A view automatic generation method and system for indoor soft furnishing design

Abstract

The application discloses a kind of view automatic generation method and system for indoor soft decoration, it is related to visual technology field, the method includes by the material attribute database established in advance, obtain the aging characteristic data of multiple soft materials under different time nodes, material aging process and surface wear details are classified and stored, obtain the initialized material aging characteristic set, according to the initialized material aging characteristic set, obtain the visual effect data set after environmental adaptability adjustment, according to the visual effect data set after environmental adaptability adjustment, using real effect restoration mapping algorithm, generate dynamic visual effect image highly consistent with actual aging process;The view automatic generation method and system for indoor soft decoration, realize the high-precision simulation of soft material aging process, provide reliable basis for product design and service life evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of visualization technology, in particular to a view automatic generation method and system for indoor soft furnishing design. BACKGROUND

[0002] In the field of interior design and soft furnishing, simulating the appearance changes of materials to predict long-term use effects is a crucial research direction. This field not only concerns the practicality of design schemes, but also directly affects users' intuitive perception of space aesthetics and durability, providing important reference for design decisions.

[0003] However, current related methods are often limited to the presentation of the initial state of materials, ignoring the dynamic evolution of material surface texture and color under the time dimension, especially in the prediction of performance under complex environmental factors, lacking depth of data support and comprehensive analysis ability, leading to difficulty in accurately evaluating the long-term effects of design schemes. The core challenge in this field is how to realistically reproduce the appearance changes of soft furnishing materials in long-term use, and the two most critical technical factors are the simulation accuracy of the material aging process and the comprehensive consideration of environmental interaction. The material aging process involves multiple details such as surface wear and color fading, and if these subtle changes cannot be accurately captured, it is difficult to restore the real use traces. This problem further extends to the influence of environmental interaction, because material aging does not occur in isolation, but is affected by external conditions such as light, humidity, and friction. If these variables cannot be integrated for dynamic simulation, the final presentation will be out of touch with reality. These two factors are closely related and together constitute the technical barrier to simulating realistic visual effects. SUMMARY

[0004] The purpose of the present application is to provide a view automatic generation method and system for indoor soft furnishing design, which accurately simulates the appearance changes of soft furnishing materials such as cloth, wood, and metal in long-term use, and combines environmental factors to build dynamic visual effects.

[0005] To achieve the above purpose, the present application provides the following technical solution: a view automatic generation method for indoor soft furnishing design, the method comprising:

[0006] By pre-establishing a material attribute database, aging characteristic data of various soft furnishing materials at different time nodes is obtained, and the material aging process and surface wear details are classified and stored to obtain an initialized material aging characteristic set;

[0007] According to the initialized material aging characteristic set, an environment adaptability adjusted visual effect data set is obtained, and according to the environment adaptability adjusted visual effect data set, a real effect restoration mapping algorithm is used to generate dynamic visual effect images highly consistent with the actual aging process.

[0008] For dynamic visual effect images, the simulation accuracy of color fading characteristics and friction use traces is verified by comparative analysis technology to determine whether it meets the preset reality standard;

[0009] If the simulation accuracy does not meet the preset reality standard, the parameters of comprehensive data analysis are adjusted by iterative optimization technology to regenerate dynamic visual effect images until the requirements of real effect restoration are met.

[0010] Preferably, the environmental adaptability adjusted visual effect data set obtained according to the initialized material aging characteristic set comprises extracting the association pattern between surface wear details and color fading characteristics by using data mining technology, and determining the key parameters of appearance change simulation in the aging process.

[0011] Preferably, the environmental adaptability adjusted visual effect data set obtained according to the initialized material aging characteristic set further comprises obtaining relevant variable data of environmental interaction, including light condition effect, humidity environmental variable and dynamic record of friction use traces, and judging the weight distribution of environmental factors on aging characteristics.

[0012] Preferably, the environmental adaptability adjusted visual effect data set obtained according to the initialized material aging characteristic set further comprises adjusting the parameter value of appearance change simulation by multivariate regression analysis technology if the weight distribution of environmental factors exceeds the preset threshold range, to obtain the corrected appearance change characteristic matrix.

[0013] Preferably, the environmental adaptability adjusted visual effect data set obtained according to the initialized material aging characteristic set further comprises obtaining the time sequence data of long-term use prediction according to the corrected appearance change characteristic matrix, and determining the visual form of the material at different time nodes in combination with the rendering requirements of dynamic visual effect.

[0014] Preferably, the environmental adaptability adjusted visual effect data set obtained according to the initialized material aging characteristic set further comprises integrating the interactive influence of light condition effect and humidity environmental variable by comprehensive data analysis technology to obtain the environmental adaptability adjusted visual effect data set for the visual form of long-term use prediction.

[0015] Preferably, the material attribute database comprises texture images, color information, wear grade and environmental response coefficient of soft furnishing materials of different material types.

[0016] Preferably, the real effect restoration mapping algorithm adopts an image generation model based on convolutional neural network for image conversion and rendering of aging characteristics.

[0017] Preferably, the contrast analysis technique includes color deviation analysis and contour retention detection based on image difference calculation at multiple time points.

[0018] A view automatic generation system for indoor soft furnishing design is used to implement the steps of the view automatic generation method for indoor soft furnishing design, and the system comprises:

[0019] A material feature extraction module is used to obtain aging feature data of various soft furnishing materials at different time nodes through a pre-established material attribute database, and the material aging process and surface wear details are classified and stored to obtain an initialized material aging feature set.

[0020] A visual effect generation module is used to perform environment adaptability adjustment based on the material aging feature set, generate a visual effect data set, and generate a dynamic visual effect image highly consistent with an actual aging process by using a real effect restoration mapping algorithm.

[0021] A reality evaluation module is used to verify color fading features and friction usage traces in the dynamic visual effect image by using a contrast analysis technique, and determine whether they meet preset reality standards.

