JP3382045B2 - Image projection system - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image projection system for projecting a document image composed of characters and figures created by a personal computer or the like onto a screen or a display board. In recent years, with the spread of software that edits text data and graphic data that represent characters and symbols together, a document image in which texts such as explanatory notes and graphics such as tables and graphs are mixed as materials for meetings and presentations. There are more and more cases of creating on a PC.

On the other hand, in conferences and presentations, a document image needs to be presented to a large number of people, and therefore a projection device such as an overhead projector device or a slide projection device is often used. In order to cope with such an application, an image projection system has been proposed in which a created document image is directly projected by a projector device.

[0003]

2. Description of the Related Art FIG. 9 shows a configuration example of a conventional image projection system. In FIG. 9, the document image created by the personal computer 410 includes character code indicating characters forming the document image, line drawing data indicating the arrangement of line segments forming a figure, and information designating the arrangement of the text and the figure. Is transmitted to the projector 420 as display data consisting of

Further, the drawing processing section 421 provided in the projector 420 is configured to create image data representing a document image based on the above-mentioned display data, and the projection processing section 422 prepares this image data. The image is projected on the screen 401 via the optical system 423. In this case, by arranging the screen 401 so as to vertically intersect with the optical axis of the optical system 423 provided in the projector 420, an image of the same quality as when the corresponding document image is displayed on the display of the personal computer 410 is obtained. , Can be enlarged and displayed on the screen 401.

[0005]

As described above, since the conventional image projection system simply projects the document image created by the personal computer 410, it is necessary to project the document image in a normal shape. , The screen 40 with respect to the optical axis of the optical system 423 of the projector 420.
It was necessary to position 1 exactly vertically.

[0006] Usually, since the user arranges the screen 401 by subjective judgment, it is very difficult to arrange the screen 401 accurately. In addition, it may be difficult to install in an accurate position due to the limitation of the venue, etc. If there is such a placement error, the optical axis and the screen 4
The projected figure is distorted according to the deviation from the normal direction of 01.

On the other hand, in the case of meetings and presentations, there are many cases where explanations are made using a lot of graphs and tables. In such a case, the misalignment of the projector 420 and the screen 401 distorts the essential graphs and tables. In addition to being unsightly, there is a possibility that the effectiveness of the presentation will be impaired. SUMMARY OF THE INVENTION It is an object of the present invention to provide an image projection system capable of correcting image deformation due to misplacement.

[0008]

FIG. 1 is a block diagram showing the principle of an image projection system according to the present invention. According to the first aspect of the invention, in the image projection system for projecting the document image composed of characters and figures on the screen 102 through the projecting means 101, a plurality of document images to be projected are arranged at predetermined positions. A reference pattern sending means 111 for sending a reference pattern composed of a marker pattern to the projection means 101, and the projection means 101.
An image capturing unit 112 for capturing a projected image of the reference pattern projected by the image capturing unit; and a position detecting unit 113 for detecting the positions of images of a plurality of marker patterns included in the projected image of the reference pattern obtained by the image capturing unit 112. , Distortion evaluation means 114 for evaluating the distortion of the projected image caused by the projection operation by the projection means 101 by collating the positions of the images of the plurality of marker patterns with the positions of the plurality of marker patterns in the reference pattern; document information representing a document image is inputted, sentence deformation corresponding to the distortion of the projected image represented by the result of evaluation by the distortion evaluating means 114
Correction means 115 for giving the writing information , and the correction means 11
Reference numeral 5 indicates the text information including text and symbols.
Text formed from a code indicating the data and attribute information
An image for displaying information and an image containing a line drawing
The separation processing unit for separating the information and the distortion evaluation unit 114
Based on the evaluation result obtained by the above, the text information
Distortion indicated by the evaluation result for the attribute information included in
And a text correction unit that performs conversion to cancel
Based on the evaluation result obtained by the evaluation means 114,
The distortion indicated by the evaluation result is applied to the image information.
A line drawing correction unit that performs conversion to cancel and a text correction
Combine the conversion result of the positive part and the conversion result of the line drawing correction part.
And a combination processing unit that restores document information
Characterized in that that.

