JP4075725B2 - Projector and projection image correction method thereof - Google Patents

Description

  The present invention relates to a projector and a projection image correction method thereof.

  A projector (projection device) is a device for displaying an image on a screen. When the optical axis of the projector is inclined with respect to the screen surface, distortion occurs in the image formed on the screen surface.

  Therefore, the projector corrects the image distortion based on the tilt angle of the optical axis with respect to the screen surface so that an image without distortion is displayed on the screen surface even if the optical axis is inclined with respect to the screen surface. Some include a correction unit (see, for example, Patent Document 1).

  A conventional correction unit reads an input video signal (video information) once after storing the video signal in a frame memory, and performs vertical and horizontal coordinate conversion of the video signal based on the tilt angle of the optical axis. Do.

When the pixel data is read by accessing the frame memory, there is an access mode in which only the column address is changed while the row address is fixed, and the row address is changed when the last column access is completed. This access mode is in accordance with the video signal scanning method. The speed at which the pixel data is read out in such an access mode is faster than the access mode in which the column address is fixed and the row address is changed, and access of the pixel data in the scanning line direction is faster.
JP 2000-122617 A (Page 4-7, FIGS. 1 and 2)

  However, if the left and right heights of the video are greatly different depending on the tilt angle of the optical axis, the row address frequently changes when coordinate conversion is performed while reading the video data from the frame memory.

  If the row address changes frequently, it takes time to access. As a result, the video signal readout timing of the correction unit is not in time, and there is a problem that the distortion of the projection image cannot be corrected well.

  The present invention has been made in view of such conventional problems, and a projector capable of correcting a projected image regardless of the inclination angle of the optical axis of the projector with respect to the screen surface, and a projection image correcting method thereof The purpose is to provide.

In order to achieve this object, a projector according to the first aspect of the present invention provides:
In a projector that projects projected light onto a screen surface to form a projected image,
Video information in which a plurality of pixel data forming an image is collected is supplied, and the projected image on the screen distorted by the tilt of the optical axis of the projected light with respect to the screen surface is subjected to coordinate conversion of each pixel data. A correction unit to correct by,
A projection unit that converts the projection image corrected by the correction unit into projection light and projects it onto the screen surface, and
The correction unit is
A storage unit that stores pixel data of the supplied video information so as to correspond to a horizontal scanning direction and a vertical scanning direction;
The pixel data is read from the storage unit, and the coordinates of the read pixel data in the vertical scanning direction are converted on the basis of the inclination angle of the optical axis with respect to the screen surface, thereby to the horizontal scanning direction coordinates of the video. A first vertical scaler that performs uniform deformation in the vertical scanning direction;
By converting the coordinate in the horizontal scanning direction of each pixel data coordinate-converted by the first vertical scaler based on the inclination angle of the optical axis with respect to the screen surface, it corresponds to the coordinate in the vertical scanning direction of the video. A horizontal scaler that deforms in the horizontal scanning direction;
The vertical scanning direction according to the horizontal scanning direction coordinate of the image is obtained by converting the vertical scanning direction coordinate of each pixel data coordinate-converted by the horizontal scaler based on the inclination angle of the optical axis with respect to the screen surface. A second vertical scaler that performs the deformation of
It is equipped with.

The first vertical scaler may be configured to convert the coordinates in the vertical scanning direction of each pixel data read from the storage unit according to the relational expression shown in Formula 4.

The horizontal scaler may be configured to convert the coordinate in the horizontal scanning direction of each pixel data coordinate-converted by the first vertical scaler according to the relational expression shown in Formula 5.

The second vertical scaler may be configured to convert the coordinates in the vertical scanning direction of each pixel data coordinate-converted by the horizontal scaler according to the relational expression shown in Equation 6.

