Specifically, this is a process of searching for the end pixel from the start pixel in the scanning direction line segment. In this case, in the conventional Bresenham method, a so-called linear search is performed in which a search is performed pixel by pixel to search for the end pixel from the start pixel in the scanning direction line segment, and whether the pixel is the end pixel is a value called an error term. And comparing this with a threshold value. However, in the linear search for each pixel, there is a problem that if the line segment in the scanning direction is long, it takes a long time to search and the processing time becomes long as a whole.

  In addition, as a modification of the presentation method, processing of line segments is processed in units of blocks, and the coordinate value and error term of the drawing start point are independently calculated for each block, thereby reducing processing time. Has also been proposed (Patent Document 1). However, this method also uses the Bresenham algorithm for line segment drawing processing in each block area, so there is a limit to shortening the processing time, and the error term of the drawing start point of the block is obtained using division. Therefore, there is a problem that the advantage of the Bresenham method is impaired.

  In order to assist the user's operation or the like, in the case of a line segment displayed so as to be superimposed on a photographed image, the line segment has a small inclination in many cases.

  It is an object of the present invention to provide a line drawing processing method and apparatus capable of shortening the drawing processing time of a line having a particularly small inclination, and an image processing apparatus to which the line drawing processing method and apparatus are applied.

  Furthermore, the present invention provides a line segment drawing processing method and apparatus for realizing line segment drawing processing in a small resource environment without performing complicated division, and a captured image processing apparatus to which the line segment drawing process is applied. With the goal.

  According to the first aspect of the present invention, a start point and an end point of a line segment are designated on a two-dimensional screen composed of a large number of pixels, and a line segment in the scanning direction of each line constituting the line segment from the start point to the end point is specified. In the line segment drawing processing method for drawing a line segment, it is determined whether or not the slope of the line segment with respect to the scanning direction is equal to or less than a predetermined value. When a binary search is performed from the start pixel of the scanning direction line segment of each line of the screen to search for the end pixel of the scanning direction line segment, and when the inclination of the line segment is not less than or equal to a predetermined value, each line of the screen constituting the line segment A search is performed pixel by pixel from the start pixel of the scanning direction line segment, and a linear search is performed to search for an end pixel of the scanning direction line segment.

  According to a second aspect of the present invention, in the line segment drawing processing method according to the first aspect, the screen scanning direction is the x direction, the x direction difference dx and the y direction difference dy of the start point and the end point of the line segment, and the line segment inclination dy / Dx is calculated to determine whether dy / dx ≦ α is satisfied, and when satisfied, a binary search is applied.

  The invention according to claim 3 is the line drawing processing method according to claim 2, wherein the predetermined value α is at least 1/8 or less.

According to a fourth aspect of the present invention, in the line segment drawing processing method according to the second or third aspect, in the binary search, the position of the pixel of interest is moved by Δx _p in the x direction, and the error _{e p = Δx p * dy /} dx calculated, the error _{e p} if the 1 ≦ _e p <2, the pixel after the mobile and end pixels in the scanning direction line, the error _{e p} is 0 _{<e p} If <1 or e _p ≧ 2, Δx _p ← 1/2 * Δx _p is set, and the pixel is further moved in the direction in which the x coordinate becomes larger or smaller, and this is repeated until 1 ≦ e _p <2. It is characterized by that.

According to a fifth aspect of the present invention, in the line segment drawing processing method according to the first aspect, the screen scanning direction is the x direction, the x direction difference dx and the y direction difference dy of the start point and the end point of the line segment are 2 ^k * dy ≦ It is determined whether dx (k is a constant) is satisfied, and when satisfied, a binary search is applied.

  The invention according to claim 6 is the line segment drawing processing method according to claim 5, wherein the constant k is a value of at least 3 or more.

According to a seventh aspect of the present invention, in the line segment drawing processing method according to the fifth or sixth aspect, in the binary search, the position of the pixel of interest is moved by Δx _p in the x direction, and The error E _p = Δx _p * dy is calculated, and if the error E _p is dx ≦ E _p <2 * dx, the pixel after the movement is set as the end pixel of the scanning direction line segment, and the error E _p is 0 <E _p If <dx or E _p ≧ 2 * dx, Δx _p ← 1/2 * Δx _p is set, and the pixel is further moved in the direction in which the x coordinate becomes larger or smaller, and this is dx ≦ E _p <2 * Δx _It repeats until it becomes _p , It is characterized by the above-mentioned.

