KR100725837B1 - Automatic exposure control method of image sensor - Google Patents

Abstract

The present invention relates to an automatic exposure control method for adjusting an exposure time and a gain in an image sensor so that an average value of brightness of an image reaches a target brightness, the method comprising: (a) dividing an entire image into a plurality of zones, Calculating a current average brightness by weighting; (b) comparing and checking whether the current average brightness is close to the target brightness; (c) calculating a target brightness to which a weight is applied by the exposure control speed value; (d) calculating an exposure amount to be applied to a next frame using the weighted target brightness value; (e) calculating an accumulation time and a gain value to be applied to the next image from the exposure amount; And (f) obtaining a modified gain to be applied to the actual gain circuit by using a gain value to be applied to the next frame.

Image sensor, exposure control

Description

TECHNICAL FIELD The present invention relates to a method of controlling exposure of an image sensor,

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view showing the relationship between an image sensor and an automatic exposure controller; Fig.

Figure 2 shows an automatic exposure control algorithm according to the present invention;

Fig. 3 is a diagram showing an entire image divided into two zones; Fig.

4A shows that the average brightness is linearly proportional to the exposure amount.

FIG. 4B is a diagram showing the final dose under the basic assumption of FIG. 4A. FIG.

5 is a diagram showing a relationship between an exposure amount, an accumulation time, and a gain;

6 is a diagram illustrating an embodiment of gain conversion;

Figures 7 and 8 are diagrams illustrating changes in frame rate through changes in frame height.

The present invention relates to an image sensor, and more particularly, to a method of controlling exposure of an image sensor.

The term " exposure " used in the present invention includes the concepts of charge integration time and gain. The charge accumulation time is the time taken to read the stored charge from the time when a pixel is reset and then begins to receive light again. The gain indicates the degree of amplification of charges generated in proportion to the accumulation time by some kind of analog / digital method. If there is enough light, adjust the exposure control by setting the gain to 1 and adjusting the charge accumulation time. However, in an environment with insufficient lighting, a bright image can not be obtained even when the exposure time is maximized. Therefore, a bright image is properly obtained by applying a gain greater than 1.

1 shows the relationship between the image sensor and the automatic exposure controller.

Referring to FIG. 1, the automatic exposure controller receives image data from the image sensor, processes the image data, and sends the accumulated accumulation time and gain to the sensor.

The present invention relates to an exposure control algorithm in which an average value of brightness of an image reaches a target value, in which a complicated exposure control function such as a fluorescent flicker and a frame height change is simply and consistently processed We suggest a method.

According to an aspect of the present invention, there is provided an automatic exposure control method for an image sensor, the method comprising: (a) dividing an entire image into a plurality of regions, calculating a current average brightness by weighting the divided regions; (b) comparing and checking the current average brightness and the target brightness; (c) calculating a target brightness to which a weight is applied by the exposure control speed value; (d) calculating an exposure amount to be applied to a next frame using the weighted target brightness; (e) calculating an accumulation time and a gain value to be applied to the next image; And (f) obtaining a modified gain for application to an actual gain circuit using a gain value to be applied to the next frame; And a control unit.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 2 illustrates an automatic exposure control algorithm according to an embodiment of the present invention, which includes an average brightness calculation step S10, an automatic exposure error checking step S20, a target brightness calculation step S30 to which a weight is applied based on the exposure control speed value, An exposure amount calculation step S40 to be applied to the next frame, a storage time and gain value calculation step S50 to be applied to the next image, and a gain conversion step S60.

The automatic exposure control method according to the present invention will be described in detail with reference to FIG. 2.

In the average brightness calculation step (S10), the entire image is divided into several zones for the first exposure control.

The example of Figure 3 shows the entire image divided into two sections.

The brightness average value (average brightness) of the center area 12 and the periphery area 14 is obtained by dividing the area by the total brightness. Then, a weight is given to the central region 12 to obtain a brightness average value.

