Night vision mode switching method and system, and camera device

Abstract

This invention discloses a night vision mode switching method, system, and camera device. The method includes: when night vision entry conditions are met, acquiring the first natural light channel value and the first infrared channel value output by a dual-channel photosensitive device with the infrared fill light off; turning on the infrared fill light and acquiring the second natural light channel value and the second infrared channel value output after turning it on; calculating a crosstalk correction coefficient, which is equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value; using this coefficient to correct the second natural light channel value to eliminate infrared crosstalk components, obtaining a corrected natural light channel value; and controlling whether to turn off the infrared fill light to exit night vision mode based on a comparison between the corrected natural light channel value and a preset night vision exit threshold. By dynamically calculating the device-specific crosstalk correction coefficient and deducting infrared crosstalk at the source, the problem of repeated erroneous switching of night vision modes caused by infrared crosstalk is solved.

Description

Technical Field

[0001] This invention relates to the field of camera technology, and in particular to a night vision mode switching method, system, and camera device. Background Technology

[0002] With the increasing popularity of security monitoring and smart homes, video cameras with night vision capabilities (such as network cameras and smart doorbells) have been widely used. These cameras can operate in daylight mode during the day or in well-lit conditions, and are equipped with infrared illuminators to automatically switch to night vision mode and activate the infrared illuminators at night or in low-light environments, thus ensuring clear imaging.

[0003] To enable automatic switching between daytime and night vision modes, camera devices typically use sensors to detect ambient light levels. When the ambient light is below a preset night vision entry threshold, the device enters night vision mode and activates the infrared fill light; when the ambient light is above a preset night vision exit threshold, the device exits night vision mode and deactivates the infrared fill light.

[0004] However, in practical applications, when the infrared fill light is turned on, the infrared light it emits can be detected by the sensor through object reflection, optical structure scattering, and other means, causing the natural light detection value to be falsely high.

[0005] At this point, the device may mistakenly determine that the ambient brightness meets the conditions for exiting night vision mode, thus prematurely turning off the infrared fill light and switching back to daytime mode. After the infrared fill light is turned off, the natural light detection value drops rapidly, and the device will re-enter night vision mode. This repeated cycle causes frequent oscillations between daytime and night vision modes, severely impacting user experience and monitoring reliability. Summary of the Invention

[0006] The purpose of this invention is to provide a night vision mode switching method, system, and camera device, which aims to improve the accuracy of night vision mode switching and avoid repeated mode switching by dynamically calculating the infrared crosstalk correction coefficient of each device and performing source correction on the natural light channel value.

[0007] To address the aforementioned technical problems, embodiments of the present invention provide a night vision mode switching method, applied to a camera device having a dual-channel light sensor and an infrared fill light. The dual-channel light sensor has a natural light detection channel and an infrared light detection channel. The natural light detection channel detects ambient natural light and outputs a natural light channel value, while the infrared light detection channel detects infrared light and outputs an infrared channel value. The method includes: When the night vision entry conditions are met, the first natural light channel value and the first infrared channel value detected by the dual-channel photosensitive device are obtained when the infrared fill light is off. Turn on the infrared fill light, and obtain the second natural light channel value and the second infrared channel value detected by the dual-channel photosensitive device after the infrared fill light is turned on; Calculate the crosstalk correction coefficient, which is equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value. The crosstalk correction coefficient is used to reflect the degree of crosstalk between infrared light and natural light detection. The crosstalk correction coefficient is used to correct the second natural light channel value to eliminate the infrared crosstalk component in the second natural light channel value, thereby obtaining the corrected natural light channel value. Based on the comparison result between the corrected natural light channel value or its converted value and the preset night vision exit threshold, control whether to turn off the infrared fill light and exit the night vision mode.

[0008] In one embodiment, after the infrared fill light is turned on and a preset stabilization time has elapsed, the second natural light channel value and the second infrared channel value are obtained; And / or, the corrected natural light channel value is calculated according to the following formula: E = D - (C - A) / K; Wherein, A is the first infrared channel value, C is the second infrared channel value, D is the second natural light channel value, K is the crosstalk correction coefficient, and E is the corrected natural light channel value.

[0009] In one embodiment, the night vision entry condition includes the natural light channel value or its converted value detected in real time by the dual-channel light sensor being lower than a preset night vision entry threshold; and the night vision entry threshold is less than the night vision exit threshold.

[0010] In one embodiment, the crosstalk correction coefficient is limited to a preset range. When the calculated crosstalk correction coefficient is less than a first threshold, the first threshold is used; when the calculated crosstalk correction coefficient is greater than a second threshold, the second threshold is used.

[0011] In one embodiment, the first threshold is greater than or equal to 14 and less than or equal to 16, and the second threshold is greater than or equal to 28 and less than or equal to 30.

[0012] In one embodiment, each time night vision mode is entered, the following steps are executed: acquiring the first natural light channel value and the first infrared channel value, turning on the infrared fill light, acquiring the second natural light channel value and the second infrared channel value, calculating the crosstalk correction coefficient, performing correction using the crosstalk correction coefficient, and controlling whether to exit night vision mode based on the corrected natural light channel value.

[0013] In one embodiment, the method further includes: pre-calibrating the dual-channel photosensitive device and establishing a conversion relationship between the original reading value output by the natural light detection channel of the dual-channel photosensitive device and the illuminance value; In the step of comparing the corrected natural light channel value or its converted value with the preset night vision exit threshold, the corrected natural light channel value is converted into an illuminance value according to the conversion relationship, and compared with the preset night vision exit threshold in illuminance units.

[0014] This application also provides a night vision mode switching system, applied to a camera device with a dual-channel light sensor and an infrared fill light, including: The acquisition module is used to acquire the first natural light channel value and the first infrared channel value detected by the dual-channel photosensitive device when the night vision entry conditions are met; the acquisition module is also used to acquire the second natural light channel value and the second infrared channel value detected by the dual-channel photosensitive device after the infrared fill light is turned on. The control system is used to control the opening and closing of the infrared supplementary light; the control system is used to calculate a crosstalk correction coefficient based on the first natural light channel value, the first infrared channel value, the second natural light channel value, and the second infrared channel value, the crosstalk correction coefficient being equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value, the crosstalk correction coefficient being used to reflect the degree of crosstalk between infrared light and natural light detection; The control system is further configured to use the crosstalk correction coefficient to correct the second natural light channel value in order to eliminate the infrared crosstalk component and obtain the corrected natural light channel value. The control system is also used to control whether to turn off the infrared fill light to exit the night vision mode based on the comparison result between the corrected natural light channel value or its converted value and the preset night vision exit threshold.