[0022] An image optimization module is used to adjust parameters of comprehensive data analysis by using an iterative optimization technique when the reality evaluation result does not meet preset standards, and regenerate a dynamic visual effect image until a real effect restoration requirement is met.

[0023] According to the technical solution, the present application has the following beneficial effects:

[0024] The view automatic generation method and system for indoor soft furnishing design obtain aging feature data by establishing a material attribute database, extract a correlation mode of surface wear and color fading features by using a data mining technique, determine key parameters of appearance change simulation, adjust parameter values by using multivariate regression analysis in combination with environmental factor influences, generate an appearance change feature matrix, generate a dynamic visual effect image by using a real effect restoration mapping algorithm based on time series data predicted for long-term use, integrate influences of environmental variables such as light and humidity, verify simulation accuracy by using a contrast analysis technique, and iteratively optimize parameters until a requirement is met. The present application realizes high-precision simulation of a soft furnishing material aging process, and provides a reliable basis for product design and service life evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 The present application is a method flowchart.

[0026] Figure 2 The present application is a system connection diagram. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0028] As shown in Figure 1 The present application provides a technical solution: a view automatic generation method for indoor soft furnishing design, which comprises:

[0029] Through a pre-established material attribute database, aging characteristic data of a plurality of soft furnishing materials at different time nodes are obtained, and the material aging process and surface wear details are classified and stored to obtain an initialized material aging characteristic set;

[0030] According to the initialized material aging characteristic set, an environment adaptability adjusted visual effect data set is obtained, and according to the environment adaptability adjusted visual effect data set, a real effect restoration mapping algorithm is used to generate a dynamic visual effect image highly consistent with the actual aging process;

[0031] For the dynamic visual effect image, the simulation accuracy of color fading characteristics and friction use traces is verified by comparative analysis technology, and it is judged whether it meets the preset reality standard;

[0032] If the simulation accuracy does not meet the preset reality standard, the parameters of comprehensive data analysis are adjusted by iteration optimization technology, and the dynamic visual effect image is regenerated until the real effect restoration requirement is met.

[0033] The view automatic generation method for indoor soft furnishing design provided by the present application is based on the collaborative processing of material attribute database, aging characteristic modeling, environment adaptability conversion and image generation algorithm, so as to realize the automatic generation of dynamic and real soft furnishing effect drawing. The specific implementation process is as follows:

[0034] Firstly, by establishing a material attribute database, a plurality of common soft furnishing materials are information modeled, covering basic information such as material name, manufacturing process, surface treatment method, typical use period, common aging performance, etc. On this basis, aging image samples at different time nodes are collected, and characteristics such as color, gloss, texture depth, crack shape and edge wear degree are quantified. The quantification method uses image analysis tools to analyze the original image at the pixel level, for example, by setting the region to calculate the color mean and standard deviation, texture distribution statistical frequency and directionality, etc., and arranging them into an aging trend model in time sequence.

[0035] The model is classified and stored as an initialized material aging feature set, which is stored in multiple dimensions according to the material (such as cloth, wood, metal, etc.) and aging factors (such as ultraviolet radiation, friction frequency, humidity influence), so as to facilitate subsequent calling. In the application scenario setting stage, the system receives user input or preset environmental parameters, including light intensity, indoor temperature and humidity, ventilation conditions, and people flow density, and adjusts the aging rate through parameter mapping rules. For example, the increase of ultraviolet intensity will significantly accelerate the color fading, so in the high light environment, the corresponding aging features will appear in advance in the image simulation. After all the parameter inputs, the environmental adaptability conversion model is processed to obtain a set of environmental adjusted visual effect data set, which contains the performance weight and change progress of each aging feature in the target environment.

[0036] Subsequently, the image mapping stage is entered. The system calls the mapping algorithm to convert the environmental adaptability data into visual image data. The mapping algorithm takes image units (such as image blocks or pixel blocks) as units, controls the changes of visual factors such as color channel, contrast, texture distribution and transparency through the corresponding relationship between aging feature values and image rendering templates. For example, if a cloth material fades by 20% of the original saturation at the 8th month, the mapping algorithm will reduce the original color channel value by 20%, and then add appropriate lightness adjustment and random texture spots to simulate usage marks. The decrease of glossiness is simulated by increasing the diffuse reflection component and reducing the specular reflection value, realizing the attenuation of image brightness and reflection effect.

[0037] The above mapping process needs to set multiple thresholds to determine whether to present the features, such as whether the cracks appear, whether the color change is perceptible, etc. These thresholds are set according to the experimental data regression results, for example, the color perceptible change threshold is set to the minimum distinguishable value of the color difference under human eye perception, which is about the change value greater than a certain interval in the unit color space; the crack presentation threshold is visualized when the image texture density exceeds a certain percentage. All thresholds can be fine-tuned and optimized according to specific material categories to ensure that the simulation image is both realistic and distinguishable.

[0038] After the image generation is completed, the system introduces a contrast analysis technique to evaluate the simulation image and the standard aging sample. This process is achieved through three steps: first, color channel comparison, which compares the simulation image with the standard image at the corresponding time node in the color space pixel by pixel, and calculates the color difference mean and standard deviation; second, texture similarity analysis, which uses image feature extraction technology to obtain edge, contour, and repeated texture information, and judges whether the overall appearance is consistent through the structural similarity index; third, wear distribution matching, which divides the surface trace position, shape, and density into geometric regions and compares them block by block to determine whether the wear position and degree are consistent with the real sample. The above three types of indicators must meet the preset comprehensive similarity threshold, and if any of them is below the standard, it is considered that the simulation accuracy is insufficient.

[0039] If the discrimination result does not meet the standard, it enters the iterative optimization stage. This stage adjusts the key control factors to improve the simulation quality, mainly including material aging rate factor, environmental adaptation weight coefficient, and image mapping parameters. The adjustment method is as follows: the aging rate factor is stretched or compressed on the corresponding interval according to the judgment result of the simulation image aging speed being too slow or too fast; the environmental adaptation weight adjusts the proportion of each environmental factor in the model according to the sensitivity deviation of the simulation image to environmental changes; the image mapping parameters such as color correction factor, gloss suppression coefficient, and texture disturbance amplitude are adjusted proportionally according to the difference between the current image and the standard sample. For example, if the image color is too bright, the color brightness control parameter is lowered; if the wear distribution is too concentrated, the disturbance range of the texture distribution control function is adjusted. After each adjustment, the image is regenerated and the comparison and analysis are repeated until all indicators meet the authenticity standard, forming the final output image.