Here, Y_mean_c and Y_mean_p mean the average brightness in the central region 12 and the average brightness in the peripheral region 14, respectively, and Y_mean means the average of the average of the central region 12 and the peripheral region 14, , And #center and #periphery mean the area (number of pixels) of the central region 12 and the peripheral region 14, respectively.

If the central region 12 is given a large weight, the central region 12 will be referred to more in calculating the exposure. That is, if the central region 12 is dark, the exposure is adjusted to illuminate the central region 12 even if the surroundings are bright. On the contrary, if the central region 12 is bright, the exposure is controlled so that the brightness of the central region 12 can be further lowered even if the surroundings are dark.

In the present invention, consideration is given to the case where only the center and the periphery are divided into two zones, and it is easily applicable even when divided into more areas.

In the automatic exposure error checking step (S20), if the value obtained by dividing the absolute value of the difference between the current average brightness and the target brightness by the target brightness is smaller than the tolerance, the automatic exposure control is not performed. This can be expressed by Equation (2).

Where Y_mean is the current average brightness calculated above, Y_target = target brightness, and Lock_threshold is the tolerance.

If the current average brightness is sufficiently close to the target brightness, automatic exposure control is not performed.

In the automatic exposure error checking step (S20), if the current average brightness is different from the target brightness, the automatic brightness control algorithm enters the target brightness calculation step to which the weight is applied by the exposure control speed value.

In the target brightness calculation step S30 to which the weight is applied by the exposure control speed value, the target brightness is calculated by weighting the exposure control speed, which means how fast the target is to be reached.

FIG. 4A shows a basic assumption that the average brightness is linearly proportional to the exposure. The higher the exposure, the higher the average brightness, and the lower the exposure, the lower the average brightness.

FIG. 4B shows the final exposure target value under the basic assumption of FIG. 4A.

According to FIG. 4B, if the current exposure amount 1, the current average brightness 2, and the target brightness 3 are known, the exposure amount 4 to be applied in order to reach the target brightness 3 is also known. That is, the final exposure amount is determined by multiplying the target brightness by the ratio of the target brightness to the current brightness.

Where N_exposure is the dose that should be applied to reach the target brightness at once, and exposure is the dose applied to the current image. However, it is better to control the speed of exposure control properly because exposure control is not so good when it is made too suddenly, and there is a possibility of even oscillation. The variable used is the exposure control speed. That is, the target brightness to which the weight is applied by the exposure control speed value is determined as follows.

Where Y_target_w is the target brightness to which the weight is applied by the exposure control speed value, and speed is the exposure control speed value. Y_target is the previous target brightness, and Y_mean is the current average brightness.

For example, if the exposure control speed value is 0.5, the intermediate value between the current average brightness and the target brightness can be set as the new target brightness. The speed of exposure control can be changed by adjusting the exposure control speed between 0 and 1.

Once the new target brightness is determined, the next exposure is calculated in the exposure calculation step to be applied to the next frame. The following exposure amount can be calculated as shown in Equation (5) by applying the above two expressions.

Where N_exposure means the amount of exposure to be applied to the next image, and exposure means the amount of exposure applied in the current image. Also, Y_target_w indicates the target brightness in consideration of the speed.

When the exposure amount to be applied to the next image is obtained, the next accumulation time and the next gain are calculated.

The next accumulation time and the next gain can be calculated as Equation (6).

Where N_Time denotes the charge accumulation time to be applied to the next image. The maximum accumulation time is one frame. If you want to increase the accumulation time, you need to increase the size of the image frame to extend the time for one frame. The maximum accumulation time specified by the user is called Max_Time. The automatic exposure controller should adjust the accumulation time within the maximum accumulation time. f_period is the period of fluorescent flicker contrast. To obtain a flicker-free image, the accumulation time should be adjusted to be an integer multiple of the flicker contrast period. M is the maximum value among all integers that satisfy both conditions where N_Time is less than or equal to max_Time and N_Time is less than or equal to N_exposure. N_exposure is the amount of exposure to be applied to the next frame calculated above, and N_gain denotes a gain value to be applied in the next frame.

