CN101482690B - Self-adapting template big-screen brightness emendation method with multi-projector combined display - Google Patents

Abstract

A method for correcting self-adapting formwork large-screen brightness displayed in combination by multiple projectors belongs to the large-screen image display field, characterized in that, the method comprises: generating a formwork; obtaining the attenuated image by multiplying the brightness value of the original image in the computer and the value in the formwork; then eliminating the high bright band in a stacked area by the mode of outputting the image to the projector. The invention provides a plurality of generated formworks in advance for the images with different brightness to select a formwork with optimum effect. The invention can self-adapting select the appropriate formwork to perform the brightness correction according to the real-time allocated brightness values of the sub images of each projector, thereby the images with various brightness all can obtain better brightness correcting effects.

Description

Adaptive template large-screen brightness correction method for multi-projector combined display

technical field

The invention relates to the brightness correction of multi-projector combination display, and realizes the large-screen high-resolution display of multi-projector combination display.

Background technique

The brightness correction of the existing multi-projector large-screen splicing display uses multiple projectors to splicing and displaying a large picture on the large screen. There is an overlapping area between adjacent projectors, that is, the corresponding projected image There are overlapping areas, and the bright bands in the overlapping areas can be eliminated by software or hardware methods, so as to realize the seamless splicing display of multi-projector large screens.

The hardware method is to use a shading plate to cover the excess light of the projectors, so that only the light of one projector can be projected in the overlapping area, and the light of other projectors is blocked. The advantage of this method is that no light can leak through the covered place. The disadvantage is that it is time-consuming and labor-intensive to manually control the position of the visor, and it is not suitable for the combination of many projectors.

The software method is to achieve brightness consistency by changing the brightness of the projector image in the frame buffer. The existing software methods mainly include edge blending, luminance consistency and luminance feedback.

Edge blending uses a function as a template to transition the overlapping areas of the projectors. Some of this function are linear functions, some are cosine functions, and some are power functions. The advantage of this method is that it is simple and easy to calculate; the disadvantage is that no matter which function is used, it cannot make the overlapping regions of all images transition uniformly.

The brightness consistency method is to first use the camera to obtain the brightness distribution of the white image displayed by each projector, and calculate the brightness of the overlapping area, that is, the sum of the brightness of each projector at this position; secondly, each projector generates a brightness Table LAM, the value of each point in the brightness table is Lmin/L, where Lmin is the minimum brightness of all projectors, and L is the actual brightness at this point; finally, each image in the frame buffer is multiplied by the value in the brightness table The template factor, and then output, so as to achieve the purpose of brightness correction. Because the method assumes that the brightness response curve of the projector is linear, but actually the brightness response curve of the projector is nonlinear, so nonlinear correction is performed. Non-linear correction is to obtain the brightness response curve of the projector, and use its inverse function to make a brightness look-up table. Before the projector outputs, first look up the table and then output. The luminance response curve can be generated by obtaining the output of each luminance of the projection with a luminance meter, or the HDR method can be used, that is, the luminance response curve of the projector can be obtained with a digital camera. The advantage of the brightness consistent method is that the brightness of the entire screen can be obtained in theory; the disadvantage is that it is expensive to use a brightness meter, and it is time-consuming and laborious to measure 255 brightness of each projector; the disadvantage is that the entire projection screen is for Consistent with the minimum output, the brightness of the entire image is reduced, wasting a lot of brightness range; the third disadvantage is that the method assumes that the brightness of the overlapping area is the sum of the brightness of each projector, and the brightness obtained by the camera is real and Evenly, but these are not true.

The method of brightness feedback is to modify the template factor of the image according to the comparison between the image acquired by the camera and the reference image. If it is too bright, it will decrease, and if it is too dark, it will increase. For the generation of reference images, non-overlapping areas are directly used in camera images, and overlapping areas are obtained by linear interpolation in non-overlapping areas. The final template with uniform brightness transition is obtained through multiple feedbacks. The advantage of this method is that in theory a uniform transition of brightness is obtained. The disadvantage is that it is greatly affected by camera errors, ambient light, and the smoothness and cleanliness of the screen.

Contents of the invention

The purpose of the present invention is to eliminate the bright bands in the overlapping area displayed by the multi-projector combination, and realize the large-screen high-resolution display of the seamless splicing of the multi-projectors.