[0040] This method can automatically and efficiently simulate the aging evolution process of soft furnishing materials in actual use, providing designers with real and reliable visual references and improving the scientificity and reliability of scheme design. At the same time, by combining environmental adaptability factors and mapping algorithms, the authenticity and dynamic response capability of image restoration are significantly improved, making the design effect more predictable and forward-looking. In addition, the iterative optimization mechanism further enhances the accuracy and stability of image simulation, ensuring that the generated rendering has full reference value in various design decisions.

[0041] According to the initialized material aging characteristic set, the environmental adaptability adjusted visual effect data set includes extracting the correlation pattern between surface wear details and color fading characteristics using data mining technology to determine the key parameters of appearance change simulation in the aging process.

[0042] The embodiment deepens the key link of generating environment adaptive visual effect dataset, focusing on identifying the internal correlation pattern between surface wear details and color fading characteristics from the initialized material aging feature set through data mining technology, and accurately extracting the key parameters required for appearance change simulation during the aging process based on this.

[0043] The specific implementation process includes the following steps:

[0044] Data preparation. The system calls the initialized aging feature set in the material attribute database, which covers the visual performance data of the material at different aging stages. Each sample entry includes at least: sample material type, time node, color value change trajectory, surface texture image, wear area morphology, observation angle and environmental background parameters, etc. These information includes both image data and structured numerical data, which need to be standardized before entering the mining process, including image format unification, resolution adjustment, numerical normalization and missing value filling, etc.

[0045] Feature extraction. The system analyzes the texture edge, wear position, color area distribution, etc. in the sample image through the image recognition module. For example, the wear area in the image is partitioned, and its area ratio, edge shape, texture density, etc. are extracted; for the color change part, the change trend of color brightness, saturation and hue in time series is extracted. The extracted features form a structured vector, and each sample is converted into descriptive data containing dozens to hundreds of dimensions, which is used for subsequent modeling.

[0046] Data mining modeling. The system uses multiple data mining techniques for analysis. The clustering method is used to divide the samples into several aging evolution pattern groups, for example, samples with rapid texture loss and obvious color change are classified into one category; then through correlation analysis, the change curves of color fading characteristics and surface wear are compared to determine whether there is a trend synchronization, driving or proportion dependence between them. After identifying the potential relationship, the system further uses the frequent pattern extraction technology to extract the repeated change patterns from large-scale samples, such as "after 6 months, the corner area appears obvious wear and the blue color decreases by more than 30% of the original value".

[0047] Key parameters were determined. The high-confidence and representative change patterns from the above analysis results were transformed into actionable control parameters. These parameters include, but are not limited to: the response coefficient of color change, the priority of wear image template selection, the gradient image overlay area, the limitation of texture change amplitude, and the aging feature activation threshold. Each parameter has its corresponding physical meaning and image representation logic, ensuring the interpretability and consistency of subsequent visual image generation. For example, if the analysis results indicate that color change in wooden furniture begins within two months of the appearance of texture fracture, a time delay parameter for color change activation is set.

[0048] The parameter set is output and passed to the image generation module. The system writes the above key parameters into the configuration file or embeds them into the image generation engine to guide the subsequent image rendering process, so that the generated image is more consistent with the actual aging process and material evolution logic in appearance.

[0049] This entire process is driven by real data, without relying on subjective settings, thus exhibiting good objectivity and adaptability, and allowing for continuous iterative optimization of model performance. By regularly updating and re-mining the sample library, the system can adapt to the simulation needs of different materials, new and old products, and changing environments, enhancing the scientific rigor and intelligence of interior design prediction.

[0050] By introducing a data mining mechanism, the relationship between wear and aging is no longer based on manual experience. Instead, it automatically learns the evolutionary patterns through a large number of real samples, improving the objectivity and accuracy of the modeling. The resulting key parameters accurately reflect the coupled evolution of various visual features of different materials during the aging process, significantly enhancing the realism and scientific rigor of the image simulation. Simultaneously, this process is highly scalable, supporting the continuous introduction of new samples to optimize the algorithm model, ensuring the system's long-term adaptability to the aging simulation needs of different materials and environments.

[0051] Based on the initial set of material aging characteristics, the visual effect dataset after environmental adaptation adjustment is obtained. It also includes key parameters for simulating appearance changes, and obtains relevant variable data on environmental interaction effects, including dynamic records of light conditions, humidity environmental variables and friction usage traces, to determine the weight distribution of environmental factors on aging characteristics.

[0052] In this embodiment, to further improve the environmental adaptability of visual simulation of material aging, the system adds a functional module for environmental interaction analysis based on the existing key parameter extraction. This module collects and analyzes the dynamic influence of variables such as light, humidity, and friction traces during the material aging process, constructing a quantitative correlation mechanism between environmental factors and aging characteristics, thereby accurately determining the weight distribution of each environmental factor in the visual change simulation. The entire implementation process is mainly divided into the following four stages:

[0053] Environment data collection phase:

[0054] The system is configured with multi-source environment data access interface, which can collect environment interaction related variables from indoor sensor devices, external environment database, user input configuration or historical sample data. Mainly including the following three types of key variables: light condition effect data, including daily sunshine duration, maximum illumination, light angle change range, light source type (such as natural light, artificial light) and intensity change curve; humidity environment variable data, including average humidity value, relative humidity daily change amplitude, humidification time period distribution and condensation frequency; Dynamic record of friction use traces, including contact area distribution, use frequency, contact duration, movement path and friction directionality. All data are stored in chronological order to form a time series variable set, which is synchronized with the material aging image sample to ensure the consistency and timeliness of modeling.

