IPC image intelligent optimization method based on Tiong operating system

Abstract

The invention discloses an IPC image intelligent optimization method based on a Tiong operating system. The method comprises the following steps: acquiring an image data stream and corresponding equipment hardware parameters; according to each parameter item in the equipment hardware parameters, matching the image data stream with various image optimization branch conditions; when it is determined that the real-time image data stream is matched with the target type of image optimization branch condition, adjusting at least one optimization parameter of the image video stream by using a target optimization processing strategy matched with the target type of image optimization branch in combination with equipment hardware parameters; according to the method, the image video stream after matching or parameter optimization is coded to obtain the target data stream, so that the captured image can be directly optimized, third-party software is not needed, on the premise of ensuring the basic attributes of the image, image defects are restored to the maximum extent, the image quality is improved, hardware parameters are taken as optimization basis, and the optimization efficiency is improved. The dependence on an image feature extraction algorithm is reduced, and the processing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the fields of image communication and image optimization technology, and in particular to an IPC image intelligent optimization method based on the Tianhong operating system. Background Technology

[0002] In snapshot scenarios, IPC (Internet Protocol Camera) devices often suffer from problems such as blurry, dark, or overexposed images due to complex ambient lighting (e.g., dim lighting, backlighting) and focus deviation, making it difficult to meet the requirement of clear and identifiable images.

[0003] Currently, the optimization of image quality captured by IPCs generally adopts a post-processing approach: that is, the image is captured and stored first, and then third-party software is used to perform offline optimization processing on the saved image. In this approach, the optimization process is independent of the capture process and relies on the algorithm support of external software (such as post-processing noise reduction, brightness adjustment, and sharpness enhancement). By performing secondary processing on the captured image file, it attempts to improve the image quality to meet recognition requirements.

[0004] However, this optimization method relies on third-party software, which increases the complexity and additional cost of system deployment. It also requires dedicated computing resources for offline processing, making it impossible to achieve real-time linkage between capture and optimization, thus reducing processing efficiency. On the other hand, since optimization occurs after capture, existing defects in the original captured image (such as loss of detail due to overexposure or severe blur) may be difficult to completely repair through post-processing due to irreversible information, resulting in limited optimization effects, difficulty in ensuring the clarity and recognizability of the image, and reduced quality of the optimized image. Summary of the Invention

[0005] This invention provides an intelligent image optimization method for IPC based on the Tianhong operating system to solve the problems of low image optimization efficiency and poor quality.

[0006] According to one aspect of the present invention, an IPC image intelligent optimization method based on the Tianhong operating system is provided, wherein the Tianhong operating system is an IPC-based embedded system, and the method includes:

[0007] Acquire the image data stream acquired by the IPC, as well as the corresponding device hardware parameters;

[0008] Based on the parameters in the device hardware parameters, the image data stream is matched with various image optimization branch conditions; among them, the image optimization branch conditions include: image blur branch condition, image underexposure branch condition, and image overexposure branch condition.

[0009] When it is determined that the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted by using the target optimization processing strategy that matches the target species image optimization branch and in combination with the device hardware parameters.

[0010] The image and video streams that have been matched and / or optimized for parameters are encoded to obtain the target data stream.

[0011] According to another aspect of the present invention, an IPC image intelligent optimization device based on the Tianhong operating system is provided, wherein the Tianhong operating system is an IPC-based embedded system, and the device includes:

[0012] The parameter acquisition module is used to acquire the image data stream collected by the IPC, as well as the device hardware parameters corresponding to the image data stream;

[0013] The condition matching module is used to match the image data stream with various image optimization branch conditions according to the parameters in the device hardware parameters; among them, the image optimization branch conditions include: image blur branch condition, image underexposure branch condition, and image overexposure branch condition;

[0014] The optimization and adjustment module is used to adjust at least one optimization parameter of the image and video stream by using a target optimization processing strategy that matches the target image optimization branch and combining it with the device hardware parameters when it is determined that the real-time image data stream matches the target image optimization branch conditions.

[0015] The image encoding module is used to encode the image and video streams that have been matched and / or optimized for parameters to obtain the target data stream.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to execute the IPC image intelligent optimization method based on the Tianhong operating system according to any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the IPC image intelligent optimization method based on the Tianhong operating system as described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of the method as described in any embodiment of the present invention.

[0020] The technical solution of this invention acquires the image data stream and corresponding device hardware parameters from the IPC, and relies on matching the hardware parameters with various optimization branch conditions such as image blurring, underexposure, and overexposure to achieve targeted identification of image problems in different scenarios. For the matched target branch, the optimization parameters are dynamically adjusted in conjunction with the hardware parameters, enabling the captured image to be directly optimized without the need for third-party software. This maximizes the repair of image defects and improves image quality while preserving the basic image attributes. Simultaneously, using hardware parameters as the optimization basis reduces reliance on complex image feature extraction algorithms, lowers the computational load of the embedded system, and improves the real-time processing efficiency of the IPC. This solution is compatible with the unified optimization logic of single-image capture streams and continuous image / video streams. By encoding and outputting the target data stream, it can meet the high requirements of single-frame quality in capture scenarios while ensuring the smoothness and consistency of the video stream, adapting to diverse application scenarios. The modular design of the branch conditions and optimization strategies facilitates subsequent expansion to add new image problem scenarios; only the corresponding logic needs to be added to achieve functional upgrades, improving maintainability.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of an IPC image intelligent optimization method based on the Tianhong operating system according to Embodiment 1 of the present invention;

[0024] Figure 2 This is a statistical information graph of autofocus statistical parameters applicable to embodiments of the present invention;

[0025] Figure 3 This is a flowchart of another intelligent IPC image optimization method based on the Tianhong operating system provided in Embodiment 2 of the present invention;

[0026] Figure 4This is a data analysis area distribution map of global brightness statistical parameters applicable to an embodiment of the present invention when the brightness is too low;

[0027] Figure 5 This is a data analysis area distribution map of global brightness statistical parameters applicable to an embodiment of the present invention when the brightness is normal;

[0028] Figure 6 This is a flowchart of another intelligent IPC image optimization method based on the Tianhong operating system provided in Embodiment 3 of the present invention;

[0029] Figure 7 This is a schematic diagram of the structure of an IPC image intelligent optimization device based on the Tianhong operating system according to Embodiment 4 of the present invention;

[0030] Figure 8 This is a schematic diagram of the structure of an electronic device that implements the IPC image intelligent optimization method based on the Tianhong operating system according to an embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0033] Example 1

[0034] Figure 1This is a flowchart illustrating an intelligent image optimization method for IPCs based on the Tianhong operating system, provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where an IPC based on the Tianhong operating system directly optimizes images captured by the IPC. This method can be executed by an intelligent image optimization device based on the Tianhong operating system, which can be implemented in hardware and / or software and is generally configured in an electronic device. The Tianhong operating system is an embedded system based on IPCs.

[0035] In this embodiment of the invention, the Tianhong operating system can be specifically understood as: an operating system based on OpenHarmony (open source HarmonyOS), which can serve as an embedded system for IPCs. This system includes IPC drivers, which are used to connect the operating system and the IPC hardware, playing a crucial role in hardware parameter acquisition and optimization method execution. The embedded system can be specifically understood as: a dedicated computer system embedded within the IPC hardware device, characterized by small size, low power consumption, and high specificity, capable of directly interacting with the device hardware. The IPC can be specifically understood as: an Internet Protocol camera, possessing image acquisition, data processing, and network transmission functions, serving as the hardware carrier for realizing intelligent image optimization.

[0036] Correspondingly, such as Figure 1 As shown, the method includes:

[0037] S110: Obtain the image data stream acquired by the IPC, as well as the device hardware parameters corresponding to the image data stream.

[0038] In this embodiment of the invention, the image data stream can be specifically understood as: a continuous sequence of image data collected by IPC, which is the processing object for image optimization, and includes information such as the image's pixels, brightness, and color.

[0039] Specifically, device hardware parameters can be understood as hardware operating parameters related to image data stream acquisition, such as analog gain, digital gain, and line exposure time, which reflect the impact of hardware status and environment on the image.

