Combination correction algorithm of static prescription lenses and dynamic add power adjustable focus module

Abstract

The application discloses a combination correction algorithm of a static prescription lens and a dynamic Add focus-adjustable module, and particularly relates to the technical field of prescription glasses optical correction and focus-adjustable optical system based on liquid crystal modulation and driving IC control, and comprises obtaining a prescription, wearing, assembling, liquid crystal and driving, wherein the prescription comprises spherical, cylindrical and axis position, the wearing comprises pupil distance and vertex distance, the assembling comprises optical center and module center, and forms a parameter set. By unifying the prescription, wearing and assembling information to the same reference surface, and combining the target wavefront, offset and readback judgment, the responsibility boundary of the static item and the dynamic item is dynamically reconstructed, so that the prescription split mismatch caused by the real wearing state change can be relatively relieved, and the target synthesized wavefront can be kept in high consistency for a long time.

Description

Technical Field

[0001] This invention relates to the field of prescription eyeglasses optical correction and adjustable focus optical systems based on liquid crystal modulation and driver IC control. More specifically, this invention relates to a combined correction algorithm of static prescription lenses and dynamic Add adjustable focus modules. Background Technology

[0002] In the field of prescription eyeglasses and adjustable focus optical correction, the mainstream practice in the industry is to provide users who have both basic refractive errors and near vision correction needs with a dynamic Add module controlled by a high-performance hybrid liquid crystal and a driver IC, on top of the static prescription lens which provides basic correction of the full aperture, including spherical, cylindrical and axis. The effect of near vision switching and wearing comfort are improved through factory calibration, temperature compensation, drive waveform optimization or mode switching smoothing. For example, in ultra-thin, low-power, modular adjustable focus glasses, the dynamic Add module is usually limited to a small-diameter area near the pupil and must meet hard constraints such as limited driving voltage, limited response time, limited power budget, and unavoidable fluctuations in interpupillary distance, vertex distance, tilt angle, wrap angle, and assembly deviation under actual wearing conditions. Under this constraint, although existing solutions can obtain a nominal synthetic wavefront in the design state or single calibration state, after continuous wear by the user, changes in posture, device temperature drift and aging, or micro-shifts in binocular assembly, observable phenomena such as shift of the clear area after switching between near and far vision, inconsistency in subjective clarity between the two eyes, increased local residual aberrations, and increasing sensitivity of switching compensation will occur. The fundamental reason is that existing technologies usually pre-fix the prescription split relationship between static basic correction terms and dynamic Add adjustment terms, and only regard high-performance hybrid liquid crystal as an actuator and driver IC as a power supply and addressing unit. They fail to reconstruct the correction responsibility boundary between static lens and dynamic Add module after the actual wearing state changes, resulting in all subsequent compensation and control being based on the drifted target synthetic wavefront. The technical problem this application aims to solve is: how to dynamically reconstruct the responsibility boundary between the static basic correction item and the dynamic Add adjustment item in the process of prescription splitting mismatch caused by changes in actual wearing status during the combined correction of static prescription lenses and dynamic Add adjustable focus modules, so as to ensure the long-term accuracy of the target synthesized wavefront. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a combined correction algorithm of static prescription lenses and dynamic Add adjustable focus modules. By unifying prescription, wearing, and assembly parameters under the same reference plane to generate a mapping table of static and dynamic items, and combining target wavefront, offset, responsibility boundary write-back, liquid crystal and drive compensation, and wavefront residual closed-loop correction, the responsibility boundary between static basic correction items and dynamic Add adjustment items is dynamically reconstructed to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a combined correction algorithm of static prescription lenses and a dynamic Add adjustable focus module, comprising: S1. Obtaining prescription, wearing, assembling, liquid crystal and driving, wherein the prescription includes spherical lens, cylindrical lens and axis, wearing includes interpupillary distance and vertex distance, and assembly includes optical center and module center, and forms a parameter set; S2. Input the parameter set into the splitting module, generate static items according to spherical lens, cylindrical lens and axis, and generate dynamic items according to target Add, pupil distance, vertex distance, optical center, module center, liquid crystal and drive, and form a mapping table; S3. Input the mapping table into the modeling module, generate a static wavefront based on the static items, generate a dynamic wavefront based on the dynamic items, generate an offset based on the coordinate difference between the optical center and the module center, and generate the target wavefront. S4. Input the target wavefront, static wavefront, dynamic wavefront, and offset into the write-back module, correct the static and dynamic items in the mapping table according to the offset, and output the corrected mapping table and target drive. S5. Input the target driver, LCD and driver into the compensation module, perform LCD compensation according to the LCD, perform driver compensation according to the driver, and output the actual Add; S6. Input the actual Add, target wavefront, and correction mapping table into the correction module. Generate the actual dynamic wavefront based on the actual Add. Superimpose the actual dynamic wavefront with the static wavefront to generate the actual wavefront. Compare the actual wavefront with the target wavefront point by point to obtain the wavefront residual. Correct the dynamic items in the correction mapping table based on the wavefront residual, and output the combined result and traceability data packet.

[0005] In a preferred embodiment, S1 includes: S1-1. Obtain the spherical lens, cylindrical lens, axis, pupil distance, vertex distance, optical center, and module center, and assemble them into the original parameter set; S1-2. Input the original parameter set into the coordinate conversion module, convert the optical center and the module center to the same reference plane according to the pupil distance and vertex distance, and output the center coordinate set. S1-3. Input the original parameter set and the center coordinate set into the parameter integration module, bind them according to the corresponding spherical lens, cylindrical lens, axis and center coordinate set, and output the parameter set.

[0006] In a preferred embodiment, S2 includes: S2-1. Input the parameter set into the static splitting module, generate static items according to spherical lens, cylindrical lens and axis, and output the static parameter group; S2-2. Input the parameter set and static parameter group into the dynamic splitting module, generate dynamic items according to the target Add, pupil distance, vertex distance, optical center, module center, liquid crystal and driver, and output the dynamic parameter group.

[0007] In a preferred embodiment, S2 further includes: S2-3. Input the static parameter set and dynamic parameter set into the boundary calculation module, calculate the allocation boundary of the static and dynamic items according to the coordinate relationship between the optical center and the module center, and output the boundary parameter set. S2-4. Input the static items, dynamic items, and boundary parameter sets into the mapping generation module, write the static items and dynamic items according to the boundary parameter sets, and output the mapping table.

[0008] In a preferred embodiment, S3 includes: S3-1. Input the mapping table into the static modeling module, generate static wavefronts based on the static items, and output the static wavefront group. S3-2. Input the mapping table into the dynamic modeling module, generate dynamic wavefronts based on the dynamic items, and output the dynamic wavefront group.

[0009] In a preferred embodiment, S3 further includes: S3-3. Input the mapping table, static wavefront group and dynamic wavefront group into the offset calculation module, generate the offset amount based on the coordinate difference between the optical center and the module center, and output the offset parameter group. S3-4. Input the static wavefront group, dynamic wavefront group, and offset parameter group into the synthesis module, generate the target wavefront based on the static wavefront group, dynamic wavefront group, and offset parameter group, and output the target wavefront.

[0010] In a preferred embodiment, S4 includes: S4-1. Input the target wavefront, static wavefront, dynamic wavefront, and offset into the evidence generation module. Generate static evidence, dynamic evidence, and offset evidence according to the difference between the target wavefront and the static wavefront, the difference between the target wavefront and the dynamic wavefront, and the offset, and write them into the evidence table. S4-2. Input the evidence table into the consistency judgment module, call the consistency verifier to cross-compare static evidence, dynamic evidence and offset evidence, generate consistency flags, conflict flags and confidence levels, and write them into the judgment table.

[0011] In a preferred embodiment, S4 further includes: S4-3. Input the decision table into the gating module, call the gating rule set to generate static write-back tokens, dynamic write-back tokens and state locks according to consistency flags, conflict flags and confidence levels, and write them into the gating table; S4-4. Input the gating table, mapping table and offset into the branch write-back module. Correct the range of static items in the mapping table according to the static write-back token, correct the range of dynamic items in the mapping table according to the dynamic write-back token, keep the unauthorized items unchanged according to the state lock, and output the branch mapping table.

[0012] In a preferred embodiment, S4 further includes: S4-5. Input the branch mapping table and decision table into the merging module, call the conflict resolver to redistribute the overlapping area of ​​static and dynamic items, perform a rollback and re-examination on the low confidence area and update the confidence, output the corrected mapping table and write it to the write-back record. S4-6. Input the correction mapping table, dynamic wavefront, and decision table into the driver generation module, generate the target driver according to the dynamic items, confidence levels, and write-back records in the correction mapping table, and write it into the driver table.

[0013] In a preferred embodiment, S5 includes: S5-1. Input the target driver and the LCD into the LCD compensation module, perform LCD compensation on the target driver according to the LCD, and output the LCD driver. S5-2. Input the LCD driver and the driver into the driver compensation module, perform driver compensation on the LCD driver according to the driver, and output the compensation driver. S5-3. Output the compensation driver to the dynamic Add module and generate the actual Add based on the compensation driver.

[0014] In a preferred embodiment, S6 includes: S6-1. Input the actual Add and correction mapping table into the dynamic correction module, generate the actual dynamic wavefront based on the actual Add, and output the actual dynamic wavefront group. S6-2. Input the actual dynamic wavefront group and the static wavefront into the wavefront synthesis module, superimpose the actual dynamic wavefront group and the static wavefront to generate the actual wavefront, and output the actual wavefront group.

[0015] In a preferred embodiment, S6 further includes: S6-3. Input the actual wavefront group and the target wavefront into the residual calculation module, compare the actual wavefront group with the target wavefront point by point to obtain the wavefront residual, and output the residual group. S6-4. Input the residual group and the correction mapping table into the mapping correction module, correct the dynamic items in the correction mapping table according to the residual group, and output the combined result and traceability data packet.