[0055] Variable and aging feature correlation modeling phase, the system adopts a feature-based modeling strategy, and establishes statistical relationship models for different types of aging performance (such as color fading, texture wear, gloss decline, etc.). Taking color fading as an example, the system extracts the channel value change trend of material color at multiple aging stages, and performs corresponding analysis with the light intensity and illumination angle data at the same period. Through time window sliding, change rate analysis and other techniques, the influence delay time and intensity influence range of light variables on color change are determined; for texture damage, the system combines humidity change record and friction path record to analyze the extreme value change of humidity fluctuation before and after crack generation and the texture fracture probability of friction concentrated area.

[0056] Environment factor weight distribution phase, after completing the variable correlation analysis, the system generates a set of weight coefficients for each aging feature, representing its sensitivity to different environmental variables. For example, for a wooden furniture, if the contribution of humidity to the formation of surface cracks is significantly higher than that of light, the system will set a higher weight for the humidity variable in the crack generation model, and the light influence weight will be correspondingly reduced. The distribution of weights refers to the frequency, change amplitude and duration of variable fluctuation and aging feature synchronous change in historical samples, and is normalized and output in percentage form to drive feature control in the image generation process.

[0057] The result is integrated with the image generation driving stage, all weight information and aging feature control parameters are embedded in the visual effect dataset template. In the image generation process, the system calls the environmental weight configuration in real time, dynamically controls the activation time, influence area and form of each feature. For example, when the simulation image needs to simulate the fabric sofa in the high humidity scene, the system will load the wear pattern template dominated by humidity in advance, and enhance the discoloration degree and superposition level of rough texture in the corner area during image generation. If there is also a high friction use record, the system will further simulate fabric pilling and fuzzy edge effect in the frequently used area, forming a dynamic image that conforms to the real environment interaction law.

[0058] This embodiment deeply integrates environmental data and material aging behavior, realizes the whole-chain modeling from external influencing conditions to visual simulation output, and effectively enhances the adaptability and credibility of simulation images to different application scenarios.

[0059] By introducing environmental interaction factors and establishing their weight relationship with appearance changes, the response ability of the visual effect dataset to real use situations is significantly enhanced, so that the simulation image is no longer a static template superposition, but an evolution process driven by environmental behavior. This mechanism can significantly improve the accuracy and reliability of the prediction image, and help designers to predict the performance trend of soft materials in different environments in the early stage, so as to optimize material selection and collocation strategy, and reduce the risk of later maintenance and replacement.

[0060] According to the initialized material aging feature set, the environment adaptability adjusted visual effect dataset also includes if the weight distribution of the environmental factor exceeds the preset threshold range, the parameter value of the appearance change simulation is adjusted through multivariate regression analysis technology, and the corrected appearance change feature matrix is obtained.

[0061] In this embodiment, the system further introduces a parameter adjustment mechanism based on multivariate regression analysis on the basis of completing the weight analysis of environmental factors. When it is detected that the influence weight of a certain environmental factor on the appearance aging feature obviously exceeds the preset reasonable range, the system will automatically correct the related simulation parameters to obtain a visual effect more consistent with the actual aging process. This process has high automation, data-driven and adaptive adjustment capabilities, mainly including the following five steps:

[0062] Weight distribution monitoring: After extracting the weight relationship between each material aging feature and environmental factors, the system compares the set of weights with the pre-set environmental weight threshold range. This threshold range is set based on historical data statistics, material performance characteristics, and industry experience. For example, in the simulation of the surface darkening process of leather-like soft furnishing materials, the system presets the weight of light intensity to be between 30% and 45%, and the humidity influence should not exceed 25%. When the detection results show that the light weight is higher than 50% and the humidity weight is lower than 10%, etc. abnormal conditions, the system determines that the current weight distribution is abnormal, and enters the next step of correction process.

[0063] Historical sample backtracking analysis: To reasonably adjust the simulation parameters, the system first retrieves a large number of historical aging sample data from the database. Each sample data contains material type, environmental variable value, aging stage label, image feature value, etc. The system filters out the comparable sample group according to the material type matching standard with the current abnormal sample, and extracts the environmental variable value and image performance index corresponding to the current aging feature to form the analysis base set.

[0064] Multivariate regression modeling: Based on the above historical sample set, the system builds a multivariate regression model to predict the reasonable parameter interval of material aging features under specific environmental variable combinations. The model considers the synergistic effect of multiple variables, such as the combined influence of light intensity, humidity fluctuation, and friction frequency on color fading. Through the regression model, the system can obtain the trend curve and fitting result of the change of aging feature with the change of each environmental variable, thereby deducing the target value of the aging simulation parameters under the current abnormal weight condition.

[0065] Parameter correction and matrix update: According to the output results of the regression model, the system adjusts the simulation parameters of the current aging feature one by one, including color channel change amplitude, texture clarity decline coefficient, wear range boundary expansion degree, etc. All adjusted parameters are organized into a new appearance change feature matrix. This matrix records the change values of each aging visual indicator in each time period, spatial region, and environmental state, and is the basic data structure for guiding subsequent image rendering and dynamic effect generation. The corrected matrix has higher physical rationality and environmental adaptability, which can significantly improve the realism of image simulation.

[0066] Simulation output and model iteration: After correction, the system transmits the updated appearance change feature matrix into the image generation module to render a new image that meets the correction results in real time. If the user or the system identifies that the simulation image still does not meet the expected standard, the system can perform iterative analysis based on the current regression model parameters, form a new regression prediction and parameter correction path, and continue until the pre-set image evaluation standard is met.

[0067] The entire process can realize automatic judgment, automatic correction and closed-loop control, ensuring that even in extreme or atypical environmental conditions, the material aging simulation image with logical consistency, physical rationality and visual realism can be output.

