CN102023456B - Light metering weight regulating method and device thereof - Google Patents

Abstract

The invention discloses a light metering weight regulating method and device. The light metering weight regulating method comprises the following steps of: firstly, dividing an input image into a plurality of light metering regions, wherein the input image comprises a plurality of pixel data, and the light metering regions correspond to a plurality of exposure weight values; secondly, judging whether the color represented by each pixel datum of the input image is skin color; thirdly, calculating the quantity of the pixel data the color of which is judged to be the skin color in each light metering region; and fourthly, according to the calculating result, regulating the exposure weight values corresponding to the light metering regions.

Description

Photometric weight adjustment method and device

Technical Field

The present invention relates to a photometric weight adjustment method and device, and more particularly, to a multi-zone photometric weight adjustment method and device.

Background

In photography, exposure is the amount of light allowed to impinge on the imaging media during the course of a photograph. The imaging medium is, for example, a film or an image sensor. The unit of exposure is the product of lux and seconds, and can be calculated from Exposure Value (EV) and ambient light in a specific area.

Nowadays, cameras can provide multiple light measurement (light measurement) modes to find the optimum exposure for a photo. In general, the metering modes include a spot metering mode, a center weighted average metering mode, an average metering mode, a local metering mode, and a multi-area metering mode.

Conventionally, when a person image is photographed, the brightness of the face in the photographing environment is usually analyzed to determine the optimal exposure time, so that the face in the photograph has the proper brightness. However, this method needs a lot of mathematical operations and image analysis on the whole image to find the position of the face. Thus, the cost and the amount of computation are increased. In addition, this method also needs to temporarily store the whole image in the memory, which results in high cost and low speed.

Disclosure of Invention

The invention relates to a light metering weight adjusting method and a device thereof, which divide an input image into a plurality of light metering areas, calculate the number of pixel data judged as skin color (skin color) in each light metering area, and adjust the exposure weight value corresponding to each light metering area according to the number. Thus, the exposure time can be determined by the exposure weight values obtained through simple operations, such as calculation quantity, thereby reducing the cost and the operation quantity. Moreover, an embodiment of the invention can process each pixel data in sequence, so that the operation speed can be effectively improved. Furthermore, the embodiment of the invention can avoid using a large-capacity memory to store the whole input image, thereby saving the area of a hardware circuit.

According to an aspect of the present invention, a method for adjusting a photometric weight includes the following steps. Firstly, an input image is divided into a plurality of light metering areas, the input image comprises a plurality of pixel data, and the light metering areas correspond to a plurality of exposure weight values. Then, it is determined whether or not the color indicated by each piece of pixel data of the input image is a skin color. Then, the number of pixel data determined as skin color in each photometric area is calculated. Then, according to the calculation result, the exposure weight values corresponding to the light metering regions are adjusted.

According to another aspect of the present invention, a metering weight adjusting apparatus is provided for dividing an input image into a plurality of metering areas and adjusting a plurality of exposure weight values corresponding to the metering areas. The input image includes a plurality of pixel data. The photometric weight adjustment device includes a determination unit, a calculation unit, and a weight adjustment unit. The judging unit is used for judging whether the color represented by each pixel data of the input image is the skin color. The calculating unit is used for calculating the number of the pixel data which are judged to be skin color in each light measuring area. The weight adjusting unit is used for adjusting the exposure weight values corresponding to the light metering regions according to the calculation result of the calculating unit.

In order to make the aforementioned and other objects of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a flowchart illustrating a method for adjusting a photometric weight according to an embodiment of the present invention.

FIG. 2 is a block diagram of a photometric weight adjustment apparatus according to an embodiment of the present invention.

Fig. 3A is a schematic diagram illustrating an image including a human face captured by an image sensing unit.

FIG. 3B shows a plurality of photometric areas of the image of FIG. 3A.

Fig. 3C is a schematic diagram illustrating an example of the distribution of the skin color spots generated by the determination unit according to the input image of fig. 3A on the plurality of light metering regions of fig. 3B.

FIG. 3D is a schematic diagram showing an example of the distribution of the flesh tone spot of FIG. 3C over the plurality of photometric areas of FIG. 3B after passing through the filtering unit.