[0016] The technical effects and advantages of this invention are as follows: 1. By unifying prescription, wearing and assembly information onto the same reference plane, and combining target wavefront, offset and write-back judgment to dynamically reconstruct the responsibility boundary of static and dynamic items, the prescription splitting mismatch caused by changes in the actual wearing state can be relatively alleviated, and the target synthetic wavefront can maintain a high consistency over a long period of time. 2. By first generating static and dynamic wavefronts, and then aligning the dynamic wavefront with the coordinate difference between the optical center and the module center to synthesize the target wavefront, the wavefront modeling process can be matched with the actual assembly relationship, thereby relatively reducing the impact of sharp area offset and local residual aberrations. 3. By setting up evidence generation, consistency determination, gating and branch write-back processes, static evidence, dynamic evidence and offset evidence are used cross-referenced for the correction of responsibility boundaries. This can make the adjustment of static and dynamic items have a clear basis, thereby relatively suppressing over-adjustment, misadjustment or confusion of responsibility in overlapping areas caused by direct compensation. 4. By sequentially passing the target drive through liquid crystal compensation and drive compensation before outputting it to the dynamic Add module, the response characteristics of the high-performance hybrid liquid crystal and the output characteristics of the driver IC can be considered simultaneously. This improves the consistency between the actual Add and the target Add within a certain range and relatively reduces drive sensitivity. 5. By converting the actual Add into the actual dynamic wavefront and superimposing it with the static wavefront to form the actual wavefront, and then comparing it point by point with the target wavefront, the wavefront residual distribution of the corresponding responsibility area can be obtained, thereby supporting the directional correction of the dynamic term and forming a closed-loop correction mechanism that can be continuously iterated. 6. By retaining the mapping table, write-back record, driver table and traceability data package in each step, the responsibility boundary correction, driver generation and residual correction processes can be associated and saved, which facilitates subsequent recalibration, after-sales correction and version inheritance, and helps maintain the continuity of combined correction results. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method steps of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0019] Refer to the instruction manual appendix Figure 1The combined correction algorithm of the static prescription lens and the dynamic Add adjustable focus module of the present invention includes: S1. Obtaining prescription, wearing, assembling, liquid crystal and driving, wherein the prescription includes spherical lens, cylindrical lens and axis, wearing includes interpupillary distance and vertex distance, and assembly includes optical center and module center, and forms a parameter set; In this specific embodiment, step S1 serves to establish a unified input basis for subsequent splitting, modeling, and write-back. Its purpose is not to separately record the spherical lens, cylindrical lens, axis, interpupillary distance, vertex distance, optical center, and module center, but rather to organize the prescription information, wearing information, and assembly information into a parameter set under the same coordinate reference, object identifier, and calling format. This allows subsequent steps to directly call the parameter set, avoiding calculation deviations caused by inconsistent parameter sources, inconsistent coordinate references, or mixing of left and right objects. To achieve this, the spherical lens, cylindrical lens, axis, interpupillary distance, vertex distance, optical center, and module center are first obtained. Then, the optical center and module center are converted to the same reference plane. Finally, the prescription field, wearing field, and the converted center coordinates are bound to the same object identifier, and the parameter set is output. This process includes the following steps: In S1-1, the spherical power, cylindrical power, axis, interpupillary distance, vertex distance, optical center, and module center are first acquired and combined to form a raw parameter set. Specifically, the spherical power, cylindrical power, and axis are taken from the user's refraction results to characterize the basic refractive state. The spherical power represents the spherical refractive power, the cylindrical power represents the amount of astigmatism, and the axis represents the direction of astigmatism. The interpupillary distance and vertex distance are taken from the fitting measurement results. The interpupillary distance represents the horizontal distance between the pupil centers, and the vertex distance represents the anterior-posterior distance from the lens wearing position to the anterior surface of the eye. The optical center is the optical center position of the static prescription lens, and the module center is the geometric center position or effective accommodation center position of the dynamic Add module. During acquisition, the spherical power, cylindrical power, axis, interpupillary distance, vertex distance, optical center, and module center under the same eye are written into the same raw record and an eye identifier is added. When processing the left and right eyes simultaneously, raw parameter sets for the left eye and the right eye are formed separately to ensure that subsequent conversion and binding are performed under the same eye and that cross-eye mixing does not occur. In addition, the liquid crystal material parameters are the material response data of the liquid crystal modulation layer in the dynamic Add module. The liquid crystal material parameters include liquid crystal material number, operating temperature, phase response slope, response delay, hysteresis, maximum phase modulation, and aging compensation coefficient. Among them, the liquid crystal material number is read from the material identifier of the dynamic Add module, the operating temperature is collected by the module temperature sensor, and the phase response slope, response delay, hysteresis, maximum phase modulation, and aging compensation coefficient are read from the factory calibration table and written into the parameter set according to the same eye identifier, the same module number, and the same collection batch. The driving parameters are the driving constraint data when the driver IC outputs electrical signals to the dynamic Add module. The driving parameters include the driver IC number, driving channel number, output voltage range, driving voltage step size, output resolution, channel gain deviation, timing delay, channel enable bit, and waveform number. Among them, the driver IC number and driving channel number are read by the driver board, the output voltage range, driving voltage step size, output resolution, channel gain deviation, and timing delay are read by the driver IC calibration table, and the channel enable bit and waveform number are read by the current driving configuration. Together with the liquid crystal material parameters, they serve as the input for dynamic item generation and compensation driving calculation. In step S1-2, the original parameter set is input into the coordinate conversion module. The optical center and module center are converted to the same reference plane according to the interpupillary distance and vertex distance, and the center coordinate set is output. Specifically, a unified reference plane is first selected. The reference plane can be any fixed plane among the lens wearing surface, the frame reference plane, or the lens main plane. A unified coordinate system is established on this reference plane, where the horizontal direction is defined as the left-right direction, the vertical direction is defined as the up-down direction, and the front-back direction is represented by the depth direction corresponding to the vertex distance. During the conversion, the optical center position and module center position in the original parameter set are read first. Then, the interpupillary distance is used to determine the horizontal correspondence between the two relative to the pupil center, and the vertex distance is used to determine the front-back position relationship when the two are projected onto the unified reference plane. When the optical center and module center are originally located on different mounting levels, their respective positions are projected onto the unified reference plane along the front-back direction. When the two are originally located on the same level, their position values ​​are kept unchanged and directly written. After conversion, the center coordinate set is output. The center coordinate set includes at least the optical center coordinates and the module center coordinates, and both are under the same reference plane and the same coordinate system, so that the coordinate difference and offset in subsequent steps have a unified value basis. In S1-3, the original parameter set and the central coordinate set are input into the parameter integration module, and bound to the corresponding values ​​of spherical lens, cylindrical lens, axis, and central coordinate set, outputting a parameter set. Specifically, the parameter integration module first reads the spherical lens, cylindrical lens, axis, pupillary distance, and vertex distance from the original parameter set, then reads the optical center coordinates and module center coordinates from the central coordinate set, and then writes them accordingly according to the same eye identifier, the same reference plane, and the same acquisition batch. Corresponding writing means writing the prescription field, wearing field, and central coordinate field under the same eye into the same parameter record, so that each record... Each parameter record includes spherical lens, cylindrical lens, axis, interpupillary distance, vertex distance, optical center coordinates, and module center coordinates. To facilitate direct calling of parameters after module splitting, the parameter set is stored in a fixed field order, preferably with the prescription field first, the wearing field in the middle, and the center coordinate field last. When there are two sets of data for the left and right eyes, the left eye parameter set and the right eye parameter set are output separately and distinguished by eye identifiers. Therefore, when subsequent steps call the parameter set, there is no need to judge the parameter source, coordinate attribution, and object correspondence again, and static item generation and dynamic item generation can be performed directly. Through the above processing, spherical lens, cylindrical lens, axis, pupillary distance, vertex distance, optical center, and module center are uniformly converted into a parameter set under the same reference plane, coordinate system, and object identifier. This eliminates the two most common problems in subsequent steps: first, coordinate difference distortion caused by the optical center and module center not being on the same reference plane; and second, splitting and modeling errors caused by the incorrect correspondence between the prescription field and the center coordinates. Based on this, subsequent static items, dynamic items, mapping tables, offsets, and target wavefronts can all be established on a unified input. In practical applications: when a pair of adjustable focus glasses is used for myopia combined with presbyopia... When a user is selected, the system first reads the spherical, cylindrical, and axis prescriptions for that single eye. Then, it measures the corresponding pupillary distance and vertex distance for that eye. The system also collects the optical center position of the static prescription lens and the module center position of the dynamic Add module. If the dynamic Add module is embedded on the front of the lens and there is a front-to-back installation difference with the static prescription lens, the optical center and module center are first projected onto a unified reference surface corresponding to the lens wearing surface to obtain the optical center coordinates and module center coordinates. Then, the center coordinates, along with the spherical, cylindrical, axis, pupillary distance, and vertex distance for the same eye, are written into the same parameter record, and the parameter set is output for direct use by subsequent module splitting.

