Simulation model construction method and device based on image features, equipment and medium

Abstract

The invention provides a simulation model construction method and device based on image features, equipment and a medium, and relates to the field of simulation, and the method comprises the steps: determining a plurality of collection positions, the collection features of each collection position, and the position correlation between each collection position and other collection positions according to the object features of a to-be-modeled object and illumination information; setting an image acquisition device at the corresponding acquisition position based on the acquisition features, and performing image acquisition to obtain a plurality of feature images of the to-be-modeled object; extracting all local contour features of the to-be-modeled object based on all the feature images, and performing contour splicing based on the local contour feature corresponding to each feature image and the position correlation to obtain an initial simulation model of the to-be-modeled object; and performing image filling on the initial simulation model based on all the feature images to obtain a simulation model of the to-be-modeled object. According to the method, illumination information can be considered when the simulation model is constructed, and the accuracy of simulation model construction is improved.

Description

Technical Field

[0001] The present invention relates to the field of simulation technology, and in particular to a method, device, equipment and medium for constructing a simulation model based on image features. Background Art

[0002] Constructing simulation models based on images can extract key information from images and quickly transform real-life scenes into digital models that can be analyzed and simulated. For example, in urban planning, by identifying satellite images, photos taken on the spot, etc., the city's architectural layout, transportation network and other conditions can be accurately restored, assisting planners in formulating more realistic urban development plans; in the medical field, it can identify medical images, build simulation models of human organs, etc., help doctors perform preoperative simulations and disease diagnosis, effectively improve the accuracy and safety of diagnosis and treatment, and provide a valuable data foundation and intuitive simulation reference for the development of many industries.

[0003] In terms of existing technologies, deep learning algorithms are its core support. Convolutional neural networks (CNNs) and their various derivative architectures have excellent image feature extraction capabilities, and can perform high-precision classification, recognition, and feature analysis on images, providing key data for simulation model construction. At the same time, open source image recognition frameworks such as TensorFlow and PyTorch have lowered the technical threshold, making it easier for developers to develop corresponding image recognition and simulation model construction applications on this basis. There are also some professional image recognition software that combines three-dimensional reconstruction technology, which can convert the recognized two-dimensional image information into a three-dimensional simulation model, further expanding the depth and breadth of applications.

[0004] However, the accuracy of image recognition in this way is easily affected by image quality. When the lighting conditions are poor, it is likely to cause recognition errors, which in turn affects the accuracy of the simulation model. Summary of the invention

[0005] The present invention provides a method, device, electronic device and storage medium for constructing a simulation model based on image features, so as to solve the problem that in the existing method of constructing a simulation model based on an image, the construction result of the simulation model is easily affected by the image quality when the lighting conditions are poor.

[0006] In a first aspect, the present invention provides a method for constructing a simulation model based on image features, comprising: Obtaining object features and lighting information of the object to be modeled, and determining multiple acquisition positions, acquisition features of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions based on the object features and lighting information; An image acquisition device is arranged at a corresponding acquisition position based on the acquisition features, and image acquisition is performed based on each image acquisition device to obtain a plurality of feature images of the object to be modeled; Extract all local contour features of the object to be modeled based on all feature images, and perform contour splicing based on the corresponding local contour features and position correlation relationship of each feature image to obtain an initial simulation model of the object to be modeled; The initial simulation model is filled with images based on all feature images to obtain a simulation model of the object to be modeled.

[0007] In a second aspect, the present invention provides a simulation model construction device based on image features, comprising: A determination module is used to obtain object features and lighting information of the object to be modeled, and based on the object features and lighting information, determine multiple acquisition positions, acquisition features of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions; A collection module, used to set an image collection device at a corresponding collection position based on the collection features, and to collect images based on each image collection device to obtain multiple feature images of the object to be modeled; A splicing module is used to extract all local contour features of the object to be modeled based on all feature images, and to perform contour splicing based on the corresponding local contour features and position correlation relationship of each feature image to obtain an initial simulation model of the object to be modeled; The filling module is used to fill the initial simulation model with images based on all feature images to obtain a simulation model of the object to be modeled.