The present invention is characterized in that it contains the following steps in sequence:

Step (1): initialization;

Step (1.1): establishing a Cartesian coordinate system for the large screen, the origin is at the lower left corner of the screen;

Step (1.2): form a projector matrix with m×n DLP projectors, m is a row, and n is a column; every two described DLP projectors are controlled by a client computer;

Step (1.3): Each client computer outputs the image composed of 8 rows and 8 columns of equidistant feature points that are uniformly distributed in each projector space to the DLP projector controlled by itself, and uses a digital camera to Taking pictures of the output of each of the DLP projectors at the same position to obtain the feature point image of each of the DLP projectors in the screen space;

Step (1.4): The server performs image processing on each feature point image of the screen space to obtain the feature points of each DLP projector in the screen space, and obtains the feature points of each DLP projector in the screen space according to the position of the feature points. , and establish a one-to-one correspondence between each projector space point and screen space point as the basis for coordinate transformation between each projector space and screen space;

Step (1.5): The server calculates a maximum displayable rectangle RECT according to the initial display area of each DLP projector in the screen space, and the actual display area of each DLP projector on the large screen is through the projection It is obtained by intersecting the initial display area of the projector with the largest rectangle RECT, generating 8 rows and 8 columns of points at equal intervals in the actual display area of each DLP projector screen space, converting these screen space points into the projector space, and generate grid mesh-i;

Step (1.6): The client computer maps the image assigned to the DLP projector controlled by itself to the geometric position corresponding to the grid mesh-i, the remaining positions are cleared with black, and the projection on the screen space is calculated The position information of the overlapping area of the projector is mapped to the projector space;

Step (2): Calculate the template value for different brightness value images according to the following steps;

Step (2.1): The client computer calculates the template value m _linear under the linear transition according to the following linear function:

m _{linear 1} =d1/(d1+d2),

_mlinear2 =d2/(d1+d2),

Wherein, m _{linear 1} is the template value of the left DLP projector, m _{linear 2} is the template value of the right DLP projector, d1 is the distance from a point in the overlapping area to the boundary of the left DLP projector, and d2 is the the distance from the point described in the overlap area section to the border of the right DLP projector;

Step (2.2): the client computer calculates the template value m _cosine under the cosine transition according to the following cosine function,

m _{cosine 1} =cos(d1/(d1+d2)),

m _{cosine 2} = 1-cos(d1/(d1+d2));

Wherein, m _{cosine 1} is the template value of the left DLP projector, m _{cosine 2} is the template value of the right DLP projector, d1 is the distance from a point in the overlapping area to the boundary of the left DLP projector, and d2 is the the distance from the point described in the overlap area section to the border of the right DLP projector;

Step (2.3): The results obtained in the step (2.1) and the step (2.2) are processed with the following nonlinear gamma correction function,

m _γ1 = (m ₁ ) ^1/γ ;

m _γ2 = (m ₂ ) ^1/γ ;

Wherein, m ₁ corresponds to the template value of the left projector under the linear transition or the cosine transition, m ₂ corresponds to the template value of the right projector under the linear transition or the cosine transition, and γ is from 1.6 to 4.8, every Take a value of 0.4 to generate 9 nonlinear gamma functions in total, corresponding to linear transition and cosine transition, and generate 18 brightness correction templates in total;

Step (2.4): The client computer takes a brightness value L every 20 brightness values between brightness values 140 and 255: 140, 160, 180, 200, 220, 240, 255, and then takes a brightness value of 100, There are 8 brightness values in total, and these brightness values are used as RGB values to form 8 pure color images with different brightness;

Step (2.5): The server distributes an image obtained in the step (2.4) to the client computer;

Step (2.6): The client computer multiplies the brightness values of the overlapped parts of the transmitted image by the brightness correction value in a template obtained in the step (2.3), respectively, to obtain the brightness values of each point after brightness correction , so that the original image of the overlapping part is attenuated, and then output to each DLP projector;

Step (2.7): For each of the 18 templates obtained in the step (2.3), repeat the step (2.6), and select the image for the step (2.5) by the method of human eye observation with the best effect a template for

Step (2.8): for each pair in the 8 images that described step (2.4) obtains, repeat described step (2.5), described step (2.6) and described step (2.7), thereby obtain for 8 images 8 optimal templates with different brightness;

Step (3): Carry out the brightness correction of the repeated area on the large screen of the self-adaptive template displayed by the combination of the m×n DLP projectors in the following steps;

Step (3.1): the server collects an image, and distributes the image to be displayed to the client computer;

Step (3.2): The client computer calculates the brightness value L1 _i of each point in the sub-image passed over,