[0015] In one embodiment, the acquisition module is used to acquire the second natural light channel value and the second infrared channel value after a preset stabilization time has elapsed after the infrared fill light is turned on; And / or, the control system is further configured to limit the crosstalk correction coefficient to a preset range, wherein when the calculated crosstalk correction coefficient is less than a first threshold, the first threshold is taken, and when the calculated crosstalk correction coefficient is greater than a second threshold, the second threshold is taken. And / or, the night vision mode switching system further includes a calibration module, which is used to pre-calibrate the dual-channel photosensitive device, establish a conversion relationship between the original reading value output by the natural light detection channel of the dual-channel photosensitive device and the illuminance value; the control system is used to convert the corrected natural light channel value into an illuminance value according to the conversion relationship, and compare it with a night vision exit threshold preset in illuminance units to determine whether to turn off the infrared fill light to exit the night vision mode.

[0016] In one embodiment, the control system calculates the crosstalk correction coefficient each time it enters night vision mode.

[0017] This application also provides a camera device, including: A dual-channel photosensitive device is used to detect ambient natural light and obtain a first natural light channel value and a second natural light channel value; it is also used to detect infrared light and obtain a first infrared channel value and a second infrared channel value. Infrared supplementary light; The aforementioned night vision mode switching system is electrically connected to the dual-channel light sensor and the infrared fill light, respectively.

[0018] The embodiments of this application have the following beneficial effects: First, by calculating the crosstalk correction coefficient and correcting the natural light channel value, the device always bases its exit judgment on the actual ambient natural light brightness, rather than being misled by artificially high readings. Therefore, the device will not mistakenly turn off the infrared supplementary light when the actual ambient brightness has not reached the required level, thus avoiding repeated oscillations that would occur when re-entering night vision mode due to a sudden drop in brightness after the lights are turned off, achieving accurate and timely switching to night vision mode.

[0019] Second, it achieves individualized adaptive correction. The embodiments of this application can acquire data on each device before and after its own infrared fill light is turned on, and dynamically calculate the crosstalk correction coefficient specific to the current device. This allows it to adapt to the individual differences of each device and overcomes the problem that a uniform compensation value is difficult to apply.

[0020] Specifically, upon entering night vision mode, the system sequentially acquires the first natural light channel value B and the first infrared channel value A when the infrared supplementary light is off, and the second natural light channel value D and the second infrared channel value C when the infrared supplementary light is on. Then, the crosstalk correction coefficient K of the current device is calculated in real time using the formula K = (C - A) / (D - B). This coefficient reflects the degree of crosstalk between infrared light and the natural light detection channel, and is obtained in real time by each device under its actual operating conditions. This effectively overcomes the problem that a uniform fixed compensation value cannot adapt to differences in different devices, structures, and assembly.

[0021] Third, it overcomes the limitations of passive avoidance strategies. This application directly corrects the natural light channel readings at the source by actively measuring and subtracting infrared crosstalk components, completely avoiding mode switching errors caused by infrared crosstalk. Specifically, the crosstalk correction coefficient K calculated in real time is substituted into the formula E = D - (C - A) / K, where (C - A) is the increment of the infrared light detection channel after the infrared illuminator is turned on, and (C - A) / K is the estimated infrared crosstalk component mixed into the natural light detection channel. After subtracting this component from the second natural light channel value D, the corrected natural light channel value E is the true ambient natural light brightness. E is compared with the night vision exit threshold, and the infrared illuminator is only turned off when the true ambient natural light brightness meets the exit condition. In this way, mode switching errors caused by infrared crosstalk are fundamentally eliminated, significantly improving the stability and accuracy of night vision mode switching.

[0022] Fourth, this application's embodiments introduce a crosstalk correction coefficient into the control loop of night vision mode switching, allowing the corrected natural light channel value to be directly compared with the night vision exit threshold. Based on the comparison result, the system actively controls the shutdown of the infrared fill light and the exit of night vision mode. In some solutions, the crosstalk coefficient is only used for image color calibration and has never been used in the closed-loop decision-making of the real-time control device's operating mode. This application breaks this technical convention, transforming the originally static image correction algorithm into a dynamic, closed-loop control basis that can participate in mode switching decisions, achieving a technological leap from image editing to light control, and demonstrating originality. Attached Figure Description

[0023] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0024] 1-Dual-channel light sensor, 2-Infrared fill light, 31-Acquisition module, 32-Control system; Figure 1 This is a flowchart illustrating the night vision mode switching method in an embodiment of this application; Figure 2 This is a circuit connection diagram of the camera device in the embodiments of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been presented in the various embodiments of the present invention to enable the reader to better understand this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments.

[0026] In the following description, certain specific details are set forth for the purpose of illustrating various disclosed embodiments in order to provide a thorough understanding of the various disclosed embodiments. However, those skilled in the art will recognize that embodiments may be practiced without one or more of these specific details. In other instances, well-known apparatuses, structures, and techniques associated with this application may not have been shown or described in detail to avoid unnecessarily obscuring the description of the embodiments.

[0027] Unless the context requires otherwise, throughout the specification and claims, the word “comprising” and its variations, such as “including” and “having”, shall be understood to have an open, inclusive meaning, that is, to be interpreted as “including, but not limited to”.

[0028] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings to provide a clearer understanding of the purpose, features, and advantages of the present invention. It should be understood that the embodiments shown in the drawings are not intended to limit the scope of the present invention, but are merely illustrative of the essential spirit of the technical solution of the present invention.

[0029] Throughout this specification, references to "an embodiment" or "an embodiment" indicate that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Therefore, the appearance of "in an embodiment" or "an embodiment" in various places throughout the specification does not necessarily refer to the same embodiment. Furthermore, a particular feature, structure, or characteristic may be combined in any manner in one or more embodiments.

[0030] The singular forms “a” and “the” used in this specification and the appended claims include plural references unless otherwise expressly stated herein. It should be noted that the term “or” is generally used to mean “and / or” unless otherwise expressly stated herein.

[0031] In the following description, in order to clearly demonstrate the structure and working method of the present invention, a number of directional terms will be used. However, terms such as "front", "back", "left", "right", "outside", "inside", "outward", "inward", "up", and "down" should be understood as convenient terms and not as limiting terms.

[0032] Embodiments of the present invention are described below with reference to the accompanying drawings.

