TWI473493B - Automatic white balance adjustment method for scanner - Google Patents

Description

Scanner automatic white balance calibration method

The present invention relates to a method of white balance automatic calibration, and more particularly to a method suitable for automatic white balance calibration of a scanner.

White balance is a very important concept in the scanning system. It can solve a series of problems of color reproduction and tone processing. The so-called white balance is to adjust the proportion of the three primary colors of red, green and blue under different lighting conditions. Blending becomes white; the object itself shows its own color by reflecting light, but if the color component of the white light source changes, it will affect the color of the reflected light, which will directly or indirectly change the color of the object itself. A chromatic aberration is formed, which is more likely to occur in the case of artificial light.

Conventional scanners usually provide a white light source to illuminate the scanned document, and obtain an image of the document by sensing the red, green, and blue color values of the reflected light. During the actual scanning process, when the background color of the file is not In white, the color of the reflected light will be affected by the background color, which makes the difficulty of identifying the content of the file increase. In order to make the image close to the color of the file content and easy to identify, it is not only necessary to perform white balance calibration, but also needs manual use. The software performs the process of removing the background color from the obtained image.

The main object of the present invention is to provide a scanner automatic white balance calibration method capable of achieving the effect of removing the background color in the document image during the scanning process and facilitating the recognition of the document image in view of the defects of the above-mentioned prior art.

To achieve the above objective, the present invention provides a scanner automatic white balance calibration method, comprising the following steps:

(1) The light source provides white light for scanning to obtain an image;

(2) capturing the red, green and blue color values of the background color of the image;

(3) calculating a coefficient value based on the detected red, green, and blue color values of the background color of the image and the calibration values of the calibration table prestored in the scanner;

(4) Adjust the brightness ratio of the red, green and blue color values in the light source according to the coefficient value.

According to the automatic white balance calibration method of the scanner of the present invention, white light is used for scanning to obtain an image, and red, green and blue color values of the background color of the image are captured, and red and green colors are detected based on the background color of the detected image. The blue color value and the calibration value of the calibration table pre-stored in the scanner calculate a coefficient value, and adjust the brightness ratio of the red, green and blue color values in the light source according to the coefficient value, so that the scanner senses according to the reflected light. The ratio of the red, green, and blue color values of the background color in the image becomes 1..1..1, that is, the image with the background color close to white is obtained, so that the effect of removing the background color in the document image can be achieved during the scanning process. This facilitates the identification of document images.

In order to explain the technical contents, structural features, objects and effects of the present invention in detail, the embodiments are described in detail below with reference to the drawings.

Referring to the first figure, the scanner automatic white balance calibration method of the present invention comprises the following steps:

S001.. The light source provides white light for scanning to obtain an image;

S002.. Capture the red, green and blue color values of the background color of the image;

S003. Calculating a coefficient value based on the detected red, green, and blue color values of the background color of the image and the calibration value of the calibration table pre-stored in the scanner;

S004.. Adjust the brightness ratio of the red, green and blue color values in the light source according to the coefficient value.

Please continue to refer to the first figure. In step S001, the light source provides red (R), green (G), and blue (B) white light with a color ratio of 1..1..1 for scanning, and red and green by sensing reflected light. In the image obtained by the blue color value, the red, green, and blue color values of each pixel image value are calculated according to the equation (1), and the equation (1) is as follows:

DigitalImage_i=

WhiteCalibration_i*(SensorValue-_i-DarkCalibration_i) ----- (1)

Where i denotes the i-th pixel of the image single color channel, DigitalImage_i is the color value of the corresponding color channel obtained by the image value (SensorValue-_i) read by the scanner, and WhiteCalibration_i is the calibration table for the i-th The white calibration value obtained by the pixels, DarkCalibration_i is the black calibration value obtained by the calibration table for the i-th pixel. It is worth mentioning that the calibration table is a uniform white calibration table pre-existing in the scanner before step S001. Therefore, the red, green, and blue color values of the calibration value, that is, the calibration value WhiteCalibrationSheetLevel (CSR, CSG, CSB), can be regarded as a fixed value, and since the calibration table is uniform white, the ratio of CSR, CSG, and CSB is 1 ..1..1.

Then, in step S002, the red, green, and blue color values UserBackgroundLevel (UR, UG, UB) of the jth pixel of the background color in the image are captured, and the image value of the jth pixel is DigitalImage_j (DIR _j , DIG _j , DIB _j ), image value read from the scanner, SensorValue-_j (SR _j , SG _j , SB _j ), white calibration value of the jth pixel WhiteCalibration_j (WR _j , WG _j , WB _j And the black calibration value DarkCalibration_j(DR _j , DG _j , DB _j ) can be obtained by the following equations (2), (3), (4) by equation (1):

DIR _j = UR=WR _j *(SR _j -DR _j ) ----- (2)

DIG _j =UG=WG _j *(SG _j -DG _j ) ----- (3)

DIB _j =UB=WB _j *(SB _j -DB _j ) ----- (4)

Next, as shown in step S003, according to the calibration table value WhiteCalibrationSheetLevel (CSR, CSG, CSB) pre-stored in the scanner and the above calculated background color red, green, blue color value UserBackgroundLevel (UR, UG, UB According to the calculation of equation (5), a set of coefficient values weighting (wtR, wtG, wtB) can be obtained, and equation (5) is as follows:

Weighting=

WhiteCalibrationSheetLevel/UserBackgroundLevel ----- (5)

The equation below can be obtained by equation (5):

wtR=CSR/UR ----- (6)

wtG=CSG/UG ----- (7)

wtB=CSB/UB ----- (8)

The image value DigitalImage_j (DIR _j , DIG _j , DIB _j ) of the jth pixel is multiplied by the coefficient value weighting (wtR, wtG, wtB) to obtain a new pixel image value, that is, the jth obtained after the calibration. The image value of the pixel is NewDigitalImage_j (NDIR _J , NDIG _J , NDIB _J ), as in equation (9)..