The invention of claim 8 is the line drawing processing method according to claim 4 or 7, the initial value of the x-direction movement amount [Delta] x _p is that more than half of the value of the number of pixels x direction of at least the screen Features.

According to a ninth aspect of the present invention, in the line segment drawing processing method according to the eighth aspect, the value of the initial value of the x-direction movement amount Δx _p is gradually decreased as the processing target line approaches the end line of the line segment. It is characterized by that.

  A tenth aspect of the present invention is characterized by a line segment drawing processing apparatus that implements the line segment drawing processing method according to any one of the first to ninth aspects.

  According to an eleventh aspect of the present invention, there is provided an image processing unit that processes a captured image captured by an imaging device, a line segment processing unit that generates a line segment, and a composite that generates a composite image in which the line segment is superimposed on the captured image. And a line segment drawing processing apparatus comprising the display means for displaying the composite image, wherein the line segment processing means comprises the line drawing processing apparatus according to claim 10.

In the present invention, when the line segment has a particularly small inclination, the process for searching for the start pixel from the start pixel to the line segment in the scanning direction of each line of the screen constituting the line segment is performed by performing a binary search instead of a linear search. Time can be shortened. Further, [Delta] x, to determine the magnitude of the slope of the line by comparing the [Delta] y, by defining a new error E _p, and unnecessary division, can drawing process of a line at a low cost small resource environment become. In addition, the further effect of this invention other than this becomes clear by description of a following example.

It is a whole block diagram of one Example of the captured image processing apparatus to which the line segment drawing process function of this invention is applied. It is a figure explaining the view of the line segment drawing processing method of this invention. It is a figure which shows the basic processing flow of the line segment drawing processing method of this invention. It is a figure which shows the specific processing image of the binary search utilized by this invention. It is a figure which shows an example of the line segment of the various inclination which can apply this invention. It is a figure which shows the whole process flow of the line segment process part of FIG. It is a figure which shows the schematic processing flow of the V _sync process in FIG. FIG. 7 is a diagram showing a schematic processing flow of H _ref processing in FIG. 6. It is a figure which shows the detailed processing flow of the binary search process in FIG. It is a figure which shows the general _| schematic process flow of V _sync process in case a process target line segment is plurality. It is a figure which shows the general _| schematic processing flow of _Href processing in case a process target line segment is plurality. It is the table | surface which compared the number of instructions of a linear search and a binary search about the line segment of various inclinations.

First, the line segment drawing processing method of the present invention will be described. Now, as shown in FIG. 2A, a line segment 1 connecting a start point A and an end point B is displayed on an xy coordinate plane (screen) composed of a large number of pixels (pixels). Let us consider a case where drawing is performed by repeatedly performing the process of obtaining the scanning direction line segment. The coordinates of the start point A are (x _s , y _s ), the coordinates of the end point B are (x _e , y _e ), and the screen scanning direction is the x direction. Slope dy / dx of the line segment 1 is represented by _{(y e -y s) / (} x e -x s). This inclination dy / dx is compared with a predetermined value α. Although α will be described later, for example, α is suitably about 1/8. Here, when dy / dx> α (the slope of the line segment to be drawn is large), an algorithm such as a conventional Bresenham method is applied to obtain the scanning line segment of each line of the screen constituting the line segment 1. Then, a so-called linear search is sequentially performed pixel by pixel from the start pixel of the scanning direction line segment to search for the end pixel. On the other hand, when dy / dx ≦ α (the slope of the line segment to be drawn is small), the end pixel is searched by a binary search with reference to the start pixel of the scanning line segment. This is a feature of the present invention. Hereinafter, the binary search will be described in detail.

When a certain point P existing on a line segment having an inclination dy / dx is shifted by one pixel in the main scanning direction (x direction), the dy / dx pixel is shifted in the sub scanning direction (y direction), but the inclination dy / dx If is small, the point will remain on the line segment. Now, when the offset [Delta] x _p pixels the point P in the main scanning direction is called sub-scanning direction of the displacement for the points (pixels) are present on a line with the error e _p. In other words, the error _{e p} is represented by _{e p = Δx p * dy /} dx (* multiplication). In row currently being processed, shifting [Delta] x _p pixels starting from the pixel (and P) in the main scanning direction of the scanning direction line, error ep is 1 ≦ e _{p <2} to become such points at that time (pixels) , The position of the end pixel of the line segment in the scanning direction.

  Note that the start pixel of the scanning direction line segment of the processing target row is uniquely determined from the position of the end pixel of the scanning direction line segment of the previous row. In addition, the start pixel of the line in the scanning direction of the first line to be drawn is the pixel of the start point of the line to be drawn (A in FIG. 2A).

