CN114998549A - Method for making three-dimensional digital model of traditional Chinese medicinal materials and computer readable medium - Google Patents
Method for making three-dimensional digital model of traditional Chinese medicinal materials and computer readable mediumDownload PDFInfo
- Publication number
- CN114998549A CN114998549ACN202210619124.2ACN202210619124ACN114998549ACN 114998549 ACN114998549 ACN 114998549ACN 202210619124 ACN202210619124 ACN 202210619124ACN 114998549 ACN114998549 ACN 114998549A
- Authority
- CN
- China
- Prior art keywords
- chinese medicinal
- model
- dimensional
- medicinal material
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
- G06T7/49—Analysis of texture based on structural texture description, e.g. using primitives or placement rules
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/593—Depth or shape recovery from multiple images from stereo images
- G06T7/596—Depth or shape recovery from multiple images from stereo images from three or more stereo images
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Image Processing (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本发明涉及图像处理技术领域，尤其是中药材三维数字重建领域，具体而言涉及一种中药材三维数字模型的制作方法与装置。The invention relates to the technical field of image processing, in particular to the field of three-dimensional digital reconstruction of Chinese medicinal materials, and in particular to a method and a device for making a three-dimensional digital model of traditional Chinese medicinal materials.
背景技术Background technique
中药是在中医药理论指导下，用于预防和治疗疾病并具有康复与保健作用的物质，主要以植物、动物、矿物等自然产物为直接或间接来源，具有产地广阔、品类繁多、来源复杂等特点。Traditional Chinese medicine is a substance that is used to prevent and treat diseases and has rehabilitation and health care effects under the guidance of traditional Chinese medicine theory. It is mainly derived from natural products such as plants, animals, and minerals. Features.
随着数字化技术飞速发展，数字化、网络化的方式已深入人心，5G、人工智能等技术与医药领域的融合不断深入，传统的检索、查阅方法和标本鉴定模式已经不能满足科研机构和社会公众对中药标本信息资源不断增长的需要。With the rapid development of digital technology, digital and networked methods have been deeply rooted in the hearts of the people, and the integration of 5G, artificial intelligence and other technologies with the medical field has continued to deepen. There is a growing need for information resources on TCM specimens.
针对中药材三维数字标本的制作，现有技术探索了采用基于切片的数字建模方式来生成，有关现有技术提出中药的数字切片库的建设，虽然能够实现中药标本的数字化，但需要将中药材做成切片，无法实现对中药材本身的形貌、药用部分的原始形态以及药用形态进行高清数字化的呈现，采用传统的三维造影和片层成像的图像融合，无法实现中药材的无损三维化；For the production of three-dimensional digital specimens of traditional Chinese medicine, the existing technology has explored the use of slice-based digital modeling to generate, and the related art proposes the construction of a digital slice library of traditional Chinese medicine, although it can realize the digitization of traditional Chinese medicine It is impossible to realize the high-definition digital presentation of the shape of the Chinese medicinal material itself, the original shape of the medicinal part, and the medicinal form. The traditional three-dimensional imaging and image fusion of slice imaging cannot realize the non-destructiveness of the Chinese medicinal material. three-dimensional;
同时，由于中药特征各异，难以形成统一的建模标准。常用中药分为12类别，包括根及根茎类、种子果实类、花类、叶类、全草类、藤木类、皮类、动物药类、矿物药类、树脂类、加工类和其他类，合计涵盖517个药材品类。每个类别、每种药材的形态各异，根茎类的药材形态与叶类的完全不一样，在建模的过程中无法进行统一的标准化。At the same time, due to the different characteristics of traditional Chinese medicines, it is difficult to form a unified modeling standard. Commonly used Chinese medicines are divided into 12 categories, including roots and rhizomes, seeds and fruits, flowers, leaves, whole plants, vines, skins, animal medicines, mineral medicines, resins, processing and others. A total of 517 medicinal material categories are covered. The shape of each type and each medicinal material is different, and the shape of the medicinal material of the rhizome is completely different from that of the leaf, so it cannot be standardized uniformly in the process of modeling.
以中药材为例，对其研究和鉴别不仅包括药用价值的部分，例如切片等，还包括对中药的鉴别部位的全面而准确地还原，因此，有必要建立全面准确的模型才能够实现关键鉴定部位的中药鉴别。中药材的三维数字化建模，与其他传统行业不同，比如建筑行业的建筑物，其表面基本是平整的或呈现出一定线性关系，而中药材的表面褶皱、纹理差异较大，必须要足够清晰和完整才能复原中药材，实现其标本的传承与科普价值。Taking Chinese herbal medicines as an example, the research and identification of them includes not only the parts of medicinal value, such as slices, etc., but also the comprehensive and accurate restoration of the identification parts of traditional Chinese medicines. Therefore, it is necessary to establish a comprehensive and accurate model to achieve the key Identification of traditional Chinese medicine for identification sites. The 3D digital modeling of Chinese medicinal materials is different from other traditional industries, such as buildings in the construction industry. Only with completeness can the Chinese medicinal materials be restored, and the inheritance and popular science value of their specimens can be realized.