[0020] S2. Input the parameter set into the splitting module, generate static items according to spherical lens, cylindrical lens and axis, and generate dynamic items according to target Add, pupil distance, vertex distance, optical center, module center, liquid crystal and drive, and form a mapping table; In this specific embodiment, step S2 further decomposes the parameter set into static items undertaken by the static prescription lens and dynamic items undertaken by the dynamic Add module, and determines the corresponding action boundaries of each, ultimately forming a mapping table that can be directly called in subsequent modeling, write-back, and correction. This process does not simply store the prescription parameters and dynamic adjustment parameters side-by-side, but first generates static items based on spherical, cylindrical, and axis positions, then generates dynamic items based on the target Add, pupillary distance, vertex distance, optical center, module center, liquid crystal, and drive. Subsequently, the allocation boundaries of the static and dynamic items are calculated based on the coordinate relationship between the optical center and the module center under the same reference plane, and the static items, dynamic items, and allocation boundaries are written into the same mapping structure with fixed fields to ensure that the static responsibility area and dynamic responsibility area can be clearly distinguished when called in subsequent steps, avoiding wavefront modeling deviations and write-back distortions caused by unclear responsibility boundaries. This implementation process includes the following steps: In S2-1, the parameter set is input into the static splitting module, which generates static items according to spherical power, cylindrical power, and axis, and outputs static parameter groups. Specifically, the static splitting module first reads the spherical power, cylindrical power, and axis from the parameter set, uses the spherical power as the basic spherical correction value, and combines the cylindrical power and axis as the astigmatism correction value. Then, it writes the above data into static items according to the design format of the static prescription lens. The static item is the basic correction data undertaken by the static prescription lens, and includes at least a spherical correction field, an astigmatism correction field, and an eye-specific identification field. During generation, it first sorts the data by eye. The corresponding spherical lens, cylindrical lens, and axis are read. The spherical lens is then directly written into the spherical correction field, and the cylindrical lens and axis are written into the astigmatism correction field. This ensures that the static items clearly represent the basic corrective responsibility undertaken by the static prescription lens for that eye. The static parameter group consists of one or more static items. When processing the left and right eyes, a left eye static parameter group and a right eye static parameter group are formed respectively. When processing a single eye, only the static parameter group for the corresponding single eye is output. In this way, the source, object, and responsibility of the static items remain clear and are not mixed with subsequent dynamic items. In S2-2, the parameter set and static parameter group are input into the dynamic splitting module. Dynamic items are generated according to the target Add, interpupillary distance, vertex distance, optical center, module center, liquid crystal, and driver, and the dynamic parameter group is output. Specifically, the dynamic splitting module first reads the target Add, using it as the basic adjustment value that the dynamic Add module needs to handle. Then, it reads the interpupillary distance, vertex distance, optical center, and module center from the parameter set to determine the corresponding action position and action area of ​​this basic adjustment value. Subsequently, it reads the liquid crystal and driver to limit the executable adjustment range within the action position and action area. The liquid crystal preferably refers to the response information of the high-performance hybrid liquid crystal in the dynamic Add module, including at least one or more of the following: response delay, phase response slope, temperature-related response value, and aging-related response value. The driver preferably refers to the drive status information of the driver IC, including at least the output voltage range, output resolution, channel gain deviation, timing delay, and channel activation. One or more of the following states are considered: When generating a dynamic item, the basic Add field of the dynamic item is first determined based on the target Add. Then, the action center and action range of the dynamic item are determined based on the interpupillary distance, vertex distance, optical center, and module center. Finally, the basic Add field and action range are subject to executability restrictions based on the liquid crystal and the driver. If the liquid crystal response range is smaller than the required range corresponding to the target Add, the action range of the dynamic item is restricted to the liquid crystal's response range. If the driver output range is smaller than the liquid crystal driver's required range, the driver association value of the dynamic item is restricted to the driver's output range. The resulting dynamic item includes at least an Add field, an action area field, a driver association field, and an eye identification field. Multiple dynamic items form a dynamic parameter group. This processing method makes the dynamic item not just a simple Add value, but a data object that simultaneously contains adjustment amount, action position, and execution constraints, thus enabling it to directly participate in subsequent boundary calculations and driver generation. In step S2-3, the static parameter set and the dynamic parameter set are input into the boundary calculation module. The module calculates the allocation boundaries of the static and dynamic items according to the coordinate relationship between the optical center and the module center, and outputs the boundary parameter set. Specifically, the boundary calculation module first reads the optical center coordinates corresponding to the static parameter set and the module center coordinates corresponding to the dynamic parameter set, both using the same reference surface coordinates converted in step S1. Then, it calculates the coordinate difference between the optical center coordinates and the module center coordinates. The coordinate difference includes at least the horizontal and vertical coordinate differences, and may further include the corresponding values ​​of the installation differences in the front and rear directions. Based on the coordinate difference, the boundary calculation module determines the responsibility boundary position between the static and dynamic items. The responsibility boundary position is preferably represented as boundary coordinates, boundary radius, and edge... The boundary width or area mask is one of the following: In the specific calculation, the center of the module can be taken as the center of the dynamic responsibility area, and the predetermined dynamic action aperture corresponding to the target Add can be used as the basic boundary size. The position of the basic boundary size is corrected by combining the coordinate difference between the optical center and the module center to obtain the dynamic responsibility area. Then, the area outside the dynamic responsibility area is determined as the static responsibility area, thus forming the allocation boundary between the static and dynamic items. If the optical center and the module center coincide, the allocation boundary is generated symmetrically with the module center as the center. If there is a coordinate difference between the two, the allocation boundary is generated by offsetting along the direction of the coordinate difference. The boundary parameter group includes at least the boundary position field, the boundary size field, and the boundary direction field, which are used to characterize the boundary relationship between the static responsibility area and the dynamic responsibility area. In S2-4, static items, dynamic items, and boundary parameter sets are input into the mapping generation module. The static and dynamic items are written according to the boundary parameter sets, and a mapping table is output. Specifically, the mapping generation module first reads the static items, dynamic items, and boundary parameter sets. Then, according to the allocation boundaries determined by the boundary parameter sets, the static items are written to the first responsibility area, and the dynamic items are written to the second responsibility area. The first responsibility area is preferably an area outside the boundary, and the second responsibility area is preferably an area inside the boundary. When the boundary parameter sets are represented using a region mask, the writing is completed directly according to the area defined by the mask. During writing, the mapping generation module stores data in a fixed field order. The fixed fields include at least the static item identifier, dynamic item identifier, boundary field, area of ​​action field, and eye identifier. The mapping table can further include driver association fields and version fields for subsequent write-back and retrospective calls. It is not simply a data concatenation but a structured record representing how static and dynamic items are allocated correction responsibility under the same reference plane. Each record contains both static and dynamic items and their corresponding boundary relationships. If there are two sets of data for the left and right eyes, a left-eye mapping table and a right-eye mapping table are generated separately. If binocular collaborative processing is required, binocular association fields can be further established based on the left and right-eye mapping tables, but the monocular mapping relationships are still stored independently. In this way, the subsequent modeling module can directly generate static and dynamic wavefronts based on the mapping table without needing to determine the responsibility area attribution of static and dynamic items again. Through the above processing, step S2 completes the continuous transformation from the parameter set to static items, dynamic items, allocation boundaries, and mapping tables. This clearly distinguishes the basic corrective responsibility borne by the static prescription lens and the accommodative responsibility borne by the dynamic Add module, and writes them into the same structure. This solves the problem that static and dynamic items only remain at the level of abstract concepts, and also solves the problem that the allocation boundaries and mapping tables lack minimum field support. Thus, it provides a directly callable responsibility mapping foundation for subsequent target wavefront generation, responsibility boundary write-back, and dynamic item correction. In practical applications: when the combined adjustable focus glasses are used by a user with myopia and presbyopia, the parameter set first reads the user's needs. The system first calculates the spherical, cylindrical, and axis positions of a user's monocular lens to generate static items handled by the static prescription lens. Then, it reads the target Add corresponding to the user's current viewing distance, along with interpupillary distance, vertex distance, optical center, module center, high-performance hybrid liquid crystal response information, and driver IC status information to generate dynamic items handled by the dynamic Add module. Subsequently, based on the coordinate difference between the optical center coordinates and the module center coordinates, it determines the offset position of the dynamic responsibility area relative to the static responsibility area and forms a boundary parameter set. Finally, it writes the static items to the region outside the boundary and the dynamic items to the region inside the boundary, generating a mapping table for subsequent modeling modules to directly use in generating static wavefronts, dynamic wavefronts, and target wavefronts.