[0008] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of the method described in the first aspect or any possible implementation manner of the first aspect.

[0009] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation of the first aspect are implemented.

[0010] The present invention provides a method, device, electronic device and storage medium for constructing a simulation model based on image features. The acquisition position, the acquisition features of each acquisition position and the position correlation between each acquisition position and the other acquisition positions are determined by the object features and illumination information of the object to be modeled, and an image acquisition device is set at the corresponding position to perform image acquisition according to the acquisition features to obtain multiple feature images of the object to be modeled, and all local contour features of the object to be modeled are extracted according to the feature images, and contour splicing is performed according to the local contour features and the position correlation to obtain an initial simulation model, and the simulation model is filled with images to obtain a simulation model. The acquisition position and the corresponding acquisition features are determined by the object features and illumination information, so that the illumination conditions of each acquisition position can be guaranteed, and contour splicing is performed by the local contour features of the feature images to first obtain the initial simulation model of the object to be modeled, and then the image is filled by the feature images to obtain the simulation model, which can improve the accuracy of the simulation model construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0012] Figure 1 It is a flowchart of a method for constructing a simulation model based on image features provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the acquisition position of the method for constructing a simulation model based on image features provided by an embodiment of the present invention; Figure 3 is a structural schematic diagram of a simulation model building device based on image features provided by an embodiment of the present invention; Figure 4 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0014] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0015] Figure 1 is a flowchart of the method for constructing a simulation model based on image features provided by an embodiment of the present invention, referring to Figure 1 , as detailed below: S110, obtaining object features and lighting information of the object to be modeled, and determining multiple acquisition positions, acquisition features of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions based on the object features and lighting information.

[0016] In some embodiments, the object to be modeled may be a building, and simulation models may be constructed for the interior and exterior of the building; or the object to be modeled may be a whole, such as a building block, and simulation models may be constructed for the exterior of the whole.

[0017] In some embodiments, object features include shape, size, color, texture, and object characteristics, etc., which are inherent properties of the object to be modeled. Object characteristics refer to whether there are important parts in the object.

[0018] Among them, lighting information covers aspects such as lighting intensity, lighting direction and lighting distribution information. The two may affect each other. Object characteristics affect lighting performance, and lighting information changes the imaging of object characteristics.

[0019] In some embodiments, a partial image of the object to be modeled may be captured at a capture position. When determining the capture position, the illumination information of the portion of the object to be modeled corresponding to the capture position needs to meet a preset condition.

[0020] It should be noted that the range of several acquisition positions can be roughly determined according to the contour of the object, and then several acquisition positions whose lighting conditions meet the preset conditions are selected from the range of these acquisition positions as the final acquisition positions. Figure 2 As shown, when determining the collection position of object A, five collection position ranges (area 1, area 2, area 3, area 4, area 5) are first determined according to the characteristics of object A, and then the specific five collection positions (position 1, position 2, position 3, position 4, position 5) are determined according to the lighting conditions and equipment characteristics.

[0021] In a possible implementation, the specific processing method of step S110 is: determine multiple initial acquisition positions based on object features, and adjust the initial acquisition positions based on lighting information to obtain multiple acquisition positions; based on the position information of each acquisition position, determine the acquisition features of the corresponding acquisition position and the position correlation relationship between the corresponding acquisition position and the remaining acquisition positions.

[0022] In some embodiments, multiple acquisition position ranges are first determined based on the shape, size and characteristics of the object (for example, whether there are positions in the object that need to be focused on to ensure the simulation effect), and when the acquisition positions are determined within these ranges, it is necessary to ensure that no part of the object to be modeled is missed during image acquisition.