FIG. 3E is a schematic diagram showing an example of the ratio of the flesh tone spots to the plurality of photometric areas.

FIG. 3F is a diagram illustrating an example of the exposure weight value set by the weight adjustment unit according to the input image of FIG. 3A according to an embodiment of the invention.

FIG. 3G shows a central weighted exposure weight value.

Fig. 4 is a schematic diagram illustrating an example of a filtering unit of the photometric weight adjusting apparatus of fig. 2.

[ description of main element symbols ]

102: image sensing unit

200: photometric weight adjustment device

210: color ratio calculation unit

220: judging unit

230: filter unit

232: judgment matrix

234: multi-input AND gate

240: computing unit

250: weight adjustment unit

P (n): pixel

K: first reference value

Z: second reference value

Im: inputting an image

MA (1) to MA (25): light metering area

Rg 1-Rg 11: register with a plurality of registers

S110 to S140: procedure step

WEV1, WEV 2: exposure weight value

Detailed Description

Referring to fig. 1, a flowchart of a method for adjusting a photometric weight according to an embodiment of the invention is shown. The method for adjusting the light weight includes the following steps. In step S110, an input image is divided into a plurality of photometric areas. The input image comprises a plurality of pixel data, and the light metering regions correspond to a plurality of exposure weight values. In step S120, it is determined whether the color indicated by each piece of pixel data of the input image is a skin color. As shown in step S130, the number of pixel data determined as skin color in each photometric area is calculated. In step S140, a plurality of exposure weight values corresponding to the light metering regions are adjusted according to the result of the calculating step.

A photometric weight adjustment apparatus that performs the photometric weight adjustment method of fig. 1 will now be described as follows. Please refer to fig. 2, fig. 3A and fig. 3B. FIG. 2 is a block diagram of a photometric weight adjustment apparatus according to an embodiment of the present invention. Fig. 3A is a schematic diagram illustrating an input image Im including a human face captured by an image sensing unit. Fig. 3B shows a plurality of light metering areas MA corresponding to the input image Im of fig. 3A.

The photometric weight adjustment device 200 includes, for example, a color ratio calculation unit 210, a determination unit 220, a filtering unit 230, a calculation unit 240, and a weight adjustment unit 250. The photometric weight adjustment device 200 is coupled to an image sensing unit 102. The metering weight adjusting apparatus 200 can divide the input image Im captured by the image sensing unit 102 into 5 × 5 metering areas MA (1) -MA (25) as shown in fig. 3B, for example, but not limited thereto, and adjust the corresponding exposure weight values.

The image sensing unit 102 includes, for example, a complementary metal-Oxide-Semiconductor (CMOS) or a Charge-coupled device (CCD). The input image Im captured by the image sensing unit 102 includes a plurality of pixels. Each pixel corresponds to a stroke of pixel data. For example, the pixel p (n) corresponds to one piece of pixel data. The pixel p (n) includes three sub-pixels, so the pixel data also includes three sub-pixel data. The number of pixels may depend on the specification (e.g., resolution) of the image sensing unit 102.

The color ratio calculating unit 210 is coupled to the image sensing unit 102, and is configured to obtain a plurality of pixel data of the input image Im from the image sensing unit 102 and calculate color components of each pixel data related to skin color. Take the pixel data of the pixel p (n) of fig. 3A as an example. In one embodiment, after obtaining the pixel data of a pixel p (n), the color ratio calculating unit 210 may calculate a first reference value K and a second reference value Z for the pixel p (n) according to the ratio between the sub-pixel data of the pixel data. For example, the two reference values K and Z are calculated as follows:

$K=\sqrt{{\left(\frac{R_{\mathrm{input}}}{G_{\mathrm{input}}}\right)}^2+{\left(\frac{R_{\mathrm{input}}}{B_{\mathrm{input}}}\right)}^2}-1.4---\left(1\right)$

$Z={\tan }^{-1}(\left(\frac{R_{\mathrm{input}}}{B_{\mathrm{input}}}\right)/\left(\frac{R_{\mathrm{input}}}{G_{\mathrm{input}}}\right))-0.7---\left(2\right)$

wherein R is_input、G_inputAnd B_inputThree sub-pixel data, which are expressed as one piece of pixel data, may be red, green, and blue sub-pixel data, respectively. The two reference values K and Z are calculated from the ratio of the pixel data. Then, the two reference values K and Z are transmitted to the determining unit 220 as reference values for skin color determination.