[0068] By constructing a dynamic threshold judgment and regression adjustment mechanism, the damage of abnormal distribution of environmental factors to the image realism can be effectively avoided. This mechanism not only enhances the adaptive ability of the system and improves the control ability of the algorithm on the simulation accuracy under extreme environmental conditions, but also ensures that the material aging simulation results have high consistency and predictability, providing more stable technical support for indoor soft decoration design evaluation.

[0069] According to the initialized material aging characteristic set, the environment adaptability adjusted visual effect data set further includes obtaining time sequence data predicted for long-term use according to the corrected appearance change characteristic matrix, combining the rendering demand of dynamic visual effect, and determining the visual form of the material at different time nodes.

[0070] The embodiment further integrates the dynamic visual rendering demand and material use cycle prediction mechanism on the basis of the corrected appearance change characteristic matrix, and constructs a dynamic modeling system that can truly simulate the visual form of the material at different use time nodes. The key technical path of the system can be divided into the following six stages:

[0071] Dynamic rendering demand identification, before starting the image simulation task, the system identifies the required dynamic visual effect type according to user demand or design task target. These requirements usually include the visual state display of the material at key nodes such as 1 month, 3 months, 6 months, 12 months, 24 months, whether the time interval granularity is once a month or once a quarter, and the rendering frame number and performance detail requirements of each stage. For example, for a commercial soft decoration effect picture for display, it may be required to generate 1 frame of image per month to continuously show the visual changes in the use process.

[0072] Prediction time axis construction, after the rendering time dimension is determined, the system constructs a prediction time axis covering the required time length. The time axis will be divided into several time nodes, and each node represents a key use cycle. The time axis not only marks the time point sequence, but also records the environmental input parameters associated with the time point, such as whether the 3rd month is in the high humidity season, whether the 9th month is exposed to high intensity sunlight, etc.

[0073] Data fusion and trend modeling, the system calls the corrected appearance change feature matrix, and simultaneously obtains the time series change data of the relevant material samples in the historical material aging database. These data include the change trend of material color value over time, the change sequence of texture damage level, the glossiness decline curve, etc. The system uses historical data to guide the change trend of the current material at each node on the prediction time axis through trend modeling, ensuring that the aging prediction conforms to the existing statistical rules.

[0074] Time series feature interpolation and dynamic parameter generation, under the guidance of the prediction model, the system performs interpolation operation on the parameters between two known time points to generate complete parameter groups corresponding to each time node. These parameters include color channel change value, texture density adjustment value, edge blur degree, and reflectivity reduction value at each time point. The interpolation process considers the influence of material physical properties, aging sensitivity, and environmental input variables, such as humidity mutation causing accelerated texture change and increased light causing premature color decay.

[0075] Multi-stage image rendering and expression, after completing parameter generation, the system calls the image rendering engine in chronological order to render the corresponding visual images at each time node. These images reflect the visual performance of the material at different aging stages, including color darkening, edge wear, pattern blurring, and surface roughness rising, ensuring that the images can truly reflect the cumulative effects of aging caused by time progression. The rendering results can be organized into a dynamic image sequence or output by stage, meeting the use requirements of different display platforms or evaluation tools.

[0076] Rendering quality check and time sequence consistency verification, to ensure the reasonableness and continuity of the dynamic image sequence in terms of vision, the system performs consistency evaluation on the images of adjacent time points after rendering. This includes whether the color change is smooth, whether the texture evolution is coherent, and whether the wear range expansion is consistent with the use logic. If a sudden change or abnormality is found at a certain node, such as color instant rebound or unnatural texture fracture, the system will backtrack the corresponding node parameters and re-execute interpolation calculation and image rendering until the set standard is met.

[0077] Through the above complete process, the embodiment realizes the output of material visual simulation images with time evolution characteristics from corrected parameters, through time modeling and dynamic rendering, providing high-precision, multi-stage, and full-process visual reference support for users.

[0078] The embodiment introduces a prediction mechanism with time dimension and a dynamic visual rendering model, making material aging simulation dynamic from static, not only showing the aging result at a specific time, but also simulating the visual change process in the complete use cycle, providing rich and visual prediction basis for indoor soft furnishing design scheme. It has important practical value in soft furnishing material selection, replacement cycle planning, and user visual expectation management.

[0079] According to the initialized material aging characteristic set, the environment adaptability adjusted visual effect data set further includes a visual form predicted for long-term use, and the environment adaptability adjusted visual effect data set is obtained by integrating the interaction of illumination condition and humidity environment variables through comprehensive data analysis technology.

[0080] The embodiment further proposes an environment factor interaction modeling mechanism on the basis of long-term use prediction visual performance modeling, aims to mine the coupling relationship between illumination condition and humidity environment variables through comprehensive data analysis technology, and accurately embeds the relationship in the simulation modeling of the material aging process, so as to improve the accuracy and dynamic adaptability of the environment adaptability adjusted visual effect data set.

[0081] The specific technical implementation process includes the following six steps:

[0082] Multi-source environment data acquisition and processing, the system first acquires long-term illumination and humidity data under the use scene of target space or material based on sensors, design input parameters, building physical models or historical environment databases. The illumination data mainly includes sunshine duration, average daily illumination, maximum incident angle and intensity distribution, and the humidity data includes average relative humidity, daily fluctuation amplitude, humid duration and periodic change characteristics. All data are recorded according to hour-level or day-level time granularity to form a time series environment variable database.

[0083] Data standardization and synchronous pairing, since illumination and humidity belong to different physical variables with different dimensions and change frequencies, the system aligns the two types of variables at the same time node through time dimension unification. For example, the average value of illumination intensity from 8 o'clock to 20 o'clock is corresponded to the average value of humidity throughout the day, and missing data is filled and abnormal values are removed, forming a unified format of environment variable table for subsequent interaction analysis.

[0084] Environment interaction behavior characteristic extraction, on the basis of standardized data, the system adopts partition statistics and joint distribution analysis method to classify and process the combined state of illumination and humidity. Specifically, the illumination intensity is divided into low, medium and high levels, and the humidity level is also divided into low, medium and high levels, thereby constructing nine typical environment combination states. The system statistically analyzes the aging performance indicators of related materials under each combination state, such as color brightness decline amplitude, texture roughness growth rate, gloss attenuation degree, etc., and extracts the joint influence characteristics of interaction variables on visual changes.