[0040] Specifically, image acquisition is first performed via IPC. The input image is preprocessed (such as basic processing like cropping, scaling, or noise reduction) to generate an image data stream to be optimized. Then, the image data stream and the corresponding device hardware parameters are obtained through the IPC interface.

[0041] S120. Based on the parameters in the device hardware parameters, match the image data stream with various image optimization branch conditions.

[0042] The image optimization branch conditions include: image blurring branch condition, image underexposure branch condition, and image overexposure branch condition.

[0043] In this embodiment of the invention, the image optimization branch condition can be specifically understood as: a standard used to determine whether there are specific quality problems in the image data stream, including three specific conditions: image blurring, underexposure, and overexposure.

[0044] Specifically, based on the various parameters included in the device hardware parameters (such as the current frame parameter reflecting a blurred state, the autofocus distance parameter, or the lens aperture value; the analog gain, digital gain, or ISO parameter reflecting an underexposed state; and the line exposure time or aperture opening parameter reflecting an overexposed state), the image data stream is compared one by one with various preset image optimization branch conditions (i.e., image blur branch conditions, image underexposed branch conditions, and image overexposed branch conditions). By determining whether the hardware parameters meet the characteristics of each branch condition (e.g., if the current frame parameter is below a preset threshold or the autofocus distance parameter exceeds a preset range, the blur condition is met; if the analog gain, digital gain, or ISO parameter is above a preset threshold, the underexposed condition is met; if the line exposure time is below a preset threshold or the aperture opening parameter is above a preset threshold, the overexposed condition is met), the quality problem of the current image data stream is determined, providing a basis for subsequent targeted optimization.

[0045] S130. When it is determined that the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted using the target optimization processing strategy that matches the target species image optimization branch and in combination with the device hardware parameters.

[0046] In this embodiment of the invention, the real-time image data stream can be specifically understood as a continuous sequence of image data being acquired and processed in real time, which is the direct object of the optimization operation. The target image optimization branch condition can be specifically understood as the condition corresponding to the specific quality problem existing in the current image (such as image blurring, underexposure, or overexposure) determined through matching. The target optimization processing strategy can be specifically understood as an optimization scheme preset for specific branch conditions, including parameter adjustment rules and target range. The optimization parameters can be specifically understood as parameters that can be adjusted to improve image quality (such as brightness compensation values ​​or quantization parameters).

[0047] Specifically, when it is determined that the real-time image data stream meets at least one target image optimization branch condition (such as blurry, too dark, or overexposed), the target optimization processing strategy corresponding to the branch condition is adopted, and at the same time, in combination with the current device hardware parameters, at least one optimization parameter in the image video stream that affects the quality problem is adjusted.

[0048] For example, if the image is blurred, adjustments can be made to the focus distance and lens aperture. If the focus distance exceeds the preset range, the lens focus parameters can be adjusted (by moving the focus distance closer or further away). If the aperture value is too large, resulting in insufficient sharpness, the aperture opening parameter can be reduced to improve the overall image clarity. If the image is too dark, adjustments can be made to the ISO and noise reduction parameters. If the ISO parameter is below the preset threshold and the noise reduction parameter is too high, the ISO parameter can be appropriately increased (to enhance the sensor's light sensitivity) and the noise reduction parameter can be decreased (to reduce detail loss). At the same time, the white balance parameter can be adjusted (shifting towards warmer tones) to improve visual brightness. If the image is overexposed, adjustments can be made to the aperture opening and ISO parameters. If the aperture opening is higher than the preset threshold, the aperture can be reduced to decrease the amount of light entering the camera. If the ISO parameter is higher than the preset threshold, the ISO parameter can be decreased (to reduce the sensor's light sensitivity) to avoid loss of detail in bright light scenes, ultimately achieving targeted image quality optimization.

[0049] Optionally, based on the above embodiments, matching the image data stream with various image optimization branch conditions according to each parameter item in the device hardware parameters may include:

[0050] The image data stream is matched with the image blur branch conditions based on the current frame parameters in the device hardware parameters;

[0051] Based on the above embodiments, when it is determined that the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted using a target optimization processing strategy that matches the target species image optimization branch and in conjunction with device hardware parameters. This adjustment may include:

[0052] When it is determined that the current frame parameter is less than the preset standard frame parameter, and the difference between the standard frame parameter and the current frame parameter is greater than the preset reduction threshold, the image data stream is determined to match the image blur branch condition.

[0053] Based on the pre-built frame quality relationship table, construct the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter that match the current frame parameters;

[0054] Within the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter, the real-time frame rate quantization parameter and the real-time I-frame quantization parameter of the image data stream are adjusted until the image data stream meets the sharpness condition.

[0055] In this embodiment of the invention, the current frame parameter (cfg_bit) can be specifically understood as: the size data of the current image frame, reflecting the image compression and quality status, which can be obtained through the command `cat / proc / umap / rc`. Here, `cat` indicates viewing the contents of the ` / proc / umap / rc` file, where `proc` is the process file system used to display the real-time running status of the kernel, hardware, and processes, `umap` is a user-space memory-mapped file used to store hardware and driver status files, and `rc` is the bitrate control file used to store running information such as video encoding bitrate, frame rate (cfg_bit), and quantization parameters (QP). The standard frame parameter can be specifically understood as: the standard value of `cfg_bit` (e.g., 38kb (kilobyte)) recorded when the image is clear after device startup, serving as a benchmark for image quality.

[0056] The frame rate quantization parameter (qp) can be understood as a parameter that controls the degree of image compression. A smaller qp value results in less compression loss and a clearer image (but a larger frame rate); conversely, a larger qp value leads to a higher compression rate and a blurrier image. An I-frame can be understood as a key frame in video encoding (such as the first frame), an independent and complete frame (not dependent on other frames), carrying the basic information of the video image, and its image quality greatly affects subsequent frames. The I-frame quantization parameter (i_qp) can be understood as a quantization parameter set for I-frames, used to control the degree of compression of I-frames, directly affecting the clarity of the I-frame.

[0057] The frame quality relationship table can be understood as: recording the effective range of quantization parameters (such as the minimum value min_qp, the maximum value max_qp, the minimum value min_i_qp, and the maximum value max_i_qp) corresponding to different frame parameters cfg_bit, which serves as the basis for parameter adjustment.

[0058] Specifically, after the device starts up and the image is clear, a clear image can be manually selected as a reference, its current frame size (cfg_bit) can be obtained, and set as the standard frame parameter. When the IPC device receives an image data stream, the current frame parameter (cfg_bit) of the image data stream can be obtained and compared with the standard frame parameter.

[0059] If the current frame parameter is greater than the standard frame parameter, then the current frame parameter is reset to the standard frame parameter. It's understandable that a current frame parameter greater than the standard frame parameter might be due to non-image quality improvement factors such as increased image complexity (e.g., a sudden increase in the number of objects and details in the scene), and does not necessarily mean the image is sharper. By resetting the parameter to ensure that the initial sharp state of the standard frame parameter is always used as a reference, interference from non-image quality factors can be avoided, ensuring the accuracy of matching image blur branch conditions. This ensures that the optimization strategy is triggered only when the frame parameter deviates from the baseline due to a genuine decrease in image quality, improving the reliability and stability of the overall optimization logic.

[0060] If the current frame parameter is less than the standard frame parameter and the difference is less than or equal to the preset reduction threshold, the quantization parameter remains unchanged. If the current frame parameter is equal to the standard frame parameter, it indicates that the image data stream is in a clear state.

[0061] If the current frame parameter is less than the standard frame parameter and the difference is greater than the preset reduction threshold, the image data stream is determined to match the fuzzy branch condition. Based on the pre-constructed frame quality relationship table (as shown in Table 1, cfg_bit=75 corresponds to max_qp=51, etc.), the effective range of the frame rate quantization parameter (between min_qp and max_qp) and the effective range of the I-frame quantization parameter (between min_i_qp and max_i_qp) corresponding to the current frame parameter are determined. Within the corresponding effective range, the frame rate quantization parameter or the I-frame quantization parameter is adjusted in a single-parameter, alternating manner (adjusting only one of the frame rate quantization parameter or the I-frame quantization parameter each time, and adjusting both alternately in sequence), until the image sharpness meets the requirements.