[0021] S3. Input the mapping table into the modeling module, generate a static wavefront based on the static items, generate a dynamic wavefront based on the dynamic items, generate an offset based on the coordinate difference between the optical center and the module center, and generate the target wavefront. In this specific embodiment, step S3 further converts the mapping table formed in step S2 into a target wavefront that can be used for write-back and correction. Its core is not to obtain static and dynamic wavefronts separately, but rather to first generate static and dynamic wavefronts based on the correction responsibilities of the static and dynamic entries in the mapping table, then generate an offset based on the coordinate difference between the optical center and the module center on the same reference plane, and finally combine the dynamic wavefront with the static wavefront after aligning its position according to the offset, thereby obtaining a target wavefront consistent with the actual assembly relationship. To ensure that evidence generation, responsibility write-back, and residual correction in subsequent steps are all based on a unified wavefront, the static wavefront, dynamic wavefront, and target wavefront in this step are all represented using the same reference plane, the same coordinate system, and the same sampling format. The preferred sampling format is one of wavefront diagram, phase matrix, or sampling point array. This implementation process includes the following steps: In step S3-1, the mapping table is input into the static modeling module, static wavefronts are generated based on the static items, and static wavefront groups are output. Specifically, the static modeling module first reads the static items, boundary fields, and eye-specific identifier fields from the mapping table. The static items are the basic correction data provided by the static prescription lens, and the boundary fields are used to define the static responsibility area corresponding to the static items. Subsequently, a computational grid for the static wavefront is established on the unified reference plane determined in step S1. The computational grid is preferably a two-dimensional sampling point array arranged at predetermined intervals, and each sampling point has fixed horizontal and vertical coordinates. During generation, the spherical correction in the static items is first used. The static phase value at each sampling point is calculated using the field and the astigmatism correction field. Then, the static phase value is written into the static responsibility area according to the boundary field. Sampling points that do not belong to the static responsibility area are left empty or written with zero value, thus forming a static wavefront. When processing the left and right eyes, the left eye static wavefront and the right eye static wavefront are generated according to the left eye static item and the right eye static item respectively, and then a static wavefront group is formed. In order to ensure that subsequent point-by-point comparison can be performed, each static wavefront in the static wavefront group adopts the same sampling interval, coordinate origin and coordinate direction as the subsequent dynamic wavefront and the target wavefront, so that the sampling points at the same position can correspond one-to-one between different wavefronts. In S3-2, the mapping table is input into the dynamic modeling module, which generates a dynamic wavefront based on the dynamic items and outputs a dynamic wavefront group. Specifically, the dynamic modeling module first reads the dynamic items, boundary fields, and eye-specific identifier fields from the mapping table, where the dynamic items include at least the Add field, the area of ​​action field, and the drive association field. Then, it establishes a computational grid for the dynamic wavefront on the same uniform reference surface and the same sampling lattice as the static wavefront. During generation, the module first determines the basic phase adjustment amount that the dynamic Add module should provide at the target line of sight based on the Add field in the dynamic items, then determines the dynamic responsibility area corresponding to this basic phase adjustment amount based on the area of ​​action field, and finally executes the basic phase adjustment amount according to the drive association field. The generated dynamic wavefront is confined within the range of the liquid crystal and the drive. The obtained dynamic phase value is then written into the dynamic responsibility area, with sampling points outside the dynamic responsibility area left empty or written with zero values, thus forming the dynamic wavefront. The dynamic wavefront preferably characterizes the phase distribution that the dynamic Add module should form under ideal alignment, without directly including positional deviations caused by assembly offsets. When processing the left and right eyes, dynamic wavefronts for the left and right eyes are generated separately and combined into a dynamic wavefront group. In this way, the dynamic wavefront and static wavefront, under the same sampling grid, respectively characterize the correction results undertaken by the static prescription lens and the dynamic Add module, providing direct input for subsequent offset calculations and wavefront synthesis. In step S3-3, the mapping table, static wavefront group, and dynamic wavefront group are input into the offset calculation module. The module generates an offset based on the coordinate difference between the optical center and the module center, and outputs the offset parameter set. Specifically, the offset calculation module first reads the optical center coordinates and module center coordinates from the mapping table. Both the optical center coordinates and module center coordinates are the center coordinates converted to the same reference plane in step S1. Then, it calculates the coordinate difference between the two, which includes at least the horizontal and vertical coordinate differences, and may further include the forward and backward offset values ​​obtained from the installation layer difference. Subsequently, the offset calculation module converts the coordinate difference into an offset corresponding to the sampling point array. The conversion method is as follows: First, the coordinate difference is converted into the displacement of the sampling points based on the spacing between the sampling points. Then, the offset direction is determined based on the direction of the coordinate difference, thus obtaining the offset that can be directly applied to the dynamic wavefront. If the coordinate difference is not an integer multiple of the sampling point spacing, a fractional offset is generated by interpolation, enabling the dynamic wavefront to be aligned at the sub-sampling level during subsequent synthesis. The offset parameter group includes at least a horizontal offset field, a vertical offset field, and an offset direction field, and may further include an interpolation identifier field to indicate whether interpolation alignment is used. This process makes the offset no longer just an abstract coordinate difference, but an execution parameter that can directly drive the dynamic wavefront to complete the position translation on the same sampling grid. Furthermore, the interpolation method is bilinear interpolation, and the fractional offset is calculated from the coordinate difference and the sampling point spacing; The offset calculation module reads the horizontal coordinate difference, vertical coordinate difference, horizontal sampling point spacing, and vertical sampling point spacing. It divides the horizontal coordinate difference by the horizontal sampling point spacing to obtain the horizontal point displacement, and divides the vertical coordinate difference by the vertical sampling point spacing to obtain the vertical point displacement. The decimal parts of both the horizontal and vertical point displacements are taken as the horizontal and vertical fractional offsets, respectively. When the target sampling point is located within the grid formed by four adjacent sampling points in the original sampling point matrix, the phase values ​​of the lower left, lower right, upper left, and upper right original sampling points are read. The horizontal fractional offset is subtracted from the vertical fractional offset, and then multiplied by the result. The obtained value is used as the weight of the lower left original sampling point. The value obtained by multiplying the horizontal fractional offset by one minus the vertical fractional offset is used as the weight of the lower right original sampling point. The value obtained by subtracting the horizontal fractional offset by one and then multiplying by the vertical fractional offset is used as the weight of the upper left original sampling point. The value obtained by multiplying the horizontal fractional offset by the vertical fractional offset is used as the weight of the upper right original sampling point. Then, the phase values ​​of the four original sampling points are multiplied by their respective weights and added together to obtain the interpolated phase value of the target sampling point. The synthesis module generates the fractionally offset dynamic wavefront according to the interpolated phase value, so that the dynamic wavefront completes the alignment of the sampling point level and fractional sampling point level according to the coordinate difference between the optical center and the module center. In S3-4, the static wavefront group, dynamic wavefront group, and offset parameter group are input into the synthesis module. The target wavefront is generated and output based on these groups. Specifically, the synthesis module first reads the static wavefront from the static wavefront group, the dynamic wavefront from the dynamic wavefront group, and the offset from the offset parameter group. Then, it aligns the dynamic wavefront on a unified reference plane and a unified sampling point matrix according to the offset parameter group. Position alignment refers to translating the entire dynamic wavefront to the sampling area corresponding to the actual position of the module center based on the horizontal and vertical offset fields. When the offset is an integer sampling point displacement, it is directly shifted and written; when the offset includes fractional displacement, interpolation is used to obtain the aligned dynamic wavefront. After position alignment is completed, the aligned wavefront is... The dynamic wavefront and static wavefront are superimposed point by point on the same sampling array to obtain the target wavefront. The target phase value at the same sampling point is formed by adding the corresponding static phase value and the aligned dynamic phase value. If a sampling point falls only in the static responsibility area and not in the dynamic responsibility area, the target phase value is the static phase value. If a sampling point falls only in the dynamic responsibility area, the target phase value is the aligned dynamic phase value. The target wavefront generated in this way includes both the basic correction result undertaken by the static prescription lens and the accommodation result that the dynamic Add module should undertake in the actual assembly position. It serves as the direct input for evidence generation and responsibility write-back in the subsequent step S4. When processing the left and right eyes, the target wavefronts for the left and right eyes are output separately and stored separately by eye identifier. Through the above processing, step S3 completes the continuous transformation from the mapping table to static wavefront, dynamic wavefront, offset, and target wavefront. This ensures that static and dynamic items are no longer merely at the responsibility allocation level, but are further implemented as wavefront data under the same reference plane, coordinate system, and sampling format. Simultaneously, by converting the coordinate difference between the optical center and the module center into an offset that can be directly applied to the dynamic wavefront, and completing position alignment before synthesis, the synthesis error caused by the spatial inconsistency between the dynamic and static wavefronts is eliminated. This provides a unified, comparable, and traceable target wavefront basis for subsequent responsibility boundary write-back and wavefront residual correction. In practical applications… In use: When a user's dynamic Add module is offset towards the nose and slightly upward relative to the static prescription lens, the static modeling module first generates a static wavefront for that eye based on the static entries in the mapping table, and then the dynamic modeling module generates a dynamic wavefront under ideal alignment based on the dynamic entries; subsequently, the offset calculation module calculates the horizontal and vertical offsets on the corresponding sampling point array based on the difference between the optical center coordinates and the module center coordinates; finally, the synthesis module first translates the dynamic wavefront to the actual assembly position according to the horizontal and vertical offsets, and then superimposes it point by point with the static wavefront to obtain the target wavefront for that eye, and uses it as the input for the subsequent write-back module.