[0085] Based on the above-mentioned extracted interaction data, a composite weight factor model under the coupling of light and humidity is constructed. This model is used to quantify the influence degree of different environmental state combinations on the visual performance of the material. For example, when it is detected that the color attenuation rate is increased by more than 30% and the crack propagation speed is increased by more than 25% under the condition of "high light plus high humidity", the system will assign a higher aging influence weight to this combined environment, and establish a mapping rule to guide the dynamic adjustment of subsequent feature parameters.

[0086] Based on the composite weight factor model, the system integrates and adjusts the visual feature parameters originally based on a single light or humidity variable to generate a new parameter control set. This parameter set includes color change response coefficient, texture fracture expansion factor, and reflectivity adjustment factor, etc. Each parameter is proportionally corrected according to the comprehensive environmental state of the current material, and finally an environmental adaptability stronger visual feature data structure is generated.

[0087] The integrated visual feature parameter set is written into the visual effect data set adjusted for environmental adaptability, which is used to guide subsequent image rendering and dynamic aging simulation. When the image generation module calls this data set, the system can automatically switch or fine-tune the material visual performance form according to the current or preset environmental state, so as to generate images that are logical and have strong realism under different environmental combinations. For example, the image generated under the "medium light and low humidity" environment has small color difference and high texture retention; while in the "high light and high humidity" scene, the system automatically enhances the color difference in dark areas and adds edge blur texture, showing high intensity aging effect.

[0088] This complete process realizes the whole chain linkage from data collection, interaction modeling, parameter adjustment to visual generation, effectively improving the response ability and simulation accuracy of aging simulation under multiple environmental conditions.

[0089] By introducing the interactive analysis mechanism of light and humidity, the fitting error problem caused by the isolated treatment of environmental factors in traditional modeling is effectively compensated. This scheme can comprehensively consider the complex effect between environmental variables when predicting the evolution of material appearance, thereby improving the credibility and detail restoration of image simulation, and providing more accurate data support for material design optimization, durability evaluation and user expectation management.

[0090] The material attribute database includes texture images, color information, wear grades and environmental response coefficients of different types of soft furnishing materials.

[0091] In this embodiment, the construction process of the material attribute database is based on comprehensive data collection and standardization processing of different types of soft furnishing materials. First, the collection of texture images uses high-definition photography equipment to shoot the material surface under standard light source, ensuring that the image details truly reflect the surface structure of the material, including its texture, glossiness and roughness. The image file is formatted and size-standardized through image processing tools to meet the reading and rendering needs of different view generation engines.

[0092] Secondly, the color information is obtained by a spectrophotometer or a high-precision image analysis system to extract the color values of the material sample under standard white light illumination. A unified color calibration tool is used during the collection process to ensure data consistency. The system binds the color data of each material sample with its texture image and stores it as part of the material attribute record to support subsequent color matching and simulation rendering in soft furnishing views.

[0093] Then, the wear grade evaluation is based on the performance test results of the material under simulated use conditions. For example, through physical tests such as friction experiments and compression experiments, the degree of wear of the material in different use cycles is evaluated, and combined with an expert scoring system, the material is divided into five grades, corresponding to "brand new", "slight wear", "moderate wear", "heavy wear" and "severe wear". This grade information will be stored together with the texture image and color information to present the real performance of the material in use in the view.

[0094] Finally, the environmental response coefficient refers to the change trend of the material's physical or visual attributes under different environmental conditions, such as whether the material's color fades or the texture changes in high temperature, high humidity or strong light environments. The method to obtain these data includes long-term testing of material samples in an environmental simulation chamber, recording their appearance and performance change curves, and converting them into a set of quantifiable numerical parameters as environmental response coefficients stored in the database. In the soft furnishing design view generation process, the system can dynamically adjust the visual effect of the material according to the user input environmental conditions to enhance the accuracy and usability of the design simulation results.

[0095] The material attribute database binds all attributes of each soft furnishing material sample to a unique identifier through an indexing mechanism and connects it with the material call interface in the view generation module. When the system constructs a design scheme according to the indoor structure, functional partition and user preferences, it can call the texture image and color data of the selected material in real time for rendering material effects; at the same time, according to the wear grade and environmental response coefficient, the display state of the material is dynamically adjusted to improve the realism and adaptability of the soft furnishing design drawing.

[0096] This embodiment realizes accurate description and calling of material characteristics by integrating the multi-dimensional attributes of soft furnishing materials, which helps to enhance the realism of virtual views and the accuracy of design decisions. In particular, the combination of texture images and color information makes the generated images closer to the actual material effects, while the wear level and environmental response coefficient helps to predict and evaluate the long-term use state of the materials, improving the feasibility and reliability of the design.

[0097] The mapping algorithm for real effect restoration adopts an image generation model based on a convolutional neural network, which is used for image conversion and rendering of aging characteristics.

[0098] In this embodiment, the mapping algorithm for real effect restoration mainly constructs an image generation model through a convolutional neural network in deep learning technology to realize image conversion and rendering of aging characteristics of soft furnishing materials. The specific implementation process of this technical solution includes five stages of training sample construction, model training, feature mapping, image generation and image rendering.

[0099] The first stage is training sample construction. In this stage, a large number of image samples of different soft furnishing materials at different use times are collected in actual scenes, covering typical image data of new materials, use for one to three years, use for three to five years, and use for more than five years. To ensure data quality, the collection process is carried out under the condition of uniform illumination and shooting angle, and the aging level label corresponding to the image is labeled as the basis data for supervised learning. Image data should cover multiple material types, including cloth, leather, wood, metal, plastic, etc., to ensure the wide applicability of the model.