L1 _i =0.299*R+0.587*G+0.114*B,

Wherein, R, G, B are respectively the pixel values of the R, G, and B channels of the point;

Step (3.3): the client computer calculates the average value L2 of the brightness of all points in the sub-image,

$\mathrm{<span\; class="notranslate">\; L</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; L</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>$

Wherein, L1 _i is the brightness value of each point of the sub-image calculated in the step (3.2), and n is the number of points in the sub-image;

Step (3.4): The client computer calculates the brightness distribution of the sub-image, and stores it in the array num, num[i] represents the number of points whose brightness value is equal to the serial number i in the sub-image, and the value of the serial number i is The value is 0 to 255; the sequence number L3 corresponding to the maximum value NL in the array num is the peak brightness of the sub-image;

Step (3.5): the client computer calculates the brightness value L of the sub-image, and when the value of NL obtained in the step (3.4) is greater than 60% of the number of points in the sub-image, the The brightness value L of the sub-image is equal to the brightness peak value L3 of the described sub-image obtained in the step (3.4), otherwise the brightness value L of the sub-image is equal to that of the described sub-image obtained in the step (3.3). Brightness average L2;

Step (3.6): The client computer automatically selects a template with the brightness closest to the brightness value L of the sub-image obtained in the step (3.5) as a correction template;

Step (3.7): The client computer multiplies the adaptive template brightness correction value obtained in the step (3.6) by the brightness value of the overlapping part to obtain the brightness value of each point after brightness correction, so that the The original image of the overlapping part is attenuated, and then output to the DLP projector corresponding to the sub-image;

Step (3.8): repeating the step (3.2) to the step (3.7) for each of the DLP projectors;

Step (3.9): the server repeats the step (3.1) to the step (3.8) after a certain time interval.

The present invention is characterized in that the DLP projectors are divided into two sequences by rows and columns, and in the row sequences, the projections of two projectors with adjacent serial numbers overlap in the horizontal direction. In the column sequence of , the projections of two projectors with adjacent serial numbers overlap vertically.

The present invention is characterized in that all the images are displayed in real time, including video images.

The present invention is characterized in that m=2 and n=2.

Compared with the prior art, the present invention has the following advantages:

1. Adaptively adopt different templates for brightness correction according to different images to be displayed, so that each image has a better brightness correction effect, thereby obtaining a good overall effect.

2. The edge fusion method is used to generate templates, which is simple and easy to operate, avoiding the use of expensive luminance meters and the heavy workload of using HDR methods.

3. The effect of adaptive template brightness correction is not affected by stains and scratches on the screen, nor is it affected by ambient light and light bands produced by special screens.

Description of drawings

Fig. 1 is a schematic diagram of the display area of the projector before correction;

Fig. 2 adaptive template brightness correction module;

Fig. 3 is the flowchart of adaptive template brightness correction module;

The schematic diagram of the process of the initialization module of Fig. 4;

Fig. 5 initialization module flowchart;

Fig. 6 is the flowchart of template generating module;

Fig. 7 is a schematic diagram of edge fusion;

Figure 8 is a schematic diagram of a linear transition;

Figure 9 is a schematic diagram of a cosine transition.

Detailed ways

The present invention is realized through the following technical solutions: m rows and n columns of projectors are used to splice and display a large picture, the image contents of adjacent projectors are all overlapped, and the overlapped areas form bright bands. By generating a template, the value in the template is used to attenuate the original image in the computer and then output to the projector to eliminate the highlight band in the overlapping area. First, 18 templates are pre-generated for images with different brightness to select the template with the best effect; secondly, 8 solid-color images with different brightness are output to each projector, and 8 optimal templates are selected; finally, According to the content of the image to be displayed, different templates are adaptively used for brightness correction. According to the brightness value of the image assigned to each projector in real time, the template with the closest brightness is selected for brightness correction, so that images of various brightness are all the same. Get a better brightness correction effect.

In the present invention, the basis of self-adaptive selection template is the average value or peak value of the brightness of the sub-image of the projector. When the number of points whose brightness is equal to the peak value of the brightness in the sub-image is greater than 60% of the number of points in the sub-image, use Brightness peak, otherwise use brightness average.

The rule for adaptively selecting templates in the present invention is the template whose brightness value is closest to the brightness average value or brightness peak value of the sub-image of the projector.

The template in the present invention is generated by edge fusion. Different templates are generated using different functions. 18 templates are pre-generated, and all the templates are generated by combining linear functions or cosine functions with different gamma functions.