[0023] After obtaining the above-mentioned collection position range, the center point of the collection position range is used as the initial collection position, and then the lighting conditions when the collection device is located at the center point of the collection range are determined according to the lighting information, and it is judged whether the lighting conditions meet the preset requirements. If the preset requirements are met, the center point is used as the collection position. If the requirements are not met, starting from the center point, the position where the lighting conditions meet the preset requirements is gradually searched outward according to a certain radius as the collection position.

[0024] Among them, the preset requirements can limit one or more characteristics of light intensity, light direction and light color. For example, the light intensity corresponding to the collection position needs to be greater than a certain set light intensity. For another example, the light intensity corresponding to the collection position needs to be greater than a certain set light intensity and the color temperature corresponding to the collection position needs to be within a certain range.

[0025] In some embodiments, the acquisition features include a minimum distance and a maximum distance between the acquisition position and the object to be modeled, and object information and lighting information of the object to be modeled that can be acquired at the acquisition position.

[0026] In some embodiments, the position correlation relationship between each acquisition position and the remaining acquisition positions includes: the distance between each of the remaining acquisition positions and the acquisition position, the direction of each of the remaining acquisition positions to the acquisition position, and other information.

[0027] By obtaining object features and lighting information to determine the collection position and the correlation between each collection position, it can adapt to object features and lighting conditions to improve image acquisition quality. Through multi-perspective comprehensive acquisition, it can take into account the position relationship, thereby improving the accuracy of model construction. Due to its flexibility, it has wide applicability and strong scalability, which is of great significance to the construction of simulation models based on image feature extraction.

[0028] S120, setting an image acquisition device at a corresponding acquisition position based on the acquisition features, and performing image acquisition based on each image acquisition device to obtain a plurality of feature images of the object to be modeled.

[0029] In some embodiments, the acquisition features can be used to determine information such as the acquisition range, image resolution, and lighting conditions that the image acquisition device needs to meet, so as to select a suitable image acquisition device for image acquisition.

[0030] In some embodiments, feature images are images acquired by a specific image acquisition device at a capture position determined based on the object features of the object to be modeled and lighting information. These images contain rich information about the object to be modeled, such as the local contour, color, texture and other features of the object. They are the basis for subsequent operations such as local contour feature extraction, contour stitching and image filling, and play a key role in building an accurate and complete simulation model. Their quality directly affects the degree to which the final simulation model restores the real object.

[0031] S130, extracting all local contour features of the object to be modeled based on all feature images, and performing contour splicing based on the local contour features corresponding to each feature image and position correlation relationship, to obtain an initial simulation model of the object to be modeled.

[0032] In some embodiments, the local contour feature mainly refers to a partial contour feature of the object to be modeled reflected in the feature image, and each feature image has a local contour feature.

[0033] In some embodiments, contour stitching refers to a process of stitching local contour features obtained at each acquisition position according to the correlation between the acquisition positions.

[0034] It should be noted that, in the local contour features corresponding to each feature image, there will be overlapping parts with the local contour features of the feature image corresponding to the adjacent acquisition position. For example, acquisition position 1 and acquisition position 2 perform image acquisition to obtain feature image 1 and feature image 2, and acquisition position 1 is adjacent to acquisition position 2, then the local contour features of feature image 1 and the local contour features of feature image 2 will have overlapping parts.

[0035] In a possible implementation, the specific processing method of step S130 is: for each feature image, perform the following steps: extract the initial local contour features of the object to be modeled based on the feature image; optimize the initial local contour features based on the acquisition features of the acquisition position corresponding to the feature image and the position correlation relationship to obtain the local contour features corresponding to the feature image.

[0036] In some embodiments, when performing initial local contour feature extraction, it is necessary to first perform grayscale and noise reduction preprocessing on the feature image, then use edge detection operators such as Sobel, Prewitt, Canny, etc. to calculate the pixel gradient amplitude and direction, screen out edge points, and then use a contour tracking algorithm to connect them into a continuous contour curve. Finally, use morphological operations or curve fitting to repair discontinuities. The comprehensive use of multiple technologies and multi-step processing can obtain accurate and complete initial local contour features.