[0100] The second stage is model training. In this stage, a multi-layer convolutional neural network model is constructed, which extracts aging-related image features such as texture, color, edge, and noise points from the input image samples layer by layer. For example, features such as color fading and fiber sparseness are extracted on cloth materials, and features such as scratches, rust spots, and gloss changes are extracted on metal materials. The neural network continuously adjusts the weight parameters through the backpropagation mechanism, so that the output image is as close as possible to the real aging image. The training process introduces residual connection and batch normalization mechanism to improve the performance of the model in image conversion accuracy and convergence speed.

[0101] The third stage is feature mapping. In the image generation task, the system inputs the material original image into the convolutional neural network model through forward propagation of the model according to the material type and the preset aging period input by the user. The model automatically identifies the texture and structure features in the initial image of the material, and matches the corresponding aging feature set based on the selected aging stage parameters. This mapping process gradually adjusts the new material image to the image in the aging state, while preserving the integrity of the original structure and visually superimposing the aging features.

[0102] The fourth stage is image generation. The convolutional neural network outputs an image with aging effects, which processes color brightness, saturation, edge sharpness, and local texture at the pixel level, presenting effects such as material fading, surface wear, or structural aging. The resolution of the generated image is consistent with the original texture Figure 1 , ensuring seamless integration during subsequent view synthesis.

[0103] The fifth stage is image rendering. The final generated aging image is mapped to the designated location in the indoor view, such as the sofa surface, curtain area, wall decoration, etc. The system automatically adjusts the image size and deformation parameters according to the geometric characteristics of the rendering target area, ensuring a natural and realistic material fitting effect. Combined with indoor lighting information for lighting rendering, the image realism is further enhanced.

[0104] In addition, to improve user experience, the system provides multiple adjustable parameter interfaces, including aging degree, aging type, and material response sensitivity, allowing users to manually fine-tune the image generation results according to design requirements. If deployed on a cloud server, the model can be trained in real-time with new samples, continuously improving its adaptability to new materials and aging patterns.

[0105] The comparative analysis technology includes color deviation analysis and contour retention detection based on multi-time point image difference calculation.

[0106] In this implementation, the core of the comparative analysis technology is to build an analysis model based on multi-time point image difference to evaluate the visual performance changes of soft furnishing materials at different usage stages, mainly including color deviation analysis and contour retention detection. The specific technical process is as follows.

[0107] First, the system establishes a time series image library by regularly collecting image data of soft furnishing materials. The collection time points can be set as before use, 1 month of use, 3 months of use, 6 months of use, and 1 year of use, etc., to ensure coverage of key nodes in the material aging process. Each collection is carried out under fixed light intensity, same angle, and background conditions to ensure comparability between images.

[0108] Next, the color deviation analysis module preprocesses the above time series images, including image size unification, denoising, color balance, etc. The system collects pixel-level color information within the image area and compares color changes at different time points to extract color shift trends. For example, the system extracts the color value change of each pixel point in the representative area of the material, forms a color shift distribution map through statistical analysis, and determines whether the material has undergone significant color fading. The color deviation value will be converted into a visual parameter, outputting, for example, a color change curve, a heat map, etc., to support designers in material selection reference.

[0109] At the same time, the contour retention detection module extracts and analyzes the structure edges in the material image. In the image at each time point, the system uses an edge detection algorithm to obtain the material texture boundary and structure contour, and matches and compares it with the contour in the initial image. The matching indicators include the change distance of the boundary position, the contour integrity, the number of corner points, etc. The system sets a contour offset threshold, and when the edge contour of the later image deviates from the original contour by more than the set threshold, it is determined that the material has deformed or worn. The result is displayed in percentage, such as a contour retention of 95% indicating minimal deformation, and a contour retention of less than 70% indicating significant deformation of the material.

[0110] To improve the scientificity and adaptability of the analysis results, the system introduces a multi-frame image averaging mechanism, that is, not less than 3 frames of images are collected at each time node, and the average result is taken as the basis for analysis, reducing the influence of single-frame image errors. At the same time, the comparative analysis module is associated with the material attribute database, and when the color deviation value or the contour retention exceeds the set threshold, the system can automatically recommend similar but more durable alternative materials to assist in optimizing the soft furnishing scheme.

[0111] In addition, the comparative analysis technology supports users to actively upload a custom image set for analysis, and the system will automatically build a difference analysis process according to the upload time sequence. The analysis results can be directly fed back to the view automatic generation module, and the aging state of the material is reflected in the design drawing in real time, providing users with more realistic and predictable design visual effects.

[0112] As shown in Figure 2 , a view automatic generation system for indoor soft furnishing design is also provided, which is used to implement the steps of the view automatic generation method for indoor soft furnishing design, and the system comprises:

[0113] A material feature extraction module is configured to obtain aging characteristic data of a plurality of soft furnishing materials at different time nodes through a pre-established material attribute database, and to store the data classified according to the material aging process and surface wear details, thereby obtaining an initialized material aging characteristic set.

[0114] A visual effect generation module is configured to perform environmental adaptability adjustment based on the material aging characteristic set, generate a visual effect data set, and generate a dynamic visual effect image highly consistent with the actual aging process by using a real effect restoration mapping algorithm.

[0115] A reality evaluation module is configured to verify color fading characteristics and friction usage traces in the dynamic visual effect image by using a comparative analysis technology, and to determine whether the dynamic visual effect image meets a preset reality standard.

[0116] An image optimization module is configured to adjust parameters of the comprehensive data analysis by an iterative optimization technique and regenerate the dynamic visual effect image until the real effect restoration requirement is met when the authenticity evaluation result does not meet the preset standard.

[0117] The technical scheme of the system mainly focuses on modeling the visual performance changes of soft furnishing materials during the aging process and automatically generating real simulation views. The implementation process is as follows:

[0118] First, the material feature extraction module is responsible for extracting the aging feature data of various soft furnishing materials at different time nodes from a pre-constructed material attribute database. The database stores data such as texture images, color information, wear and tear levels, and environmental response coefficients of the materials. The system extracts typical aging image features at each stage according to the material use cycle, such as 0 months, 3 months, 6 months, 12 months, etc., and classifies and structures the storage of texture variation, color change, and surface damage morphology. Finally, an initialized material aging feature set is formed, laying the foundation for subsequent image generation.