A projection image correction method for a projector according to a second aspect of the present invention includes:
A projection image correction method for a projector that corrects a projection image on the screen surface distorted by an inclination of an optical axis of projection light projected by the projector with respect to the screen surface,
Storing pixel data of the supplied video information so as to correspond to a horizontal scanning direction and a vertical scanning direction;
The pixel data is read, and the vertical scanning direction coordinates of the read pixel data are converted on the basis of the inclination angle of the optical axis with respect to the screen surface. Performing a uniform deformation;
By transforming the coordinate in the horizontal scanning direction of each pixel data subjected to the coordinate conversion based on the inclination angle of the optical axis with respect to the screen surface, deformation in the horizontal scanning direction according to the coordinate in the vertical scanning direction of the image is performed. Steps to do,
By transforming the coordinate in the vertical scanning direction of each pixel data obtained by the coordinate conversion based on the inclination angle of the optical axis with respect to the screen surface, the vertical scanning direction of the image is deformed in accordance with the coordinate in the horizontal scanning direction. And performing steps.

  According to the present invention, even if the inclination angle of the optical axis of the projector with respect to the screen surface is large, the projection image can be corrected appropriately.

  A projector according to an embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows the configuration of a projector according to an embodiment of the present invention.
The projector 1 according to the present embodiment includes an inclination angle detection sensor 11, a trapezoid correction unit 12, a projection light conversion device 13, and an optical mechanism unit 14.

  As shown in FIG. 2, the tilt angle detection sensor 11 detects the tilt angle θv in the vertical direction and the tilt angle θh in the horizontal direction of the optical axis of the projector 1 with respect to the screen 2.

  In FIG. 2, the projected image abcd is a quadrangular projected image formed on the screen 2 when the tilt angles θv and θh are both 0. The projection image a′b′c′d ′ is a quadrangle projection image formed on the screen 2 when the optical axis of the projector 1 is inclined with respect to the surface of the screen 2 (θv ≠ 0, θh). ≠ 0).

The tilt angle detection sensor 11 is constituted by, for example, a distance measurement sensor that detects the tilt angles θv and θh by measuring distances from a plurality of distance measuring points on the screen 2. However, the inclination angle θv may be configured to be detected using an acceleration sensor.
The tilt angle detection sensor 11 supplies the detected tilt angles θv and θh to the trapezoid correction unit 12.

  The trapezoidal correction unit 12 performs trapezoidal correction of the projected image based on the inclination angles θv and θh supplied from the inclination angle detection sensor 11. The trapezoidal correction unit 12 corrects the projected image formed on the screen 2 by performing keystone correction.

Specifically, the trapezoidal correction unit 12 cuts out the projection image p′q′r ′s ′ from the projection image a′b′c′d ′ shown in FIG.
FIGS. 3A and 3B show the input image abcd, the projected image a′b′c′d ′, and the projected image p′q′r ′s ′ cut out by the trapezoid correcting unit 12, respectively.

  As shown in FIG. 3C, the trapezoid correcting unit 12 generates a projection image pqrs by inversely transforming the cut out projection image p′q′r ′s ′, and adds video information ( The trapezoidal correction is performed by projective conversion of the video signal) over time.

The trapezoid correction unit 12 includes a trapezoid correction pre-stage unit 21 and a trapezoid correction post-stage unit 22.
The trapezoidal correction pre-stage unit 21 performs vertical scaling and horizontal scaling on the input video information, and includes a frame memory 23, a vertical scaler 24, and a horizontal scaler 25.

  The frame memory 23 stores input video information. The frame memory 23 is constituted by, for example, a DRAM (Dynamic Random Access Memory). This video information is information in which a plurality of pixel data forming a video is collected. By performing the operation of scanning the pixel data in the horizontal direction sequentially in the vertical direction, one screen is formed.

  The frame memory 23 designates the storage location of the pixel data in the horizontal direction and the vertical direction using the column address and the row address, respectively. When storing the pixel data in the horizontal direction, the frame memory 23 changes the column address while keeping the row address constant.

  When the access to the last column is completed, the frame memory 23 changes the column address in the same manner after changing the row address by one or two, and the next or next horizontal pixel data. Is stored.

In this way, the frame memory 23 stores the pixel data of the input video information so as to correspond to the horizontal scanning direction and the vertical scanning direction.
The vertical scaler 24 reads out the pixel data of the video information from the frame memory 23 and uniformly converts the coordinates in the vertical scanning direction with respect to the coordinates in the horizontal scanning direction of the video information.