According to a second aspect of the invention, in the image projection system according to the first aspect, the reference pattern sending means 11 is provided.
1 is a configuration for transmitting a reference pattern in which a marker pattern in which a plurality of line segments intersect at one point is arranged, and the position detection means 113 is based on the feature of the shape of the marker pattern,
A configuration including pattern extracting means 121 for extracting the image of the marker pattern included in the projected image of the reference pattern, and center detecting means 122 for detecting the center position of the image of the marker pattern extracted by the pattern extracting means 121. Is characterized in that.

[0010]

According to the invention of claim 1, the reference pattern sending means 11 is provided.
By the operation 1 described above, the projection unit 101 projects a reference pattern including a marker pattern serving as a reference for position determination.
By detecting the position of the image of the marker pattern from the projected image obtained in 2 and comparing it with the position of the marker pattern in the original reference pattern, the distortion caused by the projection operation by the projection unit 101 can be evaluated.

Based on this evaluation result, the correction means 115
The text correction unit and the line drawing correction unit provided in, its
After applying the deformations that can be canceled by the above-mentioned distortion to the text part and the line drawing part respectively ,
The transformation result of is combined by the combination processing unit. in this way
Then, by correcting the input document image, it is possible to correct the deformation of the image due to an arrangement error between the projection unit 101 and the screen 102 and project a normal image. According to the second aspect of the present invention, by adopting a pattern in which a plurality of line segments intersect at one point as the marker pattern, the pattern extracting unit 121 extracts the image of the marker pattern and the center detecting unit 122 detects the center thereof. By doing so, the projection position of the marker pattern can be detected easily and accurately.

[0012]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 2 shows a block diagram of an embodiment of the image projection system of the present invention. 2, the image projection system includes an image correction device 200 added to the conventional image projection system shown in FIG. 9, and the image correction device 200 corrects the display data received from the personal computer 410 as described later. The projector 42
It is configured to send to 0.

In the image correction apparatus 200 shown in FIG. 2, the display data holding section 201 holds line drawing data indicating a reference pattern as described later, and the selector 202 instructs the correction control section 203. In response to the,
The line drawing data described above is selected instead of the output of the correction calculation unit 210, and the projector 4 corresponding to the projection unit 101 is selected.
It is configured to be sent to 20.

Here, as the reference pattern, for example,
As shown in Fig. 3 (a), prepare a pattern in which marker patterns such as crosses are arranged at the positions corresponding to the four corners of the document image, and prepare the line drawing data necessary for drawing the line segments that make up these marker patterns. It may be held in the display data holding unit 201. In this way, the correction control unit 203 controls the operation of the selector 202 and sends the line drawing data held in the display data holding unit 201 to the projector 420, thereby realizing the function of the reference pattern sending unit 111 and the projector. By 420, the reference pattern shown in FIG. 3A can be projected.

The image correction apparatus 200 is equipped with an image input section 210 corresponding to the photographing means 112, and the image input section 210 is provided with the correction control section 2 described above.
It operates according to the instruction from 03, and the CCD camera 211 photographs the reference pattern projected on the screen 401,
The conversion processing unit 212 converts the video signal of the CCD camera 211 into binary image data and stores it in the frame memory 213.

Further, in FIG. 2, the projection position detecting unit 22
0 corresponds to the position detecting means 113, and corresponds to the marker pattern included in the reference pattern shown in FIG. 3A from the image data stored in the frame memory 213 as described later. The projection image is extracted, the coordinates indicating the position thereof are calculated, and the coordinates are sent to the difference calculation unit 204.

In FIG. 2, the position data holding unit 205
Holds the basic position of the marker pattern in the reference pattern described above, and the difference calculation unit 204 compares the projected position of the marker pattern received from the projected position detection unit 220 with this basic position, and for each marker pattern The difference between the projected position and the basic position is calculated and output as a result of evaluating the distortion of the projected figure.

As described above, the difference calculating unit 204 operates based on the detection result of the projection position detecting unit 220 and the basic position held in the position data holding unit 205,
The function of the distortion evaluation unit 114 can be realized, and the distortion of the projected figure can be evaluated using the positional deviation of the marker pattern as an index. Here, since the distortion of the projected figure caused by the deviation between the optical axis of the projector 420 and the normal direction of the screen 401 becomes remarkable especially around the projected figure, the marker patterns are arranged at the four corners as described above. By projecting the reference pattern described above and evaluating the deviation of the projected marker pattern from the basic position, the distortion of the projected figure can be reliably detected and evaluated accurately.