[0033] In response to the problems in the background technology, the applicant has also tried some other methods, such as using a fixed delay, setting a large hysteresis interval, or making simple compensation for the photosensitive readings to alleviate them. However, the above methods are all passive avoidance strategies and cannot fundamentally eliminate the impact of infrared crosstalk on natural light detection. Moreover, the differences in optical structure and device consistency between different devices make it difficult to apply a uniform compensation value.

[0034] Specifically, the applicant tried a fixed delay method, which involves waiting a fixed amount of time after the infrared illuminator is turned on before reading the light sensor data to avoid the light intensity fluctuations at the moment the light is turned on. However, the infrared illuminator activation characteristics of different devices vary, and the fixed delay cannot adapt to all devices. Moreover, crosstalk persists during the delay period and cannot eliminate the artificially high readings. The applicant also tried setting a hysteresis interval with a large range, that is, setting different values ​​for the night vision entry threshold and exit threshold to increase the switching buffer area. However, the hysteresis interval can only reduce the switching frequency. Once the artificially high brightness caused by infrared crosstalk exceeds the exit threshold, the device will still switch erroneously and then re-enter night vision mode due to a sudden drop in brightness, resulting in oscillation. The applicant also tried a simple compensation method for the photosensitive reading, that is, subtracting a fixed compensation amount from the natural light channel reading based on empirical values ​​or factory calibration values. However, due to the large differences in optical structure, sensor consistency, assembly errors and other factors among different devices, a uniform fixed compensation value cannot adapt to the actual crosstalk level of each device. This often results in undercompensation or overcompensation, making it difficult to guarantee a stable switching effect. These passive avoidance strategies cannot fundamentally eliminate the impact of infrared crosstalk on natural light detection, nor can they solve the technical problem that a uniform compensation value is difficult to apply to individual differences in different devices.

[0035] Embodiments of the present invention are described below with reference to the accompanying drawings.

[0036] First Implementation Method The first embodiment of the present invention relates to a night vision mode switching method, applied to a camera device having a dual-channel light sensor 1 and an infrared fill light 2. The dual-channel light sensor 1 has a natural light detection channel and an infrared light detection channel. The natural light detection channel is used to detect ambient natural light and output a natural light channel value, and the infrared light detection channel is used to detect infrared light and output an infrared channel value. The core of this implementation is as follows: when the night vision entry conditions are met, the first natural light channel value and the first infrared channel value output by the dual-channel photosensitive device 1 are obtained when the infrared fill light 2 is off; the infrared fill light 2 is turned on, and the second natural light channel value and the second infrared channel value output by the dual-channel photosensitive device 1 after the infrared fill light 2 is turned on are obtained; a crosstalk correction coefficient is calculated, which is equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value. The crosstalk correction coefficient is used to reflect the degree of crosstalk between infrared light and natural light detection; the second natural light channel value is corrected using the crosstalk correction coefficient to eliminate the infrared light crosstalk component in the second natural light channel value, thereby obtaining a corrected natural light channel value; based on the comparison result between the corrected natural light channel value or its converted value and a preset night vision exit threshold, the system controls whether to turn off the infrared fill light 2 and exit the night vision mode. In this way, each device can calculate its own crosstalk correction coefficient in real time when entering night vision mode, and use this coefficient to perform source correction on the natural light channel reading, thereby obtaining the true ambient natural light brightness and avoiding false exit from night vision mode due to infrared crosstalk. The implementation details of the night vision mode switching method of this embodiment are described below. The following content is only for ease of understanding and is not essential for implementing this solution.

[0037] The night vision mode switching method in this embodiment is as follows: Figure 1 As shown, the specific steps include: Step S101: When the natural light channel value or its converted value detected in real time by the dual-channel photosensitive device 1 is lower than a preset night vision entry threshold, it is determined that the night vision entry condition is met. The night vision entry threshold can be 0.5 lux. The first natural light channel value and the first infrared channel value output by the dual-channel photosensitive device 1 are obtained when the infrared fill light 2 is off. The night vision entry condition can be that the first natural light channel value or its converted value is lower than the preset night vision entry threshold, such as 0.5 lux. At this time, since the infrared fill light 2 is not yet turned on, the first infrared channel value is usually close to zero and can be used as a benchmark for subsequent calculations. It should be noted that the first natural light channel value and the first infrared channel value are the values ​​of two different channels output by the dual-channel photosensitive device 1 at the same time, representing the ambient natural light intensity and the ambient infrared light intensity, respectively.

[0038] Specifically, the first natural light channel value refers to the raw value output by the natural light detection channel of the dual-channel photosensitive device 1 when the infrared supplementary light 2 is off. This value represents the brightness of the current ambient natural light. The first infrared channel value refers to the raw value output by the infrared light detection channel of the dual-channel photosensitive device 1 at the same moment. Since the infrared supplementary light 2 is not yet turned on, this value is usually close to zero and can be used as a reference benchmark for ambient infrared background noise. These two values ​​together constitute the baseline data before the infrared supplementary light 2 is turned on.

[0039] Step S102: Turn on the infrared supplementary light 2 and acquire the second natural light channel value and the second infrared channel value output by the dual-channel photosensitive device 1 after the infrared supplementary light 2 is turned on. To ensure sampling accuracy, a preset stabilization time (e.g., 200 milliseconds) can be allowed after the infrared supplementary light 2 is turned on before acquiring the second natural light channel value and the second infrared channel value, to avoid the influence of light intensity fluctuations at the moment of light activation. This stabilization time can be calibrated according to the model of the infrared supplementary light 2 and the characteristics of the driving circuit. For example, for a common 850nm infrared LED, the stabilization time can be set between 100 milliseconds and 300 milliseconds. In a specific embodiment of this implementation, when the rated current of the infrared supplementary light 2 is 150mA, the stabilization time can be 150 milliseconds. At this time, the acquired second natural light channel value and the second infrared channel value best reflect the true situation after the infrared supplementary light 2 has stabilized.

[0040] The second natural light channel value refers to the original value output by the natural light detection channel of the dual-channel photosensitive device 1 after the infrared fill light 2 is turned on (after a preset stabilization time). This value includes both the actual ambient natural light component and the infrared contamination component that crosstalks into the natural light detection channel due to the reflection or scattering of infrared light emitted by the infrared fill light 2. The second infrared channel value refers to the original value output by the infrared light detection channel of the dual-channel photosensitive device 1 after the infrared fill light 2 is turned on. This value represents the detectable infrared light intensity and is used to reflect the relative strength of the infrared light emitted by the infrared fill light 2 when it is received by the sensor.