  When the vertical scaler 24 reads pixel data of video information from the frame memory 23, the storage location is specified using a column address and a row address in the same manner as the frame memory 23 stores the pixel data.

The vertical scaler 24 uses the old coordinates of the pixel data before coordinate conversion and the new coordinates of the coordinate points after coordinate conversion as (x, y0) and (x, y1), respectively, according to the relational expression shown in Equation 7, Uniform conversion of coordinates in the vertical scanning direction to coordinates in the horizontal scanning direction of information is performed.

The x coordinate corresponds to the horizontal coordinate on the screen 2 shown in FIG. 2, and the y coordinate corresponds to the vertical coordinate.
The conversion coefficients α1 to α3 are obtained by substituting a known coordinate set (a set of y0 and y1) of a plurality of pixel data of 3 sets or more, for example, 25 sets, into Equation 7 and performing regression analysis by the least square method or the like. Required by doing. By obtaining the conversion coefficients α1 to α3 in this way, information on the inclination angles θv and θh is also included in the conversion coefficients α1 to α3.

  This equation 7 does not include information on the x coordinate. That is, the vertical scaler 24 can perform vertical coordinate conversion without using position information of pixel data in the horizontal direction. This means that uniform vertical coordinate conversion is performed on the coordinates in the horizontal scanning direction. By doing so, changes in the row address of the frame memory 23 can be minimized.

The horizontal scaler 25 performs horizontal coordinate conversion corresponding to the coordinates in the vertical scanning direction on the video information that has been subjected to vertical coordinate conversion by the vertical scaler 24. The horizontal scaler 25 sets the old coordinates and the new coordinates as (x0, y) and (x1, y), respectively, and performs horizontal enlargement and reduction in accordance with the relational expression shown in Equation 8 to convert the horizontal coordinates of the video. Do.

  As in the case of α1 to α3, the conversion coefficients β1 to β8 are obtained by substituting a set of known coordinates (a set of x0, x1, and y) of a plurality of pixel data of 8 sets or more, for example, 25 sets, into Equation 8. It is obtained by performing multiple regression analysis using the least square method. By obtaining the conversion coefficients β1 to β8 in this way, the conversion coefficients β1 to β8 include information on the inclination angles θv and θh.

  The trapezoidal correction post-stage unit 22 performs vertical scaling on the video information corrected by the keystone correction pre-stage unit 21, and includes a vertical scaler 26 and a frame memory 27.

The vertical scaler 26 performs vertical coordinate conversion according to the coordinates in the horizontal scanning direction of the video information corrected by the trapezoid correction pre-stage unit 21. The vertical scaler 26 converts the old coordinates and new coordinates of the pixel data to (x, y0) and (x, y1), respectively, and converts the old coordinates of the pixel data to new coordinates according to the relational expression shown in Equation 9. Thus, the vertical coordinate conversion is performed by enlarging or reducing the image converted by the horizontal scaler 25.

  Note that the conversion coefficients γ1 to γ4 are the same as α1 to α3 and β1 to β8, and are a set of four or more, for example, 25 known pixel data sets (a set of x, y0, and y1). It is obtained by substituting into 9 and performing multiple regression analysis. By obtaining the conversion coefficients γ1 to γ4 in this way, the conversion coefficients γ1 to γ4 include information on the inclination angles θv and θh.

  The frame memory 27 stores video information that has been subjected to vertical coordinate conversion by the vertical scaler 26.

  Note that a device such as a conventional integrated circuit chip can be used for the trapezoid correction pre-stage portion 21, and a dedicated device can be used for the trapezoid correction post-stage portion 22.

  The projection light conversion device 13 converts the video information corrected by the trapezoidal correction unit 12 into projection light of the video.

  The optical mechanism unit 14 performs focus control so that an image is formed on the screen 2, and projects the projection light converted by the projection light conversion device 13 onto the screen 2.