[0062] The achievement of meeting the image sharpness requirements can be accomplished through dual-verification fine-tuning of quantization parameters and visual verification: First, the autofocus statistical parameters are obtained by calling the interface. Based on the autofocus statistical parameters, the image sharpness value is calculated using the peak method (the higher the value, the sharper the image). At the same time, it is observed whether the frame size (cfg_bit) is stable within a range close to the standard frame parameters (e.g., the fluctuation is less than the preset threshold range). Accordingly, within the preset range of the frame quality relationship table, only one parameter is fine-tuned at a time while the other parameters are fixed. The sharpness value is monitored in real time to see if it increases and reaches the peak value, and whether the frame size is stable, until the sharpness value reaches the highest level in the current scene and is in a stable state. The fine-tuned image is output, and the edges and textures of moving targets are manually checked for sharpness, and for the absence of ghosting or blurring, to ensure that the quantized sharpness is consistent with the visual sharpness. Finally, when the conditions of sharpness value reaching the peak value, frame size stabilization (the change in frame size within the preset adjustment step is less than the preset threshold) and visual sharpness are met, the image sharpness is determined to meet the requirements.

[0063] Understandably, within the empirical range defined by the frame quality relationship table (e.g., the qp value range corresponding to cfg_bit=75 is 43-51), adjustments can be made based on the magnitude of the reduction in the current frame parameters (frame size) compared to the standard frame parameters: if the reduction in frame size is greater than a preset threshold, qp is adjusted towards a higher value (e.g., towards 51) within that range (e.g., initially set to 50); if the reduction is smaller, it is adjusted towards a relatively lower value within the range (e.g., towards 43) (e.g., initially set to 45). This process is a range-level coarse adjustment based on the frame size change within the preset parameter range. The purpose is to initially improve the image quality degradation caused by the reduction in frame size by quickly matching empirical values ​​within the range. Correspondingly, if the reduction in frame size is greater than a preset threshold, the max_qp value can be appropriately increased to widen the adjustable range and improve the optimization quality.

[0064] Table 1

[0065]

[0066] Optionally, based on the above embodiments, while optimizing image quality through quantization parameter adjustment, noise reduction, color correction, and exposure control algorithms can be combined to further enhance the effect: For noise generated by excessive gain, isolated static bright and dark spots can be eliminated through Gaussian noise reduction, while temporal noise reduction can be used to compare continuous frames to preserve the edges of moving targets and eliminate random noise, thus avoiding blurring of moving targets; For graying or color cast issues that may occur after optimization, color temperature data can be obtained through the interface to determine the type of ambient light (such as a cool light environment), and color matrix parameters can be adjusted to correct the hue (such as adjusting the blue to a warm hue) and adjust the saturation to restore the true color; If there are cases where local overexposure and underexposure coexist in the image, two frames of short exposure (preserving light details) and long exposure (preserving dark details) images can be acquired and fused through HDR (High Dynamic Range) exposure control algorithms to achieve overall brightness and darkness balance in the image, ultimately achieving comprehensive optimization of image quality.

[0067] By matching the current frame parameters with the image blur branch conditions, optimization is triggered only when the current frame parameters are less than the standard frame parameters and the difference exceeds a preset threshold. This accurately identifies image blur issues and avoids resource waste or image quality fluctuations caused by indiscriminate adjustments. Simultaneously, relying on a pre-built frame quality relationship table, the effective range of frame rate quantization parameters and I-frame quantization parameters is determined. This ensures that parameter adjustments do not exceed reasonable ranges, avoid image quality anomalies, and quickly lock in the optimization direction based on empirical ranges, reducing trial-and-error costs. Then, by selectively adjusting optimization parameters, the system gradually approaches the optimal value until the image meets the sharpness requirements. This effectively improves blur issues in scenarios such as moving target capture, ensuring stable image sharpness. Furthermore, the entire optimization process uses hardware parameters as the judgment basis and adjustment benchmark, eliminating the need for complex image feature analysis. This simplifies the logic while improving response speed, adapting to the needs of real-time image acquisition and processing scenarios for IPCs.

[0068] Furthermore, based on the above embodiments, before constructing the effective range of the target default frame rate quantization parameters and the effective range of the target default I-frame quantization parameters that match the current frame parameters according to the pre-built frame quality relationship table, the following may also be included:

[0069] Acquire blurred image data streams under various alternative frame parameters acquired by IPC, and obtain the quality correlation parameters corresponding to each blurred image data stream;

[0070] Among them, the quality-related parameters include: the maximum value of the frame rate quantization parameter, the minimum value of the frame rate quantization parameter, the maximum value of the I-frame quantization parameter, and the minimum value of the I-frame quantization parameter;

[0071] For each candidate frame parameter, the optimal value of each quality-related parameter is obtained when the blurred image data stream reaches the peak image sharpness by adjusting the single parameter in turn; wherein, the peak image sharpness is calculated based on the autofocus statistics of the blurred image data stream.

[0072] A frame quality relationship table is constructed using the optimal values ​​of each candidate frame parameter and the corresponding quality association parameters.

[0073] In this embodiment of the invention, the candidate frame parameters can be specifically understood as: image frame size data with different degrees of blur (such as cfg_bit=75 and 85, etc.) adjusted manually, covering various image quality degradation scenarios. The quality-related parameters can be specifically understood as: a set of quantization parameters directly related to image sharpness, including min_qp (minimum value of frame rate quantization parameter), max_qp (maximum value of frame rate quantization parameter), min_i_qp (minimum value of I-frame quantization parameter), and max_i_qp (maximum value of I-frame quantization parameter). Single-parameter rotation adjustment can be specifically understood as: adjusting only one quality-related parameter at a time, fixing other parameters, and alternately optimizing to avoid mutual interference. The autofocus statistical parameter (be_af_stat) can be specifically understood as: parameters preset by the IPC hardware focusing mechanism, used to objectively statistically analyze the physical sharpness of the image (such as edge gradient and detail contrast). The peak image sharpness can be specifically understood as: the maximum sharpness calculated based on the be_af_stat parameter, at which point the image physical sharpness is optimal.

[0074] Specifically, by debugging, the IPC outputs image data streams with various degrees of blur, and the corresponding candidate frame parameters are obtained, as well as the quality-related parameters such as min_qp, max_qp, min_i_qp, and max_i_qp for each data stream.

[0075] Understandably, quality-related parameters have default value ranges, and these ranges are related to hardware configuration. For example, max_qp and max_i_qp are between [0, 51], min_qp is less than or equal to max_qp, and min_i_qp is less than or equal to max_i_qp. For the blurred data stream of each candidate frame parameter, a single-parameter rotation adjustment method is adopted (only one parameter is adjusted each time, while other parameters are fixed). The be_af_stat parameter (autofocus statistics parameter) is obtained through the platform interface, and the sharpness value is calculated. The adjustment is continued until the optimal value of each quality-related parameter is found at the sharpness peak. The optimal values ​​of min_qp, max_qp, min_i_qp, and max_i_qp in this state are recorded (e.g., cfg_bit=75 corresponds to the optimal values ​​43, 51, 43, and 51). The above process is repeated to obtain the optimal quality-related parameters corresponding to multiple candidate frame parameters. Finally, each candidate frame parameter is mapped one-to-one with its corresponding optimal quality association parameter, and a frame quality relationship table is constructed to provide a basis for subsequent matching of the current frame parameters and determining the effective range of quantization parameters.

[0076] Figure 2This is a statistical information chart of autofocus parameters applicable to embodiments of the present invention. The horizontal axis corresponds to the quantization parameter (or the number of iterations for parameter adjustment), and the vertical axis represents the image sharpness value calculated based on the `be_af_stat` parameter. When the parameter is in a low or high range, the sharpness value remains at a low level of around 2000, corresponding to a blurred image. As the parameter is adjusted, the sharpness value gradually increases, reaching a peak of around 7000 at approximately 800 on the horizontal axis (i.e., the marked sharp point), at which point the image's physical sharpness is optimal. If the parameter continues to be adjusted to a higher range, the sharpness value will quickly drop back, and the image will revert to a blurred state. Only when the parameter is within a specific range can the image reach its peak sharpness, providing an intuitive visual basis for subsequent image optimization through adjusting the quantization parameter.