For example, as a result of projecting the reference pattern shown in FIG. 3A, each marker pattern of the reference pattern is shown in FIG.
When projected at a position as shown in (b), the difference between the x coordinate and the y coordinate (d _1x) for each of the upper left, upper right, and lower right marker patterns with the position of the lower left marker pattern as a reference. , D _1y ), (d _2x , d _2y ),
(D _3x , d _3y ) is obtained.

By the way, in this case, for example, as shown in FIG.
Interval D between the marker pattern of the reference pattern shown in
1 is the y of the marker pattern in the projected image
It can be considered that the difference is reduced by the coordinate difference d _1x . Therefore, if the corresponding interval of the document image to be projected is enlarged by the above-mentioned difference d _1x and then projected on the screen 401 by the projector 420,
In the projected document image, the corresponding interval is considered to be the normal length.

In FIG. 2, the correction value calculation unit 230 calculates a correction value indicating the amount of correction of the character or line drawing data forming the document image, based on the evaluation result obtained as described above. It is configured to be sent to the correction calculation unit 240. In the correction value calculation unit 230, the deformation rate calculation unit 231 first determines the enlargement / reduction rate for each side of the quadrangle formed by sequentially connecting the marker patterns to, based on the difference between the marker patterns. The deformation ratios shown are obtained, and the deformation ratio at each position inside the quadrangle is calculated based on these deformation ratios.

For example, the deformation rates C _y1 and C _{y2 of} two sides parallel to the y-coordinate axis in the reference pattern are obtained from the y-coordinate difference for each of the marker pattern and the marker pattern. From the x-coordinate difference between the marker pattern and the marker pattern, the deformation rates C _x1 and C _{x2 of the} two sides parallel to the x-coordinate axis in the reference pattern are obtained.

Then, y at each position inside the quadrangle
The deformation rates in the axial direction and the x-axis direction may be obtained by interpolating the deformation rates C _y1 and C _y2 and the deformation rates C _x1 and C _x2 , respectively. Here, the text part included in the document image is
It is formed of a character code and its attribute information, and this attribute information includes the number of points indicating the size of the character. Therefore, if the number of points corresponding to each character included in the text is changed according to the transformation rate, the text portion in the image document can be easily transformed.

For example, as shown in FIG. 2, the correction value calculation unit 230 is provided with a point correction value calculation unit 232, and based on the transformation rate obtained by the transformation rate calculation unit 231, the number of points of characters at each position is calculated. A correction value is calculated, and this is used as a correction value for the text portion, and the point correction table 2
The configuration may be such that it is stored in 33. At this time, the point correction value calculation unit 232 calculates the enlargement / reduction ratio of the unit area from the deformation ratios in the x-axis direction and the y-axis direction, and calculates the correction value of the number of points based on the enlargement / reduction ratio. Good.

Further, the line drawing data is formed from the coordinates of the reference point which is the reference of the figure such as the end points of the line segments forming the graph or figure and the center of the circle, and the attribute information which indicates the type of the line. The position of each figure in the entire document image is indicated by the position of the reference point of this figure.
Therefore, by moving the positions of these reference points according to the deformation rate corresponding to the positions, the corresponding figure can be pseudo-enlarged / reduced.

For example, as shown in FIG. 2, the correction value calculation unit 230 is provided with a coordinate correction value calculation unit 234, and based on the deformation rate obtained by the deformation rate calculation unit 231, the coordinates of the reference point at each position are calculated. The correction value is calculated, and the coordinate correction value table 23
5 may be stored. In this case, as shown in FIG.
1, the display data received from the personal computer 410 is separated into a text portion and a line drawing portion, which are sent to the text correction unit 242 and the line drawing correction unit 243, respectively, and the correction results by each are combined again by the combination processing unit 244. The configuration may be such that the data is sent to the projector 420.

In FIG. 2, the text correction unit 242 is
Arrangement information indicating the position of the text is received together with the text data from the separation processing unit 241, the position of each character is obtained based on this arrangement information, and the number of points corresponding to the corresponding position is corrected from the point correction table 233 described above. The value may be searched, the number of points of the corresponding attribute information may be replaced, and the value may be transmitted to the combining processing unit 244.