[0022] S4. Input the target wavefront, static wavefront, dynamic wavefront, and offset into the write-back module, correct the static and dynamic items in the mapping table according to the offset, and output the corrected mapping table and target drive. In this specific embodiment, step S4 serves to perform a traceable responsibility boundary write-back on the static and dynamic items in the mapping table based on the target wavefront, static wavefront, dynamic wavefront, and offset obtained in step S3. This allows the corrective division of labor between the static prescription lens and the dynamic Add module to be adjusted according to the actual assembly relationship and target wavefront requirements. This process does not directly rewrite the mapping table all at once according to the offset; instead, it first converts the differences between the target wavefront and the static wavefront, the target wavefront and the dynamic wavefront, and the offset itself into referable static evidence, dynamic evidence, and offset evidence. Then, it performs a consistency judgment on the three types of evidence based on the consistency standard. The process involves generating static write-back tokens, dynamic write-back tokens, and state locks based on records, conflict markers, and confidence levels. Branch write-backs are then performed on both static and dynamic items, with overlapping areas reassigned and low-confidence areas undergoing rollback and re-examination. Finally, a target driver is generated based on the dynamic items, confidence levels, and write-back records in the correction mapping table, ensuring that subsequent compensation and correction are based on the corrected responsibility boundaries. To guarantee the entire write-back process is executable, the evidence table, decision table, gating table, correction mapping table, write-back records, and driver table all use fixed fields for storage, and each field corresponds to the eye identifier, region identifier, and sampling location. This implementation process includes the following steps: In step S4-1, the target wavefront, static wavefront, dynamic wavefront, and offset are input into the evidence generation module. Static evidence, dynamic evidence, and offset evidence are generated based on the differences between the target wavefront and the static wavefront, the differences between the target wavefront and the dynamic wavefront, and the offset, and written into the evidence table. Specifically, the evidence generation module first reads the target wavefront, static wavefront, and dynamic wavefront corresponding to the same eye, and confines them to the same reference plane, coordinate system, and sampling point matrix used in step S3. Then, it calculates the difference between the target wavefront and the static wavefront point by point to obtain a static difference set. The static difference set is used to characterize the remaining difference between the target wavefront and the static wavefront when only the static term is present at each sampling point. Subsequently, it calculates the difference between the target wavefront and the dynamic wavefront point by point to obtain a dynamic difference set. The dynamic difference set is used to characterize the remaining difference between the target wavefront and the dynamic wavefront when only the dynamic term is present at each sampling point. The remaining differences are identified; the horizontal offset field, vertical offset field, and offset direction field in the offset are then read and converted into an offset distribution corresponding to the sampling point matrix to obtain the offset influence group; static evidence is generated based on the static difference group, which includes at least the sampling position, static difference magnitude, and static difference direction; dynamic evidence is generated based on the dynamic difference group, which includes at least the sampling position, dynamic difference magnitude, and dynamic difference direction; offset evidence is generated based on the offset influence group, which includes at least the sampling position, offset direction, and offset effect intensity; then, the static evidence, dynamic evidence, and offset evidence are written into the evidence table according to the sampling position, so that each record in the evidence table corresponds to the three types of evidence data under the same sampling position; after this processing, the subsequent consistency determination no longer directly faces the abstract wavefront, but directly calls the aligned, classified, and located evidence data; In S4-2, the evidence table is input into the consistency determination module, which calls the consistency checker to perform cross-comparison of static evidence, dynamic evidence, and offset evidence, generating consistency markers, conflict markers, and confidence levels, and writing them into the determination table. Specifically, the consistency determination module first reads the static evidence, dynamic evidence, and offset evidence from the evidence table according to the sampling position, and then calls the consistency checker to perform cross-comparison of the three types of evidence. The cross-comparison includes at least directional consistency comparison, regional consistency comparison, and trend consistency comparison. Directional consistency comparison is used to determine whether the static difference direction, dynamic difference direction, and offset direction point to the same correction direction. Regional consistency comparison is used to determine whether the three types of evidence are concentrated in the same responsibility area or the same boundary neighborhood. Trend consistency comparison is used to determine whether the three types of evidence change continuously in the same direction at adjacent sampling positions. If the three types of evidence are in the above... If at least two types of evidence simultaneously meet the consistency criteria, a consistency marker is generated. If the three types of evidence show opposite results in terms of correction direction, area of ​​effect, or trend of change, a conflict marker is generated. The confidence level is determined by the number of evidences, the degree of consistency of evidence, and the number of conflicts. Specifically, the values ​​can be determined as follows: First, count the number of evidences that meet the consistency criteria in the same area. Then, count the number of evidences with conflict markers in the same area. After that, normalize the number of consistent evidences and the number of conflict evidences to obtain a confidence level value between the lowest and highest confidence levels. When the number of consistent evidences is high and the number of conflict evidences is low, the confidence level is higher. When the number of consistent evidences is low or the number of conflict evidences is high, the confidence level is lower. Finally, the consistency markers, conflict markers, and confidence levels are written into the judgment table according to the sampling location or area number to form a judgment result that can be directly called by the subsequent gating module. Furthermore, the direction consistency comparison includes execution according to the sign of the difference direction; the consistency determination module reads the static difference direction, dynamic difference direction, and offset direction at the same sampling position, takes the static difference direction as the positive or negative sign of the result of subtracting the static wavefront phase value from the target wavefront phase value, takes the dynamic difference direction as the positive or negative sign of the result of subtracting the dynamic wavefront phase value from the target wavefront phase value, and takes the offset direction as the direction sign formed by the horizontal offset field and the vertical offset field; when at least two of the static difference direction, dynamic difference direction, and offset direction point to the same boundary movement direction, the direction consistency comparison result is passed; when at least two of them point to opposite boundary movement directions, the direction consistency comparison result is failed. Furthermore, the regional consistency comparison includes execution according to the sampling location; the consistency determination module divides the area where the same sampling location is located into a static responsibility area, a dynamic responsibility area, and a boundary neighborhood. The boundary neighborhood is the set of sampling points within a distance of one sampling point on each side of the boundary line between the static responsibility area and the dynamic responsibility area; when at least two types of evidence among static evidence, dynamic evidence, and offset evidence fall into the same area, or simultaneously fall into the same boundary neighborhood, the regional consistency comparison result is passed; the trend consistency comparison is performed according to the sign sequence of adjacent sampling locations; the consistency determination module reads the static difference direction, dynamic difference direction, and offset direction of three adjacent sampling locations along the horizontal sampling direction and the vertical sampling direction, respectively. When at least two of the three adjacent sampling locations maintain the same direction sign and there is no alternation of positive and negative signs, the trend consistency comparison result is passed; In S4-3, the decision table is input into the gating module, which then calls the gating rule set to generate static write-back tokens, dynamic write-back tokens, and state locks based on consistency flags, conflict flags, and confidence levels, and writes them into the gating table. Specifically, the gating module first reads the consistency flags, conflict flags, and confidence levels from the decision table, and then calls the gating rule set to execute hierarchical gating. The gating rule set includes at least consistency pass rules, conflict blocking rules, and low confidence backoff rules. The consistency pass rule allows a sampling location or area to enter the write-back process when a consistency flag exists and no conflict flag exists. The conflict blocking rule prohibits direct write-back when a conflict flag exists at a sampling location or area. The low confidence backoff rule is used to prevent direct write-back when a conflict flag exists at a sampling location or area. When the confidence level of a location or region is lower than the average confidence level of the current write-back batch for the same eye, the location or region is marked as a region to be re-examined. Based on the above rules, if static evidence dominates in a region and meets the consistency pass rule, a static write-back token is generated; if dynamic evidence dominates in a region and meets the consistency pass rule, a dynamic write-back token is generated; if a region has a conflict marker or its confidence level is in a state to be re-examined, a state lock is generated. The static write-back token is used to authorize that static items can be corrected, the dynamic write-back token is used to authorize that dynamic items can be corrected, and the state lock is used to keep unauthorized items unchanged. Then, the static write-back token, dynamic write-back token, and state lock are written into the gating table according to the region number and eye identifier, so that each responsible area has a clear write-back authorization status. Furthermore, the consistency pass rule includes judgment based on three types of evidence at the same sampling location or within the same region. The gating module reads the directional consistency comparison result, regional consistency comparison result, trend consistency comparison result, conflict marker, and confidence level from the judgment table. When the directional consistency comparison result, regional consistency comparison result, and trend consistency comparison result are all passed, and the conflict marker is empty, the sampling location or region is recorded as a consistency pass zone. When the static evidence strength is greater than the dynamic evidence strength within the consistency pass zone, and the static difference direction and the offset direction point to the same static boundary movement direction, a static write-back token is generated. When the dynamic evidence strength is greater than the static evidence strength within the consistency pass zone, and the dynamic difference direction and the offset direction point to the same dynamic boundary movement direction, a dynamic write-back token is generated. The static evidence strength is the sum of the absolute values ​​of the static differences within the same region, and the dynamic evidence strength is the sum of the absolute values ​​of the dynamic differences within the same region. This establishes a computational correlation between the consistency pass rule and the target wavefront, static wavefront, dynamic wavefront, and offset. Furthermore, the conflict blocking rules are implemented according to three scenarios: opposite evidence direction, opposite regional attribution, and opposite trend. When the static difference direction is opposite to the dynamic difference direction at the same sampling location or within the same region, or when static evidence falls into the static responsibility area while dynamic evidence falls into the dynamic responsibility area and both point to expanding their own responsibility areas, or when the difference direction of adjacent sampling locations alternates between positive and negative, a conflict marker is generated and a state lock is generated, prohibiting direct write-back at that sampling location or region. The low confidence rollback rule determines the confidence level based on the number of consistent evidence and the number of conflicting evidence within the same region. The confidence level is calculated by dividing the number of consistent evidence by the sum of the number of consistent evidence and the number of conflicting evidence. When the confidence level is lower than the average confidence level of the current write-back batch for the same eye, the sampling location or region is marked as a re-examination area and a state lock is generated. The re-examination area rereads the evidence table to perform direction consistency comparison, regional consistency comparison, and trend consistency comparison, updates the confidence level, and then enters the gating judgment. In S4-4, the gating table, mapping table, and offset are input into the branch write-back module. The static items in the mapping table are range-corrected according to the static write-back token, and the dynamic items in the mapping table are range-corrected according to the dynamic write-back token. Unauthorized items are kept unchanged according to the state lock, and the branch mapping table is output. Specifically, the branch write-back module first reads the static write-back token, dynamic write-back token, and state lock from the gating table, then reads the static items, dynamic items, and boundary fields from the mapping table, and simultaneously reads the horizontal and vertical offset fields from the offset. For areas with static write-back tokens, the branch write-back module adjusts the scope of the static items according to the offset. Specifically, if the offset direction points to the static responsibility area, the boundary of the static item in the corresponding direction is moved towards the dynamic responsibility area. The region shrinks or expands by a predetermined step size, preferably one or more sampling point intervals, to form the corrected static responsibility area. For regions with dynamic write-back tokens, the responsibility area of ​​dynamic items is adjusted in the same way, causing the dynamic responsibility area to expand or shrink accordingly. For regions with state locks, the corresponding static and dynamic items remain unchanged, and boundary rewriting is not performed. After each correction, the corrected static item range, dynamic item range, and boundary fields are written back to the corresponding region to form a branch mapping table. In this way, the correction of static and dynamic items is no longer a unified table rewrite, but is processed separately according to different tokens, so that the static responsibility area and the dynamic responsibility area can be updated by region branch within the same mapping table. In S4-5, the branch mapping table and decision table are input into the merging module. The conflict resolver is called to redistribute the overlapping areas of static and dynamic items. Low-confidence areas are re-examined and their confidence levels are updated. The corrected mapping table is output and written back to the record. Specifically, the merging module first reads the responsibility allocation results of each region in the branch mapping table, then reads the conflict markers and confidence levels in the decision table. It first identifies the overlapping areas that are simultaneously covered by static and dynamic items after correction. The overlapping area is the region where the static responsibility area and the dynamic responsibility area overlap at the same sampling position. Subsequently, the conflict resolver is called to redistribute the overlapping area. Specifically, it first compares the static evidence strength and dynamic evidence strength corresponding to the overlapping area, then combines the offset evidence direction of the overlapping area to transfer at least a portion of the sampling positions in the overlapping area from the static item to the dynamic item, or from the dynamic item to the static item. When the dynamic evidence... When the strength of the evidence is higher than the strength of the static evidence and the offset direction points towards the module center, the responsibility of the overlapping area is preferentially assigned to the dynamic item; when the strength of the static evidence is higher than the strength of the dynamic evidence and the offset direction is far from the module center, the responsibility of the overlapping area is preferentially assigned to the static item; for low confidence areas, the merging module calls back the evidence table and decision table data corresponding to steps S4-1 and S4-2 according to the re-examination flag in the decision table to perform a rollback re-examination, that is, re-performs the evidence cross-comparison and updates the consistency flag, conflict flag and confidence, and then participates in the redistribution of the overlapping area with the updated confidence; after completing the redistribution of the overlapping area and the re-examination of the low confidence area, a corrected mapping table is obtained; at the same time, the original boundary, corrected boundary, conflict type, confidence update value and final allocation result of each responsibility area are written into the write-back record, which is used to indicate the actual modification content and its basis in this write-back process; Furthermore, the static and dynamic evidence strengths are obtained by accumulating the absolute values ​​of the differences between sampling points within the overlapping area. The conflict resolver reads the target wavefront phase value, static wavefront phase value, and dynamic wavefront phase value of each sampling point within the overlapping area. The absolute value of the difference between the target wavefront phase value and the static wavefront phase value is recorded as the static difference, and the absolute value of the difference between the target wavefront phase value and the dynamic wavefront phase value is recorded as the dynamic difference. The static difference values ​​of all sampling points within the same overlapping area are then added together to obtain the static evidence strength, and the dynamic difference values ​​of all sampling points within the same overlapping area are then added together to obtain the dynamic evidence strength. When there are null sampling points within the overlapping area, the null sampling points are not included in the accumulation. When there are no valid sampling points within the overlapping area, both the static and dynamic evidence strengths are zero. Furthermore, the redistribution of overlapping areas can be carried out according to the relationship between the static and dynamic strengths of evidence; When the strength of dynamic evidence is greater than the strength of static evidence, and the direction of the offset evidence points towards the center of the module, the sampling points in the overlapping area are assigned to the dynamic item; when the strength of static evidence is greater than the strength of dynamic evidence, and the direction of the offset evidence is far from the center of the module, the sampling points in the overlapping area are assigned to the static item; when the strength of static evidence is equal to the strength of dynamic evidence, the confidence level of the same overlapping area is read. If the confidence level meets the write-back condition, the allocation result is determined according to the direction of the offset evidence; if the confidence level does not meet the write-back condition, the overlapping area is marked as a re-examination area and the original allocation result is maintained. In S4-6, the correction mapping table, dynamic wavefront, and decision table are input into the driver generation module. The target driver is generated according to the dynamic items, confidence levels, and write-back records in the correction mapping table and written to the driver table. Specifically, the driver generation module first reads the dynamic items in the correction mapping table, which include at least the corrected Add field, the corrected scope field, and the driver association field. Then, it reads the confidence levels in the decision table, as well as the final allocation results and boundary modification information in the write-back records. When generating the target driver, the module first determines the base driver value that each responsibility area of ​​the dynamic Add module should output based on the corrected dynamic items. Then, it performs a weighted correction on the base driver value according to the confidence level of the corresponding responsibility area, ensuring that high-confidence responsibility areas maintain their current driver allocation and low-confidence responsibility areas are downgraded. The driving amplitude is low. Then, combined with the boundary modification information in the write-back record, the driving distribution of the adjacent positions of the overlapping area is processed to make the target driving smooth at the responsibility boundary and avoid driving abrupt changes in adjacent areas. If a certain area is marked as unchanged in the write-back record, the original driving association value of the area is used. If a certain area is marked as dynamic responsibility extension in the write-back record, the driving coverage is recalculated according to the extended dynamic action area. The final target driving is written into the driving table according to the area number, sampling position and eye identifier. Each record in the driving table corresponds to a dynamic responsibility area in the correction mapping table, so that the liquid crystal compensation and driving compensation in the subsequent step S5 can directly use the target driving as input. It should be further explained that the liquid crystal compensation calculation method includes: the base liquid crystal voltage equals the target phase modulation amount divided by the phase response slope; the delay compensation amount equals the response delay multiplied by the phase change rate of the current driving batch; the hysteresis compensation amount equals the hysteresis amount multiplied by the sign of the change direction of the target phase modulation amount; the liquid crystal compensation voltage equals the sum of the base liquid crystal voltage, the delay compensation amount, and the hysteresis compensation amount, and then multiplied by an aging compensation coefficient; where the target phase modulation amount is taken from the target drive, and the phase response slope, response delay, hysteresis amount, maximum phase modulation amount, and aging compensation coefficient are taken from the liquid crystal material parameters; when the target phase modulation amount is greater than the maximum phase modulation amount, the maximum phase modulation amount is used to replace the target phase modulation amount in the liquid crystal compensation calculation, and a saturation mark is recorded in the liquid crystal drive; The drive compensation calculation method includes: the channel correction voltage is equal to the liquid crystal compensation voltage divided by the channel gain deviation; the delay correction voltage is equal to the channel correction voltage plus the product of the timing delay and the channel voltage change rate; the quantization voltage is equal to the delay correction voltage divided by the drive voltage step size, rounded down, and then multiplied by the drive voltage step size; the drive code value is equal to the quantization voltage divided by the drive voltage step size, and the range of the drive code value is limited according to the output resolution; the output voltage range in the drive parameters is used to limit the upper and lower limits of the quantization voltage. When the channel enable bit is off, the compensation drive is set to zero and a disabled flag is recorded; when the channel enable bit is on, the corresponding drive waveform is called according to the waveform number to output the compensation drive. The actual Add is calculated by the compensation drive. The calculation method of actual Add includes: the actual output voltage equals the drive code value multiplied by the drive voltage step; the actual liquid crystal voltage equals the actual output voltage multiplied by the value obtained by adding the channel gain deviation; the actual phase modulation amount equals the actual liquid crystal voltage multiplied by the phase response slope; the actual Add equals the value obtained by converting the actual phase modulation amount according to the calibration ratio of the phase modulation amount of the dynamic Add module and Add; the compensation module binds the actual Add with the responsibility area number, sampling position, eye identification, drive channel number and drive batch and writes it into the actual Add record, which is called when generating the actual dynamic wavefront in step S6; Through the above processing, step S4 achieves a complete closed-loop write-back from the target wavefront, static wavefront, dynamic wavefront, and offset to the correction mapping table and target driver. This makes the responsibility boundaries of static and dynamic items no longer fixed, but can be dynamically adjusted in a partitioned, graded, and traceable manner based on evidence, consistency judgment, gating results, and conflict resolution results. Simultaneously, by including low-confidence areas in the backtesting, including overlapping areas in the redistribution, and writing the write-back results into the write-back record and driver table, it solves the three most common problems in the responsibility boundary write-back process: blind rewriting when evidence is insufficient, duplicate responsibility when overlapping areas are unclear, and the inability of the target driver to maintain consistency with the correction mapping table after boundary correction. This provides stable intermediate results for subsequent liquid crystal compensation, driver compensation, and wavefront residual correction. In practical applications: when a target wavefront and static wavefront of one eye have a persistent difference in the nasal boundary neighborhood, and the target… When the wavefront and dynamic wavefront show a consistent directional difference within the same region, and the module center shifts towards the nose relative to the optical center, the evidence generation module first generates static evidence, dynamic evidence, and shift evidence in this region and writes them into the evidence table. The consistency determination module generates a consistency marker and a higher confidence level based on the directional consistency and regional consistency of the three types of evidence within the same region. The gating module generates a dynamic write-back token accordingly. The branch write-back module expands the dynamic responsibility area of ​​this region towards the nose and forms a branch mapping table. If the expanded region overlaps with the original static responsibility area, the merging module calls the conflict resolver to reassign a portion of the sampling positions closer to the module center in the overlapping area to the dynamic item and writes the boundary modification information into the write-back record. Finally, the drive generation module generates a new target drive based on the expanded dynamic item, the updated confidence level, and the write-back record and writes it into the drive table for direct use by the subsequent compensation module.