[0077] By collecting blurred image data streams under various alternative frame parameters, covering different image quality degradation scenarios, and then using a single-parameter rotation adjustment method combined with the sharpness peak calculated from autofocus statistical parameters to determine the optimal quality correlation parameter for each frame parameter, the constructed frame quality relationship table is supported by a large amount of measured data, ensuring its accuracy and reliability. Furthermore, the optimal value not only conforms to hardware configuration constraints but also points to the optimal sharpness state, improving the adaptability of subsequent parameter adjustments. Simultaneously, the relationship table pre-stores the mapping relationship between different frame parameters and the optimal quality correlation parameter. During subsequent real-time optimization, there is no need for repeated complex multi-parameter debugging; simply matching the current frame parameter is sufficient to quickly lock the effective range of the quantization parameter, significantly reducing the computational cost and trial-and-error difficulty of real-time adjustments. In addition, the debugging and data accumulation of alternative frame parameters across multiple scenarios allows the frame quality relationship table to adapt to image optimization needs with different levels of blur, eliminating the need for re-tuning for each new scenario and enhancing the versatility and reusability of the entire optimization scheme.

[0078] S140. Encode the image and video streams that have completed matching and / or parameter optimization to obtain the target data stream.

[0079] In this embodiment of the invention, encoding can be specifically understood as: a process of data compression and format conversion of image and video streams, used to reduce data redundancy and adapt to transmission and storage requirements.

[0080] Specifically, the input image and video stream is first optimized by performing branch condition matching and / or parameter optimization. If the image quality meets the requirements and no optimization is needed, the original image and video stream is directly sent to the encoding stage. If the image has quality problems and needs to be optimized, all optimization operations such as branch condition matching and parameter adjustment are completed first, and then the optimized image and video stream is encoded.

[0081] After encoding, three target data streams can be generated: a high-definition main stream for video playback, a low-bandwidth sub-stream adapted for low-bandwidth scenarios and also used for video playback, and an image capture stream specifically for the snapshot function, to meet the usage needs of different scenarios.

[0082] The technical solution of this invention acquires the image data stream and corresponding device hardware parameters from the IPC, and relies on matching the hardware parameters with various optimization branch conditions such as image blurring, underexposure, and overexposure to achieve targeted identification of image problems in different scenarios. For the matched target branch, the optimization parameters are dynamically adjusted in conjunction with the hardware parameters, enabling the captured image to be directly optimized without the need for third-party software. This maximizes the repair of image defects and improves image quality while preserving the basic image attributes. Simultaneously, using hardware parameters as the optimization basis reduces reliance on complex image feature extraction algorithms, lowers the computational load of the embedded system, and improves the real-time processing efficiency of the IPC. This solution is compatible with the unified optimization logic of single-image capture streams and continuous image / video streams. By encoding and outputting the target data stream, it can meet the high requirements of single-frame quality in capture scenarios while ensuring the smoothness and consistency of the video stream, adapting to diverse application scenarios. The modular design of the branch conditions and optimization strategies facilitates subsequent expansion to add new image problem scenarios; only the corresponding logic needs to be added to achieve functional upgrades, improving maintainability.

[0083] Example 2

[0084] Figure 3 This is a flowchart of another intelligent image optimization method for IPC based on the Tianhong operating system provided in Embodiment 2 of the present invention. This embodiment is a refinement of the above embodiments' "matching the image data stream with various image optimization branch conditions according to each parameter item in the device hardware parameters" and "when it is determined that the real-time image data stream matches the target type of image optimization branch condition, using the target optimization processing strategy that matches the target type of image optimization branch and combining it with the device hardware parameters to adjust at least one optimization parameter of the image video stream." Figure 3 As shown, the method includes:

[0085] S310: Acquire the image data stream acquired by the IPC, as well as the device hardware parameters corresponding to the image data stream.

[0086] S320: Match the image data stream with the image over-dark branch condition based on the analog gain parameters and digital gain parameters in the device hardware parameters.

[0087] In this embodiment of the invention, the analog gain parameter (again) can be specifically understood as an analog parameter that enhances the light signal of the image sensor. The higher the value, the stronger the gain, reflecting the ambient brightness. The digital gain parameter (dgain) can be specifically understood as a parameter that amplifies the digital signal. The higher the value, the greater the amplification, used to assist in judging the ambient brightness. The analog and digital gain parameters can be obtained using the command `cat / proc / umap / isp`, where `isp` is a directory or file used to store the running status of the Image Signal Processor (ISP). The command can read the real-time parameters of the ISP module (such as gain data like `again` and `dgain`), which directly reflect information such as the current image brightness and processing status. The image too dark branch condition can be specifically understood as a preset standard for judging too dark based on the analog and digital gain parameters.

[0088] S330. When it is determined that the analog gain parameter is greater than the preset analog gain value and the digital gain parameter is greater than the preset digital gain value, the image data stream is determined to match the image over-dark branch condition.

[0089] S340. Based on the pre-built brightness compensation relationship table, construct an effective range of brightness compensation values ​​that matches the analog gain parameters and digital gain parameters.

[0090] In this embodiment of the invention, the brightness compensation value can be specifically understood as an exposure compensation parameter, with a preset value range of [0, 255], used to adjust the image brightness; the higher the value, the stronger the brightness. The brightness compensation relationship table can be specifically understood as a mapping table recording analog gain parameters, digital gain parameters, and corresponding brightness compensation value parameters, as shown in Table 2.

[0091] Table 2

[0092]

[0093] Furthermore, based on the above embodiments, before constructing the effective range of brightness compensation values ​​matching the analog gain parameters and digital gain parameters according to the pre-constructed brightness compensation relationship table, the following may also be included:

[0094] Acquire overly dark image data streams under various alternative analog gain parameters and alternative digital gain parameters acquired by IPC, and obtain the brightness compensation value corresponding to each overly dark image data stream.

[0095] For each alternative analog gain parameter and alternative digital gain parameter for an overly dark image data stream, the optimal value of the brightness compensation value is obtained when the overly dark image data stream reaches the normal brightness of the image by adjusting the brightness compensation value. The method for determining the normal brightness of the image is as follows: a data analysis area is generated based on the global brightness statistics parameters of the overly dark image data stream. When the data analysis area is within the preset normal brightness range, the image is determined to have reached the normal brightness.

[0096] A brightness compensation relationship table is constructed using each alternative analog gain parameter and alternative digital gain parameter, as well as the optimal value of the brightness compensation value corresponding to each alternative analog gain parameter and alternative digital gain parameter.

[0097] In this embodiment of the invention, the alternative analog gain parameter (again) can be specifically understood as: different analog gain values ​​set manually to cover scenes with various levels of darkness. The alternative digital gain parameter (dgain) can be specifically understood as: different digital gain values ​​set manually, combined with the alternative analog gain parameter, to simulate different dark environments. The excessively dark image data stream can be specifically understood as: image data with insufficient brightness acquired under the combination of alternative gain parameters. The global brightness statistics parameter (ae_hist1024_value) can be specifically understood as: statistical data reflecting the global brightness distribution of the image, and its clustered area can intuitively reflect the overall brightness level. The preset normal brightness range can be specifically understood as: the pre-set range of the ae_hist1024_value clustered area, and the image brightness is judged to be normal when it is within this range.

[0098] Specifically, the IPC is debugged to collect dark image data streams under different dim lighting conditions. Multiple alternative analog gain parameters (again) and alternative digital gain parameters (dgain), along with the brightness compensation values ​​(usually default values) for each data stream, are obtained. Global brightness statistics parameters (ae_hist1024_value) are acquired through the platform interface, and a data analysis region is generated. At this point, the clustered area of ​​ae_hist1024_value is within a preset dark range (e.g., 0-128). For each parameter combination of dark images, the brightness compensation value is adjusted, and the clustered area of ​​ae_hist1024_value is observed in real time. When the clustered area (the area containing the preset proportion data) enters the preset normal brightness range (e.g., 129-384), the image is considered to have reached normal brightness, and the brightness compensation value at this point is recorded as the optimal value. This process is repeated to obtain the correspondence between multiple parameter combinations and the optimal brightness compensation value.