In FIG. 2B, the processing target row is the first row (y _s ) of the line segment 1 to be drawn, the pixel at the start point A is the start pixel P of the scanning direction line segment of the row, and the main scanning direction (x It shows the relationship between the line 1 and the error e _p in the pixel P 'by shifting [Delta] x _p pixels in the direction). Error e _p is determined whether satisfy 1 ≦ e _{p <2,} if not satisfied, while reducing by half the moving amount, in accordance with the magnitude of the error e _p, or the positive x-axis the position of the pixel P ' It repeats the process of moving in the negative direction, to search for the position of the pixel P 'of the error e _p satisfies 1 ≦ e _{p <2} (binary search). If the error e _p has satisfy the 1 ≦ e _{p <2,} the position of the pixel P 'at that time as an end pixel in the scanning direction line of the line (in FIG. 2 _(b) y _s), pixel P from the pixel A The pixel group up to 'is determined as the line segment in the scanning direction of the row. In the next row, the start pixel of the scan direction line segment of the previous row is uniquely determined based on the end pixel of the scan direction line segment of the previous row, and the binary search is performed in the same manner on the basis of the start pixel. Search for a pixel. Thereafter, the same processing is repeated up to the end point B row (y _e ).

  FIG. 3 shows the basic processing flow of the line segment drawing processing method of the present invention. This processing flow is applied when the slope dy / dx of the line segment to be drawn is dy / dx ≦ α (the slope of the line segment to be drawn is small). Here, it is assumed that it is already determined that the slope of the line segment to be drawn is dy / dx ≦ α (α is, for example, 1/8).

Now, let P be the start pixel of the line in the scanning direction of the row to be processed, and x and y coordinates of the pixel P be represented by x _p and y _p . First, an initial value of the movement amount Δx _p of the pixel P in the x direction is set (step 101). The initial value is set to a factorial value of 2 in order to reduce calculation cost. The size of the initial value may be a value that is half or more of the number of pixels in the main scanning direction of the screen. For example, when the screen size is 180 (number of pixels in the main scanning direction) × 120 (number of pixels in the sub scanning direction), the initial value of Δx _p is 2 ⁷ (= 128). Note that the value of the initial value may be sequentially decreased every time the number of processed rows reaches a predetermined number, that is, as the end of the line segment is approached. This makes it possible to reduce the number of binary searches.

Next, [Delta] x _p is moved (step 102) the position of the pixel P in the x direction, the pixel after movement P (for convenience, and P ') for calculating an error _{e p} at the position (step 103). As mentioned earlier, the error _{e p} is represented by _{e p = Δx p * dy /} dx.

Next, it is determined the magnitude of the error _{e p} (step 104). Here, if 1 ≦ e _p <2, the pixel group from the pixel P to the pixel P ′ is determined as the scanning direction line segment of the row, with the pixel P ′ as the end pixel of the scanning direction line segment of the row ( Step 105), the processing of the row is completed. Then, the process moves to the next line.

On the other hand, if 0 <e _p <1, a movement amount Δx _p is set such that the pixel P ′ is moved in the direction (positive direction) in which the x coordinate increases from the current position of the pixel P ′ (step 106). At this time, the movement amount Δx _p is set to the previous half. After setting the movement amount Δx _p , the process returns to step 102. If e _p ≧ 2, a movement amount Δx _p is set so as to move the pixel P ′ from the current position of the pixel P ′ in the direction in which the x coordinate decreases (negative direction) (step 107). Also at this time, the movement amount Δx _p is set to the former half. After setting the movement amount [Delta] x _p, likewise returns to step 102. Thereafter, until 1 ≦ _e p <2, the steps 102,103,104,106 or 107 loop, proceed to 1 ≦ _e p <Step 105 if became 2.

  FIG. 4 shows a specific processing image of the processing flow of FIG. FIG. 4A shows the entire screen of the xy coordinate plane composed of a large number of pixels, and FIG. 4B shows an enlarged portion (portion for drawing a line segment). Again, the screen scanning direction is the x direction.

  Consider a case where the starting point of a line segment to be drawn is A, the end point is B, the line drawing process up to y = n−1 lines is completed, and the line drawing process on y = n lines is started. Note that the black lines in FIG. 4A indicate the drawn line segments.