[0023] S5. Input the target driver, LCD and driver into the compensation module, perform LCD compensation according to the LCD, perform driver compensation according to the driver, and output the actual Add; In this specific embodiment, step S5 further converts the target driver generated in step S4 into a compensation driver that can be actually executed by the dynamic Add module, and obtains the actual Add accordingly. This process does not directly send the target driver to the dynamic Add module; instead, it first performs liquid crystal compensation on the target driver based on the liquid crystal's response state, and then performs drive compensation on the liquid crystal-compensated driver based on the driver's output state. This ensures that the compensation driver finally output to the dynamic Add module simultaneously satisfies the response characteristics of the high-performance hybrid liquid crystal and the output characteristics of the driver IC, avoiding inconsistencies between the actual Add and the target Add due to liquid crystal response deviations or driver output deviations. To ensure a clear basis for the generation of the actual dynamic wavefront in the subsequent step S6, the liquid crystal driver, compensation driver, and actual Add are all stored according to the same eye identifier, the same responsibility area identifier, and the same call batch in this step. This implementation process includes the following steps: In step S5-1, the target driver and liquid crystal are input into the liquid crystal compensation module. Liquid crystal compensation is performed on the target driver according to the liquid crystal, and the liquid crystal driver is output. Specifically, the liquid crystal compensation module first reads the target driver, which is taken from the dynamic responsibility area driver record corresponding to the driver table in step S4, and includes at least the responsibility area identifier, driver amplitude, and driver distribution information. Then, it reads the liquid crystal, preferably the response information of the high-performance hybrid liquid crystal in the dynamic Add module, including at least one or more of the following: response delay, phase response slope, temperature-related response value, and aging-related response value. During compensation, the delay correction amount of the target driver at the current call time is first determined based on the response delay in the liquid crystal, and then based on the phase... The response slope determines the conversion relationship between the target driver and the liquid crystal phase output, and the current value of this conversion relationship is corrected based on the temperature-related response value and the aging-related response value. Then, the delay correction amount and the conversion relationship are applied to the target driver to correct the liquid crystal response of the target driver's driving amplitude and driving distribution information, thus obtaining the liquid crystal driver. If the response slope of the liquid crystal in the current responsibility area is low, the liquid crystal driving amplitude in that responsibility area is increased accordingly. If the response delay of the liquid crystal in the current temperature or aging state is large, the liquid crystal driving timing of that responsibility area is shifted forward. The final output liquid crystal driver is still saved according to the responsibility area identifier and the eye identifier, so that subsequent driving compensation can be called up area by area. In S5-2, the LCD driver and the driver input compensation module perform driver compensation on the LCD driver according to the driver and output the compensated driver. Specifically, the driver compensation module first reads the LCD driver, then reads the driver, which preferably refers to the driver IC's driver status information, including at least one or more of the following: output voltage range, output resolution, channel gain deviation, timing delay, and channel enable state. During compensation, it first determines whether the drive amplitude of each responsible area in the LCD driver is within the output range of the driver IC based on the output voltage range. When the LCD drive amplitude of a certain responsible area exceeds the current output voltage range, the drive amplitude of that responsible area is truncated to the upper or lower limit of the allowable range. Then, according to the input... The output resolution quantizes the driving amplitude of the LCD driver to correspond to the minimum output step of the driver IC. Then, based on the channel gain deviation, the driving amplitude of the LCD in different responsibility areas is corrected to ensure that the same target driver maintains consistent output across different driving channels. Next, the output sequence and enable state of the LCD driver are corrected based on timing delay and channel enable state to ensure that the driving signals of all responsibility areas can be correctly output within the executable channels of the current driver IC. After the above processing, a compensated driver is obtained. The compensated driver is the final execution driver that simultaneously meets the LCD response requirements and the driver IC output requirements, and the call result is written according to the responsibility area identifier, channel identifier, and eye identifier. In S5-3, the compensation drive is output to the dynamic Add module, and the actual Add is generated based on the compensation drive. Specifically, the compensation drive obtained in step S5-2 is first output to the corresponding drive area of ​​the dynamic Add module according to the responsibility area identifier, so that the high-performance hybrid liquid crystal in the dynamic Add module forms the actual phase adjustment corresponding to the compensation drive in each responsibility area; then, the actual Add is generated according to the correspondence between the compensation drive and the current response state of the liquid crystal. The actual Add can be generated in either of the following two ways: First, the actual Add is directly calculated based on the liquid crystal response result corresponding to the compensation drive, that is, the compensation drive is first read in each responsibility area. The driving amplitude within each region is combined with the current liquid crystal response relationship used in the liquid crystal compensation in step S5-1 to obtain the actual adjustment amount corresponding to each responsibility region, and the actual adjustment amount of each responsibility region is summarized into actual Add; secondly, the output result of the dynamic Add module after the compensation drive is measured or reconstructed, and then the actual Add is calculated from the measurement or reconstruction result; regardless of the method used, the actual Add is output according to the eye identifier and responsibility region identifier, and is used as the direct input for generating the actual dynamic wavefront in step S6, thereby ensuring that the actual Add is consistent with the aforementioned target drive, liquid crystal drive and compensation drive in terms of object attribution and responsibility region attribution; Through the above processing, step S5 achieves a continuous transition from target driving to liquid crystal driving, compensation driving, and actual Add, ensuring that the target driving generated in step S4 is no longer merely a theoretical driving mechanism, but is further implemented as a compensation driving mechanism that the dynamic Add module can actually execute. Based on this, an actual Add is generated that matches the current liquid crystal response state and the driver IC output state. Simultaneously, by processing liquid crystal compensation and driving compensation in layers, the two most common problems encountered when the target driving is directly output are solved: under-adjustment or over-adjustment caused by liquid crystal response deviation, and driving distortion or channel inconsistency caused by driver IC output limitations. This provides a clear, stable, and traceable input basis for the generation of the actual dynamic wavefront, actual wavefront, and wavefront residual in step S6. In practical applications: when When a certain dynamic responsibility region of a particular eye requires a large target Add, and the corresponding high-performance hybrid liquid crystal experiences a decrease in response slope at the current temperature, and some channels of the driver IC exhibit gain deviation, the liquid crystal compensation module first reads the target drive for that responsibility region. Based on the current liquid crystal response delay, phase response slope, and temperature-related response value, it increases the liquid crystal drive amplitude for that responsibility region and corrects the output timing. Then, the drive compensation module performs amplitude clipping, quantization, and channel correction on the liquid crystal drive based on the driver IC's output voltage range, output resolution, and channel gain deviation, forming a compensated drive. Finally, the compensated drive is output to the corresponding responsibility region of the dynamic Add module, and the actual Add is generated based on the actual phase adjustment amount formed by the compensated drive under the current liquid crystal response relationship, for subsequent correction modules to continue to use.