Further, the line drawing correction unit 243 has a separation processing unit 2
Based on the coordinates of the reference point included in the line drawing data received from 41, a corresponding correction value is retrieved from the coordinate correction value table 235 described above, and the coordinates of the new reference point obtained by adding the correction value to the coordinates of the original reference point. And replace it with
The configuration may be such that the data is sent to 44. In this way, the correction value calculation unit 230 and the correction calculation unit 240 operate to realize the function of the correction unit 115, and, for example, according to the evaluation result of the distortion of the projected figure, for example, FIG.
As shown in, it is possible to give a deformation to the document image to be projected that is canceled by the distortion generated by the projection process.

Further, at this time, the correction control unit 203 instructs the selector 202 to select the output of the image correction apparatus 200, and in response thereto, the display to which the above-mentioned modification is applied is made via the selector 202. The data is sent to the projector 420 and provided for projection processing. As described above, when the document image thus modified is projected through the projector 420, each part of the document image is modified at the modification rate corresponding to the position obtained by the modification rate calculating unit 231. Therefore, the deformation given by the correction calculation unit 240 is removed, and a projection image without distortion corresponding to the original document image can be obtained on the screen 401.

That is, the image correction device 20 described above.
By configuring the image projection system with 0, it is possible to correct the deformation of the image due to the placement error of the projector 420 or the screen 401, and a projection image without distortion can be obtained regardless of the placement of the projector 420 or the screen 401. Can be provided to users.

As a result, it becomes unnecessary to strictly adjust the position when arranging the projector 420 and the screen 401, so that the time required for setting up a venue can be greatly shortened. By the way, as described above, when the correction process is performed based on the positional deviation of the marker pattern in the projected reference pattern, the CCD camera 2
It is necessary to accurately extract the image of the marker pattern from the binary image data captured in 11.

A method of accurately extracting the image of the marker pattern from the binary image data held in the frame memory 213 and detecting the position in the above-mentioned projection position detecting section 220 will be described below. In FIG. 4, claim 2
The detailed configuration of the projection position detection unit 220 to which the invention of FIG. Further, FIG. 5 shows a flowchart showing the operation of detecting the projection position.

In the projection position detecting section 220 shown in FIG. 4, the expansion processing section 221 and the thinning processing section 222 have the steps 301,
At 302, the data stored in the frame memory 213
The value image data is configured to be expanded and thinned, respectively. In this way, by first performing expansion processing and then thinning processing, for example, as shown in FIG. 6 (a), when there is a faint portion in the captured binary image, After converting to an expanded pattern as shown in 6 (b), thinning processing is performed to connect the faint parts to reproduce a pattern without fading as shown in FIG. 6 (c). The image of the marker pattern to be processed can be provided.

Further, in FIG. 4, the line segment extraction unit 223
In step 303, a pattern in which a plurality of black pixels are continuous is extracted as a line segment from the binary image data in the frame memory 213 and is sent to the length determination unit 224. In response to this, the length determination unit 22
In step 304, first, in step 304, each length of the extracted line segments is compared with a predetermined threshold value, and based on the comparison result, each line segment is a candidate line that may be a part of a marker pattern. It is configured to determine the minutes.

At this time, the length discriminating section 224 uses a threshold value corresponding to the length of the line segment forming the marker pattern arranged in the reference pattern, and the line segment which is longer than this threshold value is a candidate. It may be determined as a line segment. in this case,
If the length is less than or equal to the threshold value, it is assumed that the corresponding pattern corresponds to dust or scratches on the screen 401, and the information about the line segment is discarded as noise. On the other hand, if the length is greater than or equal to the threshold value, the length determination unit 2
24 indicates the information about the line segment by the candidate pattern detection unit 2
25 to be sent.

Here, as shown in FIG. 3A, the marker pattern is a pattern in which two line segments intersect,
In step 305, the candidate pattern detection unit 225
May detect one of the candidate line segments obtained in step 304 that intersects another candidate line segment based on the binary image data stored in the frame memory 213.