In FIG. 4 (b), P represents the start pixel in the scanning direction line of y = n rows, x-coordinate of the pixel P, respectively y coordinates _x p, and _{y p, (x p, y} p) = (M, n). Pixel P can be uniquely determined from the position of the previous line y = n-1 of the end pixel p _e in the scanning direction line (m-1, n-1 ). First, by setting the initial value of the movement amount [Delta] x _p, move [Delta] x _p pixels x coordinate positive direction from the pixel P, the pixel P ₁ after movement, determine the error e _p, said error e _p is 1 ≦ e _{It is} determined whether _p <2 is satisfied. Since it was not satisfied here and e _p ≧ 2, next, the previous half Δx _p pixel is moved in the negative x coordinate direction from the pixel P ₁ , and the error e _p is obtained for the pixel P ₂ after the movement, said error _{e p} determines whether satisfies 1 ≦ _e p <2. Again, since e _p ≧ 2, it was still ep ≧ 2, and again moved from the pixel P _{2 in the} negative x coordinate direction by a quarter ∆x _p pixel, and the error e _p for the pixel P ₃ after the movement. look, determine whether said error _{e p} satisfies 1 ≦ _e p <2. Again, because 0 <e _p <1, this time, the pixel P ₃ moves from the pixel P _{3 in the} positive x-coordinate in the previous half by 1 / 8Δx _p pixels, and the error e _p for the pixel P ₄ after the movement look, determine whether said error _{e p} satisfies 1 ≦ _e p <2. Since we are satisfied here, the pixel P ₄ (m ′, n) is the terminal pixel of the line segment in the scanning direction of y = n rows, and the pixel group (pixels from P to P _{4) in} the range of m ≦ x <m ′. ) Is determined as a line segment in the scanning direction at y = n rows. Then, these pixel groups are filled.

Next, the processing proceeds to the y = n + 1 row, and the start pixel P (x _p , y _p ) of the scanning direction line segment is positioned at the position (m ′, line) of the line segment end pixel P4 in the previous y = n row. n), P (x _p , y _p ) = (m ′ + 1, n + 1) is determined.

The basic idea of the line segment drawing processing method of the present invention has been described above. However, the inclination and direction of the line segment to be drawn are not limited to those shown in FIG. 4, for example, FIGS. ) Is also considered. In either case, A is the start point and B is the end point. Here, in the case of FIG. 5A, in the target row, first, the start pixel of the line in the scanning direction of the row may be moved by Δx _p pixels in the negative x coordinate direction. The subsequent processing is the same as in FIG. In the case of FIGS. 5B and 5C, the main scanning direction may be the y direction.

  When the line segment has a particularly small slope, the binary search method as described above is used to search the end pixel of the scan direction line segment for each row, thereby starting the scan direction line segment as in the conventional Bresenham method. The processing time can be shortened as compared with the case where an error term is obtained pixel by pixel and the end pixel is searched (linear search).

  Next, specific examples of the line segment drawing processing method of the present invention will be described.

  FIG. 1 is a diagram showing an overall configuration of an embodiment of a captured image processing apparatus having a line segment drawing processing function of the present invention. The captured image processing apparatus includes an operation unit and the like, which are omitted in FIG.

Image data captured by the imaging device 10 such as a camera is sent to the image processing unit 20. The image data is generally color image data, but may be monochrome image data. Here, the imaging device 10 also sends out a synchronization signal V _sync indicating the start of the screen (frame) and a synchronization signal H _ref indicating the start of each row of the screen (frame) together with the image data.

The image processing unit 20 performs predetermined image processing on the image data sent from the imaging device 10. The predetermined image processing includes, for example, Bayer interpolation processing, magnification chromatic aberration correction processing, YUV conversion processing, MTF correction processing, distortion aberration correction processing, and γ correction processing. These image processes are performed in synchronization with the synchronization signals V _sync , H _{ref and the} like, and the image data of the processing result is stored in the frame memory 50. Since each image processing is generally performed sequentially, the data flow between the image processing unit 20 and the frame memory 50 is bidirectional.

The line segment processing unit 30 receives the synchronization signals V _sync and H _ref from the image processing unit 20, and executes line segment processing in synchronization with the synchronization signal. The start point / end point coordinates of the line segment are stored in the line segment memory 40 in advance. One or more line segments may be drawn. The line segment processing unit 30 reads the start point / end point coordinates of the line segment to be processed from the line memory 40 in synchronization with the synchronization signal V _sync for each frame, and the inclination of the line segment, the start pixel of the scan direction line segment, etc. After the initial value calculation is performed, the pixels of the line segment to be drawn (scan direction line segment) are determined for each row of the frame in synchronization with the synchronization signal H _ref . The result is held in the line segment memory 40. Here, the value held in the line segment memory 40 is data representing the display color of each pixel in the scanning direction line segment, and corresponds to each pixel in the scanning direction line segment when the line segment is displayed in red. Red pixel data is held at the memory address.