The determining unit 220 is configured to determine whether a color represented by each pixel data of the input image Im is a skin color. Take pixel P (n) as an example. In an embodiment, the determining unit 220 determines whether the coordinates represented by the first reference value K and the second reference value Z are within a predetermined range when determining whether the color represented by the pixel data of the pixel p (n) is a skin color. For example, the determination of whether the coordinates represented by the two reference values K and Z are within the predetermined range is as follows:

<math> <mrow> <msub> <mi>Skin</mi> <mi>Pixel</mi> </msub> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>1</mn> <mo>,</mo> <mi>if</mi> <mn>0</mn> <mo>&le;</mo> <mi>K</mi> <mo>&le;</mo> <mn>2.1</mn> <mi>and</mi> <mn>0</mn> <mo>&le;</mo> <mi>Z</mi> <mo>&le;</mo> <mn>0.35</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> <mi>otherwise</mi> </mtd> </mtr> </mtable> </mfenced> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow> </math>

wherein the predetermined range is K between 0 and 2.1And Z is in the interval between 0 and 0.35. Skin_PixelRepresents the judgment result of one pixel data by the judgment unit 220. Skin_PixelThe value of (a) is used to indicate whether the pixel data is skin tone. For the pixel data of pixel P (n), if Skin_Pixel(n) is designated as 1, indicating that the pixel data of the pixel P (n) is a skin tone. If Skin_Pixel(n) is assigned a value of 0, indicating that the pixel data for pixel P (n) is not skin tone. In other words, for each pixel data, the determining unit 220 generates a corresponding Skin_PixelAnd determining Skin by determining whether the reference values K and Z are the above-mentioned regions_PixelTo know whether each pixel data is skin color. In this example, Skin with a value of 1 is referred to_PixelThe corresponding pixel is a flesh tone spot.

In the above embodiments, the calculation method of the two reference values K and Z and the determination method of the predetermined range can be designed according to different requirements. In addition, the above description uses the color ratio calculating unit 210 to calculate two reference values K and Z of a pixel data, and the determining unit 220 determines whether the pixel data is the skin color data, which is used for illustrating the present invention, but the present invention is not limited thereto. In practice, the manner of determining the skin color may be implemented in other ways, for example, the possible value ranges of the general skin color in the red, green and blue sub-pixel data may be counted first and compared with the three sub-pixel data for determination. In short, it is a feasible embodiment of the present invention to determine whether a piece of pixel data is a skin color when the piece of pixel data is received.

Fig. 3C is a schematic diagram illustrating an example of the distribution of the skin color spots generated by the determining unit 220 according to the input image Im of fig. 3A on the plurality of light metering regions of fig. 3B. In this example, the flesh tone spots are generated according to the expressions (1) to (3). The skin color spots are included in the skin color regions F1 to F3. In this example, the skin color points in the skin color regions F1 and F2 substantially correspond to the positions of the human face in fig. 3A.

However, in practice, there may still be regions that do not correspond to the face position, such as the skin color region F3 of fig. 3C. Since skin color has a continuous character. The continuous characteristic may mean a characteristic that the number of flesh tone spots is large, the aggregated area is large, or the flesh tone spots appear continuously. Therefore, in one embodiment of the present invention, the continuous characteristic can be used to remove unnecessary information (e.g., discrete flesh tone spots) while retaining flesh tone spots closer to the face position, thereby improving the correctness of the flesh tone judgment. This function is performed by the filter unit 230, which is now described below in conjunction with FIG. 2.

The filtering unit 230 is used for determining whether the color represented by one pixel data is the skin color again according to the determination result of the determining unit 220 for one pixel data and a plurality of adjacent pixel data. Take pixel P (n) as an example. In one embodiment, if the determining unit 220 determines that the pixel p (n) is a skin tone point, the filtering unit 230 may determine again according to a plurality of previous pixel data.

