A gain adjustment method and device for preventing overexposure under low illumination

Abstract

The present invention relates to image processing technology, and discloses a gain adjustment method and device for preventing overexposure under low illumination, the method comprising: obtaining analog pixel signals and electronic quantity EQ, obtaining analog pixel signals and electronic quantity EQ through optical conversion for an optical array; gain adjustment, adjusting the gain of the acquired analog pixel signals to obtain adjusted analog pixel signals; digital pixel signal conversion, converting the analog pixel signals into digital pixel signals through an ADC module; output of digital pixel signals, outputting the converted digital pixel signals. The present invention supports secondary adjustment of the configured sensor gain (i.e., the analog signal of all pixels in a frame after photoelectric conversion, the multiple to be amplified, hereinafter represented by global gain) through the sensor end, and the AE algorithm dynamically controls the bright area threshold, while reducing the overbright area, it also ensures the image consistency of the non-overbright area; it has low power consumption and low cost.

Description

Technical Field

[0001] The invention relates to image processing technology, and in particular to a gain adjustment method and device for preventing overexposure under low illumination. Background Art

[0002] Currently, outdoor cameras generally do not support HDR due to cost or power consumption. In low-light conditions at night, when there is light in front, in order to increase the brightness of the dark area, the sensor gain is usually increased significantly, which can easily cause overexposure around the light.

[0003] Adopting HDR multi-frame integration: The cost is relatively increased: Both the sensor and the main control need to support this function. In order to process multi-frame integration, the main control needs to increase the memory space and the time efficiency is relatively increased; the power consumption increases: the main control needs to receive multiple frames to synthesize one frame. In order to achieve the same frame rate in non-HDR mode, it needs to process at least twice the frames, and the power consumption will increase significantly; there is a tailing phenomenon caused by multi-frame integration. Adopting digital wide dynamic mode: This method requires a better ISP algorithm on the main control side. It is easy to bring dark noise; it is easy to cause the face to be too pink due to the excessive pull of digital wide dynamic; it takes a lot of effort to adjust the image to achieve a better balance. When the difference between light and dark is too large, the effect is still not obvious.

[0004] For example, prior art 1: CN200680008406.8 discloses a processing method. The present invention discloses an imaging device for reducing fixed pattern noise caused by dark current or the like. Summary of the invention

[0005] The present invention aims at the high cost of adopting the HDR multi-frame integration method in the prior art, and the need for both the sensor and the main control to support this function. In order to process the multi-frame integration, the main control end needs to increase the memory space and the time efficiency is relatively increased; the high power consumption main control end needs to receive multiple frames to synthesize one frame. In order to achieve the same frame rate in the non-HDR mode, at least 2 times the frames need to be processed, and the power consumption will increase significantly; there is a tailing phenomenon caused by the integration of multiple frames. For the use of digital wide dynamic mode: this method requires a better ISP algorithm on the main control end. It is easy to bring dark noise; it is easy to cause the face to be too strong due to the digital wide dynamic pull, resulting in a pink phenomenon; it takes a lot of effort to tune the image to achieve a better balance, and when the difference between light and dark is too large, the effect is still not obvious. A gain adjustment method and device for preventing overexposure under low illumination are provided.

[0006] In order to solve the above technical problems, the present invention is solved by the following technical solutions:

[0007] A gain adjustment method for preventing overexposure under low illumination, the method comprising:

[0008] Acquisition of analog pixel signals and electronic quantity EQ, for optical arrays, analog pixel signals and electronic quantity EQ are acquired through optical conversion;

[0009] Gain adjustment: adjusting the gain of the acquired analog pixel signal to obtain an adjusted analog pixel signal;

[0010] Conversion of digital pixel signals, converting analog pixel signals into digital pixel signals through ADC modules;

[0011] The output of the digital pixel signal outputs the converted digital pixel signal.

[0012] Preferably, the gain adjustment includes adjusting the gain of the electronic quantity EQ. out ,

[0013] When the electronic quantity EQ is less than or equal to EQ 阈值 hour; ;

[0014] otherwise, ;

[0015] Among them, EQ 总 is the total amount of electrons; EQ is the real-time amount of electrons, EQ 阈值 is the threshold value of the amount of electrons,

[0016] Gain in Gain is the gain before adjustment. out is the adjusted gain, and K is the gain adjustment coefficient.

