Exposure time dynamic compensation method and device for backside CIS sensor

Abstract

This application relates to the field of scanner image acquisition technology, and in particular to a method and apparatus for dynamic compensation of exposure time of a reverse-side CIS sensor. The method is applied to a duplex automatic document feeder scanning system, which is equipped with independent CIS sensors for the front and back sides to achieve synchronous scanning of the front and back sides. The method includes: upon receiving a duplex scanning task instruction initiated by a user, obtaining the current relative luminous intensity of the reverse-side CIS sensor LED based on the current cumulative illumination time of the LED and a pre-stored attenuation curve model of the reverse-side CIS sensor LED; obtaining the compensated exposure time based on the current relative luminous intensity and pre-stored initial parameters of the reverse-side CIS sensor; and controlling the reverse-side CIS sensor LED to illuminate according to the compensated exposure time, thereby achieving exposure compensation and image quality maintenance of the reverse-side image during duplex scanning.

Description

Technical Field

[0001] This application relates to the field of scanner image acquisition technology, and in particular to a method for dynamic compensation of exposure time of a reverse-side CIS sensor, a device for dynamic compensation of exposure time of a reverse-side CIS sensor, and a duplex automatic paper feeder scanning system. Background Technology

[0002] In existing automatic document feeder (ADF) scanning systems, a dual CIS (contact image sensor) structure is typically used to achieve synchronous scanning of both sides. The front-side scanning module reuses the CIS sensor from the flatbed scanning unit, with its optical path enclosed within the device and a dedicated calibration white strip located beneath the flatbed. The system can trigger a calibration process upon each power-on or at a preset cycle, causing the front-side CIS sensor to acquire the standard brightness signal from the calibration white strip. This dynamically corrects parameter deviations caused by device aging, temperature changes, or ambient light interference, thereby ensuring brightness consistency and scanning accuracy of the front-side scanned image.

[0003] However, because the CIS sensor in the reverse scanning module needs to be placed within the ADF paper feed path, its optical surface is directly exposed to the external structural environment. Due to spatial constraints and the paper feed path, it is impossible to configure a physical calibration white strip. Therefore, the reverse CIS sensor typically relies on a one-time calibration at the factory: during the manufacturing process, a dedicated calibration program determines calibration parameters such as the initial exposure time and gain parameters, and these parameters are written into non-volatile memory. Throughout the entire lifecycle of the equipment, these initial parameters are always called for scanning control, without any further dynamic updates or compensation.

[0004] However, since the light source of the reverse-side CIS sensor is an LED device, its luminous intensity gradually decreases with the increase of cumulative illumination time, usually manifesting as a decrease in brightness or a reduction in light output stability. Because existing technologies for reverse-side CIS always control the LED illumination according to the initial exposure time, without considering the decay of LED luminous intensity, the actual light intensity incident on the surface of the medium to be scanned is lower than the design value. The intensity of the reflected light signal received by the image sensor subsequently decreases, making the scanned image generally darker, with reduced contrast, and even resulting in the loss of detail. Image quality continuously deteriorates over time, making it difficult to meet the requirements for long-term stable scanning. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, this application provides a method and apparatus for dynamic compensation of exposure time of a reverse CIS sensor, which solves the technical problem that the existing reverse CIS sensor uses one-time calibration parameters and cannot be dynamically adjusted according to the cumulative lighting time decay of LEDs, resulting in decreased brightness, reduced contrast and deterioration of image quality over time.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted in this application include:

[0009] In a first aspect, embodiments of this application provide a method for dynamic compensation of exposure time of a reverse-side CIS sensor, applied to a duplex automatic paper feeder scanning system. The duplex automatic paper feeder scanning system is equipped with independent CIS sensors for the front and reverse sides to achieve synchronous scanning of both sides. The method includes:

[0010] Upon receiving a user-initiated bi-sided scanning task instruction, the relative luminous intensity of the reverse CIS sensor LED is obtained based on the current cumulative lighting time of the reverse CIS sensor LED and the pre-stored attenuation curve model of the reverse CIS sensor LED. The cumulative lighting time of the reverse CIS sensor LED is: during the execution of historical bi-sided scanning tasks, the actual lighting time of the reverse CIS sensor LED in each bi-sided scanning task is recorded by a timer, and the total duration is obtained by accumulating the actual lighting time of the LED in each bi-sided scanning task.

[0011] The compensated exposure time is obtained based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor, wherein the initial parameters of the reverse CIS sensor include: initial exposure time T0 and initial LED luminous intensity I0.

[0012] The LED of the reverse CIS sensor is controlled to light up according to the compensated exposure time.

[0013] Optionally, in some embodiments of this application, after controlling the LED of the reverse CIS sensor to be lit according to the compensated exposure time, the method further includes:

[0014] Based on the actual illumination time of the LED of the reverse CIS sensor in the user-initiated double-sided scanning task, the cumulative illumination time in the timer is updated to obtain a new cumulative illumination time, which serves as dynamic compensation for the exposure time of the next user-initiated double-sided scanning task.

[0015] In the user-initiated double-sided scanning task, the actual lighting time of the LED on the reverse side of the CIS sensor is determined by the product of the number of pages scanned in the user-initiated double-sided scanning task and the pre-acquired single-page scanning time.

[0016] Optionally, in some embodiments of this application, the compensated exposure time is obtained based on the current relative luminous intensity and pre-stored initial parameters of the reverse CIS sensor, including:

[0017] The product of the initial exposure time T0 and the initial luminous intensity I0 of the LED is taken as the target luminous flux Q;

[0018] The compensated exposure time is obtained by dividing the target luminous flux Q by the current relative luminous intensity.

[0019] Optionally, in some embodiments of this application, the attenuation curve model of the reverse CIS sensor LED is expressed according to the following formula:

[0020] I(t) = I0 × (1 - k × t), where t is the cumulative lighting time, k is the preset attenuation coefficient, and I(t) is the relative luminous intensity of the LED of the reverse CIS sensor when the cumulative lighting time is t.

[0021] Optionally, in some embodiments of this application, after N consecutive user-initiated bi-directional scanning tasks are completed, the method further includes:

[0022] The average brightness value of the scanned images corresponding to N double-sided scanning tasks is obtained respectively, and the average brightness value of the N scanned images is statistically analyzed to obtain the average brightness value, which is used as the actual brightness value (Lactual).

[0023] The actual brightness value Lactual is compared with the preset target brightness threshold Ltarget to obtain the brightness deviation ΔL;

[0024] When the brightness deviation ΔL exceeds the preset brightness tolerance threshold, the attenuation coefficient k is updated according to the following formula to obtain the updated attenuation coefficient k′, and the updated attenuation coefficient k′ is stored for use in the dynamic compensation calculation of exposure time in subsequent user-initiated double-sided scanning tasks.