For example, referring to fig. 4, a schematic diagram of an example of the filtering unit 230 of the photometric weight adjusting device 200 of fig. 2 is shown. The filter unit 230 includes a 1 × 11 decision matrix 232 and a multi-input and gate 234. The decision matrix 232 includes 11 registers Rg 1-Rg 11, and the multi-input AND gate 234 determines the output thereof according to the contents of the 11 registers. The registers Rg 1-Rg 11 store the judgment results of the pixels P (n) and the first ten pixels (not shown), that is, the Skin generated by the judgment unit according to the formula (3)_Pixel(n-1)～Skin_Pixel(n-10) and Skin_PixelThe value of (n). In this example, if the pixel P (n) and its first ten pixels are determined to be Skin-color dots (i.e., Skin)_Pixel(n) and Skin_Pixel(n-1)～Skin_Pixel(n-10) are all 1), then the multiple-input AND gate 234 will output a value of 1 Skin_{Pixel_f}(n) to indicate that the pixel data of the pixel P (n) is indeed a flesh tone spot. On the contrary, if any of the pixel P (n) and its first ten pixels is not determined as the flesh tone spot (i.e. the flesh tone spot Skin)_Pixel(n) and Skin_Pixel(n-1)～Skin_Pixel(n-10) anyA pen value is not 1), the filtering unit 230 outputs Skin with a value of 0 for the pixel p (n) originally determined as a Skin-color point_{Pixel_f}(n) and re-determining that the pixel P (n) is not a flesh tone spot.

Fig. 3D is a schematic diagram showing an example of the distribution of the skin-color dots of fig. 3C on the plurality of photometric areas of fig. 3B after passing through the filtering unit 230. In the example shown in fig. 3C, the skin color dots in the skin color region F3 can be removed and the skin color dots in the regions F1 and F2 with continuous characteristics can be retained after being judged again by the filtering unit 230. Thus, the accuracy of skin color judgment can be improved.

Please continue to refer to fig. 2. In an embodiment, the calculating unit 240 is configured to calculate a ratio of the number of pixel data determined as skin color in each light metering region to the number of all pixel data covered by each light metering region. The weight adjusting unit 250 is used for adjusting the exposure weight values corresponding to the light metering regions according to the proportions calculated by the calculating unit 240. The calculation unit 240 calculates the ratio, for example, as follows:

<math> <mrow> <msub> <mi>W</mi> <mi>skin</mi> </msub> <mo>=</mo> <mfrac> <msubsup> <mi>Skin</mi> <mi>pixel</mi> <mo>&prime;</mo> </msubsup> <msub> <mi>Block</mi> <mi>pixel</mi> </msub> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow> </math>

wherein, Skin_pixel' represents the number of flesh tone spots in this photometric area; biock_pixelDenotes the number of all pixel data covered by the photometric area, and W_skinRepresenting the number of flesh tone spots as a proportion of the number of all pixel data covered by this photometric area.

Take fig. 3D as an example.The calculating unit 240 may calculate the corresponding ratio W in each of the 25 photometric areas MA (1) to MA (25) according to the formula (4)_skin(1)～W_skin(25) The results are shown in FIG. 3E. Then, the weight adjustment unit 250 adjusts the weight according to the ratio W calculated by the calculation unit 240_skin(1)～W_skin(25) The exposure weight values corresponding to the light measurement regions MA (1) to MA (25) are adjusted.

When adjusting the exposure weight value corresponding to the light metering region, the weight adjusting unit 250 may set a higher exposure weight value to the light metering region with a higher proportion, and may also provide an exposure value reference near the low weight region, so as to avoid the excessive exposure for the face region and make the whole image inconsistent. Referring to fig. 3F, a schematic diagram of an example of the exposure weight value WEV2 set by the weight adjusting unit 250 according to the input image Im of fig. 3A according to an embodiment of the invention is shown. The plurality of exposure weight values WEV2(1) to WEV2(25) set by the weight adjustment means 250 for the light metering regions MA (1) to MA (25) are, for example, indicated by the numerical values of the regions. The light metering regions (such as light metering regions MA (12) and MA (9)) containing more skin color spots are set with higher exposure weight values. Thus, the exposure weight values provided for the regions in the present embodiment correspond to the proportion of the skin color points in the regions of the image, so that the accuracy of skin color exposure can be increased.