[0017] As a preferred method, the AE algorithm calculates the threshold EQ of the electronic amount in real time 阈值 include:

[0018] When the image brightness reaches the set target brightness, the threshold EQ of the electronic amount is dynamically adjusted according to the overly bright area in the image. 阈值 :When the electron quantity threshold EQ 阈值 When it is lowered, the overly bright area is reduced, and the threshold of the electronic amount EQ 阈值 Continue to reduce the threshold of the electronic volume EQ 阈值 Stop lowering;

[0019] When the electron quantity threshold EQ 阈值 When the over-brightness area does not increase, the threshold of the electronic quantity EQ 阈值 Increase, otherwise the electronic volume threshold EQ 阈值 Stop increasing.

[0020] As a preference: judging the digital pixel signal, judging the converted digital pixel signal, and when the digital pixel signal is greater than a set threshold, feeding back the adjustment signal to the gain adjustment.

[0021] Preferably, the gain adjustment includes: gain conversion, when there is a digital pixel signal, the optical array obtains the analog pixel signal through optical conversion to perform gain conversion;

[0022] The analog pixel signal after gain conversion is converted into a digital pixel signal through the ADC module; when the digital pixel signal is greater than the set threshold, the gain conversion is further performed, otherwise the digital pixel signal is output.

[0023] As a preference, the gain conversion is performed by a search method.

[0024] In order to solve the above technical problems, the present invention also provides a gain adjustment device for preventing overexposure under low illumination, which comprises:

[0025] An acquisition module for analog pixel signals and electronic quantity EQ, which acquires analog pixel signals and electronic quantity EQ through optical conversion for the optical array;

[0026] A gain adjustment module adjusts the gain of the acquired analog pixel signal to obtain an adjusted analog pixel signal;

[0027] A digital pixel signal conversion module converts the analog pixel signal into a digital pixel signal through an ADC module;

[0028] The output module of the digital pixel signal outputs the converted digital pixel signal.

[0029] Preferably, it also includes: a digital pixel signal judgment module, which judges the converted digital pixel signal, and when the digital pixel signal value is greater than a set numerical threshold, feeds back the adjustment signal to the gain adjustment.

[0030] The present invention has significant technical effects due to the adoption of the above technical solution:

[0031] The present invention increases the configured global gain by a secondary adjustment at the sensor end, and the AE algorithm dynamically controls the bright area threshold to ensure image consistency in non-overbright areas while reducing overbright areas; it has low power consumption and low cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the AE threshold calculation adjustment of the present invention.

[0033] Figure 2 It is a flow chart of analog gain secondary adjustment of the present invention.

[0034] Figure 3 It is a flow chart of iterative adjustment of analog gain of the present invention.

[0035] Figure 4 It is the sensor analog gain flow chart of the present invention.

[0036] Figure 5 It is a graph of the output gain after adjustment of the present invention.

[0037] Figure 6 It is a sensor Bayer format (RGGB) pixel array diagram of the present invention. DETAILED DESCRIPTION

[0038] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0039] Example 1

[0040] A gain adjustment method for preventing overexposure under low illumination. Figure 1 The methods include:

[0041] Acquisition of analog pixel signals and electronic quantity EQ, for optical arrays, analog pixel signals and electronic quantity EQ are acquired through optical conversion;

[0042] Gain adjustment: adjusting the gain of the acquired analog pixel signal to obtain an adjusted analog pixel signal;

[0043] Conversion of digital pixel signals, converting analog pixel signals into digital pixel signals through ADC modules;

[0044] The output of the digital pixel signal outputs the converted digital pixel signal.

[0045] Gain adjustment includes adjusting the gain of the electronic EQ out ,

[0046] When the electronic quantity EQ is less than or equal to EQ 阈值 hour; ;

[0047] otherwise, ;

[0048] Among them, EQ 总 is the total amount of electrons; EQ is the real-time amount of electrons, EQ 阈值 is the threshold value of the amount of electrons,

[0049] Gain in Gain is the gain before adjustment. out is the adjusted gain, and K is the gain adjustment coefficient.

[0050] Figures 2 to 4 The sensor end adds a secondary gain adjustment module, which is similar to a slider. The gain after secondary adjustment is determined according to the amount of photoelectrons in the potential well. That is, for analog pixels that reach a certain amount of electrons, secondary adjustment is performed on the set global gain before amplification.