  The synthesizing unit 60 inputs pixel data representing the display color of each pixel in the scanning direction line segment held in the line segment memory 40 and image data held in the frame memory 50 and inputs the line segment on the image. To superimpose. The display device 70 displays the captured image output from the synthesis unit 60 and the superimposed image of the line segment on the monitor. For example, the line segment is displayed in red on the captured image. In addition, it is also possible to display by color classification for each type of line segment.

  Here, the feature of the present invention resides in the line segment processing unit 30. The line segment processing unit 30 includes, for example, a microprocessor, and executes line segment drawing processing using an assembler language or the like. Hereinafter, the line segment drawing process in the line segment processing unit 30 will be described in detail.

FIG. 6 shows an overall processing flow of the line processing unit 30. The line segment drawing processing in the line segment processing unit 30 needs to be performed in synchronization with the processing on the image processing unit 20 side. Therefore, the line segment processing unit 30 waits for a synchronization signal from the image processing unit 20 (step 201). When a synchronization signal is received from the image processing unit 20, it is determined whether the synchronization signal is a V _sync signal or a H _ref signal (step 202). The V _sync signal is a synchronization signal indicating the start of the screen (frame), and the _Href signal is a signal indicating the start of each row of the screen. When the V _sync signal is received, the V _sync process is executed (step 203), and when the H _ref signal is received, the H _ref process is executed (step 204). Thereafter, the process returns to step 201. Here, in the V _sync processing, preprocessing for line segment drawing processing is mainly executed. In the H _ref process, a line drawing process in the scanning direction for each row of the frame is executed.

FIG. 7 is a schematic process flow of the V _sync process. When the V _sync signal indicating the start of the screen (frame) is turned on (step 301), the line segment processing unit 30 calculates an initial value for line segment drawing processing (step 302). The initial value is the slope of the line segment to be drawn, the coordinates of the start pixel of the first line in the scanning direction, and the like. The line segment processing unit 30 reads the start point coordinates (x _s , y _s ) and end point coordinates (x _e , y _e ) from the line segment memory 40 to calculate the slope dy / dx and the start point. The coordinates (x _s , y _s ) are set as the start pixel coordinates of the line segment in the scanning direction of the first line to be drawn. Then, initial values such as the slope of the line segment to be drawn and the start pixel coordinates are stored in the line segment memory 40 (step 303).

Here, instead of the line of the slope _{dy / dx, dx = x e} -x s, dy = y e -y s
Is stored in the line segment memory 40. Thereby, it is not necessary to calculate dy / dx, and division can be omitted.

FIG. 8 is a schematic process flow of the H _ref process. When the H _ref signal indicating the start of each row of the screen (frame) is turned on (step 401), the line segment processing unit 30 reads an initial value from the line segment memory 40 (step 402). Then, it is determined whether the slope dy / dx of the line segment is equal to or smaller than a predetermined value α (step 403). Here, it is determined whether 2 ^k * dy ≦ dx so that division is not required. That is, dx is compared with a value obtained by shifting dy by k bits, and if dx is equal or large, it can be determined that the slope of the line segment is small with respect to the main scanning direction. For example, α = 1/8 corresponds to the case of k = 3.

When the slope dy / dx of the line segment is dy / dx ≦ α, that is, 2 ^k * dy ≦ dx, a search is performed from the start pixel of the line segment in the scanning direction by the binary search process. Starting, the end pixel of the scanning direction line segment is obtained (step 404). Then, the pixel data is stored in the line segment memory 40 with the start pixel and the end pixel of the line segment in the scanning direction and the pixels sandwiched between them as line drawing pixels (step 406). For example, pixel data such as red is stored at a corresponding memory address in the line segment memory 40. Further, the coordinate value of the position shifted by one pixel in the scanning direction from the scanning direction position of the end pixel of the scanning direction line segment is similarly stored in the line segment memory 40. On the other hand, when the slope dy / dx of the line segment is dy / dx> α, that is, when 2 ^k * dy> dx, the Bresenham method or the like is applied and the line of interest is subjected to the linear search process. A search is performed pixel by pixel from the start pixel of the scanning direction line to obtain a terminal pixel of the scanning direction line (step 405). Then, the process proceeds to Step 406.