In another embodiment, the calculating unit 240 is configured to calculate a number of pixel data determined as Skin color in each photometric area, for example, Skin 'of formula (4)'_PixelAs shown. The weight adjusting unit 250 is used for adjusting the exposure weight values corresponding to the light metering regions according to the quantity calculated by the calculating unit 240. Then, the weight adjusting unit 250 adjusts the exposure weight values corresponding to the photometric areas MA (1) to MA (25) according to the number calculated by the calculating unit 240. Similarly, the weight adjustment unit 250 may also set a higher exposure weight value to a higher number of photometry areas, and the result may also be shown in fig. 3F.

The above description is given by taking the calculation unit 240 as an example to calculate the ratio or number of the skin-color spots, but the invention is not limited thereto. It is possible to implement the present invention as long as a reference value that is meaningful for the skin color statistic value in each photometric range can be calculated from the skin color point. In the example shown in fig. 3B, the input image Im is divided into 25 light metering regions MA (1) to MA (25), and the number of light metering regions is not limited to the number shown in the present embodiment. The number of light metering regions should be designed according to the user's needs and practical application.

As can be seen from the above description, the metering weight adjusting device can adjust the exposure weight value of each metering region by calculating the number or ratio of flesh tone spots (i.e., pixel data determined as flesh tones), and accordingly determine the exposure time. Therefore, the photometric weight adjustment device of the present embodiment can determine the exposure weight value through simple operations, such as calculating the number or ratio. Thus, the cost can be reduced and the calculation amount can be reduced.

In addition, in another embodiment, the weight adjusting unit 250 may adjust the exposure weight value as follows. Taking fig. 3A as an example, it is assumed that the pixel data of the input image Im is obtained in accordance with a plurality of first exposure weight values WEV1(1) to WEV1(25) corresponding to the light metering regions MA (1) to MA (25), which are, for example, the center-weighted exposure weight values shown in fig. 3G. At this time, the weight adjusting unit 250 is configured to generate a plurality of second exposure weight values WEV2(1) -WEV 2(25) corresponding to the photometric areas MA (1) -MA (25) according to the calculation result (the number of the above-mentioned skin color dots or the ratio thereof) of the calculating unit 240, which is shown in fig. 3F, for example. Then, the weight adjusting means 250 determines the exposure weight values corresponding to the light metering regions MA (1) to MA (25) based on the first exposure weight values WEV1(1) to WEV1(25) and the second exposure weight values WEV2(1) to WEV2(25) corresponding to the light metering regions MA (1) to MA (25). In practice, the exposure weight values of the photometric areas MA (1) to MA (25) can be obtained by appropriately calculating the two exposure weight values WEV1 and WEV 2. For example, the exposure weight values of the respective light metering regions MA (1) to MA (25) can be obtained by adding or taking the maximum value of the exposure weight values WEV1 and WEV2 in correspondence with each other.

In an embodiment, the photometric weight adjustment device 200 can further increase the operation speed, as described below. In an implementation, referring to fig. 2, the color ratio calculating unit 210 may sequentially obtain a plurality of pixel data of the input image Im from the image sensing unit 102. In other words, the color ratio calculating unit 210 obtains one pixel data of the input image Im at a time. Then, the calculating unit 240 performs a calculating operation according to the photometric area to which each pixel data belongs. Thus, the photometric weight adjustment device 200 can sequentially process each pixel data, and the operation speed of the photometric weight adjustment device 200 can be effectively increased. Therefore, the present embodiment can achieve substantially real-time exposure correction, and can be applied to dynamic video technology. In addition, since the photometric weight adjustment apparatus 200 can avoid the use of a large-capacity memory device, the area of a hardware circuit can be saved.