The present invention generates a template every 20 brightness values between 140 and 255, and only takes a template with a brightness value of 100 if the brightness is lower than 140, because images with low brightness are not picky about the template. Finally, 8 optimal templates are obtained for 8 luminances.

FIG. 1 is a schematic diagram of display areas 15 , 16 , 17 , 18 of four projectors 11 , 12 , 13 , 14 on a large screen. In order for the four projectors to splice and display a large picture, each projector only displays a part of the image of the large picture, and the projections between the adjacent projectors of the sub-image content overlap, and these overlapping areas will be highlighted bring. The purpose of the present invention is to eliminate these highlight bands and realize the seamless splicing of multiple projectors for large-screen display.

The present invention mainly has three modules: an initialization module, a template generation module and an adaptive template brightness correction module.

Fig. 2 is a schematic diagram of the process of the adaptive template brightness correction module. There is a server and two computer clients in the figure, each client computer controls two projectors, and there are four projectors 21, 22, 23, 24 in total. Fig. 3 is a flowchart of an adaptive template brightness correction module. The server collects an image, and the content of the sub-image to be displayed is distributed to two client computers, and the client computer calculates the brightness value of the sub-image.

The brightness value L1 of a certain point in the sub-image,

L1＝0.299*R+0.587*G+0.114*B

Among them, R, G, and B are the pixel values of the R, G, and B channels at this point, respectively.

Calculate the average brightness of the image, and count the brightness distribution.

The brightness value L of the sub-image is equal to the average or peak value of the brightness of all points in the sub-image. When the number of points whose brightness is equal to the brightness peak value in the sub-image is greater than 60% of the number of points in the sub-image, use the brightness peak value, otherwise Use the brightness average;

The client computer automatically selects a template with the brightness closest to the brightness value L as the correction template; the client computer multiplies the obtained adaptive template brightness correction value by the brightness value of the overlapping part to obtain the brightness correction value of each point The brightness value is used to attenuate the original image of the overlapped part, and then output to each of the DLP projectors.

After brightness correction, the four projectors spliced to display a rectangular area, as shown in the thick black frame in Figure 2. The actual display area of each projector on the screen is shown as 25, 26, 27, 28 in Fig. 2, which is the intersection area between the original display area and the thick black frame.

Fig. 4 is a schematic diagram of the process of initializing the module. Figure 5 is a flowchart of the initialization module. A projector matrix is composed of m×n DLP projectors, and two adjacent projectors are controlled by a client computer. Let each projector output 8 rows and 8 columns of equally spaced feature points, take pictures with a digital camera, get the feature point image of each projector, perform image processing on the acquired image, and compare the screen space points with the projector The points in the space establish a one-to-one correspondence, which is used as the basis for the conversion between the screen space and the projector space. Obtain the area to be displayed by the projector on the large screen by intersecting with the largest rectangle RECT, map the area to be displayed on the large screen to the projector space, and establish a grid mesh-i in the projector space. Map the image assigned to the projector to the geometric position corresponding to the grid mesh-i, and clear other positions with black. Get the location information of the overlapping area of the projector in the screen space, and map it to the projector space.

Fig. 6 is a flow chart of the template generation module. In the overlapping area of each projector, 18 different functions are used to generate the template, and the calculation method of the template will be described in detail later. Finally, the server sends 8 solid-color images with different brightness to the computer client, and the client uses 18 different templates for brightness attenuation, and selects the optimal template for different brightness through human observation.

The present invention uses an edge fusion method to eliminate the highlight bands in the overlapping areas. Use different functions to get different templates, all of which use distance as an independent variable. For any point in the overlapping area, the closer it is to which projector, the greater the value in the template. In Figure 7, two projectors are taken as an example. d1 is the distance from a point in the overlapping area to the boundary of the left projector, and d2 is the distance from the same point to the boundary of the right projector. In a linear transition, the template value of the left projector is

m ₁ =d1/(d1+d2),

The template value for the right beamer is

m ₂ =d2/(d1+d2).

Figure 8 is a schematic diagram of a linear transition between two projectors. For cosine transition, the template value of the left projector

m ₁ =cos(d1/(d1+d2)),

The template value for the right beamer is

m ₂ =1-cos(d1/(d1+d2)).