[0099] Finally, each alternative analog gain parameter and alternative digital gain parameter is associated with its corresponding optimal brightness compensation value, and a brightness compensation relationship table is constructed to provide a basis for subsequent matching of real-time gain parameters and determination of the effective range of brightness compensation values.

[0100] Figure 4 This is a data analysis area distribution map of global brightness statistical parameters applicable to an embodiment of the present invention when the brightness is too low. Figure 5 This is a data analysis region distribution map of global brightness statistical parameters applicable to an embodiment of the present invention when the brightness is normal. With frequency as the vertical axis and pixel value as the horizontal axis, it presents the clustered region of global brightness statistical parameters (ae_hist1024_value). Figure 4 Concentrated within the 0-128 range on the horizontal axis, the frequency values ​​on the vertical axis fluctuate but the overall brightness distribution is dark, intuitively reflecting the global brightness characteristics of an overly dark image. Figure 5 The clustering area of ​​the brightness statistics parameters is mainly distributed in the horizontal axis range of 128-384, while the peak frequency of the vertical axis is more consistent with the distribution pattern of the preset normal brightness range. This shows the global brightness characteristics of the image after brightness adjustment to reach a normal state. The two are compared to provide a visual basis for judging whether the image brightness is normal.

[0101] By collecting overly dark image data streams under various alternative analog and digital gain parameter combinations, covering different dim scenes, and using the objective criterion of whether the data analysis area of ​​the global brightness statistics parameter is within the preset normal brightness range, the optimal brightness compensation value corresponding to each parameter combination is determined. This approach ensures that the constructed brightness compensation relationship table is supported by a large amount of measured data, guaranteeing the accuracy of subsequent brightness compensation value adjustments. Furthermore, by pre-storing the mapping relationship between different gain parameter combinations and the optimal brightness compensation value, subsequent real-time optimization does not require repeated complex parameter adjustments. It only needs to match the current gain parameter to quickly lock the effective range of brightness compensation value, significantly reducing the computational cost and trial-and-error time of real-time adjustments. At the same time, the optimal brightness compensation value is determined separately for each alternative gain parameter combination and relies on objective judgment criteria, which can accurately adapt to the brightness requirements of corresponding dim scenes, avoiding over- or under-adjustment issues. The coverage and data accumulation of multiple scene parameter combinations also allow the brightness compensation relationship table to adapt to different degrees of overly dark scenes, eliminating the need for complete re-adjustment for new scenes, expanding the applicability of the optimization solution, and improving implementation efficiency.

[0102] S350. Within the valid range of the brightness compensation value, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition.

[0103] In this embodiment of the invention, the brightness condition can be specifically understood as: judging the standard of normal image brightness by whether the global brightness statistics parameter (ae_hist1024_value) is within a preset range, combined with subjective observation.

[0104] Specifically, when the device receives an image data stream, it obtains the analog gain parameter (again) and digital gain parameter (dgain) through the device's hardware parameters. These are compared with the image over-dark branch condition (analog gain parameter greater than a preset analog gain value, and digital gain parameter greater than a preset digital gain value). If they match, it is determined that the current environment is dim, and the image data stream matches the over-dark branch condition. The brightness compensation value needs to be adjusted. Based on a pre-built brightness compensation relationship table, the effective range of brightness compensation values ​​corresponding to the current analog gain parameter (again) and digital gain parameter (dgain) is determined. Within this range, the brightness compensation value of the image data stream is adjusted, and the automatic exposure algorithm uses this value as a benchmark to calculate the brightness of the current dim scene. If the ambient brightness decreases, causing "again" and "dgain" to increase and the image brightness to be insufficient, the compensation value is further increased, and the brightness is recalculated until the image meets the brightness condition.

[0105] The criteria for determining whether an image meets the brightness requirements can be as follows: Determine the effective range based on the empirical compensation values ​​corresponding to the current `again` and `dgain` values ​​in the brightness compensation relationship table. This range can be expanded to ±2-4, centered on the corresponding empirical compensation value in the table (e.g., when `again=16384` and `dgain=1024` correspond to compensation=74, the effective range is set to 72-76). Fine-tuning is then performed within this range. During fine-tuning, it is necessary to observe in real-time whether the clustered area of ​​`ae_hist1024_value` (global brightness statistics parameter) is within the normal range, while also considering subjective judgment of whether the image brightness is appropriate. When the clustered area of ​​the brightness statistics parameter is normal and the image brightness is observed to be at a moderate level, the image can be determined to meet the brightness requirements.

[0106] Understandably, if the actual obtained "again" or "dgain" does not match the table value (e.g., "again=18000"), then refer to the compensation values ​​(74 and 84) corresponding to adjacent "again" values ​​(e.g., 16384 and 32786) in the table, and take the range of these two values ​​(74-84) as the valid range of the current parameter. Then, within this range, combine brightness statistics and subjective observation to fine-tune the specific value.

[0107] Optionally, based on the above embodiments, further optimization is achieved through digital gain adjustment, local area brightness adjustment, and intelligent fill light algorithms: Digital gain adjustment enables low-noise digital gain for dark areas of the target region, amplifying the signal in that area by a preset gain factor (the gain value is adjusted in real-time according to dgain to avoid exceeding the hardware noise threshold). Simultaneously, a bilateral filtering algorithm suppresses the amplified noise, preserving facial skin texture and eliminating color particles. Digital gain is not enabled for potentially overexposed areas such as window edges with faint light, preventing further overexposure. For local area brightness adjustment, a threshold segmentation algorithm first divides the image into dark and normal areas based on brightness values, then performs targeted adjustments, increasing the slope of the brightness curve for dark areas separately, while keeping the brightness of normal areas unchanged. When the IPC device is equipped with a fill light, the intelligent fill light algorithm detects that the target area (such as a person's face) is continuously in a dark area (e.g., the continuous frame rate is greater than the preset frame rate) and the ambient light brightness does not show an increasing trend (the change in ambient light brightness is less than the preset threshold). At the same time, the ISP adjusts the exposure time of the fill light area in real time to concentrate the fill light on the target area. If the brightness of the target area meets the standard after the fill light is applied, the fill light is turned off. Otherwise, the fill light power is gradually increased to achieve fine brightening of the dark area and ultimately optimize the dim image into a normal brightness image.

[0108] S360: Encode the image and video streams that have completed matching and / or parameter optimization to obtain the target data stream.

[0109] The technical solution of this invention acquires the image data stream collected by the IPC and the corresponding device hardware parameters. Based on the analog gain and digital gain parameters in the device hardware parameters, the image data stream is matched with the image over-dark branch condition to achieve targeted identification of over-dark image scenes. When it is determined that the analog gain parameter is greater than a preset analog gain value and the digital gain parameter is greater than a preset digital gain value, it is determined that the image data stream matches the image over-dark branch condition. According to a pre-constructed brightness compensation relationship table, a valid range of brightness compensation values ​​matching the analog gain and digital gain parameters is constructed. Within the valid range of brightness compensation values, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition. Based on the pre-constructed brightness compensation relationship table, the brightness compensation value matching the current... The effective range of brightness compensation values, matched with analog and digital gains, eliminates the need for blind adjustments, significantly shortening parameter adjustment cycles and improving optimization efficiency. Adjusting brightness compensation values ​​within the effective range can target underexposure issues with targeted brightness enhancement while avoiding overexposure risks due to excessive compensation through range constraints, ensuring that the final image brightness remains stable within the required range and enhancing the reliability of the optimization effect. Furthermore, optimization is based on the device's own hardware parameters, combined with the embedded system interface for parameter adjustment, eliminating the need for external computing resources. This better suits the hardware characteristics of IPC devices, reducing system load and ensuring device operational stability. It allows captured images to be directly optimized without the need for third-party software, maximizing the repair of image defects and improving image quality while preserving basic image attributes.

[0110] Example 3

[0111] Figure 6 This is a flowchart of another intelligent image optimization method for IPC based on the Tianhong operating system provided in Embodiment 3 of the present invention. This embodiment is a refinement of the above embodiments' "matching the image data stream with various image optimization branch conditions according to each parameter item in the device hardware parameters" and "when it is determined that the real-time image data stream matches the target type of image optimization branch condition, using the target optimization processing strategy that matches the target type of image optimization branch and combining it with the device hardware parameters to adjust at least one optimization parameter of the image video stream." Figure 6 As shown, the method includes:

[0112] S610: Acquire the image data stream acquired by the IPC, as well as the device hardware parameters corresponding to the image data stream.