In the photometry weight adjusting method and the photometry weight adjusting device disclosed in the above embodiments of the present invention, an input image is divided into a plurality of photometry areas, the number of pixels determined as skin color in each photometry area is calculated, and the exposure weight value corresponding to each photometry area is adjusted accordingly. Therefore, the cost and the calculation amount can be reduced. Moreover, an embodiment of the invention can process each pixel data in sequence, so that the operation speed can be effectively improved. In addition, the embodiment of the invention can avoid using a large-capacity memory to store the whole input image, thereby saving the area of a hardware circuit.

While the invention has been described with reference to a preferred embodiment, it is not intended to be limited thereto. Various modifications and alterations may be made by those skilled in the art without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.

Claims (16)

1. A photometric weight adjustment method, comprising:

dividing an input image into a plurality of light metering areas, wherein the input image comprises a plurality of pixel data, and the light metering areas correspond to a plurality of exposure weight values;

judging whether the color represented by each pixel data of the input image is the skin color;

calculating the number of the pixel data judged as skin color in each light measuring region; and

adjusting the exposure weight values corresponding to the light metering regions according to the result of the calculating step;

wherein, the step of judging whether the color represented by the pen pixel data is the skin color comprises the following steps:

calculating a first reference value and a second reference value according to the proportion among a plurality of sub-pixel data of one pixel data; and

judging whether the coordinates represented by the first reference value and the second reference value are in a preset range;

the step of calculating a first reference value and a second reference value according to the ratio of the plurality of sub-pixel data of one pixel data includes calculating the first reference value K and the second reference value Z as follows:

$K=\sqrt{{\left(\frac{R_{\mathrm{input}}}{G_{\mathrm{input}}}\right)}^2+{\left(\frac{R_{\mathrm{input}}}{B_{\mathrm{input}}}\right)}^2}-1.4---\left(1\right)$

$Z={\tan }^{-1}(\left(\frac{R_{\mathrm{input}}}{B_{\mathrm{input}}}\right)/\left(\frac{R_{\mathrm{input}}}{G_{\mathrm{input}}}\right))-0.7---\left(2\right)$

wherein R is_input、G_inputAnd B_inputThree sub-pixel data, which are respectively red, green, and blue sub-pixel data, are represented as the one piece of pixel data.

2. The photometric weight adjustment method of claim 1, further comprising:

sequentially obtaining the pixel data of the input image from an image sensing unit;

wherein the image sensing unit is a CMOS or CCD.

3. The photometric weight adjustment method of claim 1 wherein the step of determining whether the color represented by the pen pixel data is a skin color further comprises:

and judging whether the color represented by the pixel data is the skin color or not again according to the judgment result of one pixel data and a plurality of adjacent pixel data in the judging step.

4. The photometric weight adjustment method of claim 3 wherein the neighboring pixel data are previous pixel data of the pixel data.

5. The photometry weight adjustment method of claim 1, wherein, in the step of calculating the number of the pixel data determined as a skin color in each photometry area, a proportion of the number of the pixel data determined as a skin color in each photometry area to the number of all pixel data covered by each photometry area is also calculated.

6. The photometric weight adjustment method according to claim 1, wherein in the step of adjusting the exposure weight values corresponding to the photometric areas, a higher exposure weight value is set to a greater number of the photometric areas.

7. The light metering weight adjustment method of claim 1, wherein the pen pixel data of the input image is obtained according to a plurality of first exposure weight values corresponding to the light metering regions, and the step of adjusting the exposure weight values of the light metering regions comprises:

generating a plurality of second exposure weight values corresponding to the light metering regions according to the result of the calculating step; and

and determining the exposure weight values corresponding to the light metering regions according to the first exposure weight values and the second exposure weight values corresponding to the light metering regions.

8. The method of claim 1, wherein the step of determining whether the coordinates represented by the first reference value and the second reference value are within a predetermined range comprises the steps of determining whether the coordinates represented by the first reference value K and the second reference value Z are within a predetermined range as follows:

<math> <mrow> <msub> <mi>Skin</mi> <mi>Pixel</mi> </msub> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>1</mn> <mo>,</mo> <mi>if</mi> <mn>0</mn> <mo>&le;</mo> <mi>K</mi> <mo>&le;</mo> <mn>2.1</mn> <mi>and</mi> <mn>0</mn> <mo>&le;</mo> <mi>Z</mi> <mo>&le;</mo> <mn>0.35</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> <mi>otherwise</mi> </mtd> </mtr> </mtable> </mfenced> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow> </math>

wherein the predetermined range is an interval where K is between 0 and 2.1 and Z is between 0 and 0.35, and Skin_PixelRepresenting the result of the determination for one pixel data.