[0025] k′=k+α×(ΔL / Ltarget)×(1 / t1), where t1 is the cumulative lighting time of the LED of the reverse CIS sensor after completing the N consecutive user-initiated double-sided scanning tasks, and α is a preset correction coefficient.

[0026] Optionally, in some embodiments of this application, the average brightness value of the scanned image is calculated according to the following steps:

[0027] A preset central region located in the middle of the width and height of the scanned image is selected as the statistical region, wherein the width and height of the central region are both less than half of the corresponding dimension of the scanned image.

[0028] Perform grayscale histogram statistics on the pixels within the statistical region to obtain the pixel quantity distribution corresponding to each grayscale level;

[0029] Based on the pixel quantity distribution corresponding to each gray level, determine the proportion of each gray level pixel quantity in the total pixel quantity;

[0030] Based on the aforementioned proportion distribution, pixels whose grayscale values ​​are below a preset lower percentile threshold and above a preset upper percentile threshold are removed.

[0031] The average brightness value of the scanned image is obtained by averaging the gray values ​​of the remaining pixels after removal.

[0032] Optionally, in some embodiments of this application, the preset correction coefficient α is set in segments based on the cumulative illumination time t1 of the reverse CIS sensor LED after completing the N consecutive user-initiated double-sided scanning tasks, wherein:

[0033] When 0 ≤ t1 < T1, α = α1;

[0034] When T1≤t1<T2, α=α2;

[0035] When t1≥T2, α=α3;

[0036] Where α1>α2>α3, and T1 and T2 are preset time thresholds.

[0037] Optionally, in some embodiments of this application, α1 is 0.1, α2 is 0.07, α3 is 0.04, T1 is 100 hours, and T2 is 500 hours.

[0038] Secondly, this embodiment also provides a dynamic compensation device for the exposure time of a reverse-side CIS sensor, applied to a duplex automatic paper feeder scanning system. The duplex automatic paper feeder scanning system is equipped with independent CIS sensors for both the front and reverse sides to achieve synchronous scanning of both sides. The dynamic compensation device for exposure time includes:

[0039] The LED luminous intensity acquisition module is used to obtain the current relative luminous intensity of the reverse CIS sensor LED based on the current cumulative lighting time of the reverse CIS sensor LED and the pre-stored decay curve model of the reverse CIS sensor LED when receiving a double-sided scanning task instruction initiated by the user. The cumulative lighting time of the reverse CIS sensor LED is obtained by accumulating statistics through a timer during the execution of historical double-sided scanning tasks.

[0040] The compensated exposure time acquisition module is used to acquire the compensated exposure time based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor, wherein the initial parameters of the reverse CIS sensor include: initial exposure time T0 and initial luminous intensity I0 of the LED.

[0041] The CIS control module is used to control the LED of the reverse CIS sensor to light up according to the compensated exposure time.

[0042] Thirdly, this embodiment also provides a duplex automatic paper feeder scanning system, including:

[0043] The scanning unit is used to perform double-sided scanning tasks;

[0044] The reverse CIS sensor LED exposure time dynamic compensation device described in the second aspect is used to control the reverse CIS sensor LED in the scanning unit to light up according to the compensated exposure time in order to complete the double-sided scanning task initiated by the user.

[0045] (III) Beneficial Effects

[0046] The method and apparatus for dynamic compensation of exposure time of the reverse CIS sensor provided in this application, upon receiving a user-initiated bi-sided scanning task instruction, obtains the current relative luminous intensity of the reverse CIS sensor LED based on the current cumulative lighting time of the LED and a pre-stored attenuation curve model of the LED. Specifically, a timer records the actual lighting time of the LED in each bi-sided scanning task, and the total lighting time is accumulated to reflect the cumulative working time of the LED. Based on this, a compensated exposure time is obtained based on the current relative luminous intensity and pre-stored initial parameters of the reverse CIS sensor (including initial exposure time T0 and initial LED luminous intensity I0). The reverse CIS sensor LED is then controlled to light up according to the compensated exposure time, so that the lighting time of the LED is no longer fixed at the initial exposure time T0, but is adjusted according to changes in relative luminous intensity.

[0047] Because the luminous intensity of the LED in the reverse-side CIS sensor gradually decreases with increasing cumulative illumination time, if the illumination control is always based on the initial exposure time T0, the actual luminous effect will be lower than the design state after the initial luminous intensity I0 changes, affecting the scanning effect. This application obtains the current relative luminous intensity and accordingly calculates the compensated exposure time, enabling the reverse-side CIS sensor LED to continue illuminating according to the compensated exposure time even when the luminous intensity decreases, thereby achieving dynamic compensation for changes in the luminous intensity of the reverse-side CIS sensor LED. Therefore, this application can achieve simultaneous scanning of both sides in a duplex automatic document feeder scanning system, while ensuring that the exposure control parameters of the reverse-side CIS sensor adjust with changes in cumulative illumination time during equipment use, improving scanning stability, delaying the decline in scanning effect caused by LED attenuation, and improving overall scanning quality. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating a method for dynamic compensation of exposure time of a reverse-side CIS sensor according to an embodiment of this application.

[0049] Figure 2 This is a flowchart illustrating a method for completing N consecutive user-initiated double-sided scanning tasks according to an embodiment of this application.

[0050] Figure 3 This is a schematic diagram of the structure of a dynamic compensation device for exposure time of a reverse-side CIS sensor according to an embodiment of this application. Detailed Implementation

[0051] To better explain and facilitate understanding of this application, the following detailed description of the application is provided in conjunction with the accompanying drawings and specific embodiments.

[0052] In related technologies, existing exposure control methods for addressing the brightness attenuation problem of the reverse-side CIS sensor in a duplex automatic document feeder scanning system can be broadly categorized as follows:

[0053] The first type is a control scheme based on fixed exposure time or factory calibration parameters. This scheme typically performs a one-time calibration at the factory based on the initial exposure time T0 and the initial LED luminous intensity I0. In all subsequent duplex scanning tasks, the LED of the reverse CIS sensor is driven to light up according to the preset exposure time, without dynamically correcting the LED's decay state. Although the implementation is simple, the LED of the reverse CIS sensor inevitably experiences light decay during long-term use. Its actual luminous intensity gradually decreases with the increase of cumulative lighting time. When the exposure time remains constant, the image signal strength will continuously weaken, leading to a darker reverse image, increased noise, or grayscale distortion. Especially in duplex synchronous scanning scenarios, inconsistent brightness between the front and back sides is likely to occur, affecting the stability and consistency of scanning quality.