NewDigitalImage_j=DigitalImage_j*weighting ----- (9)

And from equations (2) to (9) can be obtained:

NDIR _J = DIR _j * wtR = UR* (CSR/UR)= CSR

NDIG _J = DIG _j * wtG = UG* (CSG/UG) = CSG

NDIB _J = DIB _j * wtB = UB* (CSB/UB) = CSB

Therefore, the ratio between NDIR _J , NDIG _J , and NDIB _J is the same as that of CSR, CSG, and CSB, both of which are 1..1..1, which is the red, green, and blue color values of the original image background color multiplied by After the coefficient value weighting (wtR, wtG, wtB), the ratio of the red, green, and blue color values of the background color in the image obtained by the rescan can be changed to 1..1..1.

Finally, in step S004, the scanner adjusts the brightness ratio of the red, green, and blue color values in the light source according to the coefficient value weighting (wtR, wtG, wtB), so that the background color of the image obtained by the reflected light is red, green, The blue color value ratio becomes 1..1..1, that is, the image with the background color close to white is obtained, and the effect of removing the background color can be achieved during the scanning process, which facilitates the subsequent recognition of the image.

The above principle will be further described below in conjunction with the embodiments.

Referring to the second figure, the scanner automatic white balance calibration apparatus 100 for performing the scanner automatic white balance calibration method of the present invention comprises an LED light source driver 101, an LED light source 102, an image sensor 103, an adjustment mechanism 104, a storage unit 105, and Processing mechanism 106. The LED light source driver 101 drives the LED light source 102 to emit white light having a red (R), green (G), and blue (B) color value ratio of 1..1..1 on the scanned document. The image sensor 103 senses the background color of the document according to the reflected light, and captures the red, green, and blue color values of the background color, and transmits them to the adjustment mechanism 104. After the adjustment mechanism 104 reads the calibration table value in the storage unit 105, it calculates a red, green, and blue color value of the background color to obtain a coefficient value.

In this embodiment, the calibration table is uniform white with red, green, and blue color values (230, 230, 230). Thereafter, the adjusting mechanism 104 controls the LED light source driver according to the coefficient value, and then adjusts the light emitted by the LED light source, and then scans again, so that the image sensor 103 detects the background color red, green, and blue according to the reflected light. The ratio of the color value is 1..1..1, that is, the image with the background color close to white is obtained, so the effect of removing the background color in the document image can be achieved during the scanning process, and the subsequent recognition of the document image is facilitated, and finally the result is obtained. The image is passed to processing unit 106 for subsequent processing.

In summary, the scanner automatic white balance calibration method provides white light for scanning by using a light source to obtain an image, and after capturing the red, green, and blue color values of the background color of the image, based on the detected background color of the image. , green, blue color value and the calibration value of the calibration table pre-stored in the scanner, calculate a coefficient value, adjust the brightness ratio of the red, green and blue color values in the light source according to the coefficient value, so that the scanner senses according to the reflected light In the image of the image, the ratio of the red, green, and blue color values of the background color becomes 1..1..1, that is, the image with the background color close to white is obtained, so that the background color in the document image can be removed during the scanning process. The effect is convenient for the recognition of document images.

100. . . Scanner automatic white balance calibration device

101. . . LED light source driver

102. . . LED light source

103. . . Image sensor

104. . . Adjustment mechanism

105. . . Storage unit

106. . . Processing organization

S001. . . step 1

S002. . . Step 2

S003. . . Step 3

S004. . . Step 4

The first figure is a flow chart of the automatic white balance calibration method of the authoring scanner.

The second figure is an architectural diagram of an embodiment of the automatic white balance calibration method of the authoring scanner.

S001. . . step 1

S002. . . Step 2

S003. . . Step 3

S004. . . Step 4

Claims (7)

A scanner automatic white balance calibration method includes the following steps:
(1) The light source provides white light for scanning to obtain an image;
(2) capturing the red, green and blue color values of the background color of the image;
(3) calculating a coefficient value based on the detected red, green, and blue color values of the background color of the image and the calibration values of the calibration table prestored in the scanner;
(4) Adjust the brightness ratio of the red, green and blue color values in the light source according to the coefficient value. The scanner automatic white balance calibration method according to claim 1, wherein the coefficient value of the step (3) is calculated as:
Weighting=WhiteCalibrationSheetLevel/UserBackgroundLevel, where weighting represents the coefficient value, WhiteCalibrationSheetLevel represents the calibration value of the calibration table, and UserBackgroundLevel represents the red, green, and blue color values of the background color of the image in step (2). The scanner automatic white balance calibration method according to claim 2, wherein the coefficient value is calculated to make the red, green, and blue color value ratios of the background color of the image become 1..1..1. The scanner automatic white balance calibration method according to claim 3, wherein the calculation method is:
NewDigitalImage_j=DigitalImage_j*weighting, where j represents the jth pixel of the background color of the image in step (1), DigitalImage_j represents the image value of the jth pixel, weighting represents the coefficient value, and NewDigitalImage_j represents the calibration. The image value of the jth pixel. The scanner automatic white balance calibration method according to claim 1, wherein the step (1) further comprises: pre-storing the calibration table into the scanner. The scanner automatic white balance calibration method according to claim 1, wherein the calibration table is uniform white, and the red, green, and blue color values of the calibration value can be regarded as a fixed value, and the ratio of the three is 1 ..1..1. The scanner automatic white balance calibration method according to claim 1, wherein the ratio of the red, green and blue color values of the white light in the step (1) is 1..1..1.