[0119] Subsequently, the visual effect generation module adjusts the environmental matching visual effect data set according to the above aging feature set and combined with specific environmental conditions (including lighting, humidity, temperature, etc. parameters). The system uses a convolutional neural network-based image generation model to map the feature data to dynamic images simulating the aging effect. The visual performance of the real restored material in the natural aging or external wear process, such as color fading, material loss of luster, surface cracks, etc., makes the generated indoor view closer to the real use state.

[0120] Then, the authenticity evaluation module automatically verifies the generated dynamic visual effect image. The system uses image difference analysis method to compare the consistency of color shift and contour retention between the generated image and the real shot image in the database, focusing on analyzing whether the color fading characteristics and friction use traces conform to the actual aging law of the material. The evaluation result is quantified by a scoring mechanism, and the system compares the evaluation result with the preset authenticity standard to determine whether the image effect meets the standard.

[0121] If the image effect does not meet the authenticity requirement, the image optimization module will start the iterative optimization process. The system adjusts the image generation parameters, aging model weights, environmental response values, and other key variables based on the previous evaluation results, and regenerates new dynamic visual images. The optimization process will continue to iterate until the evaluation result meets the real effect restoration standard, thereby ensuring that the final output soft furnishing view image has high fidelity and visual persuasiveness.

[0122] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A method for automatically generating views for interior soft furnishing design, characterized in that, The method includes: By using a pre-established material property database, we obtain aging characteristic data of various soft furnishing materials at different time points, classify and store them according to the material aging process and surface wear details, and obtain an initial set of material aging characteristics. Based on the initialized set of material aging features, a visual effect dataset after environmental adaptation adjustment is obtained. Based on the visual effect dataset after environmental adaptation adjustment, an image conversion and rendering mapping algorithm based on a convolutional neural network image generation model is used to generate dynamic visual effect images that are highly consistent with the actual aging process. For dynamic visual effect images, comparative analysis technology is used to verify the simulation accuracy of color fading characteristics and friction marks, and to determine whether they meet the preset authenticity standards. If the simulation accuracy does not meet the preset realism standard, the aging rate factor, environmental adaptation weight coefficient, and image mapping parameters are adjusted to analyze and adjust the data on light conditions, humidity environmental variables, and friction traces, and then the dynamic visual effect image is regenerated until the requirements for realistic effect reproduction are met.

2. The method for automatically generating views for interior soft furnishing design according to claim 1, characterized in that: The process of obtaining the environmentally adaptive visual effect dataset based on the initialized material aging feature set includes extracting the correlation patterns between surface wear details and color fading features using data mining techniques based on the initialized material aging feature set, and determining the key parameters for simulating appearance changes during the aging process.

3. The method for automatically generating views for interior soft furnishing design according to claim 2, characterized in that: The process of obtaining the visual effect dataset after environmental adaptation adjustment based on the initialized set of material aging characteristics also includes key parameters for simulating appearance changes, acquiring relevant variable data on environmental interaction effects, including dynamic records of light conditions, humidity environmental variables, and friction usage traces, and determining the weight distribution of environmental factors on aging characteristics.

4. The method for automatically generating views for interior soft furnishing design according to claim 3, characterized in that: The process of obtaining the visual effect dataset after environmental adaptation adjustment based on the initialized material aging feature set also includes adjusting the parameter values ​​of the appearance change simulation through multivariate regression analysis if the weight distribution of environmental factors exceeds a preset threshold range, thereby obtaining a corrected appearance change feature matrix.

5. The method for automatically generating views for interior soft furnishing design according to claim 4, characterized in that: The process of obtaining the environmentally adaptive visual effect dataset based on the initialized material aging feature set also includes obtaining long-term use prediction time-series data based on the corrected appearance change feature matrix and the rendering requirements of dynamic visual effects, to determine the visual performance of the material at different time points.

6. The method for automatically generating views for interior soft furnishing design according to claim 5, characterized in that: The environmentally adaptive visual effect dataset obtained based on the initialized set of material aging characteristics also includes visual performance predicted for long-term use. The dataset is obtained by integrating the interaction between lighting conditions and humidity environmental variables through comprehensive data analysis technology.

7. The method for automatically generating views for interior soft furnishing design according to claim 1, characterized in that: The material property database includes texture images, color information, wear levels, and environmental response coefficients of soft furnishing materials of different material types.

8. The method for automatically generating views for interior soft furnishing design according to claim 1, characterized in that: The mapping algorithm for restoring the realistic effect adopts an image generation model based on a convolutional neural network, which is used to perform image conversion and rendering of aging features.

9. The method for automatically generating views for interior soft furnishing design according to claim 1, characterized in that: The comparative analysis technique includes color deviation analysis and contour preservation detection based on multi-time point image difference calculation.

10. An automatic view generation system for interior soft furnishing design, used to implement the method according to any one of claims 1-9, characterized in that, The system includes: The material feature extraction module is used to obtain aging feature data of various soft furnishing materials at different time points through a pre-established material property database, and to classify and store the material aging process and surface wear details to obtain an initialized set of material aging features. The visual effects generation module is used to make environmental adaptation adjustments based on the material aging feature set, generate a visual effects dataset, and use an image conversion and rendering mapping algorithm based on a convolutional neural network image generation model to generate dynamic visual effects images that are highly consistent with the actual aging process. The authenticity assessment module is used to verify the color fading features and friction marks in the dynamic visual effect image through comparative analysis technology, and to determine whether it meets the preset authenticity standards. The image optimization module is used to analyze and adjust the data on light conditions, humidity environmental variables, and friction traces when the authenticity assessment results do not meet the preset standards. It adjusts the aging rate factor, environmental adaptation weight coefficient, and image mapping parameters, and regenerates dynamic visual effect images until the requirements for realistic effect reproduction are met.

Images