Figure 9 is a schematic diagram of the cosine transition, and it can be found that the cosine function has no sudden change and the transition is smoother. What used in the present invention is a DLP projector, and the relation of the pixel value of projector and brightness is nonlinear gamma function relation, so also will carry out gamma correction to it,

m _γ1 = (m ₁ ) ^1/γ ; m _γ2 = (m ₂ ) ^1/γ ,

Among them, m ₁ is the template value of the left projector under linear transition or cosine transition, and m ₂ is the template value of right projector under linear transition or cosine transition.

Because γ is unknown, we take γ from 1.6 to 4.8, take a γ value every 0.4, and generate 9 γ functions in total. The present invention superimposes 9 different γ functions on the linear transition and the cosine transition respectively to generate templates, and there are 9*2=18 templates in total.

Take a brightness value L every 20 brightness values between brightness values 140 and 255: 140, 160, 180, 200, 220, 240, 255, and then take a brightness value of 100, a total of 8 brightness values, use these brightness values As RGB values, 8 solid-color images with different brightness are formed, and these images are distributed to each client computer, and the brightness correction is performed with the 18 templates generated earlier, and the template with the best effect is taken as the brightness template. Finally, the optimal templates of 8 luminances are obtained, and the template generation functions corresponding to each luminance are shown in Table 1.

brightness 255 240 220 200 180 160 140 100 gamma value 2.0 2.8 3.2 3.6 4.8 4.0 2.5 2.8 Linear transition Cosine transition linear linear linear linear cosine cosine linear cosine

Table 1 Optimal templates for each brightness

Claims (4)

1. the self-adapting template big-screen brightness emendation method of multi-projector combination demonstration is characterized in that, contains following step successively:

Step (1): initialization;

Step (1.1): set up a rectangular coordinate system for described large-screen, initial point is in the lower left corner of this screen;

Step (1.2): form a projecting apparatus matrix with m * n DLP projecting apparatus, m is row, and n is row; Per two described DLP projecting apparatus are controlled with a client computer;

Step (1.3): the image that described each client computer is formed the equally spaced characteristic point of equally distributed 8 row, 8 row in each projecting apparatus space outputs to respectively in the DLP projecting apparatus of oneself controlling, the output of described each DLP projecting apparatus is taken pictures respectively in identical position with digital camera, obtain the characteristic point image of described each DLP projecting apparatus at screen space;

Step (1.4): server carries out image processing to described each characteristic point image of screen space and obtains the characteristic point of described each DLP projecting apparatus at screen space, obtain the initial display of described each DLP projecting apparatus according to characteristic point position at screen space, and the point of the point in each projecting apparatus space and screen space is set up one-to-one relationship, as the foundation of Coordinate Conversion between each projecting apparatus space and the screen space;

Step (1.5): described server calculates the shown rectangle RECT of a maximum in the initial display of screen space according to described each DLP projecting apparatus, each DLP projecting apparatus actual zone that will show on large-screen is to ask to hand over by the initial display of this projecting apparatus and maximum rectangle RECT to obtain, equidistantly generate the point of 8 row, 8 row in the actual displayed zone in each DLP projecting apparatus screen space, the point of these screen spaces is transformed in the projecting apparatus space, and generating mesh mesh-i;

Step (1.6): described client computer handle is distributed to the image mapped of the DLP projecting apparatus of oneself controlling to the corresponding geometric position of described grid mesh-i, remove with black all the other positions, calculate the positional information of the overlapping region of the above projecting apparatus of screen space, and be mapped in the projecting apparatus space;

Step (2): calculate stencil value according to the following steps at different brightness value images;

Step (2.1): described client computer is pressed the stencil value m linearity under the following linear function calculating linear transitions:

m _{Linear 1}=d1/ (d1+d2),

m _{Linear 2}=d2/ (d1+d2),

Wherein, m _{Linear 1}Be the stencil value of left side DLP projecting apparatus, m _{Linear 2}Be the stencil value of the right DLP projecting apparatus, d1 is the distance that a bit arrives DLP projecting apparatus border, the described left side in the part of described overlay region, and d2 is the distance that a bit arrives DLP projecting apparatus border, described the right described in the part of described overlay region;

Step (2.2): described client computer is by the stencil value m under the transition of following cosine function calculating cosine _Cosine,

m _{Cosine 1}=cos (d1/ (d1+d2)),

m _{Cosine 2}=1-cos (d1/ (d1+d2));

Wherein, m _{Cosine 1}Be the stencil value of left side DLP projecting apparatus, m _{Cosine 2}Be the stencil value of the right DLP projecting apparatus, d1 is the distance that a bit arrives DLP projecting apparatus border, the described left side in the part of described overlay region, and d2 is the distance that a bit arrives DLP projecting apparatus border, described the right described in the part of described overlay region;