[0041] For ease of understanding, the meanings of the various natural light channel-related values ​​involved in this application are explained below: The natural light channel value is a higher-level concept, generally referring to the original value output by the natural light detection channel of the dual-channel photosensitive device 1 at any time. This value is the original digital quantity without any correction.

[0042] The first natural light channel value refers to the original value output by the natural light detection channel of the dual-channel photosensitive device 1 when the infrared supplementary light 2 is off, denoted as B. This value represents the ambient natural light brightness before the infrared supplementary light 2 is turned on, and is the reference value for subsequent crosstalk correction.

[0043] The second natural light channel value refers to the original value output by the natural light detection channel of the dual-channel photosensitive device 1 after the infrared supplementary light 2 is turned on, denoted as D. This value includes two parts: the true ambient natural light component and the spurious component introduced by infrared light crosstalk, which is the object of subsequent correction.

[0044] The corrected natural light channel value, denoted as E, is the value obtained by correcting the second natural light channel value D using the crosstalk correction coefficient K. Its calculation formula is E = D - (C - A) / K, where A is the first infrared channel value and C is the second infrared channel value. This value eliminates the influence of infrared crosstalk components and accurately reflects the brightness of the ambient natural light. Its physical meaning is the same as the first natural light channel value B. Theoretically, E is approximately equal to B and is used to compare with the night vision exit threshold to determine whether to exit night vision mode.

[0045] The infrared channel value is a higher-level concept, generally referring to the original value output by the infrared light detection channel of the dual-channel photosensitive device 1 at any time. This value represents the intensity of infrared light that can be detected.

[0046] The first infrared channel value refers to the original value output by the infrared light detection channel of the dual-channel photosensitive device 1 when the infrared fill light 2 is off, denoted as A. Since the infrared fill light 2 is not yet turned on, this value is usually close to zero and can be used as a reference benchmark for ambient infrared background noise.

[0047] The second infrared channel value, denoted as C, refers to the original value output by the infrared light detection channel of the dual-channel photosensitive device 1 after the infrared fill light 2 is turned on. This value represents the relative intensity of the infrared light emitted by the infrared fill light 2 when it is received by the dual-channel photosensitive device 1. Subtracting the first infrared channel value A yields the effective increment of infrared light ΔI = C - A, which is used to calculate the crosstalk correction coefficient K and the corrected natural light channel value E.

[0048] Step S103: Calculate the crosstalk correction coefficient K. The crosstalk correction coefficient K is equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value, i.e., K = (C - A) / (D - B), where A is the first infrared channel value, B is the first natural light channel value, C is the second infrared channel value, and D is the second natural light channel value. This crosstalk correction coefficient K reflects the degree of crosstalk between infrared light and the natural light detection channel: the smaller the K value, the more severe the contamination of the natural light channel by the same infrared light intensity; the larger the K value, the lighter the crosstalk. The reciprocal of K, 1 / K, represents the spurious increment caused by each unit of infrared light intensity in the natural light channel; therefore, K can be used as the basis for correction.

[0049] To prevent sudden environmental changes, such as external light source interference at the moment the light is turned on, from causing abnormalities in the calculated crosstalk correction coefficient K, the K value can be limited and set within a preset range, specifically from a first threshold to a second threshold. Specifically, when the calculated crosstalk correction coefficient K exceeds the preset range, it is directly replaced with a preset boundary value to avoid abnormal K values ​​contaminating subsequent correction results. For example, the crosstalk correction coefficient K of a normal device is usually stable between 15 and 29, i.e., the preset range is 15 to 29. If, at the moment the infrared supplementary light 2 is turned on, a sudden situation occurs where a car headlight directly shines on the sensor, causing the second natural light channel value D to be abnormally high, thus significantly increasing the denominator (DB), then due to the sudden environmental change, K becomes less than 15. In this case, K is forced to take the first threshold, i.e., 15. Another scenario that could lead to an abnormally high K value is as follows: When the infrared fill light 2 is turned on, the user simultaneously begins to draw the curtains. As the ambient natural light gradually decreases, the system first executes step S101 to read the first natural light channel value B. At this time, the curtains are not yet fully drawn, and the first natural light channel value B reflects the intermediate brightness during the curtain-drawing process. Subsequently, step S102 is executed to read the second natural light channel value D. At this time, the curtains are fully drawn, and the second natural light channel value D reflects the lower brightness after the curtains are drawn. Because the ambient natural light changes between the two samplings due to the drawing of the curtains, the denominator (DB) becomes abnormally small, resulting in a calculated K value that is greater than the normal range (e.g., greater than 29). In this case, if no amplitude limiting is performed, this abnormal K value will be used for subsequent correction, leading to inaccurate correction of the natural light channel value. Therefore, when the calculated K is greater than the second threshold (e.g., 29), K is also forced to be the second threshold, i.e., 29.

[0050] The actual value calculated in this instance is only used when the K value falls within the range of 15 to 29. This limiting process ensures that even if a single sample is affected by external interference, the subsequent correction results in the entire night vision mode will not be severely deviated due to abnormal K values, thereby improving the robustness and stability of the system.

[0051] It should be noted that the crosstalk correction coefficient K of normal equipment is typically stable between 15 and 29. Considering individual differences between different devices and the influence of environmental factors, the first threshold can be selected from any value between 14 and 16, such as 14, 15, or 16; the second threshold can be selected from any value between 28 and 30, such as 28, 29, or 30. In a specific example of this embodiment, 15 is used when the calculated K is less than 15, and 29 is used when the calculated K is greater than 29.

[0052] Step S104: The second natural light channel value D is corrected using the crosstalk correction coefficient K to eliminate the infrared crosstalk component, resulting in the corrected natural light channel value E. The correction formula is: E = D - (C - A) / K. Where (C) is the increment of the infrared light detection channel after the infrared supplementary light 2 is turned on; (C) / K is the estimated infrared crosstalk component mixed into the natural light detection channel. After subtracting this component from the second natural light channel value D, the remaining E is the true ambient natural light brightness.

[0053] Step S105: Based on the comparison between the calibrated natural light channel value E or its converted value and the preset night vision exit threshold, control whether to turn off the infrared fill light 2 and exit night vision mode. If E or the calibrated and converted illuminance value is greater than the night vision exit threshold (e.g., 2 lux), it indicates that the actual ambient natural light is bright enough, and the infrared fill light 2 should be turned off and the mode switched back to daytime mode; otherwise, the night vision mode is maintained. In this way, since the comparison object is the calibrated actual ambient natural light brightness, rather than the artificially high reading value after infrared contamination, the night vision mode is only exited when the environment actually becomes brighter, thus completely avoiding erroneous switching caused by infrared crosstalk.