Since the amount of exposure (N_exposure) includes the accumulation time (N_time) and the gain (N_gain), if the exposure amount becomes larger than the maximum accumulation time, the gain is generated from the remaining value obtained by subtracting the maximum accumulation time from the exposure amount. If you want to get more exposure than the maximum accumulation time, you have to increase the gain.

When adjusting the accumulation time, flicker of light should be considered. It is already known that when the accumulation time is arbitrarily set in an environment in which an AC light (fluorescent lamp) is present, exposure time may be determined only by an integral multiple of the lightness and darkness of the light.

The accumulation time must be less than the maximum exposure time and must be smaller than the exposure amount, since the accumulation time must be set to an integer multiple of the flicker period to obtain a flicker-free image, which must be an integral multiple of the flicker period. For flicker-free illumination, set the flicker period to 1. Then the accumulation time is 1, 2, 3, 4, ... Which is proportional to the exposure amount.

If the accumulation time is set to an integer multiple of the flicker period, a phenomenon of flickering may occur during exposure control due to a large difference in brightness between the (N) period exposure image and the (N + 1) period or N-1 period exposure image . In some cases, the proper brightness can not be obtained. In order to avoid this, it is necessary to change the gain as well as the accumulation time to obtain a smooth image.

FIG. 5 is a graph showing the relationship between the exposure amount to be applied to the next image, the accumulation time and the gain, and the gain to be applied in the next frame can be calculated through the exposure amount and the accumulation time to be applied to the next image.

5, the gain value to be applied in the next frame will be described in detail by intervals of the accumulation time to be applied to the next image.

1 section is a section where the exposure amount gradually increases from 0, and the exposure amount is smaller than one flicker period. At this time, the exposure amount is the accumulation time. In other words, when the environment is too bright, if the exposure time is given as 1 flicker period in order to eliminate the flicker, the overexposure bright image appears, so that the exposure time is an integer multiple of f_period and the flicker is generated. However, And controls it in a direction to appropriately adjust it. At this time, the gain is fixed to 1 times.

2, the exposure amount is equal to or greater than one flicker period (1 * f_period), smaller than two flicker periods (2 * f_period), and the accumulation time is fixed to one flicker period (1 * f_period). At this time, the gain becomes the ratio of the exposure amount to the accumulation time. In this case, since the accumulation time is one flicker period (1 * f_period), it can be calculated as follows.

Therefore, the gain is greater than or equal to 1 and less than 2.

In section 3, the exposure amount is equal to or greater than 2 flicker periods (2 * f_period) and less than 3 flicker periods (3 * f_period), and the accumulation time is fixed to 2 flicker periods (2 * f_period). In this case, the gain can be calculated as previously calculated, and the gain is greater than or equal to 1 and less than 1.5. That is, Equation 8 is obtained.

In this way, it is possible to set the accumulation time and gain in the same way from 4 to 6 intervals.

In section 7, the exposure amount is equal to or greater than the 5 flicker period (5 * f_period), and the accumulation time becomes 5 flicker period (5 * f_period). In this case, the gain can be calculated as shown in Equation (9).

After calculating the next gain in the same manner as described above, the gain conversion step is entered.

FIG. 6 shows an embodiment of the gain conversion, in which the linear gain is converted into a nonlinear gain.

The gain value (N_gain) to be applied in the next frame obtained above is a linear value. That is, if the gain is 2.5, the gain is 2.5 times. However, when realizing a gain with a semiconductor circuit, it is often implemented as an exponential type or another type of gain curve. Therefore, in order to apply it to the actual gain circuit, it is necessary to convert the linear gain obtained in the previous step to the final gain value (C_gain).

Exposure variables (exposure and time) defined above are based on the pixel clock cycle. That is, exposure = 100 is equivalent to exposure during reading of 100 pixels. Therefore, when the size of the image frame is given as (frame_width, frame_height), the maximum accumulation time is (frame_width x frame_height).

The factors determining the frame rate of video are frame width, frame height, and clock speed. Therefore, there are ways to change the frame rate by changing the previous frame width, frame height, and clock speed.