[0037] In a possible implementation, the specific processing method of step S130 is: for each feature image, perform the following steps: obtain the illumination feature of the feature image based on the acquisition feature of the acquisition position corresponding to the feature image, compare the illumination feature with the standard illumination feature, and obtain the illumination correction value of the feature image; determine the neighborhood image corresponding to the feature image based on the position correlation relationship of the feature image; optimize the initial local contour feature of the feature image based on the neighborhood image and the illumination correction value to obtain the local contour feature corresponding to the feature image.

[0038] In some embodiments, it is necessary to obtain illumination features from the acquisition features of the corresponding acquisition position of the feature image, including information such as illumination intensity, direction, and color, such as by recording through an acquisition device or determining specific values ​​through an illumination analysis algorithm. Then, it is compared with pre-set standard illumination information, which is based on ideal modeling conditions or a large amount of experimental data. Then, the illumination correction value is calculated based on the comparison result, such as determining the intensity adjustment value when the illumination intensity deviates, adjusting the angle when the illumination direction does not match the calculated direction, and obtaining the illumination correction value when the illumination color affects the color of the object. These correction values ​​are used for subsequent contour feature adjustments.

[0039] The above-mentioned lighting correction values ​​include lighting intensity correction value, lighting direction correction value and lighting color correction value. The lighting intensity correction value is the interpolation between the lighting intensity corresponding to the acquisition position and the standard lighting intensity. The lighting direction correction value is the difference between the lighting direction corresponding to the acquisition position and the vertical lighting direction. The lighting color correction value needs to construct a standard lighting color model (RGB lighting color model) according to the color characteristics and modeling targets of the object to be modeled, and calculate the difference between the lighting color corresponding to the acquisition position and the lighting color corresponding to the standard lighting color model.

[0040] In some embodiments, for a certain feature image, it is necessary to determine the neighborhood image of the feature image based on the positional correlation between the acquisition position and other positions. The positional correlation covers information such as relative distance and angle. For example, according to rules such as distance threshold and angle range, images that are neighbors of each other are screened out from all acquired images. These neighborhood images contain information related to the object part of the current image at different viewing angles, and can be used to adjust the initial local contour features of the feature image. For example, acquisition position 1 is adjacent to acquisition position 2, and there is an overlapping part in the acquisition range of acquisition position 1 and acquisition position 2. When the local contour features of acquisition position 1 are obtained, the initial local contour features of acquisition position 1 can be adjusted by the initial local contour features of the overlapping part of acquisition position 2 and acquisition position 1, to supplement some incomplete lines or adjust the errors caused by the acquisition direction.

[0041] It should be noted that when optimizing the initial local contour features, it is necessary to first optimize according to the illumination correction value and then optimize according to the neighborhood features.

[0042] In a possible implementation, the specific processing method of step S130 is: randomly select a feature image as the stitching starting point feature image, and determine the stitching order of each feature image based on the position correlation relationship of the stitching starting point feature image; based on the stitching order, perform contour stitching on the corresponding local contour features of each feature image in turn to obtain the initial simulation model of the object to be modeled.

[0043] In some embodiments, the splicing starting point feature image is randomly determined. For example, the object A to be modeled has five feature images, namely feature image 1, feature image 2, feature image 3, feature image 4, and feature image 5. These five feature images can all be the splicing starting point feature images.

[0044] In some embodiments, the splicing order is determined by the position correlation relationship, from which the adjacent feature images of the splicing starting point feature image and the angle information and distance information of each feature image can be obtained. First, the splicing order of each adjacent feature image of the splicing starting point feature image is determined according to the distance, and then after the adjacent feature images of the splicing starting point feature image are spliced, the remaining adjacent images of each adjacent image of the splicing starting point feature image are determined, and the splicing order is determined in turn according to the splicing order of the adjacent images of the splicing starting point feature image and the distance between the adjacent images of each splicing starting point feature image and its remaining adjacent images.