发明内容SUMMARY OF THE INVENTION
本发明实施例目的在于提供一种中药材三维数字模型的制作方法，针对于中药领域的特殊性，尤其是当前中药材的三维数字标本制作模型和方法的缺失，从三维数字标本的差异度、完整性和清晰度出发，提出中药材三维数字模型的制作方法，在模型差异度、完整性和清晰度的基础上进行模型反馈，再经由纹理贴图的迭代修正，输出高质量的中药材三维模型，其中模型差异度是指三维重建的模型与真实物体之间的差异化程度，差异度值越小，还原度越高，差异度值越大，还原度越低。The purpose of the embodiments of the present invention is to provide a method for making a three-dimensional digital model of traditional Chinese medicinal materials, aiming at the particularity of the field of traditional Chinese medicine, especially the lack of the current three-dimensional digital sample making model and method of traditional Chinese medicinal materials, from the difference degree of three-dimensional digital samples, Starting from completeness and clarity, a method for making 3D digital models of Chinese herbal medicines is proposed. Model feedback is carried out on the basis of model difference, integrity and clarity, and then iterative correction of texture maps is used to output high-quality 3D models of Chinese herbal medicines. , where the model difference degree refers to the degree of difference between the 3D reconstructed model and the real object. The smaller the difference degree value, the higher the restoration degree, and the larger the difference degree value, the lower the restoration degree.
为实现上述目的，本发明实施例的第一方面提出一种中药材三维数字模型的制作方法，包括以下步骤：In order to achieve the above object, a first aspect of the embodiments of the present invention proposes a method for making a three-dimensional digital model of Chinese medicinal materials, including the following steps:
获取在同一高度和背景下拍摄的中药材的多角度图像，构造图像序列；Obtain multi-angle images of Chinese herbal medicines taken at the same height and background, and construct image sequences;
基于图像序列进行三维重建，获得中药材的点云空间模型；Perform 3D reconstruction based on image sequences to obtain the point cloud space model of Chinese herbal medicines;
对所述中药材的点云空间模型进行网格化，生成网格，获得所述中药材的初始模型；Gridding the point cloud space model of the Chinese medicinal material to generate a grid to obtain the initial model of the Chinese medicinal material;
在所述初始模型的基础上，生成纹理，获得有纹理的中药材的三维解析模型；On the basis of the initial model, a texture is generated to obtain a three-dimensional analytical model of the Chinese medicinal material with texture;
基于所述图像序列，根据中药材特性，计算所述中药材的三维解析模型的综合质量因子；以及Based on the sequence of images, according to the properties of the Chinese medicinal material, calculate the comprehensive quality factor of the three-dimensional analytical model of the Chinese medicinal material; and
基于综合质量因子与预设要求的比较，获得最终的中药材的三维解析模型并输出。Based on the comparison of the comprehensive quality factor and the preset requirements, the final three-dimensional analytical model of the Chinese herbal medicine is obtained and output.
可选地，所述基于图像序列进行三维重建，包括以下过程：Optionally, the three-dimensional reconstruction based on the image sequence includes the following processes:
对所述多角度图像提取特征值，进行稀疏重建；extracting feature values from the multi-angle image, and performing sparse reconstruction;
基于所述图像序列以及稀疏重建获得的拍摄参数，进行稠密重建，获得中药材的点云空间模型；其中，所述拍摄参数包括相机内参以及相机外参。Based on the image sequence and the shooting parameters obtained from the sparse reconstruction, dense reconstruction is performed to obtain a point cloud space model of the Chinese medicinal material; wherein the shooting parameters include camera internal parameters and camera external parameters.
可选地，所述对所述多角度图像提取特征值，进行稀疏重建，包括以下过程：Optionally, the extraction of feature values from the multi-angle image to perform sparse reconstruction includes the following processes:
基于所述多角度图像提取特征值，进行所述多角度图像之间的特征点匹配，确定所述多角度图像之间的匹配关系，从而获得极几何图和拍摄参数，完成稀疏重建。Extract feature values based on the multi-angle images, perform feature point matching between the multi-angle images, and determine the matching relationship between the multi-angle images, thereby obtaining a polar geometry map and shooting parameters, and completing sparse reconstruction.
可选地，所述基于所述图像序列以及所述拍摄参数，进行稠密重建，包括：Optionally, performing dense reconstruction based on the image sequence and the shooting parameters, including:
以所述图像序列为基础，按顺序为其中的每一个图像选择领域图像，构造立体图像对；On the basis of the image sequence, a domain image is selected for each of the images in sequence, and a stereo image pair is constructed;
计算每一个图像的深度图；Calculate the depth map of each image;
基于所述立体图像对的领域图像将所述每一个图像的深度图进行融合，完成稠密重建。The depth map of each image is fused based on the domain images of the stereo image pair to complete dense reconstruction.