  The binary search process and the linear search process actually start from the start line of the line segment to be drawn (the first line of the line segment) and end at the end line. When the process is the first line of the line segment, the pixel corresponding to the start point coordinates of the process target line segment stored as the initial value in the line segment memory 40 is used as the start pixel of the first line in the scanning direction. In each subsequent row, the pixels corresponding to the coordinate value of the position shifted by one pixel in the scanning direction from the scanning direction position of the terminal pixel of the previous line in the scanning direction line segment, which are sequentially stored in the line segment memory in step 406. Is used as the start pixel of the line in the scanning direction of the row.

In FIG. 8, the slope determination of the line segment in step 403 is executed only when the H _ref signal is first turned on, that is, only when the processing target line returns to the top line of the screen. Thereafter, step 403 may be skipped and the process may immediately proceed to step 404 or 405. Further, it is also possible to execute line segment inclination determination in the initial value calculation in step 302 of FIG. 7 and store the determination result in the line segment memory 40.

FIG. 9 shows a specific processing flow for the binary search process (step 404) of FIG. This is a process flow basically the same as that of FIG. 3 before, by defining a new error E _p in place of the error e _p, is obtained by eliminating the need for division of the slope dy / dx. Further, the line segment to be drawn is a line segment as shown in FIG.

The start pixel of the scanning direction line segment of the row to be processed is represented by P, and the x coordinate and y coordinate of the pixel P are represented by x _p and y _p . First, an initial value of the movement amount Δx _p of the pixel P in the x direction is set (step 501). Here, the initial value of [Delta] x _p is ^{2 7.} As described above, the initial value of Δx _p is preferably set to a value that is half or more of the number of pixels in the main scanning direction of the screen. Alternatively, the value may be sequentially decreased every time a predetermined number of processed rows are reached.

Next, the position of the pixel P is moved by Δx _p in the x direction (step 502). Then, an error at the pixel P ′ is calculated in order to check whether or not the pixel P after the movement (for convenience, P ′) is the terminal pixel of the line in the scanning direction of the row (step 503).

As described above, the error is expressed as e _p = Δx _p * dy / dx. It said error e _p is 1 ≦ e _{p <2} to become such points (pixels), the end pixel in the scanning direction line of the row. However, division is required to obtain the value of dy / dx. Here, in order to define 2 ^k * dy ≦ dx for the determination of the inclination of the previous line segment and to eliminate the need for division, here, a new error E _p = dx * e _p = Δx _p * e is substituted for the error e _p. Define _y . By obtaining a point such said error E _p is dx ≦ E _{p <2} * dx (in pixels), it is the end pixel in the scanning direction line of the row. By thus defining the error E _p, it does not require the value of the slope dy / dx, without division can be obtained end pixel in the scanning direction line.

Next, the magnitude of the error E _p, or to terminate the binary search processing, determines whether to continue (step 504). Here, if dx ≦ E _p <2 * dx, the pixel group from the pixel P to the pixel P ′ is determined as the scanning direction line segment of the row, with the pixel P ′ as the terminal pixel of the scanning direction line segment of the row. (Step 505). If 0 <E _p <dx, a movement amount Δx _p is set such that the pixel P ′ is moved in the direction (positive direction) in which the x coordinate increases from the current position of the pixel P ′ (step 506). At this time, the movement amount Δx _p is set to the previous half. After setting the movement amount Δx _p , the process returns to Step 502. If E _p ≧ 2 * dx, a movement amount Δx _p is set so as to move the pixel P ′ from the current position of the pixel P ′ in the direction in which the x coordinate decreases (negative direction) (step 507). Also at this time, the movement amount Δx _p is set to the previous half. After setting the movement amount [Delta] x _p, likewise returns to step 502.

  Up to now, the processing target line segment has been described as one, but in general, the number of line segments to be superimposed on an image is not necessarily one. The line drawing process in the line processing unit 30 when the processing target line is expanded to a plurality will be described below. The overall processing flow is the same as in FIG.

FIG. 10 is a processing flow when the line processing unit 30 receives the V _sync signal. It is assumed that the total number of processing target line segments and the coordinates of the start point and end point of each line segment are stored in advance in the line segment memory 40.

  First, n is initially set to 1 (step 601). Next, for the nth line segment, the coordinates of the start point and end point are read from the line segment memory 40 (step 602), and the line property (LP) is calculated (step 603). Here, various values frequently used in the line segment drawing process are referred to as line properties (LP).

Specific values of the line property (LP)
A value indicating the direction of inclination: StepX
Y coordinate of the end point: y _e
X direction difference between start and end points: dx
Difference in Y direction between start point and end point: dy
X coordinate of line drawing point: D _raw X
Y coordinate of line drawing point: D _raw Y
and so on.