[0045] The stitching order is illustrated below through an example. Assume that the object A to be modeled has five feature images, namely feature image 1, feature image 2, feature image 3, feature image 4, and feature image 5. The stitching starting point image is feature image 1. The adjacent feature images of feature image 1 are feature image 2 and feature image 5. The distance between the acquisition positions of feature image 1 and feature image 2 is 30m, and the distance between the acquisition positions of feature image 1 and feature image 5 is 25m. The remaining adjacent images of feature image 2 are feature image 3, and the remaining adjacent images of feature image 5 are feature image 4. Then the stitching order is: feature image 5, feature image 2, feature image 4, feature image 3.

[0046] It should be noted that the initial simulation model of the object to be modeled only has contour information but no image information or color information.

[0047] By fully extracting the initial local contour features through multiple feature images, the image information can be fully utilized to ensure the integrity of the contour features. The initial local contour features are optimized by considering the position correlation relationship to make the contour more accurate. Based on this, the contour splicing is performed to obtain the initial simulation model, which not only improves the efficiency of model construction, but also enhances the accuracy and stability of the model, laying a solid foundation for subsequent image filling and the generation of the final high-quality simulation model, and effectively improving the overall effect and practicality of the simulation model construction based on image feature extraction.

[0048] S140, performing image filling on the initial simulation model based on all feature images to obtain a simulation model of the object to be modeled.

[0049] In some embodiments, image filling refers to extracting color information for each feature image to obtain the color information of each small area, and then filling the image of the initial simulation model based on the acquisition position and the color information of each small area to enrich the model details and surface texture.

[0050] In a possible implementation, the specific processing method of step S140 is: based on each feature image, partially fill the initial simulation model to obtain the initial filling result corresponding to each feature image; based on the position correlation relationship between the corresponding acquisition position of each feature image and the remaining acquisition positions, correct the initial filling result of the corresponding feature image to obtain the filling result; based on all the filling results and the initial simulation model, obtain the simulation model of the object to be modeled.

[0051] In some embodiments, color information of a corresponding part of the object to be modeled can be obtained based on each feature image, and image filling can be performed on this part in the initial simulation model based on the color information to obtain an initial filling result.

[0052] In some embodiments, since the image acquisition devices at each acquisition position are different, there will be color differences in the acquisition results of the image acquisition devices. The color difference between the image acquisition devices can be obtained by the color difference of the overlapping parts between each feature image and the feature images of its adjacent areas. The initial filling result can be corrected for the color difference based on the color difference.

[0053] It should be noted that when chromatic aberration correction is performed on a certain feature image, the chromatic aberration correction matrix can be determined according to the chromatic aberration between the feature image and each adjacent feature image and the number of adjacent images of the feature image. The chromatic aberration correction matrix is:

[0054] in, For the The chromatic aberration correction matrix of the feature image, For the The first feature images adjacent to feature images, For the The number of images adjacent to a feature image, is the red correction parameter, is the green correction parameter, is the blue correction parameter, For the The area of ​​the object to be modeled collected by the feature image, For the The first The area of ​​the object to be modeled collected by the feature images adjacent to the feature images, For the In the feature image, The adjacent feature image The image and The average red depth of the overlapping area of ​​the feature images, For the The adjacent feature image In the image, The adjacent feature image The image and The average red depth of the overlapping area of ​​the feature images, For the In the feature image, The adjacent feature image The image and The average green depth of the overlapping area of ​​the feature images, For the The adjacent feature image In the image, The adjacent feature image The image and The average green depth of the overlapping area of ​​the feature images, For the In the feature image, The adjacent feature image The image and The average blue depth of the overlapping area of ​​the feature images, For the The adjacent feature image In the image, The adjacent feature image The image and The average blue depth of the overlapping area of ​​the feature images.

[0055] In a possible implementation, the specific processing method of step S140 is: for each feature image, perform the following steps: extract color information based on the feature image to obtain the initial regional color information of the feature image; adjust the initial regional color information based on the corresponding acquisition features of the feature image to obtain the regional color information of the feature image; and fill part of the image of the initial simulation model based on the corresponding acquisition position and regional color information of the feature image.