[0054] The second type is a compensation scheme based on periodic manual calibration or maintenance reminders. This scheme sets a maintenance cycle or scan count threshold, prompting the user to calibrate or replace the reverse CIS sensor LED when preset conditions are met, thereby restoring image brightness. While this method can maintain the long-term performance of the device to some extent, its compensation behavior is lagging and cannot accurately reflect the current changes in luminous intensity based on the actual cumulative lighting time of the LED. Furthermore, the decay rate of LEDs varies significantly under different usage frequencies, and relying solely on time cycles or scan count thresholds is insufficient to precisely match the actual decay level, easily leading to over-maintenance or under-maintenance, thus affecting the user experience.

[0055] Therefore, the dynamic compensation method for exposure time of the reverse-side CIS sensor provided in this application embodiment is applied to a duplex automatic paper feeder scanning system configured with independent CIS sensors for the front and back sides to achieve synchronous scanning of the front and back sides. Upon receiving a user-initiated duplex scanning task command, the system first accumulates the cumulative illumination time of the LED based on the actual illumination time of the reverse-side CIS sensor LED recorded by a timer during each historical duplex scanning task execution. Then, combined with a pre-stored attenuation curve model of the reverse-side CIS sensor LED, the relative luminous intensity of the current LED is obtained. Further, based on the current relative luminous intensity and pre-stored initial parameters of the reverse-side CIS sensor (including initial exposure time T0 and initial LED luminous intensity I0), the compensated exposure time is calculated, and the reverse-side CIS sensor LED is controlled to illuminate according to the compensated exposure time.

[0056] The technical solution of this application introduces precise statistics on the cumulative LED lighting time at the data level, and combines this with a decay curve model to quantitatively evaluate the current relative luminous intensity. This enables dynamic identification and pre-compensation of the light decay state of the reverse CIS sensor, avoiding the brightness mismatch problem caused by fixed exposure time. Furthermore, this compensation process is completed before the scanning task starts, without relying on image content for feedback adjustment. This ensures both the real-time performance of dual-sided synchronous scanning and improves compensation accuracy and stability. Even under long-term, high-frequency use, the brightness of the reverse image remains consistent with that of the front image, ensuring stable and reliable overall scanning quality.

[0057] To better understand the above technical solutions, exemplary embodiments of this application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application can be understood more clearly and thoroughly, and that the scope of this application can be fully conveyed to those skilled in the art.

[0058] Figure 1 This is a flowchart illustrating a method for dynamic compensation of exposure time for a reverse-side CIS sensor according to an embodiment of this application. This method is applied to a duplex automatic document feeder scanning system, which is equipped with independent CIS sensors for both the front and back sides to achieve synchronous scanning of both sides. Figure 1 As shown, the exposure time dynamic compensation method for the reverse-side CIS sensor includes:

[0059] S1, upon receiving a user-initiated double-sided scanning task instruction, based on the current cumulative lighting time of the reverse CIS sensor LED and the pre-stored attenuation curve model of the reverse CIS sensor LED, the current relative luminous intensity of the reverse CIS sensor LED is obtained. The cumulative lighting time of the reverse CIS sensor LED is: during the execution of historical double-sided scanning tasks, the actual lighting time of the reverse CIS sensor LED in each double-sided scanning task is recorded by a timer, and the total duration is obtained by accumulating the actual lighting time of the LED in each double-sided scanning task.

[0060] Specifically, upon receiving a user-initiated double-sided scanning task command, the system does not immediately illuminate the LED on the reverse CIS sensor according to a fixed exposure time. Instead, it retrieves the execution records of historical double-sided scanning tasks. By setting a timer during each double-sided scanning task, the actual illumination time of the reverse CIS sensor LED is recorded. For example, if the reverse LED illuminates for 3.2 seconds in one double-sided scanning task and 5.6 seconds in another, the system accumulates the actual illumination times from all historical double-sided scanning tasks to obtain the cumulative illumination time of the LED. Assuming the scanning equipment operates for one year, the cumulative illumination time reaches 1200 hours. This value accurately reflects the aging degree of the LED, rather than being a rough estimate based on the equipment's power-on time or the number of scans, thus improving the accuracy of attenuation assessment.

[0061] After obtaining the current cumulative illumination time, the current relative luminous intensity is obtained based on the pre-stored decay curve model of the LED in the reverse CIS sensor. This decay curve model can be established through aging tests performed on the LED during the factory manufacturing process, for example, by measuring the luminous intensity change curves of the LED at different illumination time points under standard temperature and drive current conditions. If the initial luminous intensity of the LED is set to I0, and the cumulative illumination time is 1200 hours, the current luminous intensity is found to be 0.88I0 according to the decay curve model, thus determining the current relative luminous intensity. This model-based mapping method transforms aging information in the "time dimension" into quantitative parameters in the "luminous intensity dimension," providing a reliable basis for subsequent exposure compensation.

[0062] S2, based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor, obtain the compensated exposure time, wherein the initial parameters of the reverse CIS sensor include: initial exposure time T0 and initial LED luminous intensity I0;

[0063] For example, suppose the scanning device is factory-set to an initial exposure time T0 of 10ms, and a standard grayscale image can be obtained under the condition of initial LED luminous intensity I0. When the current relative luminous intensity drops to 0.88I0, in order to keep the image photosensitive energy basically constant, the compensated exposure time is calculated based on the exposure energy while maintaining consistent signal output.

[0064] S3, control the LED of the reverse CIS sensor to light up according to the compensated exposure time. This ensures that the reverse image can still obtain light energy input close to the initial state even when the LED luminous intensity decreases.

[0065] This embodiment introduces a precise statistical mechanism for the cumulative lighting time of the LED before the double-sided scanning task is executed, so that the aging state of the LED can be quantitatively expressed, avoiding the error caused by rough judgment based solely on the number of scans or the years of use. At the same time, through a pre-established decay curve model, the time decay characteristics are converted into the current relative luminous intensity, realizing feedforward prediction of the light decay state, rather than relying on image results for post-correction, thereby improving the certainty and stability of compensation.

[0066] Specifically, after controlling the LED of the reverse CIS sensor to light up according to the compensated exposure time, the method further includes:

[0067] Based on the actual illumination time of the LED of the reverse CIS sensor in the user-initiated double-sided scanning task, the cumulative illumination time in the timer is updated to obtain a new cumulative illumination time, which serves as dynamic compensation for the exposure time of the next user-initiated double-sided scanning task.