9. A metering weight adjusting device is used for dividing an input image into a plurality of metering areas and adjusting a plurality of exposure weight values corresponding to the metering areas, wherein the input image comprises a plurality of pixel data, and the metering weight adjusting device comprises:

a judging unit, for judging whether the color represented by each pixel data of the input image is skin color;

a calculating unit for calculating a number of the pixel data determined as skin color in each of the light measuring regions;

a weight adjusting unit for adjusting the exposure weight values corresponding to the light metering regions according to the calculation result of the calculating unit; and

a color ratio calculating unit for calculating a first reference value and a second reference value according to the ratio of a plurality of sub-pixel data of a pixel data;

when the judging unit judges whether the color represented by the pixel data is the skin color, whether the coordinates represented by the first reference value and the second reference value are in a preset range is judged;

wherein, the color ratio calculating unit calculates the first reference value K and the second reference value Z in the following way:

$K=\sqrt{{\left(\frac{R_{\mathrm{input}}}{G_{\mathrm{input}}}\right)}^2+{\left(\frac{R_{\mathrm{input}}}{B_{\mathrm{input}}}\right)}^2}-1.4---\left(1\right)$

$Z={\tan }^{-1}(\left(\frac{R_{\mathrm{input}}}{B_{\mathrm{input}}}\right)/\left(\frac{R_{\mathrm{input}}}{G_{\mathrm{input}}}\right))-0.7---\left(2\right)$

wherein R is_input、G_inputAnd B_inputThree sub-pixel data, which are respectively red, green, and blue sub-pixel data, are represented as the one piece of pixel data.

10. The photometric weight adjustment device of claim 9, wherein the color scale calculation unit is coupled to an image sensing unit for sequentially obtaining the pixel data of the input image from the image sensing unit;

wherein the image sensing unit is a CMOS or CCD.

11. The photometric weight adjustment device of claim 9, further comprising:

and the filtering unit is used for judging whether the color represented by the pixel data is the skin color again according to the judgment result of the judging unit on one pixel data and a plurality of adjacent pixel data.

12. The photometric weight adjustment device of claim 9 wherein the neighboring pixel data are previous pixel data of the pixel data.

13. The photometric weight adjustment device according to claim 9 wherein the calculation unit further calculates a ratio of the number of the pixel data determined to be a skin color in each photometric area to the number of all pixel data covered by each photometric area.

14. The metering weight adjustment device of claim 9, wherein the weight adjustment unit sets a higher exposure weight value to a greater number of the metering regions when adjusting the exposure weight values corresponding to the metering regions.

15. The photometric weight adjustment device of claim 9, wherein the pen pixel data of the input image is obtained according to a plurality of first exposure weight values corresponding to the photometric areas, and the weight adjustment unit is configured to:

generating a plurality of second exposure weight values corresponding to the light metering regions according to the calculation result of the calculation unit; and

and determining the exposure weight values corresponding to the light metering regions according to the first exposure weight values and the second exposure weight values corresponding to the light metering regions.

16. The photometric weight adjustment device of claim 9, wherein the determining unit determines whether the coordinates represented by the first reference value K and the second reference value Z are within a predetermined range by:

<math> <mrow> <msub> <mi>Skin</mi> <mi>Pixel</mi> </msub> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>1</mn> <mo>,</mo> <mi>if</mi> <mn>0</mn> <mo>&le;</mo> <mi>K</mi> <mo>&le;</mo> <mn>2.1</mn> <mi>and</mi> <mn>0</mn> <mo>&le;</mo> <mi>Z</mi> <mo>&le;</mo> <mn>0.35</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> <mi>otherwise</mi> </mtd> </mtr> </mtable> </mfenced> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow> </math>

wherein,the predetermined range is an interval where K is between 0 and 2.1 and Z is between 0 and 0.35, and Skin_PixelRepresenting the result of the determination for one pixel data.

Images