[0113] S620: Match the image data stream with the image overexposure branch condition based on the line exposure time in the device hardware parameters.

[0114] In this embodiment of the invention, the line exposure time (line value) can be specifically understood as: the exposure time of each row of pixels of the image sensor, which can be obtained by the command `cat / proc / umap / isp`. The smaller the value, the shorter the exposure time. The image overexposure branch condition can be specifically understood as: an overexposure judgment criterion preset based on the line exposure time (line value).

[0115] S630. When it is determined that the line exposure time is less than the preset exposure time value, the image data stream is matched with the image overexposure branch condition.

[0116] S640. Based on the pre-built exposure compensation relationship table, construct an effective range of brightness compensation values ​​that matches the line exposure time.

[0117] In this embodiment of the invention, the exposure compensation relationship table can be specifically understood as a mapping table between the exposure time (line value) and the corresponding brightness compensation value (compensation), as shown in Table 3. The smaller the brightness compensation value, the lower the image brightness, which is used to suppress overexposure.

[0118] Table 3

[0119]

[0120] Furthermore, based on the above embodiments, before constructing the effective range of brightness compensation values ​​matching the line exposure time according to the pre-constructed exposure compensation relationship table, the following may also be included:

[0121] Acquire overexposed image data streams from multiple alternative line exposure times captured by IPC, and obtain the brightness compensation value corresponding to each overexposed image data stream.

[0122] For each alternative exposure time, the optimal value of the brightness compensation is obtained when the overexposed image data stream reaches the normal brightness of the image by adjusting the brightness compensation value.

[0123] An exposure compensation relationship table is constructed using the optimal values ​​of each alternative row exposure time and the corresponding brightness compensation values.

[0124] In this embodiment of the invention, the alternative line exposure time (line value) can be specifically understood as: different line exposure durations set manually to simulate various overexposure scenarios. The overexposed image data stream can be specifically understood as: image data with excessive brightness acquired under the alternative line exposure time, whose global brightness statistics parameter (ae_hist1024_value) is clustered in the high brightness range (e.g., 896-1024).

[0125] Specifically, overexposed image data streams are acquired via IPC under different overbright environments to obtain corresponding alternative line exposure times (line values). Simultaneously, global brightness statistics parameters (ae_hist1024_value) are obtained through the platform interface. For each alternative line exposure time with an overexposed image (where ae_hist1024_value clusters in the overbright region), the brightness compensation value is adjusted. The data analysis area generated by the global brightness statistics parameter is observed in real time. When the clustered area enters the preset normal brightness range, the brightness compensation value at this point is recorded as the optimal brightness compensation value to achieve normal image brightness. This process is repeated to obtain multiple sets of parameter correspondences. Finally, each alternative line exposure time is associated and integrated with its corresponding optimal brightness compensation value to construct an exposure compensation relationship table, providing a basis for subsequent matching of real-time line exposure times and determining the effective range of brightness compensation values.

[0126] By collecting overexposed image data streams under various alternative line exposure times, covering different degrees of overexposed scenarios, and using the data analysis area of ​​global brightness statistics parameters as the criterion to determine whether it is within the preset normal range, the optimal brightness compensation value for each scenario is determined, and an exposure compensation relationship table is constructed. This ensures that subsequent brightness compensation value adjustments are accurately matched with the degree of overexposedness. When processing overexposed images in real time, there is no need to repeat the entire process of parameter debugging; only the current line exposure time needs to be matched to quickly lock the effective range of brightness compensation values, significantly reducing the computational cost and trial-and-error time of real-time optimization and improving processing efficiency. At the same time, the optimal brightness compensation value is determined separately for each alternative line exposure time (corresponding to different overexposed scenarios), which can accurately adapt to the overexposed suppression requirements of the corresponding scenario and avoid insufficient or excessive compensation. The accumulation of data from multiple scenarios also allows the exposure compensation relationship table to adapt to overexposed environments with different brightness levels. When encountering new scenarios, there is no need to carry out complete debugging again; only the compensation values ​​corresponding to similar parameters in the table need to be referenced for quick adaptation, improving the reusability and applicability of the solution.

[0127] S650. Within the valid range of the brightness compensation value, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition.

[0128] Specifically, the line exposure time (line value) from the device hardware parameters is obtained and compared with the image overexposure branch condition (line exposure time less than the preset exposure time value). If they match, it is determined that the current environment is backlit, and the image data stream matches the overexposure branch condition. Based on a pre-built exposure compensation relationship table, the effective range of brightness compensation values ​​corresponding to the current line exposure time is determined. Within this range, the image brightness is reduced by decreasing the brightness compensation value of the image data stream (reducing the brightness compensation value will simultaneously shorten the exposure time). At the same time, the clustering area of ​​the global brightness statistics parameter (ae_hist1024_value), whether it is within the preset range, and whether the subjective observation of the image brightness is appropriate are considered to determine whether the brightness condition is met. Within the effective range, the brightness compensation value is adjusted until the image meets the brightness condition, and the overexposed image is adjusted to a normal brightness image.

[0129] Optionally, based on the above embodiments, further optimization can be achieved through automatic exposure control, local contrast enhancement, or dynamic backlight adjustment algorithms: Automatic exposure control first divides the image into highlight and shadow areas. A sensor collects the current ambient brightness distribution in real time and compares it with a preset target brightness (129-384 range). Exposure time (e.g., extending the line) and gain parameters (e.g., reducing again and dgain) are dynamically adjusted. Brightness distribution is detected for each frame to ensure the overall brightness approaches the target range, avoiding overall imbalance. Local contrast enhancement can employ the Retinex (retina-cortex) algorithm to separate the image's brightness layers and details. The system uses a threshold segmentation method to mark highlight and shadow areas. For shadow areas, the brightness curve slope is increased and contrast enhanced to emphasize facial features. For highlight areas, the brightness curve slope is reduced to restore scene details. Simultaneously, the boundary between light and shadow is smoothed to avoid brightness banding. Through dynamic backlight adjustment, edge detection and region growing algorithms are used to identify target backlight areas (such as regularly shaped, brightly lit windows with brightness significantly higher than the surrounding environment). Low exposure parameters (line exposure time and brightness compensation value) are set individually for these target backlight areas, while maintaining exposure parameters for other areas to avoid excessive brightness suppression. If other areas change (such as character movement), the individual settings for the target backlight area are automatically canceled. Ultimately, this eliminates overexposure in the target backlight areas while ensuring moderate brightness in other areas, resulting in a clear image with natural overall brightness transitions.

[0130] In a specific example, an Internet Protocol (IPC) image intelligent optimization system for cameras based on the Tianhong operating system includes an image input module, a video processing subsystem, an IPC image intelligent optimization module, and an encoding output module.

[0131] The image input module is used to input image data streams from the device, which are then transmitted to the video processing subsystem. After processing by the video processing subsystem, the image data stream is input to the IPC image intelligent optimization module. The IPC image intelligent optimization module has autonomous judgment and scene-based optimization functions. It can first determine whether the image needs optimization. If no optimization is needed, the image data is directly input to the encoding output module. If optimization is needed, it identifies the current environmental scene (blurred, too dark, or overexposed) based on a trained model, performs targeted image optimization, and then inputs the image to the encoding output module. Finally, the encoding output module generates three data streams. The main stream and sub-stream are used for video display and playback, and the image capture stream is used as image data for the capture function. The modules work together to realize a complete process from image input, intelligent optimization to multi-purpose data stream output.

[0132] S660: Encode the image / video stream that has been matched and / or optimized for parameters to obtain the target data stream.