[0054] To better understand the technical effects of this embodiment, a specific application scenario is described below. The night vision entry threshold is set to 0.5 lux, and the night vision exit threshold is set to 2 lux. Assume a camera is already in night vision mode. At this time, the infrared fill light 2 is on, and the actual ambient natural light brightness is 0.4 lux, which is less than the night vision entry threshold of 0.5 lux. Without calibration, due to infrared light crosstalk, the natural light channel value sensed by the dual-channel photosensitive device 1 changes to a brightness value of 3 lux. 3 lux is greater than the preset night vision exit threshold of 2 lux, so the camera exits night vision mode and turns off the infrared fill light 2. When the infrared fill light 2 is off, the natural light channel value sensed by the dual-channel photosensitive device 1 changes back to a brightness value of 0.4 lux, and the camera re-enters night vision mode. This cycle repeats, creating an oscillation.

[0055] The method described in this embodiment first calculates the crosstalk correction coefficient K, then calculates the corrected natural light channel value E based on the crosstalk correction coefficient K, and converts the corrected natural light channel value E into the actual brightness value of the current ambient natural light, which is 0.4 lux. Comparing the actual brightness value of the ambient natural light (0.4 lux) with the night vision exit threshold of 2 lux, 0.4 is less than 2, therefore the device continues to remain in night vision mode and will not experience an erroneous switch.

[0056] It is important to emphasize that in this embodiment, each time the camera switches from daytime mode to night vision mode, steps S101 to S105 are executed completely once. In this way, the crosstalk correction coefficient K always reflects the latest crosstalk characteristics of the device under its current operating state, thus adapting to parameter drift during long-term use due to equipment aging, temperature changes, loose assembly, etc., and ensuring correction accuracy.

[0057] Additionally, it should be noted that lux is a unit of illuminance, used to represent the luminous flux of visible light received per unit area, denoted by lx. In this embodiment, lux is used to quantify the brightness of ambient natural light; for example, the night vision entry threshold and night vision exit threshold are preset in lux. Typically, the ambient illuminance of a moonless night sky is approximately 0.001 lux, a full moon night is approximately 0.2 lux, dusk is approximately 10 lux, and normal indoor lighting is approximately 100 to 500 lux. The first natural light channel value, the second natural light channel value, and the corrected natural light channel value mentioned in this embodiment are all original values ​​of the dual-channel light sensor 1. Through a pre-calibrated conversion relationship, they can be converted into illuminance values ​​in lux, and then compared with the night vision entry threshold and night vision exit threshold in lux. It should be noted that the conversion relationship between the original values ​​of the dual-channel light sensor 1 and lux varies depending on the device model and individual differences, and usually needs to be determined through calibration experiments. For example, if a certain type of dual-channel photosensitive device 1 outputs a first natural light channel value of 130946 under a 300 lux standard light source, then its conversion factor is approximately 436.5, with the original value corresponding to 1 lux.

[0058] The calibration process can be implemented as follows: Place the camera device in a standard light source box, where the illuminance of the light source can be precisely adjusted. Sequentially set multiple known illuminance values, such as 0.1 lux, 1 lux, 10 lux, 50 lux, 100 lux, 200 lux, 300 lux, and 500 lux, covering the range from extremely low illuminance to daytime illuminance. At each illuminance point, after the light source stabilizes, read the raw value output from the natural light detection channel of the dual-channel photosensitive device 1. This raw value is the natural light channel value output by the natural light detection channel. Multiple samples can be taken at each illuminance point, and the average value is calculated to eliminate random noise. Record the collected data, with each set including the illuminance value and the corresponding raw reading of the natural light detection channel, establishing a mapping relationship between the raw reading of the natural light detection channel and the illuminance value. For example, if the original reading of the natural light detection channel is 130946 at 300 lux and 43650 at 100 lux, then the conversion factor can be calculated to be approximately 436.5 for the original reading corresponding to 1 lux. This mapping relationship can be stored as a lookup table or fitting formula in the device's non-volatile memory. In subsequent use, the control system 32 quickly calculates the corresponding illuminance value based on the currently acquired original reading of the corrected natural light channel value using the lookup table or formula, and then compares it with the night vision exit threshold in lux. If the corrected natural light channel value is directly compared with the original digital threshold, the night vision exit threshold (e.g., 2 lux) can also be converted in reverse to the corresponding original digital threshold. The calibration process can be completed on the camera equipment production line, with each device calibrated independently to eliminate individual device differences. After calibration, the camera equipment can accurately convert the corrected natural light channel value into the true illuminance, ensuring the accuracy of the night vision exit judgment.

[0059] The technical effects of the embodiments of this application are described below.

[0060] To address the shortcomings of the aforementioned passive avoidance strategies such as fixed delay, hysteresis interval, and simple compensation, this application achieves improvements through the following technical means, with the following technical effects: Firstly, addressing the issue that fixed delay methods cannot accommodate the differences in infrared lamp activation characteristics across different devices, this application does not rely on a fixed waiting time. Instead, it directly measures the actual infrared light intensity and natural light channel increment in a stable state after the infrared supplementary light 2 is turned on. Since the crosstalk correction coefficient is calculated based on the device's own real-time data, this application can obtain an accurate crosstalk ratio regardless of how the infrared lamp activation characteristics of different devices change, thus completely eliminating the sensitivity of fixed delays to individual device differences.

[0061] Secondly, addressing the issue that the hysteresis interval method can only reduce the switching frequency but not eliminate erroneous switching, this application corrects the natural light channel readings at the source, comparing the corrected actual ambient natural light brightness with the night vision exit threshold. Because the correction process eliminates the artificially high component caused by infrared crosstalk, the comparison result always reflects the true ambient brightness. Therefore, even if the artificially high reading caused by infrared crosstalk exceeds the exit threshold, the correction will not trigger erroneous switching, thus fundamentally eliminating the root cause of oscillation, rather than merely reducing the oscillation frequency.