[0024] S6. Input the actual Add, target wavefront and correction mapping table into the correction module, generate the actual dynamic wavefront based on the actual Add, superimpose the actual dynamic wavefront with the static wavefront to generate the actual wavefront, compare the actual wavefront with the target wavefront point by point to obtain the wavefront residual, correct the dynamic items in the correction mapping table based on the wavefront residual, and output the combined result and traceability data packet. In this specific embodiment, step S6 is used to perform closed-loop correction on the current combined correction result based on the actual Add obtained in step S5, and write the correction result back to the correction mapping table, thereby forming a combined result and traceability data package that can be used for further iteration. This process does not only record the actual Add, but first generates an actual dynamic wavefront corresponding to the actual output of the current dynamic Add module based on the actual Add, then superimposes the actual dynamic wavefront with the static wavefront to obtain the actual wavefront, and then compares the actual wavefront with the target wavefront point by point under the same reference plane, the same coordinate system and the same sampling point matrix to obtain the wavefront residual that can directly characterize the current deviation distribution. Finally, the dynamic items in the correction mapping table are directionally corrected based on the wavefront residual, and the correction result and related process fields are written into the traceability data package so that the correction result can be directly inherited when called in the future. To ensure consistency between the steps, the actual dynamic wavefront, actual wavefront, target wavefront and wavefront residual in this step all adopt the same sampling format as in step S3, and all maintain a corresponding relationship with the eye identifier, responsibility area identifier and calling batch. This implementation process includes the following steps: In S6-1, the actual Add and the correction mapping table are input into the dynamic correction module. The actual dynamic wavefront is generated based on the actual Add, and the actual dynamic wavefront group is output. Specifically, the dynamic correction module first reads the actual Add output in step S5, storing it according to eye identifier and area of ​​responsibility identifier. Then, it reads the correction mapping table output in step S4, which includes at least dynamic items, boundary fields, area of ​​action fields, and eye identifier fields. During generation, the dynamic items and area of ​​action fields corresponding to the actual Add are extracted from the correction mapping table according to the eye identifier. Then, a computational grid for the actual dynamic wavefront is established on the unified reference surface and unified sampling lattice used in step S3. Subsequently, the actual Add is used as the actual adjustment amount for the current area of ​​responsibility, and the actual adjustment amount is... The action area corresponding to the dynamic item is written, and the distribution boundary of the actual dynamic wavefront is limited by the boundary field in the correction mapping table, so that the actual dynamic wavefront forms a valid phase value only within the current dynamic responsibility area. Sampling points that do not belong to the current dynamic responsibility area are kept null or written with zero values. If there are multiple dynamic responsibility areas for a single eye, the actual phase distribution of the corresponding area is generated according to the actual Add of each responsibility area, and then written and merged into the actual dynamic wavefront of that eye according to the same sampling point matrix. When processing the left and right eyes, the actual dynamic wavefronts of the left eye and the right eye are formed respectively, and then they are combined into an actual dynamic wavefront group. In this way, the actual Add is converted into an actual dynamic wavefront that is consistent with the current responsibility boundary, consistent with the current eye, and can directly participate in superposition and comparison. In S6-2, the actual dynamic wavefront group and static wavefront are input into the wavefront synthesis module. The actual dynamic wavefront group and static wavefront are superimposed to generate the actual wavefront, and the actual wavefront group is output. Specifically, the wavefront synthesis module first reads the actual dynamic wavefront in the actual dynamic wavefront group, and then reads the static wavefront generated and retained in step S3-1. The static wavefront and the actual dynamic wavefront are under the same reference plane, the same coordinate system, and the same sampling point matrix. During superposition, the actual dynamic wavefront and static wavefront of the same eye are matched according to the eye identifier. Then, the static phase value and the actual dynamic phase value at the same sampling position are superimposed point by point to obtain the actual wavefront value at that sampling position. If a certain sampling position is only located at... In the static responsibility area, the actual wavefront value is the phase value of the static wavefront; if a sampling position is located in the dynamic responsibility area, the actual wavefront value is the phase value of the superposition of the static wavefront and the actual dynamic wavefront; if a sampling position is simultaneously located in the corrected overlap area, the corresponding superposition value is still written according to the final responsibility allocation result in the corrected mapping table; after completing the point-by-point superposition for one eye, the actual wavefront of that eye is output; when processing the left and right eyes simultaneously, the actual wavefronts of the left and right eyes are output respectively, and they are combined into an actual wavefront group; after this processing, the actual wavefront group directly represents the true combined correction result under the combined action of the current static prescription lens, the current dynamic Add module output, and the current responsibility boundary; In step S6-3, the actual wavefront group and the target wavefront are input into the residual calculation module. The actual wavefront group and the target wavefront are compared point by point to obtain the wavefront residuals, and the residual group is output. Specifically, the residual calculation module first reads the actual wavefront in the actual wavefront group, and then reads the target wavefront output and retained in step S3-4. The actual wavefront and the target wavefront must be located on the same reference plane, the same coordinate system, and the same sampling point matrix. If the sampling start point, sampling interval, or region number of the two are inconsistent, they are first aligned according to the sampling format in step S3, and then point-by-point comparison is performed. During point-by-point comparison, the actual wavefront value and the target wavefront value are read one by one according to the eye identifier and sampling position, and the difference between the two is calculated at each sampling position to obtain the wavefront residual value at that sampling position. The wavefront residual values ​​retain both the magnitude and direction of the residuals to distinguish whether the actual wavefront is larger or smaller than the target wavefront. After calculating the difference for all sampling locations, the wavefront residual values ​​at each sampling location are summarized according to the responsibility area identifier to form a residual group. The residual group includes at least the sampling location field, residual magnitude field, residual direction field, responsibility area identifier field, and eye identifier field. If multiple consecutive sampling locations within a responsibility area show residuals in the same direction, an additional area continuity identifier is written into the residual group to determine whether the responsibility area belongs to a local deviation or an overall deviation during subsequent dynamic item correction. In this way, the wavefront residual is no longer a single total error, but is refined into residual distribution data that can directly point to the specific responsibility area and specific sampling location. In S6-4, the residual group and correction mapping table are input into the mapping correction module. The module corrects the dynamic items in the correction mapping table based on the residual group and outputs the combined result and traceability data packet. Specifically, the mapping correction module first reads the residual distribution of each responsibility area in the residual group, then reads the dynamic items, boundary fields, and drive association fields in the correction mapping table. Afterwards, it maps the residual group to the correction mapping table area by area according to the eye identifier and responsibility area identifier. During correction, it first determines the main distribution direction of the residual group within a certain responsibility area: when the residual direction at most sampling positions within the responsibility area indicates that the actual wavefront is lower than the target wavefront, it increases the Add allocation value in the dynamic item of the responsibility area or expands the effective area of ​​the responsibility area; when the residual direction at most sampling positions within the responsibility area indicates that the actual wavefront is higher than the target wavefront, it decreases the Add allocation value in the dynamic item of the responsibility area or shrinks the effective area of ​​the responsibility area; when the residuals are mainly concentrated in the responsibility boundary neighborhood, it prioritizes correcting the boundary fields corresponding to the dynamic items. Alternatively, it can drive the associated fields to make the connection between the responsibility area and adjacent responsibility areas smoother; if a responsibility area has a regional continuity identifier in the residual group, the dynamic items of the responsibility area are uniformly corrected according to the overall correction method; if a responsibility area only has residuals at local sampling positions, the dynamic item fields corresponding to that locality are adjusted only according to the local correction method; after completing the dynamic item correction of all responsibility areas, the combined result after this round of correction is obtained, and the combined result includes at least the updated dynamic items and the updated correction mapping table; at the same time, the eye identifier, responsibility area identifier, actual Add, actual wavefront identifier, target wavefront identifier, residual group identifier, and corrected dynamic item identifier corresponding to this round of call are written into the traceability data packet, so that the correction trajectory of this round can be directly read during subsequent recalibration, rewrite, or after-sales correction; if the left and right eyes are processed at the same time, the left eye traceability data packet and the right eye traceability data packet are formed respectively, or stored separately in the same traceability data packet by the eye identifier field; Through the above processing, step S6 completes the closed-loop correction process from actual Add to actual dynamic wavefront, actual wavefront, wavefront residual, and dynamic term correction. This allows the responsibility boundary and compensation driver formed in steps S4 and S5 to accept reverse correction from the actual output results, thus avoiding the entire combined correction process remaining at the level of single modeling and single compensation. Simultaneously, by comparing the actual wavefront and target wavefront point-by-point on a unified sampling basis and directly mapping the residual set to the dynamic term in the correction mapping table, three key problems are solved: the actual output results cannot be accurately written back to the dynamic responsibility area, the wavefront deviation cannot be located to the specific responsibility area, and the correction results cannot be continuously inherited to the next round of calls. This makes the combined results iterable and the traceability data packets inheritable. In practical applications: when a certain After a user completes a near / far handover, the dynamic Add module outputs the actual Add according to the compensation drive. However, if the user still experiences local undercorrection in the vicinity of the module center, the dynamic correction module first generates the corresponding actual dynamic wavefront based on the actual Add and correction mapping table of the responsible area. Then, the wavefront synthesis module superimposes the actual dynamic wavefront with the static wavefront to obtain the actual wavefront. Subsequently, the residual calculation module compares the actual wavefront with the target wavefront point by point on the same sampling array and finds that there are continuous residuals in the same direction in the vicinity of the module center. Finally, the mapping correction module performs directional correction on the Add allocation value and the area of ​​action in the dynamic item of the corresponding responsible area based on the continuous residuals, and writes the actual Add, actual wavefront, residual distribution and corrected dynamic item of this round into the traceability data packet for direct use in the next round of combined correction.