[0068] In the user-initiated double-sided scanning task, the actual lighting time of the LED on the reverse side of the CIS sensor is determined by the product of the number of pages scanned in the user-initiated double-sided scanning task and the pre-acquired single-page scanning time.

[0069] Specifically, after the completion of this duplex scanning task, the cumulative illumination time in the timer is updated based on the actual illumination time of the LED of the reverse CIS sensor during this duplex scanning task. The actual illumination time is not obtained through additional real-time hardware timing sampling, but is determined by multiplying the number of pages scanned by the pre-acquired single-page scanning time. Since in a duplex automatic paper feeder scanning system, the scanning process for each page corresponds to a fixed number of scan lines and sensor drive timing, and under stable equipment parameters, the single-page scanning time can be pre-acquired and stored during the factory or calibration stage, thus serving as a stable time reference.

[0070] For example, assuming the pre-acquired single-page scan time is 1.8 seconds, when a user initiates a double-sided scan task, scanning a total of 50 pages, the theoretical actual lighting time of the LED on the reverse CIS sensor in this task is 50 × 1.8 seconds = 90 seconds. If the cumulative lighting time recorded in the timer is 1200 hours, after the task is completed, this 90 seconds is converted to hours and added to the original cumulative lighting time to obtain a new cumulative lighting time, for example, updated to 1200.025 hours. This new cumulative lighting time will be used as the input parameter for obtaining the current relative luminous intensity when the user initiates the next double-sided scan task, thereby achieving continuous tracking of the decay state. In this way, the cumulative lighting time is not a one-time static parameter, but dynamically increases with each scan task, forming a cyclical mechanism of "scan task execution - actual lighting time calculation - cumulative time update - next compensation call". In this way, each exposure time dynamic compensation is based on the latest LED usage data, avoiding compensation lag or error expansion due to the cumulative time not being updated for a long time.

[0071] Preferably, the compensated exposure time is obtained based on the current relative luminescence intensity and pre-stored initial parameters of the reverse CIS sensor, including:

[0072] The product of the initial exposure time T0 and the initial luminous intensity I0 of the LED is taken as the target luminous flux Q. Since the light energy obtained per pixel during scanning imaging is approximately proportional to the product of the light source luminous intensity I and the exposure time T, it can be expressed as: Q≈I×T. During the equipment's factory calibration phase, the initial exposure time T0 is the optimal exposure parameter determined through grayscale testing and image quality calibration under the condition of the initial LED luminous intensity I0. Therefore, the product of T0 and I0 reflects the target luminous flux Q required to achieve the standard image brightness at that time, i.e.: Q=I0×T0. This target luminous flux Q can be regarded as the target light energy.

[0073] The compensated exposure time is obtained by dividing the target luminous flux Q by the current relative luminous intensity. After the device has been running for a period of time, the current relative luminous intensity of the LED is obtained based on the cumulative lighting time and attenuation curve model. For example, if the current relative luminous intensity is 0.85I0, then to keep the target luminous flux Q constant, the following condition must be met: Compensated exposure time = Q ÷ Current relative luminous intensity.

[0074] In this embodiment, the attenuation curve model of the reverse-side CIS sensor LED is expressed according to the following formula:

[0075] I(t) = I0 × (1 - k × t), where t is the cumulative lighting time, k is the preset attenuation coefficient, and I(t) is the relative luminous intensity of the LED of the reverse CIS sensor when the cumulative lighting time is t.

[0076] Specifically, by using the expression of the decay curve model of the LED in the reverse CIS sensor, the light decay characteristics during the LED aging process are quantitatively described in the form of a mathematical model, enabling a calculable mapping of the LED's luminous intensity change with the cumulative lighting time. From a physical perspective, during long-term operation, the luminous intensity of an LED typically decreases gradually over time due to factors such as chip material aging, changes in packaging thermal stress, and phosphor decay. Within a certain operating range, this decay process can be approximated as a linear change; therefore, a linear decay model of I(t) = I0 × (1 - k × t) is used, which is both consistent with engineering practice and easy to implement. The decay coefficient k can be obtained through factory aging tests or batch sample statistical experiments. For example, under standard operating current and ambient temperature conditions, continuous lighting tests are conducted on the LED, recording the luminous intensity at different time points, and the value of k is determined through linear fitting.

[0077] For example, assuming the initial luminous intensity I0 of the LED is normalized to 1.0, and tests show that after 2000 hours of cumulative lighting time, the luminous intensity drops to 0.90, this can be explained by the formula:

[0078] 0.90 = 1.0 × (1 - k × 2000), solving for k gives k = 0.00005 (unit: hourly). When the device has run for a cumulative lighting time t = 1000 hours, the current luminous intensity can be calculated using the model:

[0079] I(1000) = 1.0 × (1 - 0.00005 × 1000) = 1.0 × (1 - 0.05) = 0.95, meaning the current luminous intensity is 95% of the initial luminous intensity. Therefore, 0.95I0 can be used as the current relative luminous intensity for subsequent dynamic compensation calculations of exposure time.

[0080] Furthermore, when the cumulative lighting time increases to 1500 hours, then:

[0081] I(1500)=1.0×(1-0.00005×1500)=1.0×(1-0.075)=0.925. It can be seen that as the cumulative lighting time t increases, I(t) shows a linear decreasing trend, which can continuously reflect the LED light decay process rather than discrete jumps, thus ensuring the smoothness of the compensation process.

[0082] In this embodiment, by constructing a clear attenuation curve model, the LED light decay problem is transformed from empirical judgment to mathematical expression, so that the current luminous intensity I(t) can be directly calculated based on the cumulative lighting time, avoiding reliance on image feedback or manual calibration for estimation, and improving the accuracy and repeatability of attenuation assessment.

[0083] Optionally, in some embodiments of this application, the attenuation coefficient k is a preset parameter, which can be configured differently according to different batches of LEDs or different specifications of CIS sensors. When using different models of LEDs, only the value of k needs to be recalibrated without changing the overall control logic, thus enhancing the versatility of the system. Finally, this attenuation curve model is closely integrated with the aforementioned dynamic compensation formula for exposure time that ensures "the target luminous flux Q remains constant". By accurately obtaining the current luminous intensity through I(t), and then combining it with Q=I0×T0 to back-calculate the exposure time, a complete closed-loop mathematical system of "cumulative time - luminous intensity - exposure time" is formed. This ensures that the entire compensation process is based on a unified physical model, with clear logic, explicit cause and effect, and stable and reliable engineering implementation.

[0084] like Figure 2 As shown, in some embodiments of this application, after N consecutive user-initiated bi-directional scanning tasks are completed, the method further includes:

[0085] The average brightness value of the scanned images corresponding to N double-sided scanning tasks is obtained respectively, and the average brightness value of the N scanned images is statistically analyzed to obtain the average brightness value, which is used as the actual brightness value (Lactual).