As described above, from the candidate line segments discriminated by the length discriminator 224, those that do not intersect with other candidate line segments are excluded as noise, and the intersecting candidate line segments are excluded. By discriminating the combination as a candidate pattern, the candidate patterns are narrowed down, and the intersection discriminating unit 22
It can be subjected to the treatment of 6. In FIG. 4, the intersection determination unit 226 is configured to determine whether or not the candidate pattern is an image of a marker pattern, based on the intersection angle of the candidate line segments that form each candidate pattern.

For example, when the cross-shaped marker pattern as shown in FIG. 3A is adopted, the crossing angle is 90 degrees even in the image of the marker pattern. Therefore, as shown in FIG. In step 306, the intersection determination unit 226 may obtain the inner product of the direction vectors indicating the directions of the respective candidate line segments, and in step 307, detect the candidate pattern whose inner product is the numerical value “0” as the image of the marker pattern. .

In this way, by performing the above-described discrimination processing based on the intersection angles of the line segments in the candidate pattern, it is possible to discriminate the pattern having all the features of the marker pattern, and the processing by the centroid detection unit 227. Can be used for In response to this, the center of gravity detector 227
In step 308, the center of gravity of the corresponding pattern is obtained, and the obtained center of gravity position is sent to the correction value calculation unit 230 as the position of the image of the marker pattern.

Here, in a pattern in which a plurality of line segments intersect, the position of the center of gravity of the pattern coincides with the position of the intersection of the line segments, so the position of the center of gravity can be taken as the position of the marker pattern. Further, in the case of obtaining the barycentric position in this manner, since the information of all the pixels forming the pattern can be used, it is less likely to be affected by noise as compared with the case where the intersection of the line segment itself is detected. The position of the marker pattern can be detected with high accuracy.

As described above, the length discriminating section 224, the candidate pattern discriminating section 225, the intersection discriminating section 226, and the center of gravity detecting section 22.
7 sequentially realizes the functions of the pattern extracting means 121 and the center detecting means 122 described in claim 4, and the characteristic of the marker pattern is used to reliably and accurately determine the image position of the marker pattern. Can be detected.

Further, a marker pattern as shown in FIG. 3 (a) is projected while being rotated around each intersection, and the projected figure is photographed at regular intervals to obtain a marker from the binary image data. It is also possible to detect the position of the pattern. In this case, as shown in FIG.
The image processing apparatus 200 shown in FIG.
In response to an instruction from the correction control unit 203, the rotation processing unit 206 generates a pattern in which each of the marker patterns is rotated by a predetermined angle based on the display data held in the display data holding unit 201. It suffices to create display data representing the data and send it to the projector 420 via the selector 202.

Further, in the projection position detection unit 250 provided in place of the projection position detection unit 220, the black pixel detection unit 251 receives the binary value stored in the frame memory 213 in response to the instruction from the correction control unit 203. A black pixel is detected from the image data, and the frequency at which each pixel in one frame is detected as a black pixel is stored in the frequency holding unit 252.

Here, by the operation of the rotation processing unit 206 described above, the marker pattern is rotated once at a predetermined time while changing the angle little by little, and the correction control unit 203 synchronizes with the change in the angle of the marker pattern to black. If the pixel detection unit 251 is instructed to perform a detection operation, the frequency holding unit 252 displays the frequency distribution of black pixels corresponding to each marker pattern as shown in FIG. A circular distribution pattern with a widened portion is retained. However, FIG.
In the above, the degree of black pixel frequency is shown by changing the density of knitting.

Therefore, if the pattern extraction unit 253 detects the corresponding distribution pattern from the frequency holding unit 252 described above, the frequency distribution corresponding to the marker pattern can be detected almost certainly, and the detected distribution pattern can be obtained. Based on the center of gravity, the center of gravity of the distribution pattern can be accurately detected by detecting the center of gravity.

In this case, the black pixel detection unit 251 operates in response to an instruction from the correction control unit 203, and the pattern extraction unit 253 operates based on the detection result. The function of the pattern extraction unit 121 can be fulfilled, an image of the marker pattern can be extracted based on the characteristics of the marker pattern, and the image of the marker pattern can be provided to the processing of the center of gravity detection unit 254 corresponding to the center detection unit 122.

As another example of the pattern extracting means 121, a method of recognizing a marker pattern as a spatial frequency change pattern can be considered.