[0062] It should be noted that, based on the source correction technology, the embodiments of this application can also set a hysteresis interval, i.e., the night vision entry threshold is less than the night vision exit threshold, for example, the night vision entry threshold is 0.5 lux and the night vision exit threshold is 2 lux. However, unlike simply relying on a large hysteresis interval to passively avoid false handovers, the hysteresis interval in this application is only used as an auxiliary means, and the hysteresis interval can be set relatively small. This is because the core of this application is that source correction eliminates most of the artificially high components caused by infrared crosstalk, making the corrected ambient natural light brightness close to the true value. At this time, even if a small hysteresis interval is set, such as setting the night vision entry threshold to 0.5 lux and the night vision exit threshold to 2 lux, false handovers will not be triggered due to artificially high crosstalk. In contrast, relying solely on a large hysteresis range to passively avoid erroneous handovers often requires setting an even larger hysteresis range (e.g., an entry threshold of 0.3 lux, an exit threshold of 5 lux, and a difference of 4.7 lux) to barely reduce the oscillation frequency without source correction. However, this can cause the device to remain in night vision mode for extended periods even when the actual ambient brightness is sufficient, impacting the user experience. Therefore, this application employs a combined approach of source correction as the primary method and hysteresis range as a secondary method. This retains the advantage of the hysteresis range in preventing minor fluctuations near the critical point while fundamentally eliminating false triggers caused by artificially high crosstalk through source correction. This allows for the use of a smaller hysteresis range while ensuring handover accuracy, further enhancing the user experience.

[0063] Secondly, to address the issue that a uniform fixed compensation value in simple compensation methods cannot adapt to individual differences among different devices, this application adopts a dynamic calculation method. Each device, when entering night vision mode, can calculate the crosstalk correction coefficient in real time based on its current detection data, rather than relying on a factory-calibrated empirical value. This coefficient is specific to the current device and can adapt to changes in crosstalk levels caused by different optical structures, sensor batches, and assembly tolerances, thereby avoiding under-compensation or over-compensation and ensuring the calibration accuracy of each device.

[0064] Finally, this application embodiment is the first to apply the crosstalk correction technology to the closed-loop control of night vision mode switching. This allows the corrected natural light channel value to be directly compared with the night vision exit threshold, and based on the comparison result, it actively controls the shutdown of the infrared fill light 2 and the exit of the night vision mode, solving the long-standing problem of repeated mode oscillations caused by infrared crosstalk in existing night vision switching technologies. Furthermore, this solution has undergone extensive engineering optimizations for night vision switching scenarios, including preset stabilization time to avoid light intensity fluctuations upon light activation, a K-value protection range to prevent interference from sudden environmental changes, and calibration conversions to facilitate threshold comparison.

[0065] In summary, this application transforms night vision mode switching from passive avoidance to active adaptation through active measurement, dynamic calculation, and source correction, fundamentally solving the problems of incorrect mode switching and repeated oscillations caused by infrared crosstalk. At the same time, it achieves self-adaptation to individual device differences, improving the stability and reliability of night vision mode switching.

[0066] Second Implementation Method The second embodiment of the present invention relates to a night vision mode switching system, such as... Figure 2 As shown, this is applied to a camera device having a dual-channel light sensor 1 and an infrared fill light 2. The night vision mode switching system of this embodiment includes: an acquisition module 31 and a control system 32.

[0067] The acquisition module 31 is used to acquire the first natural light channel value and the first infrared channel value output by the dual-channel photosensitive device 1 when the night vision entry conditions are met. The acquisition module 31 is also used to acquire the second natural light channel value and the second infrared channel value output by the dual-channel photosensitive device 1 after the infrared fill light 2 is turned on. The acquisition module 31 can be a data acquisition interface of the camera device's main controller, connected to the dual-channel photosensitive device 1 via an I2C or SPI bus, to read the register values ​​inside the dual-channel photosensitive device 1 in digital form.

[0068] In this embodiment, the dual-channel photosensitive device 1 is an integrated chip that integrates a natural light detection channel and an infrared light detection channel. The natural light detection channel detects ambient natural light and outputs a natural light channel value, specifically a first natural light channel value and a second natural light channel value. The infrared light detection channel detects infrared light and outputs an infrared channel value, specifically a first infrared channel value and a second infrared channel value. For example, the dual-channel photosensitive device 1 can be a dual-channel ambient light sensor chip of model HX3205 or HX3221. These chips internally encapsulate two photodiodes with different spectral response characteristics and their readout circuits, enabling simultaneous output of a natural light channel value corresponding to visible light intensity and an infrared channel value corresponding to infrared light intensity.

[0069] Of course, in other embodiments, the dual-channel light sensing device 1 can also consist of two independent sensors, one of which is dedicated to detecting ambient natural light and outputting natural light channel values, and the other is dedicated to detecting infrared light and outputting infrared channel values. The two independent sensors can each be selected from device models suitable for their respective spectral responses; for example, a TSL2591 can be used for the natural light sensor, and a VEML6075 for the infrared sensor. They can be laid out and connected on the circuit board according to their respective datasheets. Whether using a single-chip integrated solution or a dual-chip discrete solution, the acquisition functions of the natural light channel values ​​and infrared channel values ​​required by this application can be achieved.

[0070] The control system 32 is used to control the turning on and off of the infrared fill light 2; the control system 32 is used to calculate a crosstalk correction coefficient based on the first natural light channel value, the first infrared channel value, the second natural light channel value, and the second infrared channel value, wherein the crosstalk correction coefficient is equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value, and the crosstalk correction coefficient is used to reflect the degree of crosstalk of infrared light to natural light detection; wherein, the control system 32 is also used to use the crosstalk correction coefficient to correct the second natural light channel value to eliminate the infrared crosstalk component therein, thereby obtaining a corrected natural light channel value; the control system 32 is also used to control whether to turn off the infrared fill light 2 to exit the night vision mode based on the comparison result between the corrected natural light channel value or its converted value and a preset night vision exit threshold.

[0071] The control system 32 can be a microcontroller (MCU), such as an STM32F103 ARM processor, which integrates an A / D converter, a timer, and GPIO ports. One GPIO port is connected to the driver circuit of the infrared fill light 2, and another GPIO port is connected to the enable terminal of the infrared fill light 2. The control system 32 performs the aforementioned calculations and decisions based on the data read from the acquisition module 31.

[0072] In another embodiment, the control system 32 can also be implemented by a system-on-a-chip (SOC), such as the Hi3516 series or SSC338Q commonly used in the security monitoring field. This type of SOC integrates a processor core, image signal processing unit (ISP), general-purpose input / output interface (GPIO), and I2C / SPI controller. By running firmware, it can simultaneously realize the data reading function of the acquisition module 31 and the calculation, decision-making, and control functions of the control system 32.