[0086] The actual brightness value Lactual is compared with the preset target brightness threshold Ltarget to obtain the brightness deviation ΔL;

[0087] When the brightness deviation ΔL exceeds the preset brightness tolerance threshold, the attenuation coefficient k is updated according to the following formula to obtain the updated attenuation coefficient k′, and the updated attenuation coefficient k′ is stored for use in the dynamic compensation calculation of exposure time in subsequent user-initiated double-sided scanning tasks.

[0088] k′=k+α×(ΔL / Ltarget)×(1 / t1), where t1 is the cumulative lighting time of the LED of the reverse CIS sensor after completing the N consecutive user-initiated double-sided scanning tasks, and α is a preset correction coefficient.

[0089] For example, after N consecutive user-initiated double-sided scanning tasks are completed, the average brightness value of the reverse scanned images corresponding to these N double-sided scanning tasks is first obtained. The average brightness value can be obtained by averaging the pixel grayscale values ​​of each scanned image. For example, if five consecutive scanning tasks are completed, and the average brightness values ​​of the reverse scanned images are 120, 118, 121, 119, and 122 respectively (assuming the grayscale value range is 0~255), then these five average brightness values ​​are statistically calculated to obtain the average brightness Lactual = (120 + 118 + 121 + 119 + 122) / 5 ≈ 120. This average brightness value can be regarded as a representative indicator of the current actual imaging brightness of the reverse CIS sensor.

[0090] Next, the actual brightness value Lactual is compared with the preset target brightness threshold Ltarget to obtain the brightness deviation ΔL. Assuming the target brightness threshold Ltarget is 125, the brightness deviation ΔL = Ltarget - Lactual = 125 - 120 = 5. This deviation reflects situations where the actual scanning brightness is lower than expected due to LED attenuation, equipment temperature changes, or other factors.

[0091] When the brightness deviation ΔL exceeds the preset brightness tolerance threshold (e.g., set to 2 grayscale units), the system updates the attenuation coefficient k according to the feedback correction formula, so as to more accurately reflect the actual attenuation characteristics of the LED in the next round of exposure time calculation. The formula is as follows:

[0092] k′=k+α×(ΔL / Ltarget)×(1 / t1);

[0093] Where t1 is the cumulative lighting time of the LED on the reverse side of the CIS sensor after completing N consecutive scanning tasks, and α is a preset correction coefficient used to adjust the magnitude of the correction. For example, assuming the current attenuation coefficient k = 0.00005, the cumulative lighting time t1 = 1200 hours, and α = 0.1, substituting ΔL = 5 and Ltarget = 125 in the example above, we get:

[0094] k′=0.00005+0.1×(5 / 125)×(1 / 1200)≈0.00005333;

[0095] Therefore, the attenuation coefficient, after fine-tuning, more closely reflects the actual LED attenuation state. The next dynamic compensation for exposure time will be calculated using the updated k′, thus adding closed-loop correction capability to the feedforward model. By statistically analyzing the average brightness of N consecutive scanning tasks instead of the brightness of a single scan, the impact of occasional image brightness fluctuations on the correction is reduced, improving the stability and reliability of the correction. Furthermore, the introduction of a preset correction coefficient α controls the magnitude of each correction, avoiding overcompensation due to excessive brightness fluctuations, making the closed-loop adjustment smooth and gradually approaching the optimal state. Finally, this mechanism does not rely on manual intervention and can automatically and adaptively optimize during long-term equipment operation, ensuring long-term stability of the reverse image brightness and improving the overall image consistency and user experience of the duplex scanning system. Therefore, by obtaining the average brightness, calculating the brightness deviation, and updating the attenuation coefficient k after N consecutive scanning tasks, the real-time performance and accuracy of the compensation calculation are guaranteed, and the adaptive capability to LED aging and environmental factors is enhanced, thereby achieving more precise, stable, and reliable dynamic compensation for exposure time.

[0096] In the practical application of this embodiment, the average brightness value of the scanned image is calculated according to the following steps:

[0097] A predetermined central region located in the middle of both the width and height of the scanned image is selected as the statistical region. The width and height of this central region are both less than half the corresponding dimension of the scanned image. For example, if the scanned image size is 1200×1600 pixels, a rectangular region with a width of 500 pixels and a height of 700 pixels can be selected as the central region. The reason for selecting the image center region as the statistical region is that uneven lighting, mechanical obstruction, or paper warping may exist at the edges of the paper or the scanned edges, while the central region usually best represents the average brightness characteristics of the scanned image.

[0098] Perform grayscale histogram statistics on the pixels within the statistical region to obtain the pixel quantity distribution corresponding to each grayscale level;

[0099] Assuming the scanned image is an 8-bit grayscale image with grayscale values ​​ranging from 0 to 255, the number of pixels with each grayscale value within a region is counted to obtain a grayscale histogram. For example, there are 5000 pixels with a grayscale value of 120, 5100 pixels with a grayscale value of 121, and so on.

[0100] Based on the pixel quantity distribution corresponding to each gray level, the proportion of each gray level's pixel quantity in the total pixel quantity is determined; for example, if the total number of pixels in the central region is 350,000, then the proportion of pixels with a gray value of 120 is approximately 5000 / 350000≈1.43%. Through the proportion distribution, the contribution of each gray level to the composition of image brightness can be quantified.

[0101] Based on the aforementioned proportion distribution, pixels with grayscale values ​​below a preset lower percentile threshold and above a preset upper percentile threshold are removed. Specifically, pixels with grayscale values ​​below the preset lower percentile threshold and above the preset upper percentile threshold, such as the lower 5 percentile and the upper 5 percentile, are removed. For example, using the aforementioned example, if pixels with grayscale values ​​less than 50 or greater than 200 constitute a very small proportion and may belong to noise or paper defect areas, these pixels will be removed to reduce the impact of abnormal grayscale values ​​on the average brightness calculation.

[0102] The average brightness value of the scanned image is obtained by averaging the gray values ​​of the remaining pixels after removal. Specifically, averaging the gray values ​​of the remaining pixels after removal involves summing the gray values ​​of pixels in the central region after removing extreme gray values ​​and dividing by the number of pixels to obtain the average brightness value of the scanned image. For example, if the total number of remaining pixels after removing extreme pixels is 330,000 and the total gray value is 39,600,000, then the average brightness value L = 39,600,000 / 330,000 ≈ 120.