StepX is a value indicating the direction of inclination (direction of “/” or “\”). With this value, the direction from the start point to the end point of the line segment can be known. Used when updating D _raw X. y _e is used when the straight line drawing process ends. dx and dy are used when determining the magnitude of the slope of the straight line. D _raw X and D _raw Y represent the current X coordinate and Y coordinate of the line in the scanning direction of the line of interest in the line currently being drawn, and the values are updated during the drawing. A line segment is drawn along a point (pixel) indicated by the coordinates (D _raw X, D _raw Y).

  Here, the line property (LP) is calculated as follows.

StepX = (Signal value of end point X coordinate-start point X coordinate) (1 or -1)
y _e = Y coordinate of the end point
dx = | End point X coordinate-Start point X coordinate |
dy = | Y coordinate of the end point−Y coordinate of the start point |
D _raw X = X coordinate of the start point of the line segment
D _raw Y = the Y coordinate of the start point of the line segment Note that Step X, y _e , dx and dy are treated as constants through the drawing process of the line segment to be processed, but D _raw X and D _raw Y are used in the LP update process described later. When the value is changed. D _raw X and D _raw Y set in step 603 are initial values.

Next, the line property (LP) calculated in step 603 exists in the line segment memory 40. Then, it is determined whether or not the line properties of all the processing target line segments have been calculated (step 605). If there are unprocessed lines, n is incremented by 1 (step 606), and the process returns to step 602. The line property of the line segment to be processed is calculated, and the V _sync process ends.

FIG. 11 shows a processing flow when the line processing unit 30 receives the H _ref signal. First, n is initially set to 1 (step 701). Next, for the n-th line segment, the current line property (LP) is read from the line segment memory 40 (step 702). Next, it is determined whether or not D _raw Y matches the currently processed row (sF) (step 703). If sF ≠ D _raw Y, since it is not necessary to start the line segment drawing process for the line currently being processed, the process goes to step 710 as it is.

On the other hand, if sF = D _raw Y, the line segment drawing process for the line currently being processed is started. First, the magnitude of the inclination of the line segment is determined (step 704). Here, as described above, the determination is performed using 2 ^k * dy ≦ dx (k is a constant) instead of dy / dx ≦ α. That is, if 2 ^k * dy ≦ dx, it can be determined that the line segment has a particularly small inclination with respect to the main scanning direction. If 2 ^k * dy ≦ dx, a line segment in the scanning direction of the row currently being processed is obtained by the binary search process (step 705). This is the same as FIG. That is, x _p in FIG. 9 may be replaced with D _raw X. When 2 ^k * dy> dx, the line segment in the scanning direction of the line currently being processed is obtained by linear search processing using the conventional Bresenham method or the like (step 706). Then, the pixel data (for example, red data) of the pixel group corresponding to the scanning direction line segment of the processing target row obtained in step 705 or 706 is written to the corresponding address of the line segment memory 40 (step 707).

Next, it is determined whether or not D _raw Y matches the Y coordinate y _e of the end point of the line segment (step 708). If D _raw Y = _{y e,} it goes to step 710. If D _raw Y ≠ y _e , the values of D _raw X and D _raw Y are updated as follows for the line segment drawing process in the next row of the line segment (step 709).
D _raw X ← D _raw X + Step X
D _raw Y ← D _raw Y + 1
These values become the coordinate values of the start pixel of the line segment in the scanning direction of the next row in the line segment. After the update, it is stored in the line segment memory 40.

  Thereafter, it is determined whether or not all the processing target line segments have been processed (step 710). If there is an unprocessed line segment, n is incremented by 1 (step 711), and the process returns to step 702. Thereafter, the processing in steps 702 to 711 is repeated by the number of processing target line segments.

Next, a specific value of the threshold α of the line segment inclination dy / dx used when the binary search process and the linear search process are divided into cases will be described. FIG. 12 shows a comparison of the number of instructions when drawing by linear search processing and binary search processing for various slope line segments. Here, the normal Bresenham algorithm was used for the linear search. Also, the screen size is 180 (the main scanning direction pixel number) × 120 (number of sub-scanning direction pixel), a binary search process was 2 ⁷ the initial value of [Delta] x _p. The number of instructions means the maximum number of instructions among the number of instructions required for line drawing processing of each row. From FIG. 12, it is understood that the binary search process may be applied when the slope dy / dx of the line segment is 1/8 = 0.125 or less. As described above, α = 1/2 ^k , and α = 1/8 corresponds to the case of k = 3. Therefore, when 2 ³ * dy ≦ dx, the binary search process may be applied. In this case, it is possible to determine whether the slope of the line segment is particularly small by simply comparing the value obtained by bit-shifting dy with dx, and division is not necessary.