[0133] The technical solution of this invention acquires the image data stream collected by the IPC and the corresponding device hardware parameters. Based on the line exposure time in the device hardware parameters, the image data stream is matched with image overexposure branch conditions to achieve targeted identification of image problems in overexposure scenes. When it is determined that the line exposure time is less than a preset exposure time value, it is determined that the image data stream matches the image overexposure branch condition. According to a pre-built exposure compensation relationship table, a valid range of brightness compensation values ​​matching the line exposure time is constructed. Within the valid range of brightness compensation values, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition. Based on the pre-built exposure compensation relationship table, the valid range of brightness compensation values ​​matching the current line exposure time can be directly located, reducing blind parameter adjustment. The device's targeting capability ensures that brightness compensation adjustments are always focused on resolving overexposure issues, significantly improving optimization efficiency. Adjusting the brightness compensation value within the effective range not only reduces exposure by lowering the compensation value to suppress overexposure but also avoids abnormal brightness caused by insufficient or excessive compensation through range constraints, ensuring stable image brightness within a reasonable range and guaranteeing the reliability of the optimization effect. Simultaneously, optimization is driven by the device's own exposure time parameters, combined with embedded system interfaces for parameter adjustment. This eliminates the need for external computing resources, aligns with IPC hardware characteristics, reduces system load, and ensures stable and efficient operation of the device in overexposure scenarios. It allows captured images to be directly optimized without third-party software, maximizing the repair of image defects and improving image quality while preserving basic image attributes.

[0134] Example 4

[0135] Figure 7 This is a schematic diagram of the structure of an IPC image intelligent optimization device based on the Tianhong operating system provided in Embodiment 4 of the present invention. Figure 7As shown, the device includes: a parameter acquisition module 710, a condition matching module 720, an optimization and adjustment module 730, and an image encoding module 740, wherein:

[0136] The parameter acquisition module 710 is used to acquire the image data stream collected by the IPC, as well as the device hardware parameters corresponding to the image data stream;

[0137] The condition matching module 720 is used to match the image data stream with various image optimization branch conditions according to the parameters in the device hardware parameters; among which, the image optimization branch conditions include: image blur branch condition, image underexposure branch condition, and image overexposure branch condition;

[0138] The optimization and adjustment module 730 is used to adjust at least one optimization parameter of the image video stream by using a target optimization processing strategy that matches the target image optimization branch and combining it with the device hardware parameters when it is determined that the real-time image data stream matches the target image optimization branch conditions.

[0139] Image encoding module 740 is used to encode the image video stream that has been matched and / or parameter optimized to obtain the target data stream.

[0140] The technical solution of this invention acquires the image data stream and corresponding device hardware parameters from the IPC, and relies on matching the hardware parameters with various optimization branch conditions such as image blurring, underexposure, and overexposure to achieve targeted identification of image problems in different scenarios. For the matched target branch, the optimization parameters are dynamically adjusted in conjunction with the hardware parameters, enabling the captured image to be directly optimized without the need for third-party software. This maximizes the repair of image defects and improves image quality while preserving the basic image attributes. Simultaneously, using hardware parameters as the optimization basis reduces reliance on complex image feature extraction algorithms, lowers the computational load of the embedded system, and improves the real-time processing efficiency of the IPC. This solution is compatible with the unified optimization logic of single-image capture streams and continuous image / video streams. By encoding and outputting the target data stream, it can meet the high requirements of single-frame quality in capture scenarios while ensuring the smoothness and consistency of the video stream, adapting to diverse application scenarios. The modular design of the branch conditions and optimization strategies facilitates subsequent expansion to add new image problem scenarios; only the corresponding logic needs to be added to achieve functional upgrades, improving maintainability.

[0141] Based on the above embodiments, the condition matching module 720 is specifically used for:

[0142] The image data stream is matched with the image blur branch conditions based on the current frame parameters in the device hardware parameters;

[0143] Based on the above embodiments, the optimized and adjusted module 730 is specifically used for:

[0144] When it is determined that the current frame parameter is less than the preset standard frame parameter, and the difference between the standard frame parameter and the current frame parameter is greater than the preset reduction threshold, the image data stream is determined to match the image blur branch condition.

[0145] Based on the pre-built frame quality relationship table, construct the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter that match the current frame parameters;

[0146] Within the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter, the real-time frame rate quantization parameter and the real-time I-frame quantization parameter of the image data stream are adjusted until the image data stream meets the sharpness condition.

[0147] Optionally, based on the above embodiments, the optimization and adjustment module 730 may include: a fuzzy acquisition unit, a fuzzy optimization unit, and a frame quality table unit, wherein:

[0148] The fuzzy acquisition unit is used to acquire fuzzy image data streams under various alternative frame parameters acquired by IPC before constructing the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter that match the current frame parameters according to the pre-built frame quality relationship table, and to acquire the quality association parameters corresponding to each fuzzy image data stream respectively; wherein, the quality association parameters include: the maximum value of the frame rate quantization parameter, the minimum value of the frame rate quantization parameter, the maximum value of the I-frame quantization parameter, and the minimum value of the I-frame quantization parameter;

[0149] The fuzzy optimization unit is used to obtain the optimal value of each quality-related parameter when the fuzzy image data stream reaches the peak image sharpness by adjusting the single parameter in turn for each candidate frame parameter of the fuzzy image data stream; wherein, the peak image sharpness is calculated based on the autofocus statistical parameters of the fuzzy image data stream.

[0150] The frame quality table unit is used to construct a frame quality relationship table using each candidate frame parameter and the optimal values ​​of the quality association parameters corresponding to each candidate frame parameter.

[0151] Based on the above embodiments, the condition matching module 720 is further configured to:

[0152] Based on the analog gain parameters and digital gain parameters in the device hardware parameters, the image data stream is matched with the image over-dark branch condition;

[0153] Based on the above embodiments, the optimized and adjusted module 730 is further used for:

[0154] When it is determined that the analog gain parameter is greater than the preset analog gain value and the digital gain parameter is greater than the preset digital gain value, the image data stream is determined to match the image over-dark branch condition.

[0155] Based on the pre-built brightness compensation relationship table, construct an effective range of brightness compensation values ​​that matches the analog gain parameters and digital gain parameters;

[0156] Within the valid range of the brightness compensation value, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition.

[0157] Optionally, based on the above embodiments, the optimization and adjustment module 730 may include: an over-darkness acquisition unit, an over-darkness optimization unit, and a brightness compensation unit, wherein:

[0158] The dark image acquisition unit is used to acquire dark image data streams under various alternative analog gain parameters and alternative digital gain parameters acquired by IPC before constructing an effective range of brightness compensation values ​​that match the analog gain parameters and digital gain parameters according to a pre-built brightness compensation relationship table, and to acquire the brightness compensation value corresponding to each dark image data stream.

[0159] The over-darkness optimization unit is used to adjust the brightness compensation value for each candidate analog gain parameter and candidate digital gain parameter for over-dark image data streams to obtain the optimal value of the brightness compensation value when the over-dark image data stream reaches the normal brightness of the image. The method for determining the normal brightness of the image is as follows: a data analysis area is generated based on the global brightness statistics parameters of the over-dark image data stream. When the data analysis area is within the preset normal brightness range, the image is determined to have reached the normal brightness.

[0160] The brightness compensation unit is used to construct a brightness compensation relationship table using each alternative analog gain parameter and alternative digital gain parameter, as well as the optimal value of the brightness compensation value corresponding to each alternative analog gain parameter and alternative digital gain parameter.

[0161] Based on the above embodiments, the condition matching module 720 is further configured to:

[0162] Based on the line exposure time in the device hardware parameters, the image data stream is matched with the image overexposure branch condition;

[0163] Based on the above embodiments, the optimized and adjusted module 730 is further used for:

[0164] When the line exposure time is determined to be less than the preset exposure time value, the image data stream is determined to match the image overexposure branch condition;

[0165] Based on the pre-built exposure compensation relationship table, construct an effective range of brightness compensation values ​​that matches the row exposure time;

[0166] Within the valid range of the brightness compensation value, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition.

[0167] Optionally, based on the above embodiments, the optimization and adjustment module 730 may include: an overexposure acquisition unit, an overexposure optimization unit, and an exposure compensation unit, wherein:

[0168] The overexposure acquisition unit is used to acquire overexposure image data streams under various alternative line exposure times collected by IPC before constructing an effective range of brightness compensation values ​​matching the line exposure time according to a pre-built exposure compensation relationship table, and to acquire the brightness compensation value corresponding to each overexposure image data stream.