Next, the operation of the projector 1 according to the embodiment will be described.
The tilt angle detection sensor 11 detects tilt angles θv and θh of the optical axis of the projector 1 with respect to the surface of the screen 2 and supplies the detected tilt angles θv and θh to the trapezoid correction unit 12.
The trapezoidal correction unit 12 performs trapezoidal correction of input video information based on the inclination angles θv and θh supplied from the inclination angle detection sensor 11.

  First, the frame memory 23 of the keystone correction pre-stage unit 21 designates a storage location of pixel data of input video information using a column address and a row address. When storing the pixel data in the horizontal direction, the frame memory 23 changes the column address while keeping the row address constant.

  When the access to the last column is completed, the frame memory 23 changes the column address in the same manner after changing the row address by one or two, and the next or next horizontal pixel data. Is stored.

The vertical scaler 24 reads out the pixel data of the video information from the frame memory 23 and performs uniform vertical coordinate conversion on the coordinates in the horizontal scanning direction according to the relational expression shown in Equation 7.
For example, the vertical scaler 24 performs vertical coordinate conversion on the input video information shown in FIG. 4 (1) and outputs video information as shown in FIG. 4 (2).

  In this case, no deformation is performed such that the left and right heights are different. That is, the vertical scaler 24 performs all the same enlargement and reduction processing in the horizontal direction of the video. As a result, the image converted by the vertical scaler 24 is symmetrical as shown in FIG. 4B, and the left and right heights are the same.

  When the vertical scaler 24 performs such processing, the video information is processed without being affected by the x coordinate (horizontal coordinate). That is, the vertical scaler 24 can perform vertical coordinate conversion without being restricted by the access speed of the frame memory 23 and without being restricted by image deformation.

  The horizontal scaler 25 performs horizontal coordinate conversion corresponding to the coordinates in the vertical scanning direction on the image as shown in FIG. 4B in which the vertical coordinate conversion is performed by the vertical scaler 24 according to the relational expression shown in Equation 8. Do.

  The horizontal scaler 25 outputs video information as shown in FIG. 4C by performing horizontal coordinate conversion.

  The vertical scaler 26 of the trapezoidal correction post-stage unit 22 performs vertical coordinate conversion according to the coordinate in the horizontal scanning direction on the video as shown in FIG.

The vertical scaler 26 stores video information as shown in FIG. 4D in the frame memory 27 by performing vertical coordinate conversion.
When the vertical scaler 26 performs such vertical coordinate conversion before storing the video information in the frame memory 27, the trapezoidal correction unit 12 is less likely to be restricted by the access speed of the frame memory 27.

  The trapezoidal correction unit 12 sequentially reads out the video information shown in FIG. 4D from the frame memory 27 and supplies it to the projection light conversion device 13.

  When the tilt angle θh is large in horizontal keystone correction or diagonal keystone correction, the right and left heights of the corrected quadrangle are greatly different. With a conventional projector, it is difficult to correct the keystone in such a case. However, in the projector 1 according to the present embodiment, such trapezoidal correction is possible.

  The projection light conversion device 13 converts the video information corrected by the trapezoid correction unit 12 into projection light of the video. The optical mechanism unit 14 performs focus control so that a projection image is formed on the screen 2, and projects the projection light converted by the projection light conversion device 13 onto the screen 2. A projection image without distortion is formed on the screen 2.

  As described above, according to the present embodiment, the trapezoid correction unit 12 includes the trapezoid correction pre-stage unit 21 and the trapezoid correction post-stage unit 22, and the vertical scaler 24 of the trapezoid correction pre-stage unit 21 is used for the input video. Uniform enlargement or reduction in the vertical direction relative to the horizontal direction. Further, the vertical scaler 26 of the trapezoid correction post-stage unit 22 performs vertical reduction according to the horizontal coordinates of the video corrected by the keystone correction pre-stage unit 21 before writing the video information to the frame memory 27. .

  Therefore, even if the inclination angles θv and θh are increased, the change in the row address of the frame memory 23 is small. Therefore, the vertical scaler 24 can perform the vertical coordinate conversion without being restricted by the access speed, and the keystone correction. The unit 12 can appropriately correct the projection image regardless of the tilt angles θv and θh. For this reason, the trapezoidal correction range in the horizontal direction and the oblique direction can be greatly expanded.