[0056] In some embodiments, when extracting color information, it is necessary to first select a suitable color space, such as RGB, CMYK or HSV, etc. The RGB color space is used in this application, and then the color information is obtained by extracting the color value of each pixel based on the pixel point and statistically analyzing the color distribution.

[0057] In some embodiments, illumination information can be obtained based on the collected features, and illumination information can be compared with standard illumination information to obtain illumination correction value, the specific calculation method has been described above and will not be repeated here. The initial regional color information can be adjusted based on the illumination correction value to obtain regional color information.

[0058] In some embodiments, the corresponding area of ​​the regional color information corresponding to the feature image on the initial simulation model can be determined according to the acquisition position, and the initial simulation model can be partially filled with the image according to the corresponding area.

[0059] The image is partially filled using the feature image to obtain an initial filling result, and the color difference correction matrix is ​​calculated based on the positional correlation between the corresponding acquisition position and the remaining acquisition positions, as well as the color information corresponding to each acquisition position. The area factor is taken into account in the calculation process, and the initial filling result is corrected according to the color difference correction matrix to obtain the filling result. Based on all the filling results, a simulation model can be obtained, which can make full use of the feature image information, effectively improve the model details and accuracy, make the model closer to the appearance of the real object, enhance the authenticity and practicality of the model, and improve the construction quality of the simulation model.

[0060] The acquisition position, the acquisition feature of each acquisition position, and the position correlation between each acquisition position and the other acquisition positions are determined by the object features and illumination information of the object to be modeled, and the image acquisition device is set at the corresponding position to perform image acquisition according to the acquisition features, so as to obtain multiple feature images of the object to be modeled, and all local contour features of the object to be modeled are extracted according to the feature images, and the contours are spliced ​​to obtain the initial simulation model, and the simulation model is filled with images to obtain the simulation model. The acquisition position and the corresponding acquisition features are determined by the object features and illumination information, so as to ensure the illumination conditions of each acquisition position, and the contours are spliced ​​by the local contour features of the feature images, so as to first obtain the initial simulation model of the object to be modeled, and then the image is filled by the feature images to obtain the simulation model. In the filling process, the color difference is corrected according to the color difference of different acquisition devices, so as to comprehensively improve the construction effect of the simulation model. The acquisition elements are determined by combining the object features and illumination information to ensure the image quality; the contour feature extraction is optimized to make it more accurate; the feature images and positional relationships are used for splicing and filling to improve the efficiency and accuracy, and a high-quality model with strong adaptability is obtained.

[0061] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0062] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0063] Figure 3 : is a schematic diagram of the structure of a simulation model building device based on image features provided by an embodiment of the present invention. For the sake of convenience, only the parts related to the embodiment of the present invention are shown, which are described in detail as follows: Reference Figure 3 , the simulation model building device 3 based on image features includes: A determination module 31 is used to obtain object features and illumination information of the object to be modeled, and determine multiple acquisition positions, acquisition features of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions based on the object features and illumination information; The acquisition module 32 is used to set an image acquisition device at a corresponding acquisition position based on the acquisition features, and perform image acquisition based on each image acquisition device to obtain multiple feature images of the object to be modeled; A splicing module 33 is used to extract all local contour features of the object to be modeled based on all feature images, and to perform contour splicing based on the local contour features corresponding to each feature image and the position correlation relationship, so as to obtain an initial simulation model of the object to be modeled; The filling module 34 is used to perform image filling on the initial simulation model based on all feature images to obtain a simulation model of the object to be modeled.

[0064] In a possible implementation, the determination module 31 is specifically used to: determine multiple initial acquisition positions based on object features, and adjust the initial acquisition positions based on lighting information to obtain multiple acquisition positions; based on the position information of each acquisition position, determine the acquisition features of the corresponding acquisition position and the position correlation relationship between the corresponding acquisition position and the remaining acquisition positions.