[0073] It is not difficult to see that this embodiment is a system implementation corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and will not be repeated here to reduce repetition. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the first embodiment.

[0074] It is worth noting that all modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. For example, the acquisition module 31 and the control system 32 can be integrated into the same microcontroller chip, and their respective functions can be implemented through software programs. Furthermore, in order to highlight the innovative aspects of this invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by this invention, but this does not mean that other units do not exist in this embodiment.

[0075] Third Implementation Method The third embodiment of the present invention relates to a night vision mode switching system. The third embodiment is largely the same as the second embodiment, with the main difference being that in the second embodiment, the acquisition module 31 and the control system 32 are described independently; while in the third embodiment, specific optional technical features are further defined. Specifically, the acquisition module 31 is used to acquire the second natural light channel value and the second infrared channel value after a preset stabilization time has elapsed after the infrared fill light 2 is turned on. The control system 32 is also used to limit the crosstalk correction coefficient to a preset range; when the calculated crosstalk correction coefficient is less than a first threshold, the first threshold is used; when the calculated crosstalk correction coefficient is greater than a second threshold, the second threshold is used.

[0076] The night vision mode switching system also includes a calibration module, which is used to pre-calibrate the dual-channel light sensor 1 and establish a conversion relationship between the original reading value of the natural light channel value output by the natural light detection channel of the dual-channel light sensor 1 and the illuminance value; the control system 32 is used to convert the corrected natural light channel value into an illuminance value according to the conversion relationship, and compare it with a night vision exit threshold preset in illuminance units to determine whether to turn off the infrared fill light 2 to exit the night vision mode.

[0077] The calibration module is used to pre-calibrate the dual-channel light sensor 1. Since the natural light channel value detected by the natural light detection channel of the dual-channel light sensor 1 is a raw digital value (e.g., an integer between 0 and 65535), the proportional relationship between the raw reading and the actual physical illuminance (lux) varies between different devices and even for the same device under different environments. In order to meaningfully compare the calibrated natural light channel value (also a raw digital value) with a preset night vision exit threshold (e.g., 2 lux) in illuminance units, a conversion relationship between the raw reading of the natural light detection channel and the illuminance value must be established.

[0078] The calibration process is typically performed during the camera equipment manufacturing stage or upon initial use: the equipment is placed in a standard light source box, and multiple known illuminance values ​​(such as 0.1 lux, 1 lux, 10 lux, 100 lux, 300 lux, etc.) are set sequentially. The raw readings of the natural light channel values ​​detected by the natural light detection channel of the dual-channel photosensitive device 1 at each illuminance point are recorded. A mapping relationship between the raw readings and illuminance values ​​is obtained through linear regression or piecewise interpolation, for example, 1 lux for every 436 raw readings. This conversion relationship is stored in the device's non-volatile memory in the form of a table or formula. During night vision mode operation, the control system 32 first converts the corrected natural light channel value into an illuminance value in lux according to the pre-stored conversion relationship, and then compares it with a preset night vision exit threshold in illuminance units. If the converted illuminance value is greater than the night vision exit threshold, it is determined that the infrared supplementary light 2 should be turned off and the night vision mode should be exited; otherwise, the night vision mode is maintained. The calibration results can be stored for a long time. Even if the equipment is powered off or restarted, the conversion relationship remains valid and there is no need to repeat the calibration.

[0079] In specific implementations, besides converting the corrected natural light channel value into an illuminance value for comparison, an equivalent alternative can be used: a preset night vision exit threshold (in illuminance) can be converted in reverse using the same conversion relationship to the corresponding original digital threshold, and then the corrected natural light channel value can be directly compared with this original digital threshold. These two methods are mathematically equivalent and both fall within the scope of this application.

[0080] The relevant technical details mentioned in the above embodiments are still valid in this embodiment, and the technical effects that can be achieved in the above embodiments can also be achieved in this embodiment. In order to reduce repetition, they will not be repeated here.

[0081] Fourth Implementation Method A fourth embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the method described in the first embodiment.

[0082] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory, random access memory, magnetic disk, or optical disk.

[0083] The fifth embodiment of the present invention relates to a camera device, such as... Figure 2 As shown, it includes: a dual-channel light sensor 1, an infrared fill light 2, and a night vision mode switching system as described in the second or third embodiment.

[0084] The dual-channel light sensor 1 is an integrated chip that integrates a natural light detection channel and an infrared light detection channel. For example, the dual-channel light sensor 1 can be a dual-channel ambient light sensor chip of model HX3205 or HX3221. This chip internally encapsulates two photodiodes with different spectral response characteristics and their readout circuits, and can simultaneously output a natural light channel value corresponding to visible light intensity and an infrared channel value corresponding to infrared light intensity. Specifically, the dual-channel light sensor 1 is used to detect ambient natural light and output a first natural light channel value and a second natural light channel value; it is also used to detect infrared light and output a first infrared channel value and a second infrared channel value. The first natural light channel value and the first infrared channel value are the output values ​​of the dual-channel light sensor 1 when the infrared supplementary light 2 is off, and the second natural light channel value and the second infrared channel value are the output values ​​of the dual-channel light sensor 1 after the infrared supplementary light 2 is turned on (preferably after a preset stabilization time).

[0085] The infrared fill light 2 is used to emit infrared light for illumination in night vision mode. Its emission wavelength can be 850nm or 940nm, and the driving current and settling time can be determined according to the actual device selection. For example, the settling time is about 150 milliseconds when the rated current is 150mA.

[0086] The night vision mode switching system is electrically connected to the dual-channel light sensor 1 and the infrared fill light 2, respectively. Specifically, the night vision mode switching system includes an acquisition module 31 and a control system 32 (e.g., Figure 2(As shown). The acquisition module 31 is connected to the dual-channel photosensitive device 1 via an I2C or SPI bus, and is used to read the first natural light channel value, the first infrared channel value, the second natural light channel value, and the second infrared channel value output by the dual-channel photosensitive device 1. The control system 32 is connected to the driving circuit of the infrared fill light 2 via a GPIO port, and is used to control the turning on and off of the infrared fill light 2, and to perform the calculation of the crosstalk correction coefficient, the acquisition of the corrected natural light channel value, and the comparison decision with the night vision exit threshold. The specific functions and implementation methods of the night vision mode switching system have been described in detail in the second or third embodiment. In order to reduce repetition, they will not be repeated here. However, this embodiment can be implemented in conjunction with the previous embodiments, and the relevant technical details mentioned in the previous embodiments are still valid in this embodiment.