[0048]

As described above, according to the present invention, distortion generated during projection is evaluated based on a binary image obtained by photographing a projected reference pattern, and the distortion is canceled. Since necessary and sufficient deformation can be added to the document image to be projected, it is possible to project a document image without distortion regardless of the arrangement of the projection means and the screen.

Thus, it is possible to save the trouble of adjusting the arrangement of the screen, and it is possible to always project a normal document image and provide it to the user.

[Brief description of drawings]

FIG. 1 is a principle block diagram of an image projection system of the present invention.

FIG. 2 is a configuration diagram of an embodiment of the image projection system of claim 1;

FIG. 3 is a diagram illustrating an image correction process.

FIG. 4 is a detailed configuration diagram of a projection position detection unit.

FIG. 5 is a flowchart showing a detection operation of a projection position.

FIG. 6 is an explanatory diagram of expansion processing and thinning processing.

FIG. 7 is a configuration diagram of another embodiment of the image projection system of the present invention.

FIG. 8 is a diagram showing an example of a frequency distribution pattern.

FIG. 9 is a diagram showing a configuration example of a conventional image projection system.

[Explanation of symbols]

101 Projection means 102, 401 screen 111 Reference pattern sending means 112 means of photography 113 Position detecting means 114 distortion evaluation means 115 Correction means 121 pattern extraction means 122 Center Detection Means 201 Display data holding unit 202 selector 203 Correction control unit 204 Difference calculation unit 205 Position data holding unit 206 Rotation processing unit 210 Image input section 211 CCD camera 212 conversion processing unit 213 frame memory 220,250 Projection position detector 221 Expansion processing unit 222 Thinning processing unit 223 Line segment extractor 224 Length discriminator 225 Candidate pattern detector 226 Intersection discrimination unit 227,254 Center of gravity detector 230 Correction value calculation unit 231 Deformation rate calculation unit 232 point correction value calculation unit 233 point correction value table 234 Coordinate correction value calculation unit 235 Coordinate correction value table 240 Correction calculator 241 Separation processing unit 242 Text correction unit 243 Line drawing correction unit 244 Combine processing unit 251 Black pixel detector 252 frequency storage 253 pattern extractor 410 PC 420 projector 421 Drawing processing unit 422 Projection processing unit 423 optical system

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G03B 21/00-21/30 H04N 5/74

Claims (2)

(57) [Claims] 1. An image projection system for projecting a document image composed of characters and figures on a screen via a projecting means, wherein the document image to be projected comprises a plurality of marker patterns arranged at respective predetermined positions. Pattern transmitting means for transmitting the reference pattern to the projecting means, photographing means for photographing the projected image of the reference pattern projected by the projecting means, and projected image of the reference pattern obtained by the photographing means By detecting the position of the image of the plurality of marker patterns included in the position detection means, by collating the position of the image of the plurality of marker patterns and the position of the plurality of marker patterns in the reference pattern, Distortion evaluation means for evaluating the distortion of the projected image caused by the projection operation by the projection means; Document information representing the document image projection object is input, and a correction means for providing a deformation corresponding to the distortion of the projected image represented by the evaluation result of the distortion evaluating means on said document information, said correction means, said document Information, text data containing letters and symbols
Information that is formed from the code that indicates the attribute and attribute information
Information and image information for displaying images including line drawings.
A separation processing unit that separates into the distribution, based on the evaluation results obtained by the distortion evaluating means
The attribute information included in the text information,
Perform conversion to cancel the distortion shown in the evaluation results.
Based on the evaluation result obtained by the text correction unit and the distortion evaluation means.
Is shown in the evaluation result for the image information.
To the line drawing correction unit that performs conversion for canceling the distortion caused by the text correction unit and the conversion result by the text correction unit and the line drawing correction unit.
Combined processing to restore the document information by synthesizing with the conversion result by
An image projection system characterized by being configured with a section . 2. The image projection system according to claim 1, wherein the reference pattern sending means is configured to send a reference pattern in which a marker pattern in which a plurality of line segments intersect at one point is arranged, and the position detecting means comprises: Pattern extraction means for extracting the image of the marker pattern included in the projected image of the reference pattern based on the characteristic of the shape of the marker pattern; and the center position of the image of the marker pattern extracted by the pattern extraction means. An image projection system comprising: a center detecting unit for detecting the.