[0103] This embodiment selects the central region between the image width and height as the statistical area, effectively avoiding the influence of brightness anomalies caused by paper edges or scanning mechanical obstructions on the average value. This allows the calculation results to more accurately reflect the actual imaging effect of the LED of the reverse CIS sensor. Subsequently, percentile threshold removal is applied to the pixel grayscale values ​​within the statistical area to eliminate occasional noise, stains, or light spots that interfere with the average brightness, improving the robustness and stability of brightness measurement. Based on grayscale histogram statistics and proportion distribution analysis, this method establishes a rigorous statistical foundation, ensuring good repeatability of the average brightness value and avoiding the influence of local anomalies on single-point measurements or the overall image mean. At the same time, statistical analysis is performed only on the central region, and extreme grayscale pixels are removed, balancing computational accuracy and efficiency. This allows for the rapid acquisition of stable average brightness values, meeting the requirements of real-time or near-real-time closed-loop adjustment in duplex scanning systems.

[0104] Specifically, the preset correction coefficient α is set in segments based on the cumulative illumination time t1 of the LED of the reverse CIS sensor after completing the N consecutive user-initiated double-sided scanning tasks, wherein:

[0105] When 0 ≤ t1 < T1, α = α1; when 0 ≤ t1 < T1, the LED is in the initial stage of use and decays rapidly, so α = α1. A larger α value enhances the feedback correction and quickly responds to brightness deviations, allowing the exposure time adjustment to promptly approach the target brightness. Assuming T1 is 500 hours and α1 is 0.1, then in the stage where the LED's cumulative lighting time is 300 hours, if the measured brightness deviation ΔL is large after N consecutive scans, a higher α1 will be used to quickly correct the decay coefficient k, thereby rapidly compensating for the brightness decrease caused by the initial decay.

[0106] When T1 ≤ t1 < T2, α = α2; when T1 ≤ t1 < T2, the LED enters the mid-term use stage, and the light decay rate tends to level off, α = α2. The value of α is less than α1, but still sufficient for necessary brightness correction. For example, if T2 is 1500 hours, and α2 is 0.07, then if the cumulative lighting time is 1000 hours, a brightness deviation ΔL will appear. The magnitude of the feedback correction is moderate, avoiding overcompensation that could cause brightness fluctuations, and ensuring the stability of dynamic compensation for exposure time.

[0107] When t1≥T2, the LED enters the long-term use stage, and the decay rate is low. α=α3, and the value of α is further reduced, for example, to 0.04, in order to reduce the feedback correction amplitude, make the exposure time adjustment more gradual, and avoid excessive correction after LED aging, which would cause brightness fluctuations.

[0108] Where α1>α2>α3, and T1 and T2 are preset time thresholds.

[0109] By segmenting the preset correction coefficient α according to the cumulative LED lighting time t1, this embodiment realizes the adaptive adjustment of closed-loop exposure time compensation, which not only ensures the timeliness of brightness correction, but also maintains long-term stability and reliability, and improves the consistency of scanned image brightness and the overall system performance.

[0110] α1 is 0.1, α2 is 0.07, α3 is 0.04, T1 is 100 hours, and T2 is 500 hours.

[0111] In some other embodiments of this application, it should be noted that after the completion of a historical duplex scanning task, image feature extraction is performed on the reverse scan image generated by each historical duplex scanning task. This feature extraction is performed after the duplex scanning task is completed, so it does not affect the exposure control of the current scan. The feature extraction content of each duplex historical scanning task includes four aspects: First, the historical average brightness value, which is calculated by statistically analyzing the grayscale values ​​of all pixels in the scanned image to determine the average level of the overall image brightness, reflecting the overall brightness and darkness characteristics of the page; Second, the brightness standard deviation, which represents the dispersion of the image grayscale distribution, reflecting the uniformity or fluctuation of the page brightness variation; Third, the proportion of black pixels on the page, which is calculated by statistically analyzing the proportion of pixels below a set grayscale threshold to the total number of pixels, reflecting the coverage of dark areas on the page; Fourth, the texture complexity index, which is calculated by statistically analyzing the local grayscale changes or edge gradients of the image, reflecting the complexity of the page structure and content. These feature parameters together constitute the image feature parameter set of each historical scanning task.

[0112] After feature extraction, the feature parameters of historical scanned images are associated and stored with the number of pages scanned, document type identifier, and task completion timestamp of the corresponding duplex scanning task. The number of pages scanned represents the total number of pages in this batch of duplex scanning tasks; the document type identifier can be the document category selected by the user in the scanning interface or a document type code preset by the system; and the task timestamp records the specific time the task was completed. This information is stored in a structured format in the historical scan feature database, providing a basis for predictive analysis of subsequent scanning tasks.

[0113] When the device receives a new duplex scanning task instruction, it first reads the number of pages to be scanned, document type identifier, and scan control parameter set for the current task before starting scanning. Scan control parameters include scan resolution, scan mode (color or grayscale), paper size, and whether image enhancement processing is enabled. Based on this, a filtering operation is performed in the historical scan feature database to select the historical records most similar to the current task. Specifically, the filtering first requires that the document type of the historical record matches the current task, and then further filters records with scanned page numbers close to the current task, i.e., the difference between the historical scanned page number and the current task's scanned page number is within a preset threshold range. Only historical records that simultaneously meet the conditions of document type matching and page number proximity, and whose number exceeds the preset minimum sample size, can constitute a "similar historical task set." The number of historical records in this set is sufficiently large to ensure the representativeness and reliability of the statistical analysis.

[0114] For image feature parameters in a set of similar historical tasks, statistical analysis is performed to generate a predicted content feature vector. Specifically, a weighted average of the historical average brightness values ​​in the set is calculated, where the weights can be set according to the proximity of the task times; that is, the closer the historical record is to the current task time, the greater the weight, thus obtaining the predicted average brightness value. The texture complexity index is obtained by arithmetically averaging the values ​​of each historical record in the set, resulting in a predicted texture complexity parameter. Subsequently, the predicted average brightness value and the predicted texture complexity parameter are normalized to eliminate the influence of different feature dimensions, ultimately yielding a predicted content feature vector, which characterizes the estimated state of the current scanning task in terms of brightness and detail complexity.

[0115] During the exposure time calculation stage, the relative luminous intensity of the current LED is first calculated based on the cumulative LED lighting time. This parameter reflects the LED's attenuation. Then, the predicted content feature vector is combined with the current relative luminous intensity of the LED and historical calibration data to correct the target luminous flux. Specifically, when the predicted brightness is low or the page texture complexity is high, the target luminous flux is appropriately increased to ensure image clarity and detail capture; when the brightness is high or the texture complexity is low, no additional luminous flux is needed. The predicted brightness and predicted texture complexity are multiplied by the brightness correction weight and texture correction weight determined during factory calibration, respectively, and then both are superimposed on the initially calibrated target luminous flux to obtain the corrected target luminous flux. The specific calculation formula is as follows:

[0116] Q′=Q×[1+c1×(1-Lnorm)+c2×Cnorm], where c1 and c2 are preset coefficients; Lnorm is the normalized predicted average brightness value; Cnorm is the normalized predicted texture complexity parameter; Q=I0×T0.