In carrying out the present invention, various forms are conceivable and the present invention is not limited to the above embodiment.
For example, in the above-described embodiment, the trapezoid correction unit 12 is configured by sequentially connecting the keystone correction pre-stage unit 21 and the keystone correction post-stage unit 22. However, on the contrary, the trapezoidal correction unit 12 can also be configured by sequentially connecting the trapezoidal correction post-stage unit 22 and the trapezoidal correction pre-stage unit 21.

It is a block diagram which shows the structure of the projector which concerns on embodiment of this invention. It is explanatory drawing for demonstrating the positional relationship of the projector and screen of FIG. It is explanatory drawing which shows operation | movement of the trapezoid correction | amendment part of FIG. It is explanatory drawing which shows operation | movement of the vertical scaler and horizontal scaler of the trapezoid correction | amendment part of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Projector, 2 ... Screen, 11 ... Inclination angle detection sensor, 12 ... Trapezoid correction | amendment part, 13 ... Projection light conversion device, 14 ... Optical mechanism part

Claims (5)

In a projector that projects projected light onto a screen surface to form a projected image,
Video information in which a plurality of pixel data forming an image is collected is supplied, and the projected image on the screen distorted by the tilt of the optical axis of the projected light with respect to the screen surface is subjected to coordinate conversion of each pixel data. A correction unit to correct by,
A projection unit that converts the projection image corrected by the correction unit into projection light and projects it onto the screen surface, and
The correction unit is
A storage unit that stores pixel data of the supplied video information so as to correspond to a horizontal scanning direction and a vertical scanning direction;
The pixel data is read from the storage unit, and the coordinates of the read pixel data in the vertical scanning direction are converted on the basis of the inclination angle of the optical axis with respect to the screen surface. A first vertical scaler that performs uniform deformation in the vertical scanning direction;
By converting the coordinate in the horizontal scanning direction of each pixel data coordinate-converted by the first vertical scaler based on the inclination angle of the optical axis with respect to the screen surface, it corresponds to the coordinate in the vertical scanning direction of the image. A horizontal scaler that deforms in the horizontal scanning direction;
The vertical scanning direction according to the horizontal scanning direction coordinate of the image is obtained by converting the vertical scanning direction coordinate of each pixel data coordinate-converted by the horizontal scaler based on the inclination angle of the optical axis with respect to the screen surface. A second vertical scaler for performing the deformation of
A projector characterized by that. The first vertical scaler is configured to convert the coordinates in the vertical scanning direction of each pixel data read from the storage unit according to the relational expression shown in Formula 1.
The projector according to claim 1.

The horizontal scaler is configured to convert the coordinate in the horizontal scanning direction of each pixel data coordinate-converted by the first vertical scaler according to the relational expression shown in Formula 2.
The projector according to claim 1, wherein the projector is a projector.

The second vertical scaler is configured to convert the coordinates in the vertical scanning direction of each pixel data coordinate-converted by the horizontal scaler according to the relational expression shown in Equation 3.
The projector according to claim 1, wherein the projector is a projector.

A projection image correction method for a projector that corrects a projection image on the screen surface distorted by an inclination of an optical axis of projection light projected by the projector with respect to the screen surface,
Storing pixel data of the supplied video information so as to correspond to a horizontal scanning direction and a vertical scanning direction;
The pixel data is read, and the vertical scanning direction coordinates of the read pixel data are converted on the basis of the inclination angle of the optical axis with respect to the screen surface. Performing a uniform deformation;
By transforming the coordinate in the horizontal scanning direction of each pixel data subjected to the coordinate conversion based on the inclination angle of the optical axis with respect to the screen surface, deformation in the horizontal scanning direction according to the coordinate in the vertical scanning direction of the image is performed. Steps to do,
By transforming the coordinate in the vertical scanning direction of each pixel data obtained by the coordinate conversion based on the inclination angle of the optical axis with respect to the screen surface, the vertical scanning direction of the image is deformed in accordance with the coordinate in the horizontal scanning direction. And performing steps,
A projection image correction method for a projector, characterized in that:

Images