Needless to say, the threshold value α of the slope of the line segment and the optimum value of 1/2 ^k , which are judgment criteria for dividing the binary search process and the linear search process into cases, are not limited to the above specific examples.

  Moreover, although the example which superimposes and displays a line segment on a captured image was shown as concrete embodiment of this invention, application of the line segment drawing processing method of this invention is not limited to this.

DESCRIPTION OF SYMBOLS 10 Imaging device 20 Image processing part 30 Line segment process part 40 Line segment memory 50 Frame memory 60 Composition part 70 Display apparatus

JP 2003-317108 A

Claims (11)

Specify the start and end points of a line segment on a two-dimensional screen composed of a large number of pixels, and draw the line segment from the start point toward the end point by finding the line segment in the scanning direction of each line that makes up the line segment. In the line segment drawing processing method to
Determine whether the slope of the line segment with respect to the scanning direction is less than or equal to a predetermined value,
When the slope of the line segment is equal to or less than a predetermined value, a binary search is performed from the start pixel of the scan direction line segment of each screen row constituting the line segment to search for the end pixel of the scan direction line segment,
If the slope of the line segment is not less than or equal to a predetermined value, a search is performed pixel by pixel from the start pixel of the scanning direction line segment of each row constituting the line segment, a linear search is performed, and the end pixel of the scanning direction line segment Explore
A line segment drawing method characterized by the above. The screen scanning direction is the x direction, the x direction difference dx and the y direction difference dy of the start point and end point of the line segment,
The line segment drawing processing method according to claim 1, wherein a slope dy / dx of the line segment is calculated to determine whether dy / dx ≦ α is satisfied, and when the condition is satisfied, a binary search is applied.   3. The line segment drawing processing method according to claim 2, wherein the predetermined value [alpha] is at least 1/8 or less. The binary search, and the position of the pixel of interest currently is [Delta] x _p moved in the x-direction, the error _{e p = Δx p * dy /} dx of the pixel after movement is calculated, the error _{e p} is 1 ≦ _{e p} If <2, the pixel after the movement is the end pixel of the scanning direction line segment,
If the error _{e p} is 0 _<e p <1 or _{e p} ≧ 2, as _{Δx p ← 1/2 * Δx} p, is further moved in the direction of the pixel x-coordinate becomes large or small, 1 ≦ it Repeat until e _p <2.
The line segment drawing method according to claim 2 or 3, The screen scanning direction is the x direction, the x direction difference dx and the y direction difference dy of the start point and end point of the line segment,
2. The line segment drawing processing method according to claim 1, wherein it is determined whether or not 2 ^k * dy ≦ dx (k is a constant) and a binary search is applied when the condition is satisfied.   6. The line segment drawing method according to claim 5, wherein the constant k is a value of at least 3 or more. In the binary search, the position of the pixel of interest is moved by Δx _p in the x direction, and an error E _p = Δx _p * dy in the pixel after the movement is calculated, and the error E _p is dx ≦ E _p <2 If * dx, the pixel after the movement is the end pixel of the scanning direction line segment,
If the error E _p is 0 <E _p <dx or E _p ≧ 2 * dx, Δx _p ← 1/2 * Δx _p is set, and the pixel is further moved in the direction in which the x coordinate becomes larger or smaller. Repeat until dx ≦ E _p <2 * Δx _p ,
The line segment drawing method according to claim 5 or 6. The initial value of the x-direction movement amount [Delta] x _p is the line drawing processing method according to claim 4 or 7, characterized in that more than half of the value of the number of pixels x direction of at least the screen. slowly line drawing processing method according to claim 8, wherein the reducing according to the value of the initial value of the x-direction movement amount [Delta] x _p, processed line approaches the line of the end point of the line segment.   10. A line segment drawing processing apparatus, wherein the line segment drawing processing method according to claim 1 is implemented. Image processing means for processing a picked-up image captured by the image pickup device, line segment processing means for generating a line segment, combining means for generating a composite image in which the line segment is superimposed on the picked-up image, and the composite image In a line drawing processing device comprising display means for displaying,
The captured image processing apparatus according to claim 10, wherein the line segment processing unit includes the line segment drawing processing apparatus according to claim 10.

Images