[0117] The revised target luminous flux takes into account LED physical attenuation, page brightness, and texture complexity, thereby providing adaptive exposure control for scanned documents with different content.

[0118] In other words, in some embodiments of this application, the corrected target luminous flux can also be used to calculate the compensated exposure time, i.e., compensated exposure time = Q′ ÷ current relative luminous intensity. Through this processing method, the determination of the compensated exposure time simultaneously considers the LED attenuation state, the calibrated luminous flux, and the predicted document brightness and texture complexity, achieving pre-estimation of the complexity of the new scanning task content and adaptive exposure control, thereby improving the consistency of image brightness and the stability of scanning quality between different batches of scanning tasks.

[0119] In addition, this embodiment also provides a dynamic compensation device for the exposure time of a reverse-side CIS sensor, applied to a duplex automatic paper feeder scanning system. The duplex automatic paper feeder scanning system is equipped with independent CIS sensors for both the front and reverse sides to achieve synchronous scanning of both sides. See [link to documentation]. Figure 3 The exposure time dynamic compensation device includes:

[0120] The LED luminous intensity acquisition module is used to obtain the current relative luminous intensity of the reverse CIS sensor LED based on the current cumulative lighting time of the reverse CIS sensor LED and the pre-stored decay curve model of the reverse CIS sensor LED when receiving a double-sided scanning task instruction initiated by the user. The cumulative lighting time of the reverse CIS sensor LED is obtained by accumulating statistics through a timer during the execution of historical double-sided scanning tasks.

[0121] The compensated exposure time acquisition module is used to acquire the compensated exposure time based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor, wherein the initial parameters of the reverse CIS sensor include: initial exposure time T0 and initial luminous intensity I0 of the LED.

[0122] The CIS control module is used to control the LED of the reverse CIS sensor to light up according to the compensated exposure time.

[0123] The dynamic exposure time compensation device for the reverse CIS sensor provided in this embodiment, through the LED luminous intensity acquisition module, can obtain the relative luminous intensity of the reverse CIS sensor LED in real time based on the current cumulative lighting time of the LED and the pre-stored decay curve model when receiving the double-sided scanning task command initiated by the user. This accurately reflects the LED usage status and light decay, provides reliable input data for exposure time compensation, and avoids the decrease in image brightness caused by LED decay.

[0124] Secondly, the compensation exposure time acquisition module calculates the compensated exposure time based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor (including the initial exposure time T0 and the initial luminous intensity I0 of the LED), realizing dynamic adjustment of the exposure time under different LED conditions, so that the image brightness is stable and consistent for a long time during the scanning process, thus improving the scanning quality.

[0125] In addition, the CIS control module can precisely control the LEDs of the reverse CIS sensor to light up based on the acquired compensated exposure time, ensuring that image acquisition can be completed with optimized exposure time for each scanning task, thereby improving the reliability and repeatability of the scanning system.

[0126] In summary, this exposure time dynamic compensation device achieves closed-loop adjustment and adaptive compensation by acquiring the relative luminous intensity of the LED in real time, dynamically calculating the compensation exposure time, and precisely controlling the LED lighting. This effectively ensures the brightness stability and imaging quality of the scanned image from the reverse side CIS sensor in the duplex automatic paper feeder scanning system.

[0127] This application also provides a duplex automatic paper feeder scanning system, including:

[0128] The scanning unit is used to perform double-sided scanning tasks;

[0129] The reverse CIS sensor LED exposure time dynamic compensation device described in the embodiment is used to control the reverse CIS sensor LED in the scanning unit to light up according to the compensated exposure time in order to complete the double-sided scanning task initiated by the user.

[0130] In summary, the dynamic compensation method and apparatus for exposure time of the reverse-side CIS sensor in this application addresses the attenuation problem of the LED of the reverse-side CIS sensor over time in a duplex automatic paper feeder scanning system by constructing a complete dynamic closed-loop compensation mechanism. Upon receiving a user-initiated duplex scanning task command, the method first obtains the relative luminous intensity of the LED based on the current cumulative lighting time of the reverse-side CIS sensor LED and a pre-stored attenuation curve model, accurately reflecting the LED's light decay state and fundamentally overcoming the impact of brightness decrease on scanned image quality during long-term use. Subsequently, by combining the initial parameters of the reverse-side CIS sensor (initial exposure time T0, initial LED luminous intensity I0), the compensated exposure time is calculated, achieving real-time dynamic adjustment of the LED lighting time. This ensures that the target luminous flux is obtained for each scanning task, maintaining the brightness stability of the scanned image. After each duplex scanning task is completed, this method further updates the cumulative lighting time, enabling the dynamic compensation of exposure time to continuously adapt to changes in the LED's usage state, avoiding compensation deviations caused by the loss or underutilization of historical lighting data. This method uses the product of the initial exposure time and the initial luminous intensity of the LED as the target luminous flux, and divides it by the current relative luminous intensity to obtain the compensated exposure time. This method is computationally simple and efficient, and allows for segmented corrections for different cumulative illumination stages. To further improve the accuracy of brightness closed-loop control, after N consecutive double-sided scanning tasks, the method calculates the average brightness value of the scanned image and compares it with a preset target brightness threshold to calculate the brightness deviation ΔL. When the deviation exceeds the tolerance threshold, the LED attenuation coefficient k is updated based on the brightness deviation, cumulative illumination time, and a preset correction coefficient α, thereby dynamically optimizing the exposure time for subsequent scanning tasks. The method used to calculate the average brightness of the scanned image, by selecting the image center region, performing grayscale histogram statistics, percentile removal, and averaging, eliminates interference from edges, extreme values, and noise, improving the robustness and stability of brightness measurement while maintaining computational efficiency, thus meeting the real-time closed-loop adjustment requirements of the scanning system. Furthermore, the preset correction coefficient α is set in segments based on the cumulative LED lighting time t1. A larger correction coefficient α1 is used in the initial stage (0 ≤ t1 < T1) to quickly respond to early brightness decline; a medium correction coefficient α2 is used in the middle stage (T1 ≤ t1 < T2) to balance adjustment speed and stability; and a smaller correction coefficient α3 is used in the later stage (t1 ≥ T2) to avoid over-correction causing brightness fluctuations, ensuring that the dynamic compensation of exposure time maintains high precision and consistency throughout the entire LED lifecycle. For example, segmented settings of α1=0.1, α2=0.07, α3=0.04, T1=100 hours, and T2=500 hours enable adaptive closed-loop adjustment of the light decay rate for different usage stages.