[0065] In a possible implementation, the stitching module 33 is specifically used to: for each feature image, perform the following steps: extract the initial local contour features of the object to be modeled based on the feature image; optimize the initial local contour features based on the acquisition features of the acquisition position corresponding to the feature image and the position correlation relationship, and obtain the local contour features corresponding to the feature image.

[0066] In a possible implementation, the stitching module 33 is further used to: for each feature image, perform the following steps: obtain the illumination feature of the feature image based on the acquisition feature of the acquisition position corresponding to the feature image, compare the illumination feature with the standard illumination feature, and obtain the illumination correction value of the feature image; determine the neighborhood image corresponding to the feature image based on the position correlation relationship of the feature image; optimize the initial local contour feature of the feature image based on the neighborhood image and the illumination correction value, and obtain the local contour feature corresponding to the feature image.

[0067] In a possible implementation, the stitching module 33 is also used to: randomly select a feature image as a stitching starting point feature image, and determine the stitching order of each feature image based on the position correlation relationship of the stitching starting point feature image; based on the stitching order, perform contour stitching on the corresponding local contour features of each feature image in turn to obtain an initial simulation model of the object to be modeled.

[0068] In one possible implementation, the filling module 34 is specifically used to: based on each feature image, partially fill the initial simulation model to obtain the initial filling result corresponding to each feature image; based on the positional correlation between the corresponding acquisition position of each feature image and the remaining acquisition positions, correct the initial filling result of the corresponding feature image to obtain the filling result; based on all the filling results and the initial simulation model, obtain the simulation model of the object to be modeled.

[0069] In a possible implementation, the filling module 34 is also used to: for each feature image, perform the following steps: extract color information based on the feature image to obtain the initial regional color information of the feature image; adjust the initial regional color information based on the corresponding acquisition features of the feature image to obtain the regional color information of the feature image; and fill part of the image of the initial simulation model based on the corresponding acquisition position and regional color information of the feature image.

[0070] Figure 4 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 4 As shown, the electronic device 4 of this embodiment includes: a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer program 42, the steps in the above-mentioned various embodiments of the method for constructing a simulation model based on image features are implemented, for example Figure 1 Alternatively, when the processor 40 executes the computer program 42, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Figure 3 Functionality of the modules / units 31 to 34 shown.

[0071] Exemplarily, the computer program 42 may be divided into one or more modules / units, one or more modules / units are stored in the memory 41 and executed by the processor 40 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 42 in the electronic device 4. For example, the computer program 42 may be divided into Figure 3 Modules / units 31 to 34 are shown.

[0072] The electronic device 4 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will appreciate that Figure 4 It is only an example of the electronic device 4 and does not constitute a limitation of the electronic device 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0073] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0074] The memory 41 may be an internal storage unit of the electronic device 4, such as a hard disk or memory of the electronic device 4. The memory 41 may also be an external storage device of the electronic device 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the electronic device 4. The memory 41 is used to store computer programs and other programs and data required by the electronic device. The memory 41 may also be used to temporarily store data that has been output or is to be output.

[0075] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0076] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0077] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0078] In the embodiments provided by the present invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0079] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0080] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0081] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned simulation model construction method based on image features can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium.

[0082] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for constructing a simulation model based on image features, characterized in that: include: Acquire object features and lighting information of the object to be modeled, and determine multiple acquisition positions, acquisition features of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions based on the object features and the lighting information; An image acquisition device is arranged at a corresponding acquisition position based on the acquisition feature, and image acquisition is performed based on each image acquisition device to obtain a plurality of feature images of the object to be modeled; Extracting all local contour features of the object to be modeled based on all feature images, and performing contour splicing based on the local contour features corresponding to each feature image and position correlation relationship, to obtain an initial simulation model of the object to be modeled; The initial simulation model is image filled based on all feature images to obtain a simulation model of the object to be modeled.