Step (2.3): the result that described step (2.1) and described step (2.2) are obtained handles with following non-linear γ correction function again,

m _γ1＝(m ₁) ^1/γ；

m _γ2＝(m ₂) ^1/γ；

Wherein, m ₁The stencil value of left side projecting apparatus under corresponding described linear transitions or the described cosine transition, m ₂Corresponding described linear transitions or described cosine transition be the stencil value of the right projecting apparatus down, and γ from 1.6 to 4.8, gets a value every 0.4, and symbiosis becomes 9 non-linear gamma functions, and corresponding linear transitions and cosine transition, symbiosis become 18 gamma correction templates;

Step (2.4): described client computer is got a brightness value L every 20 brightness between brightness value 140 and 255: 140,160,180,200,220,240,255, get a brightness value 100 again, totally 8 brightness values, as as rgb value, constitute the pure color image of 8 different brightness with these brightness values;

Step (2.5): described server is distributed to described client computer to an image that obtains in the described step (2.4);

Step (2.6): described client computer multiply by intensity correction values in the template that described step (2.3) obtains to the brightness value of the lap of the image that passes over respectively, obtain the brightness value of each point behind gamma correction, so that the original image of described lap decays, in each DLP projecting apparatus that outputs to again;

Step (2.7): each in 18 templates that described step (2.3) is obtained, repeating said steps (2.6) is selected at a best template of the image effect of described step (2.5) by the method for eye-observation;

Step (2.8): each pair in 8 sub-pictures that described step (2.4) is obtained, repeating said steps (2.5), described step (2.6) and described step (2.7), thus obtain 8 optimum templates at 8 different brightness;

Step (3): the gamma correction that carries out the above repeat region of self-adapting template big-screen of described m * n DLP projecting apparatus combination demonstration according to the following steps;

Step (3.1): described collection of server one sub-picture, the image that will show, distribute to described client computer;

Step (3.2): described client computer calculates the brightness value L 1 that passes every bit in the subimage of coming _i,

L1 _i＝0.299*R+0.587*G+0.114*B，

Wherein, R, G, B are respectively R, the G of described point, the pixel value of B passage;

Step (3.3): described client computer calculates the mean value L2 of all some brightness in the described subimage,

Wherein, L1 _iBe the brightness value of the described subimage every bit that calculates in the described step (3.2), n is the number of described subimage mid point;

Step (3.4): described client computer calculates the Luminance Distribution of described subimage, leaves among the array num num[i in] brightness value equals the number of the point of sequence number i in the described subimage of expression, and the value of sequence number i is 0 to 255; Maximum NL corresponding sequence number L3 among the array num is the brightness peak of described subimage;

Step (3.5): described client computer calculates the brightness value L of described subimage, when the value of the NL that obtains in the described step (3.4) greater than the number of described subimage mid point 60% the time, the brightness value L of described subimage equals the brightness peak L3 of the described subimage that obtains in the described step (3.4), otherwise the brightness value L of described subimage equal the described subimage that obtains in the described step (3.3) average brightness L2;

Step (3.6): described client computer automatically selects the template of the immediate brightness of brightness value L of a described subimage that obtains with described step (3.5) as calibration template;

Step (3.7): the adaptive template intensity correction values that described client computer obtains described step (3.6) multiply by the brightness value of described lap, obtain the brightness value of each point behind gamma correction, so that the original image of described lap is decayed, output to again in the DLP projecting apparatus of described subimage correspondence;

Step (3.8): described each DLP projecting apparatus repeating said steps (3.2) is arrived described step (3.7);

Step (3.9): described server is after certain time interval, and repeating said steps (3.1) is to described step (3.8).

2. the self-adapting template big-screen brightness emendation method that multi-projector combination according to claim 1 shows is characterized in that, described DLP projecting apparatus is divided into two sequences by row and column, in described capable sequence, there be the overlapping of horizontal direction in the projection of two projecting apparatus that sequence number is adjacent separately, in described row sequence, there be the overlapping of vertical direction in the projection of two projecting apparatus that sequence number is adjacent separately.

3. the self-adapting template big-screen brightness emendation method that multi-projector combination according to claim 1 shows is characterized in that described all images is to show in real time, comprising video image.

4. the self-adapting template big-screen brightness emendation method that multi-projector combination according to claim 1 shows is characterized in that m=2, n=2.

Images