[0131] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0132] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first and second features are in direct contact, or that they are in indirect contact through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0133] In the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0134] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for dynamic compensation of exposure time of a reverse-side CIS sensor, characterized in that, An application in a duplex automatic document feeder scanning system, wherein the duplex automatic document feeder scanning system is equipped with independent CIS sensors for the front and back sides to achieve synchronous scanning of the front and back sides, the method includes: Upon receiving a user-initiated bi-sided scanning task instruction, the relative luminous intensity of the reverse CIS sensor LED is obtained based on the current cumulative lighting time of the reverse CIS sensor LED and the pre-stored attenuation curve model of the reverse CIS sensor LED. The cumulative lighting time of the reverse CIS sensor LED is: during the execution of historical bi-sided scanning tasks, the actual lighting time of the reverse CIS sensor LED in each bi-sided scanning task is recorded by a timer, and the total duration is obtained by accumulating the actual lighting time of the LED in each bi-sided scanning task. The compensated exposure time is obtained based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor, wherein the initial parameters of the reverse CIS sensor include: initial exposure time T0 and initial LED luminous intensity I0. The LED of the reverse CIS sensor is controlled to light up according to the compensated exposure time.

2. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 1, characterized in that, After controlling the LED of the reverse CIS sensor to be lit according to the compensated exposure time, the method further includes: Based on the actual illumination time of the LED of the reverse CIS sensor in the user-initiated double-sided scanning task, the cumulative illumination time in the timer is updated to obtain a new cumulative illumination time, which serves as dynamic compensation for the exposure time of the next user-initiated double-sided scanning task. In the user-initiated double-sided scanning task, the actual lighting time of the LED on the reverse side of the CIS sensor is determined by the product of the number of pages scanned in the user-initiated double-sided scanning task and the pre-acquired single-page scanning time.

3. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 1 or 2, characterized in that, The compensated exposure time is obtained based on the current relative luminescence intensity and pre-stored initial parameters of the reverse CIS sensor, including: The product of the initial exposure time T0 and the initial luminous intensity I0 of the LED is taken as the target luminous flux Q; The compensated exposure time is obtained by dividing the target luminous flux Q by the current relative luminous intensity.

4. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 3, characterized in that, The attenuation curve model of the reverse-side CIS sensor LED is expressed by the following formula: I(t) = I0 × (1 - k × t), where t is the cumulative lighting time, k is the preset attenuation coefficient, and I(t) is the relative luminous intensity of the LED of the reverse CIS sensor when the cumulative lighting time is t.

5. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 4, characterized in that, After N consecutive user-initiated bi-directional scanning tasks are completed, the method further includes: The average brightness value of the scanned images corresponding to N double-sided scanning tasks is obtained respectively, and the average brightness value of the N scanned images is statistically analyzed to obtain the average brightness value, which is used as the actual brightness value (Lactual). The actual brightness value Lactual is compared with the preset target brightness threshold Ltarget to obtain the brightness deviation ΔL; When the brightness deviation ΔL exceeds the preset brightness tolerance threshold, the attenuation coefficient k is updated according to the following formula to obtain the updated attenuation coefficient k′, and the updated attenuation coefficient k′ is stored for use in the dynamic compensation calculation of exposure time in subsequent user-initiated double-sided scanning tasks. k′=k+α×(ΔL / Ltarget)×(1 / t1), where t1 is the cumulative lighting time of the LED of the reverse CIS sensor after completing the N consecutive user-initiated double-sided scanning tasks, and α is a preset correction coefficient.

6. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 5, characterized in that, The average brightness value of the scanned image is calculated according to the following steps: A preset central region located in the middle of the width and height of the scanned image is selected as the statistical region, wherein the width and height of the central region are both less than half of the corresponding dimension of the scanned image. Perform grayscale histogram statistics on the pixels within the statistical region to obtain the pixel quantity distribution corresponding to each grayscale level; Based on the pixel quantity distribution corresponding to each gray level, determine the proportion of each gray level pixel quantity in the total pixel quantity; Based on the aforementioned proportion distribution, pixels whose grayscale values ​​are below a preset lower percentile threshold and above a preset upper percentile threshold are removed. The average brightness value of the scanned image is obtained by averaging the gray values ​​of the remaining pixels after removal.

7. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 6, characterized in that, The preset correction coefficient α is set in segments based on the cumulative illumination time t1 of the LED of the reverse CIS sensor after completing the N consecutive user-initiated double-sided scanning tasks, wherein: When 0 ≤ t1 < T1, α = α1; When T1≤t1<T2, α=α2; When t1≥T2, α=α3; Where α1>α2>α3, and T1 and T2 are preset time thresholds.

8. The method for dynamic compensation of exposure time of the reverse-side CIS sensor according to claim 7, characterized in that, α1 is 0.1, α2 is 0.07, α3 is 0.04, T1 is 100 hours, and T2 is 500 hours.

9. A dynamic compensation device for exposure time of a reverse-side CIS sensor, characterized in that, This is applied to a duplex automatic document feeder scanning system, which is equipped with independent CIS sensors for the front and back sides to achieve synchronous scanning of both sides. The dynamic exposure time compensation device includes: The LED luminous intensity acquisition module is used to obtain the current relative luminous intensity of the reverse CIS sensor LED based on the current cumulative lighting time of the reverse CIS sensor LED and the pre-stored decay curve model of the reverse CIS sensor LED when receiving a double-sided scanning task instruction initiated by the user. The cumulative lighting time of the reverse CIS sensor LED is obtained by accumulating statistics through a timer during the execution of historical double-sided scanning tasks. The compensated exposure time acquisition module is used to acquire the compensated exposure time based on the current relative luminous intensity and the pre-stored initial parameters of the reverse CIS sensor, wherein the initial parameters of the reverse CIS sensor include: initial exposure time T0 and initial luminous intensity I0 of the LED. The CIS control module is used to control the LED of the reverse CIS sensor to light up according to the compensated exposure time.

10. A double-sided automatic paper feeder scanning system, characterized in that, include: The scanning unit is used to perform double-sided scanning tasks; The reverse CIS sensor LED exposure time dynamic compensation device according to claim 9 is used to control the reverse CIS sensor LED in the scanning unit to light up according to the compensated exposure time, so as to complete the double-sided scanning task initiated by the user.

Images