[0051] Example 2

[0052] Based on Example 1, Figure 5 In this embodiment, the threshold value of the electron quantity EQ 阈值 Through real-time calculation and adjustment by AE algorithm. When adjusting the sensor gain through the AE (auto exposure) module of ISP, the AE algorithm evaluates the average brightness of the picture on the one hand, and the size of the over-bright area on the other hand, to dynamically adjust the range of the bright area that needs secondary gain adjustment, that is, to reduce the over-bright area of ​​the actual light, while allowing a smooth transition process for the brightness around the light;

[0053] AE algorithm calculates the threshold EQ of adjusting the electronic amount in real time 阈值 include:

[0054] When the image brightness reaches the set target brightness, the threshold EQ of the electronic amount is dynamically adjusted according to the overly bright area in the image. 阈值 :When the electron quantity threshold EQ 阈值 When it is lowered, the overly bright area is reduced, and the threshold of the electronic amount EQ 阈值 Continue to reduce the threshold of the electronic volume EQ 阈值 Stop lowering;

[0055] When the electron quantity threshold EQ 阈值 When the over-brightness area does not increase, the threshold of the electronic quantity EQ 阈值 Continue to increase, otherwise the threshold of the electronic volume EQ 阈值 Stop increasing.

[0056] N-fold gain: the gain configuration calculated in real time by the AE algorithm (the gain changes dynamically depending on the environment), which is the same for all pixels; electron quantity: the amount of electrons stored in the potential well after photoelectric conversion (i.e. after exposure), which is measured by the percentage of the maximum storable amount;

[0057] Set threshold: The AE algorithm adjusts the image in real time to ensure that pixels that are not too bright are not affected. Only pixels that exceed the threshold setting are adjusted at a given gain.

[0058] If the threshold is set to 75%, the amount of electrons is 80%, and the gain is set to 16 times, the output gain after adjustment is: 12.8 times; if the threshold is set to 75%, the amount of electrons is 80%, and the gain is set to 32 times, the output gain after adjustment is 25.6 times.

[0059] Example 3

[0060] On the basis of the above-mentioned embodiment, the digital pixel signal of this embodiment is judged by judging the converted digital pixel signal, and when the digital pixel signal is greater than the set threshold, the adjustment signal is fed back to the gain adjustment.

[0061] Gain adjustment includes: gain conversion, when there is a digital pixel signal, the optical array obtains the analog pixel signal through optical conversion to perform gain conversion;

[0062] The analog pixel signal after gain conversion is converted into a digital pixel signal through the ADC module; when the digital pixel signal is greater than the set threshold, the gain conversion is further performed, otherwise the digital pixel signal is output.

[0063] The gain conversion is performed by searching.

[0064] Example 4

[0065] Based on the above embodiment, this embodiment is a gain adjustment device for preventing overexposure under low illumination. Figure 6 In the image processing, each pixel is a single color photoelectric conversion (i.e. one-dimensional). After ADC conversion, it needs to be interpolated by the de-massage module of the ISP algorithm, and the interpolation is calculated by other nearby color pixels. In this way, if different gain adjustments are made for each pixel, it is easy to cause deviations in the interpolated RGB three-channel data, causing color anomalies. It includes an acquisition module for analog pixel signals and electronic quantity EQ, which acquires analog pixel signals and electronic quantity EQ through optical conversion for the optical array;

[0066] A gain adjustment module adjusts the gain of the acquired analog pixel signal to obtain an adjusted analog pixel signal;

[0067] A digital pixel signal conversion module converts the analog pixel signal into a digital pixel signal through an ADC module;

[0068] The output module of the digital pixel signal outputs the converted digital pixel signal.

[0069] It also includes: a digital pixel signal judgment module, which judges the converted digital pixel signal, and when the digital pixel signal value is greater than the set numerical threshold, feeds back the adjustment signal to the gain adjustment.