2. The method for constructing a simulation model based on image features according to claim 1, characterized in that: The step of extracting all local contour features of the object to be modeled based on all feature images includes: For each feature image, perform the following steps: Extracting initial local contour features of the object to be modeled based on the feature image; Based on the acquisition features of the acquisition positions corresponding to the feature image and the position correlation relationship, the initial local contour features are optimized to obtain the local contour features corresponding to the feature image.

3. The method for constructing a simulation model based on image features according to claim 2, characterized in that: The step of optimizing the initial local contour feature for each feature image based on the acquisition features of the acquisition position corresponding to the feature image and the position correlation relationship to obtain the local contour feature corresponding to the feature image includes: For each feature image, perform the following steps: Obtaining a lighting feature of the feature image based on a collection feature of a collection position corresponding to the feature image, and comparing the lighting feature with a standard lighting feature to obtain a lighting correction value of the feature image; Based on the position correlation relationship of the feature image, determine the neighborhood image corresponding to the feature image; Based on the neighborhood image and the illumination correction value, the initial local contour features of the feature image are optimized to obtain the local contour features corresponding to the feature image.

4. The method for constructing a simulation model based on image features according to claim 1, characterized in that: The step of performing contour stitching based on the local contour features and position correlation relationship corresponding to each feature image to obtain an initial simulation model of the object to be modeled includes: Randomly selecting a feature image as a splicing starting feature image, and determining a splicing order of each feature image based on a position correlation relationship of the splicing starting feature image; Based on the stitching sequence, the local contour features corresponding to each feature image are stitched in turn to obtain an initial simulation model of the object to be modeled.

5. The method for constructing a simulation model based on image features according to claim 1, characterized in that: The determining, based on the object features and the illumination information, a plurality of acquisition positions, an acquisition feature of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions includes: Determining a plurality of initial acquisition positions based on the object features, and adjusting the initial acquisition positions based on the illumination information to obtain a plurality of acquisition positions; Based on the position information of each acquisition position, the acquisition characteristics of the corresponding acquisition position and the position correlation relationship between the corresponding acquisition position and the remaining acquisition positions are determined.

6. The method for constructing a simulation model based on image features according to claim 1, characterized in that: The step of performing image filling on the initial simulation model based on all feature images to obtain a simulation model of the object to be modeled includes: Based on each characteristic image, partially fill the initial simulation model to obtain an initial filling result corresponding to each characteristic image; Based on the position correlation relationship between the acquisition position corresponding to each feature image and the remaining acquisition positions, the initial filling result of the corresponding feature image is corrected to obtain a filling result; Based on all the filling results and the initial simulation model, a simulation model of the object to be modeled is obtained.

7. The method for constructing a simulation model based on image features according to claim 6, characterized in that: The partially filling the initial simulation model based on each feature image comprises: For each feature image, perform the following steps: Extracting color information based on the feature image to obtain initial region color information of the feature image; Based on the corresponding acquisition features of the feature image, the initial regional color information is adjusted to obtain the regional color information of the feature image; Based on the corresponding acquisition position of the feature image and the regional color information, the initial simulation model is partially filled with the image.

8. A simulation model construction device based on image features, characterized in that: include: A determination module, used to obtain object features and illumination information of the object to be modeled, and determine multiple acquisition positions, acquisition features of each acquisition position, and a position correlation relationship between each acquisition position and other acquisition positions based on the object features and illumination information; A collection module, used to set an image collection device at a corresponding collection position based on the collection feature, and to collect images based on each image collection device to obtain a plurality of feature images of the object to be modeled; A splicing module, used for extracting all local contour features of the object to be modeled based on all feature images, and performing contour splicing based on the local contour features corresponding to each feature image and the position correlation relationship, so as to obtain an initial simulation model of the object to be modeled; A filling module is used to perform image filling on the initial simulation model based on all feature images to obtain a simulation model of the object to be modeled.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for constructing a simulation model based on image features as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for constructing a simulation model based on image features as described in any one of claims 1 to 7 above are implemented.

Images