[0169] The overexposure optimization unit is used to obtain the optimal value of the brightness compensation value when the overexposed image data stream reaches the normal brightness of the image by adjusting the brightness compensation value for each alternative line exposure time.

[0170] The exposure compensation unit is used to construct an exposure compensation relationship table by using the optimal values ​​of the exposure time for each alternative row and the corresponding brightness compensation values ​​for each alternative row exposure time.

[0171] The IPC image intelligent optimization device based on the Tianhong operating system provided in this embodiment of the invention can execute the IPC image intelligent optimization method based on the Tianhong operating system provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0172] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0173] Example 5

[0174] Figure 8 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0175] like Figure 8 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0176] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0177] Processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the IPC image intelligent optimization method based on the Tianhong operating system, namely:

[0178] Acquire the image data stream acquired by the IPC, as well as the corresponding device hardware parameters;

[0179] Based on the parameters in the device hardware parameters, the image data stream is matched with various image optimization branch conditions; among them, the image optimization branch conditions include: image blur branch condition, image underexposure branch condition, and image overexposure branch condition.

[0180] When it is determined that the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted by using the target optimization processing strategy that matches the target species image optimization branch and in combination with the device hardware parameters.

[0181] The image and video streams that have been matched and / or optimized for parameters are encoded to obtain the target data stream.

[0182] In some embodiments, the IPC image intelligent optimization method based on the Tianhong operating system can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the IPC image intelligent optimization method based on the Tianhong operating system described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the IPC image intelligent optimization method based on the Tianhong operating system by any other suitable means (e.g., by means of firmware).

[0183] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0184] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0185] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0186] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0187] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0188] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0189] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0190] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for intelligent optimization of IPC images based on the Tianhong operating system, characterized in that, The Tianhong operating system is an IPC-based embedded system, and the method includes: Acquire the image data stream acquired by the IPC, as well as the corresponding device hardware parameters; Based on the parameters in the device hardware parameters, the image data stream is matched with various image optimization branch conditions; among them, the image optimization branch conditions include: image blur branch condition, image underexposure branch condition, and image overexposure branch condition. When it is determined that the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted by using the target optimization processing strategy that matches the target species image optimization branch and in combination with the device hardware parameters. The image and video streams that have been matched and / or optimized for parameters are encoded to obtain the target data stream.

2. The method according to claim 1, characterized in that, Based on the parameters in the device hardware specifications, the image data stream is matched with various image optimization branch conditions, including: The image data stream is matched with the image blur branch conditions based on the current frame parameters in the device hardware parameters; When determining whether the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted using a target optimization processing strategy that matches the target species image optimization branch and in conjunction with device hardware parameters, including: When it is determined that the current frame parameter is less than the preset standard frame parameter, and the difference between the standard frame parameter and the current frame parameter is greater than the preset reduction threshold, the image data stream is determined to match the image blur branch condition. Based on the pre-built frame quality relationship table, construct the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter that match the current frame parameters; Within the effective range of the target default frame rate quantization parameter and the effective range of the target default I-frame quantization parameter, the real-time frame rate quantization parameter and the real-time I-frame quantization parameter of the image data stream are adjusted until the image data stream meets the sharpness condition.

3. The method according to claim 2, characterized in that, Before constructing the effective range of the target default frame rate quantization parameters and the effective range of the target default I-frame quantization parameters that match the current frame parameters based on the pre-built frame quality relationship table, the following steps are also included: Acquire blurred image data streams under various alternative frame parameters acquired by IPC, and obtain the quality correlation parameters corresponding to each blurred image data stream; Among them, the quality-related parameters include: the maximum value of the frame rate quantization parameter, the minimum value of the frame rate quantization parameter, the maximum value of the I-frame quantization parameter, and the minimum value of the I-frame quantization parameter; For each candidate frame parameter, the optimal value of each quality-related parameter is obtained when the blurred image data stream reaches the peak image sharpness by adjusting the single parameter in turn; wherein, the peak image sharpness is calculated based on the autofocus statistics of the blurred image data stream. A frame quality relationship table is constructed using the optimal values ​​of each candidate frame parameter and the corresponding quality association parameters.

4. The method according to claim 1, characterized in that, Based on the parameters in the device hardware specifications, the image data stream is matched with various image optimization branch conditions, including: Based on the analog gain parameters and digital gain parameters in the device hardware parameters, the image data stream is matched with the image over-dark branch condition; When determining whether the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted using a target optimization processing strategy that matches the target species image optimization branch and in conjunction with device hardware parameters, including: When it is determined that the analog gain parameter is greater than the preset analog gain value and the digital gain parameter is greater than the preset digital gain value, the image data stream is determined to match the image over-dark branch condition. Based on the pre-built brightness compensation relationship table, construct an effective range of brightness compensation values ​​that matches the analog gain parameters and digital gain parameters; Within the valid range of the brightness compensation value, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition.

5. The method according to claim 4, characterized in that, Before constructing the effective range of brightness compensation values ​​matching the analog and digital gain parameters based on the pre-built brightness compensation relationship table, the following steps are also included: Acquire overly dark image data streams under various alternative analog gain parameters and alternative digital gain parameters acquired by IPC, and obtain the brightness compensation value corresponding to each overly dark image data stream. For each alternative analog gain parameter and alternative digital gain parameter for an overly dark image data stream, the optimal value of the brightness compensation value is obtained when the overly dark image data stream reaches the normal brightness of the image by adjusting the brightness compensation value. The method for determining the normal brightness of the image is as follows: a data analysis area is generated based on the global brightness statistics parameters of the overly dark image data stream. When the data analysis area is within the preset normal brightness range, the image is determined to have reached the normal brightness. A brightness compensation relationship table is constructed using each alternative analog gain parameter and alternative digital gain parameter, as well as the optimal value of the brightness compensation value corresponding to each alternative analog gain parameter and alternative digital gain parameter.

6. The method according to claim 1, characterized in that, Based on the parameters in the device hardware specifications, the image data stream is matched with various image optimization branch conditions, including: Based on the line exposure time in the device hardware parameters, the image data stream is matched with the image overexposure branch condition; When determining whether the real-time image data stream matches the target species image optimization branch conditions, at least one optimization parameter of the image video stream is adjusted using a target optimization processing strategy that matches the target species image optimization branch and in conjunction with device hardware parameters, including: When the line exposure time is determined to be less than the preset exposure time value, the image data stream is determined to match the image overexposure branch condition; Based on the pre-built exposure compensation relationship table, construct an effective range of brightness compensation values ​​that matches the row exposure time; Within the valid range of the brightness compensation value, the brightness compensation value of the image data stream is adjusted until the image data stream meets the brightness condition.

7. The method according to claim 6, characterized in that, Before constructing the effective range of brightness compensation values ​​matching the line exposure time based on the pre-built exposure compensation relationship table, the following steps are also included: Acquire overexposed image data streams from multiple alternative line exposure times captured by IPC, and obtain the brightness compensation value corresponding to each overexposed image data stream. For each alternative exposure time, the optimal value of the brightness compensation is obtained when the overexposed image data stream reaches the normal brightness of the image by adjusting the brightness compensation value. An exposure compensation relationship table is constructed using the optimal values ​​of each alternative row exposure time and the corresponding brightness compensation values.

8. An IPC image intelligent optimization device based on the Tianhong operating system, characterized in that, The Tianhong operating system is an IPC-based embedded system, and the device includes: The parameter acquisition module is used to acquire the image data stream collected by the IPC, as well as the device hardware parameters corresponding to the image data stream; The condition matching module is used to match the image data stream with various image optimization branch conditions according to the parameters in the device hardware parameters; among them, the image optimization branch conditions include: image blur branch condition, image underexposure branch condition, and image overexposure branch condition; The optimization and adjustment module is used to adjust at least one optimization parameter of the image and video stream by using a target optimization processing strategy that matches the target image optimization branch and combining it with the device hardware parameters when it is determined that the real-time image data stream matches the target image optimization branch conditions. The image encoding module is used to encode the image and video streams that have been matched and / or optimized for parameters to obtain the target data stream.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the IPC image intelligent optimization method based on the Tianhong operating system as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the IPC image intelligent optimization method based on the Tianhong operating system as described in any one of claims 1-7.

Images