[0087] The camera device of this embodiment can dynamically calculate its own exclusive crosstalk correction coefficient and deduct infrared crosstalk from the source, thereby avoiding repeated false switching of night vision mode and improving the stability and accuracy of mode switching of the camera device in low-light environment.

[0088] The preferred embodiments of the present invention have been described in detail above, but it should be understood that, if necessary, aspects of the embodiments can be modified to utilize aspects, features, and concepts from various patents, applications, and publications to provide other embodiments.

[0089] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing the present invention, and in practical applications, various changes in form and detail may be made without departing from the spirit and scope of the present invention.

Claims

1. A night vision mode switching method, characterized in that, An imaging device with a dual-channel photosensitive device and an infrared fill light is used. The dual-channel photosensitive device has a natural light detection channel and an infrared light detection channel. The natural light detection channel is used to detect ambient natural light and output a natural light channel value. The infrared light detection channel is used to detect infrared light and output an infrared channel value. The device includes: When the night vision entry conditions are met, the first natural light channel value and the first infrared channel value detected by the dual-channel photosensitive device are obtained when the infrared fill light is off. Turn on the infrared fill light, and obtain the second natural light channel value and the second infrared channel value detected by the dual-channel photosensitive device after the infrared fill light is turned on; Calculate the crosstalk correction coefficient, which is equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value. The crosstalk correction coefficient is used to reflect the degree of crosstalk between infrared light and natural light detection. The crosstalk correction coefficient is used to correct the second natural light channel value to eliminate the infrared crosstalk component in the second natural light channel value, thereby obtaining the corrected natural light channel value. Based on the comparison result between the corrected natural light channel value or its converted value and the preset night vision exit threshold, control whether to turn off the infrared fill light and exit the night vision mode.

2. The night vision mode switching method according to claim 1, characterized in that, After the infrared fill light is turned on and a preset stabilization time is elapsed, the second natural light channel value and the second infrared channel value are obtained. And / or, the corrected natural light channel value is calculated according to the following formula: E = D - (C - A) / K; Wherein, A is the first infrared channel value, C is the second infrared channel value, D is the second natural light channel value, K is the crosstalk correction coefficient, and E is the corrected natural light channel value; And / or, the night vision entry condition includes the natural light channel value or its converted value detected in real time by the dual-channel light sensor being lower than a preset night vision entry threshold; and the night vision entry threshold is less than the night vision exit threshold.

3. The night vision mode switching method according to claim 1, characterized in that, The crosstalk correction coefficient is limited to a preset range. When the calculated crosstalk correction coefficient is less than a first threshold, the first threshold is used. When the calculated crosstalk correction coefficient is greater than a second threshold, the second threshold is used.

4. The night vision mode switching method according to claim 3, characterized in that, The first threshold is greater than or equal to 14 and less than or equal to 16, and the second threshold is greater than or equal to 28 and less than or equal to 30.

5. The night vision mode switching method according to claim 1, characterized in that, Each time night vision mode is entered, the following steps are executed: acquiring the first natural light channel value and the first infrared channel value, turning on the infrared fill light, acquiring the second natural light channel value and the second infrared channel value, calculating the crosstalk correction coefficient, obtaining the corrected natural light channel value using the crosstalk correction coefficient, and controlling whether to exit night vision mode based on the corrected natural light channel value.

6. The night vision mode switching method according to claim 1, characterized in that, Also includes: The dual-channel photosensitive device is calibrated in advance to establish a conversion relationship between the original reading value and the illuminance value output by the natural light detection channel of the dual-channel photosensitive device; In the step of comparing the corrected natural light channel value or its converted value with the preset night vision exit threshold, the corrected natural light channel value is converted into an illuminance value according to the conversion relationship, and compared with the preset night vision exit threshold in illuminance units.

7. A night vision mode switching system, characterized in that, Applications include camera equipment with dual-channel light sensors and infrared fill lights, including: The acquisition module is used to acquire the first natural light channel value and the first infrared channel value detected by the dual-channel photosensitive device when the night vision entry conditions are met; the acquisition module is also used to acquire the second natural light channel value and the second infrared channel value detected by the dual-channel photosensitive device after the infrared fill light is turned on. The control system is used to control the opening and closing of the infrared supplementary light; the control system is used to calculate a crosstalk correction coefficient based on the first natural light channel value, the first infrared channel value, the second natural light channel value, and the second infrared channel value, the crosstalk correction coefficient being equal to the difference between the second infrared channel value and the first infrared channel value divided by the difference between the second natural light channel value and the first natural light channel value, the crosstalk correction coefficient being used to reflect the degree of crosstalk between infrared light and natural light detection; The control system is further configured to use the crosstalk correction coefficient to correct the second natural light channel value in order to eliminate the infrared crosstalk component and obtain the corrected natural light channel value. The control system is also used to control whether to turn off the infrared fill light to exit the night vision mode based on the comparison result between the corrected natural light channel value or its converted value and the preset night vision exit threshold.

8. The night vision mode switching system according to claim 7, characterized in that, The acquisition module is used to acquire the second natural light channel value and the second infrared channel value after a preset stabilization time has elapsed after the infrared fill light is turned on; And / or, the control system is further configured to limit the crosstalk correction coefficient to a preset range, wherein when the calculated crosstalk correction coefficient is less than a first threshold, the first threshold is taken, and when the calculated crosstalk correction coefficient is greater than a second threshold, the second threshold is taken. And / or, the night vision mode switching system further includes a calibration module, which is used to pre-calibrate the dual-channel photosensitive device, establish a conversion relationship between the original reading value output by the natural light detection channel of the dual-channel photosensitive device and the illuminance value; the control system is used to convert the corrected natural light channel value into an illuminance value according to the conversion relationship, and compare it with a preset night vision exit threshold in illuminance units to determine whether to turn off the infrared fill light and exit the night vision mode.

9. The night vision mode switching system according to claim 7, characterized in that, The control system calculates the crosstalk correction coefficient each time it enters night vision mode.

10. A camera device, characterized in that, include: A dual-channel light sensor is used to detect ambient natural light and obtain the values ​​of the first natural light channel and the second natural light channel. It is also used to detect infrared light and obtain the values ​​of the first infrared channel and the second infrared channel; Infrared supplementary light; The night vision mode switching system as described in any one of claims 7 to 9, wherein the night vision mode switching system is electrically connected to the dual-channel light sensor and the infrared fill light, respectively.

Images