Figures 7 and 8 show a change in frame rate through a change in frame height.

The entire variable height frame can be divided into an active window, which is the part where the image is actually seen, and a timing frame, which is an invisible part.

An exposure amount smaller than or equal to the width of the set basic frame can be adjusted within the basic frame without changing the size (area) of the frame. However, if the maximum accumulation time is set larger than the frame width x frame height value, if the exposure amount is larger than the frame width, the accumulation time should be longer than the basic frame area. However, The dose does not maintain the indicated accumulation time. Therefore, in this case, you should adjust the frame height to a larger value to increase the time it takes to read one frame.

If the dose continues to increase and becomes greater than the specified maximum accumulation time, the area of the frame is fixed to the extent that it can accommodate the maximum accumulation time, and the remaining dose is used to form the gain value. The gain value is expressed by Equation (10).

According to FIG. 8, in the section a, the exposure amount is smaller than the frame area, and the frame rate is fixed to the maximum value (the frame height is minimum). In the section b, the exposure amount is equal to or larger than the frame area, It is a small value. In the section b, the frame height value changes according to the exposure amount, thereby changing the frame rate.

In section c, the exposure rate is greater than or equal to the maximum accumulation time, and the frame rate is fixed at the minimum value (frame height is maximum), and does not change.

You can change the frame rate by changing the frame width in the same way as changing the frame height.

Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit of the invention.

The automatic exposure control method according to the present invention has an effect that the image sensor has a proper brightness by using a combination of proper exposure time and gain, thereby improving the image of the image sensor.

In addition, the present invention controls only a single variable called "exposure amount " to easily implement complicated algorithms such as flicker removal of a fluorescent lamp and frame rate change.

Claims (8)

An exposure control method for an image sensor, (a) dividing an entire image into a plurality of zones, weighting each of the divided zones to calculate a current average brightness; (b) comparing and checking the current average brightness and the target brightness; (c) calculating a target brightness to which a weight is applied by the exposure control speed value; (d) calculating an exposure amount to be applied to a next frame using the target brightness value; (e) calculating an accumulation time and a gain value to be applied to the next frame by the exposure amount; And (f) obtaining a modified gain for application to an actual gain circuit using a gain value to be applied to the next frame. The method of claim 1, wherein step (a) (Average brightness) of the center area and the periphery area is obtained as a value obtained by dividing the area by the total brightness and then a weight is given to the center area to obtain a brightness average value. A method of controlling automatic exposure in an image sensor. 2. The method of claim 1, wherein step (b) Wherein the exposure control is not performed if the ratio of the difference between the average brightness and the target brightness is smaller than the tolerance rate. 2. The method of claim 1, wherein step (c)

(Where Y_target_w is the target brightness to which the weight is applied based on the exposure control speed value, speed is the exposure control speed value, Y_target is the target target brightness, and Y_mean is the current average brightness) Is calculated by the following equation: < EMI ID = 1.0 > 2. The method of claim 1, wherein step (d)

(Where exposure is the exposure applied for the current image, N_exposure is the exposure to be applied to the next image, Y_mean is the value representing the brightness of the current image, and Y_target is the value representing the target brightness). Is calculated by the following equation: < EMI ID = 1.0 > 6. The method of claim 5, wherein in step (d)

(Where speed is a variable that specifies the rate of exposure control and has a value between 0 and 1). Is calculated by the following equation: < EMI ID = 1.0 > 2. The method of claim 1, wherein step (e)

(Where N_Time denotes the accumulation time to be applied to the next image, M denotes the maximum possible integer value, f_period denotes the period of the flicker lightness, and N_gain denotes the gain value to be applied in the next frame , And N_exposure is the amount of exposure to be applied to the next frame calculated above.) Is calculated by the following equation: < EMI ID = 1.0 > 8. The method of claim 7, Wherein when the calculated N_time is greater than one frame time, the frame duration is extended by increasing the height of the frame, thereby securing the exposure time.

Images