[0025] Working principle: First, the user's prescription information, wearing information, and lens fitting information are unified into a parameter set under the same coordinate reference. Then, the complete correction task is split into two parts: one part is handled by the static prescription lens, forming a static item, and the other part is handled by the dynamic Add module, forming a dynamic item. The boundaries of the areas each is responsible for are further determined, generating a mapping table. Next, based on the mapping table, static and dynamic wavefronts are generated separately. Then, combined with the actual offset relationship between the optical center and the module center, a target wavefront that more closely matches the actual fitting state is obtained. Then, instead of directly driving the dynamic Add module, the system first generates evidence based on the target wavefront, static wavefront, dynamic wavefront, and offset to determine which areas should adjust the static item and which areas should adjust the dynamic item. The dynamic terms are adjusted, and the responsibility boundary is written back to generate a new target driver. During driver execution, the process is divided into two layers: liquid crystal compensation and driver compensation, so that the final output is closer to the real state of the high-performance hybrid liquid crystal and the driver IC. Finally, the actual Add, actual dynamic wavefront, and actual wavefront are obtained based on the actual output, and compared with the target wavefront point by point to calculate the residual. Then, the dynamic terms and mapping table are corrected in turn, and a traceability data packet is generated. In other words, this scheme does not end after a one-time calculation, but is a closed-loop process of "acquisition - splitting - modeling - writing back - compensation - correction". The result of the previous step directly determines the input of the next step, and the correction of the next step will in turn correct the previous responsibility allocation. Therefore, it can continuously approach a more accurate combined correction result. For example, in a pair of ultra-thin modular adjustable glasses, if the user has both myopia and astigmatism, as well as presbyopia, the static prescription lens is responsible for correcting the basic refractive errors such as myopia and astigmatism, while the dynamic add module is only responsible for supplementing the focus when viewing near objects. However, in actual wear, the dynamic add module may not perfectly coincide with the optical center of the static prescription lens, and it is also affected by temperature changes, liquid crystal response changes, and driver IC output deviations. Therefore, the system first converts the prescription, pupillary distance, vertex distance, optical center, and module center onto the same reference plane, calculates which area each static lens and dynamic module should be responsible for, and then generates the target wavefront. If it is found that the actual assembly position of the dynamic module is slightly off, the system will... Through evidence assessment and responsibility boundary write-back, some areas are reassigned from static items to dynamic items, or vice versa. Then, a new target driver is generated, and after compensation based on the liquid crystal state and driver state, it is output to the dynamic Add module. The actual imaging result seen by the user is then converted into the actual wavefront and compared with the original target wavefront. If a certain area is still undercorrected or overcorrected, the system continues to correct the dynamic item of that area and records it. In this way, the glasses do not work in a fixed way in actual use, but will continuously fine-tune the division of labor between the static lens and the dynamic Add module according to assembly deviations, device states, and actual output results, making near-far switching more accurate, wearing more stable, and more suitable for engineering mass production and subsequent calibration.

[0026] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A combined correction algorithm of static prescription lenses and dynamic Add adjustable focus module, characterized in that, include: S1. Obtaining prescription, wearing, assembling, liquid crystal and driving, wherein the prescription includes spherical lens, cylindrical lens and axis, wearing includes interpupillary distance and vertex distance, and assembly includes optical center and module center, and forms a parameter set; S2. Input the parameter set into the splitting module, generate static items according to spherical lens, cylindrical lens and axis, and generate dynamic items according to target Add, pupil distance, vertex distance, optical center, module center, liquid crystal and drive, and form a mapping table; S3. Input the mapping table into the modeling module, generate a static wavefront based on the static items, generate a dynamic wavefront based on the dynamic items, generate an offset based on the coordinate difference between the optical center and the module center, and generate the target wavefront. S4. Input the target wavefront, static wavefront, dynamic wavefront, and offset into the write-back module, correct the static and dynamic items in the mapping table according to the offset, and output the corrected mapping table and target drive. S5. Input the target driver, LCD and driver into the compensation module, perform LCD compensation according to the LCD, perform driver compensation according to the driver, and output the actual Add; S6. Input the actual Add, target wavefront, and correction mapping table into the correction module. Generate the actual dynamic wavefront based on the actual Add. Superimpose the actual dynamic wavefront with the static wavefront to generate the actual wavefront. Compare the actual wavefront with the target wavefront point by point to obtain the wavefront residual. Correct the dynamic items in the correction mapping table based on the wavefront residual, and output the combined result and traceability data packet.

2. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 1, characterized in that: S1 includes: S1-1. Obtain the spherical lens, cylindrical lens, axis, pupil distance, vertex distance, optical center, and module center, and assemble them into the original parameter set; S1-2. Input the original parameter set into the coordinate conversion module, convert the optical center and the module center to the same reference plane according to the pupil distance and vertex distance, and output the center coordinate set. S1-3. Input the original parameter set and the center coordinate set into the parameter integration module, bind them according to the corresponding spherical lens, cylindrical lens, axis and center coordinate set, and output the parameter set.

3. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 2, characterized in that: S2 includes: S2-1. Input the parameter set into the static splitting module, generate static items according to spherical lens, cylindrical lens and axis, and output the static parameter group; S2-2. Input the parameter set and static parameter group into the dynamic splitting module, generate dynamic items according to the target Add, pupil distance, vertex distance, optical center, module center, liquid crystal and driver, and output the dynamic parameter group.

4. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 3, characterized in that: S2 further includes: S2-3. Input the static parameter set and dynamic parameter set into the boundary calculation module, calculate the allocation boundary of the static and dynamic items according to the coordinate relationship between the optical center and the module center, and output the boundary parameter set. S2-4. Input the static items, dynamic items, and boundary parameter sets into the mapping generation module, write the static items and dynamic items according to the boundary parameter sets, and output the mapping table.

5. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 4, characterized in that: S3 includes: S3-1. Input the mapping table into the static modeling module, generate static wavefronts based on the static items, and output the static wavefront group. S3-2. Input the mapping table into the dynamic modeling module, generate dynamic wavefronts based on the dynamic items, and output the dynamic wavefront group.

6. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 5, characterized in that: S3 further includes: S3-3. Input the mapping table, static wavefront group and dynamic wavefront group into the offset calculation module, generate the offset amount based on the coordinate difference between the optical center and the module center, and output the offset parameter group. S3-4. Input the static wavefront group, dynamic wavefront group, and offset parameter group into the synthesis module, generate the target wavefront based on the static wavefront group, dynamic wavefront group, and offset parameter group, and output the target wavefront.

7. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 6, characterized in that: S4 includes: S4-1. Input the target wavefront, static wavefront, dynamic wavefront, and offset into the evidence generation module. Generate static evidence, dynamic evidence, and offset evidence according to the difference between the target wavefront and the static wavefront, the difference between the target wavefront and the dynamic wavefront, and the offset, and write them into the evidence table. S4-2. Input the evidence table into the consistency judgment module, call the consistency verifier to cross-compare static evidence, dynamic evidence and offset evidence, generate consistency flags, conflict flags and confidence levels, and write them into the judgment table.

8. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 7, characterized in that: S4 further includes: S4-3. Input the decision table into the gating module, call the gating rule set to generate static write-back tokens, dynamic write-back tokens and state locks according to consistency flags, conflict flags and confidence levels, and write them into the gating table; S4-4. Input the gating table, mapping table and offset into the branch write-back module. Correct the range of static items in the mapping table according to the static write-back token, correct the range of dynamic items in the mapping table according to the dynamic write-back token, keep the unauthorized items unchanged according to the state lock, and output the branch mapping table.

9. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 8, characterized in that: S4 further includes: S4-5. Input the branch mapping table and decision table into the merging module, call the conflict resolver to redistribute the overlapping area of ​​static and dynamic items, perform a rollback and re-examination on the low confidence area and update the confidence, output the corrected mapping table and write it to the write-back record. S4-6. Input the correction mapping table, dynamic wavefront, and decision table into the driver generation module, generate the target driver according to the dynamic items, confidence levels, and write-back records in the correction mapping table, and write it into the driver table.

10. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 9, characterized in that: S5 includes: S5-1. Input the target driver and the LCD into the LCD compensation module, perform LCD compensation on the target driver according to the LCD, and output the LCD driver. S5-2. Input the LCD driver and the driver into the driver compensation module, perform driver compensation on the LCD driver according to the driver, and output the compensation driver. S5-3. Output the compensation driver to the dynamic Add module and generate the actual Add based on the compensation driver.

11. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 10, characterized in that: S6 includes: S6-1. Input the actual Add and correction mapping table into the dynamic correction module, generate the actual dynamic wavefront based on the actual Add, and output the actual dynamic wavefront group. S6-2. Input the actual dynamic wavefront group and the static wavefront into the wavefront synthesis module, superimpose the actual dynamic wavefront group and the static wavefront to generate the actual wavefront, and output the actual wavefront group.

12. The combined correction algorithm of static prescription lens and dynamic Add adjustable focus module according to claim 11, characterized in that: S6 further includes: S6-3. Input the actual wavefront group and the target wavefront into the residual calculation module, compare the actual wavefront group with the target wavefront point by point to obtain the wavefront residual, and output the residual group. S6-4. Input the residual group and the correction mapping table into the mapping correction module, correct the dynamic items in the correction mapping table according to the residual group, and output the combined result and traceability data packet.

Images