Claims

1. A gain adjustment method for preventing overexposure under low illumination, the method comprising: Acquisition of analog pixel signals and electronic quantity EQ, for optical arrays, analog pixel signals and electronic quantity EQ are acquired through optical conversion; Gain adjustment, through the ISP's AE automatic exposure module to calculate the threshold EQ of the electron quantity in real time 阈值 Thus, the gain is adjusted; and then the adjusted analog pixel signal is obtained; Conversion of digital pixel signals, converting analog pixel signals into digital pixel signals through ADC modules; The output of the digital pixel signal is to output the converted digital pixel signal; the gain is adjusted by comparing the electronic quantity EQ with the electronic quantity threshold EQ 阈值 The size of the gain is adjusted to obtain the adjusted gain Gain out ; When the electronic quantity EQ is less than or equal to EQ 阈值 hour; Gain out =Gain in ; Otherwise, Gain out =K*Gain in ; Among them, EQ 阈值 is the electronic quantity threshold; EQ is the real-time electronic quantity; Gain in Gain is the gain before adjustment. out is the adjusted gain, K is the gain adjustment coefficient, and its value range is (0,1); the threshold value of the electron quantity EQ 阈值 Real-time calculation and adjustment through AE algorithm; AE algorithm calculates and adjusts the threshold EQ of electronic quantity in real time 阈值 include: When the image brightness reaches the set target brightness, the threshold EQ of the electronic amount is dynamically adjusted according to the overly bright area in the image. 阈值 :When the electron quantity threshold EQ 阈值 When it is lowered, the overly bright area is reduced, and the threshold of the electronic amount EQ 阈值 Continue to reduce the threshold of the electronic volume EQ 阈值 Stop lowering; When the electron quantity threshold EQ 阈值 When the over-brightness area does not increase, the threshold of the electronic quantity EQ 阈值 Increase, otherwise the electronic volume threshold EQ 阈值 Stop increasing.

2. The gain adjustment method for preventing overexposure under low illumination according to claim 1, characterized in that: The digital pixel signal is judged, and the converted digital pixel signal is judged. When the digital pixel signal is greater than the set threshold, the adjustment signal is fed back to the gain adjustment.

3. The gain adjustment method for preventing overexposure under low illumination according to claim 1, characterized in that: Gain adjustment includes: gain conversion, when there is a digital pixel signal, the optical array obtains the analog pixel signal through optical conversion to perform gain conversion; The analog pixel signal after gain conversion is converted into a digital pixel signal through the ADC module; when the digital pixel signal is greater than the set threshold, the gain conversion is further performed, otherwise the digital pixel signal is output.

4. The gain adjustment method for preventing overexposure under low illumination according to claim 1, characterized in that: The gain conversion is performed by searching.

5. A gain adjustment device for preventing overexposure under low illumination, characterized in that: include: An acquisition module for analog pixel signals and electronic quantity EQ, which acquires analog pixel signals and electronic quantity EQ through optical conversion for the optical array; The gain adjustment module calculates the threshold EQ of the electron quantity in real time through the AE automatic exposure module of the ISP 阈值 Thus, the gain is adjusted; and then the adjusted analog pixel signal is obtained; A digital pixel signal conversion module converts the analog pixel signal into a digital pixel signal through an ADC module; An output module for digital pixel signals, outputting the converted digital pixel signals; The output of the digital pixel signal is to output the converted digital pixel signal; the gain is adjusted by comparing the electronic quantity EQ with the electronic quantity threshold EQ 阈值 The size of the gain is adjusted to obtain the adjusted gain Gain out ; When the electronic quantity EQ is less than or equal to EQ 阈值 hour; Gain out =Gain in ; Otherwise, Gain out =K*Gain in ; Among them, EQ 阈值 is the electronic quantity threshold; EQ is the real-time electronic quantity; Gain in Gain is the gain before adjustment. out is the adjusted gain, K is the gain adjustment coefficient, and its value range is (0,1); AE algorithm calculates the threshold EQ of adjusting electronic quantity in real time 阈值 include: When the image brightness reaches the set target brightness, the threshold EQ of the electronic amount is dynamically adjusted according to the overly bright area in the image. 阈值 :When the electron quantity threshold EQ 阈值 When it is lowered, the overly bright area is reduced, and the threshold of the electronic amount EQ 阈值 Continue to reduce the threshold of the electronic volume EQ 阈值 Stop lowering; When the electron quantity threshold EQ 阈值 When the over-brightness area does not increase, the threshold of the electronic quantity EQ 阈值 Increase, otherwise the electronic volume threshold EQ 阈值 Stop increasing.

6. A gain adjustment device for preventing overexposure under low illumination according to claim 5, characterized in that: Also includes: The digital pixel signal judgment module judges the converted digital pixel signal, and when the digital pixel signal value is greater than the set numerical threshold, the adjustment signal is fed back to the gain adjustment.

Images