可选地，所述计算每一个图像的深度图，包括：Optionally, the calculating the depth map of each image includes:
根据所述相机内参以及所述极几何图的极线计算每一个图像的每一个像素上的深度值；Calculate the depth value on each pixel of each image according to the camera intrinsic parameter and the epipolar line of the polar geometry map;
根据所述相机外参，确定每一个像素的前后关系，并根据所述前后关系，舍弃后方被遮挡的像素并保留前方未被遮挡的像素，获得每一个图像的深度图。According to the camera extrinsic parameters, the front-to-back relationship of each pixel is determined, and according to the front-to-back relationship, the rear occluded pixels are discarded and the front unoccluded pixels are retained to obtain a depth map of each image.
可选地，所述计算所述中药材三维解析模型的综合质量因子，包括：Optionally, the calculation of the comprehensive quality factor of the three-dimensional analytical model of the Chinese medicinal material includes:
以下述方式获得中药材三维解析模型的综合质量因子F(x)：The comprehensive quality factor F(x) of the three-dimensional analytical model of Chinese medicinal materials is obtained in the following way:
F(x)＝αf1(x)+βf2(x)+γf3(x)F(x)=αf1 (x)+βf2 (x)+γf3 (x)
其中，f1(x)、f2(x)、f3(x)分别表示模型差异度、模型完整性和贴图清晰度；α、β、γ分别表示模型差异度、模型完整性和贴图清晰度的权重值，α、β、γ∈[0,1]，并且α+β+γ＝1。Among them, f1 (x), f2 (x), f3 (x) represent the model difference, model integrity and map clarity, respectively; α, β, γ represent the model difference, model integrity and map clarity, respectively degree weight value, α, β, γ∈[0,1], and α+β+γ=1.
可选地，所述模型差异度、模型完整性和贴图清晰度的权重值被设置成根据中药材的纹理、形态的复杂程度和重建还原的效果动态调整。例如，权重值根据中药材的纹理、形态的复杂程度和重建还原的效果而定，对于形态较为复杂的中药材，模型完整性为主要影响因素，模型完整性权重值β∈[0.5,1)，如威灵仙中药材；对于纹理比较复杂的中药材，贴图的清晰度为主要影响因素，贴图清晰度权重值γ∈[0.5,1)，如羌活中药材；对于纹理和形态均比较复杂的中药材，模型完整性和贴图清晰度的权重值β、γ∈(0,0.4]，如黄连(味莲)中药材。Optionally, the weight values of the model difference degree, model integrity and map clarity are set to be dynamically adjusted according to the texture of the Chinese medicinal material, the complexity of the shape and the effect of reconstruction and restoration. For example, the weight value is determined according to the texture of the Chinese medicinal material, the complexity of the shape and the effect of reconstruction and restoration. For the Chinese medicinal material with complex shape, the model integrity is the main influencing factor, and the model integrity weight value β∈[0.5,1) , such as Weilingxian Chinese medicinal materials; for Chinese medicinal materials with complex textures, the clarity of the map is the main factor, and the weight of the map clarity is γ∈[0.5,1), such as Qianghuo Chinese medicinal materials; for the texture and shape are more complex The weight values of model integrity and map clarity are β, γ∈(0, 0.4], such as Coptis chinensis (flavor lotus) Chinese herbal medicine.
可选地，所述基于综合质量因子与预设要求的比较，输出中药材的三维解析模型，包括：Optionally, based on the comparison between the comprehensive quality factor and the preset requirements, output a three-dimensional analytical model of the Chinese medicinal material, including:
基于综合质量因子达到预设要求，则输出中药材的三维解析模型；否则，修正所述中药材的三维解析模型，直到综合质量因子达到预设要求；Based on the comprehensive quality factor reaching the preset requirement, output the three-dimensional analytical model of the Chinese medicinal material; otherwise, correct the three-dimensional analytical model of the Chinese medicinal material until the comprehensive quality factor reaches the predetermined requirement;
在达到预设修正结束条件时，如果综合质量因子仍未达到预设要求，则选择综合质量因子评估最高值对应的模型作为最终输出的中药材三维解析模型。When the preset correction end condition is reached, if the comprehensive quality factor still does not meet the preset requirements, the model corresponding to the highest value of the comprehensive quality factor evaluation is selected as the final output three-dimensional analytical model of traditional Chinese medicinal materials.
可选地，所述修正所述中药材的三维解析模型，包括：Optionally, the revising the three-dimensional analytical model of the Chinese medicinal material includes:
对于中药材三维解析模型中的每一点(u,v)，根据所述图像序列中第i张图像中包含的具有相同尺寸的点(u’,v’)的光照强度、光照方向和灰度值进行修正，细化纹理图贴；i＝1，2，3，…， N；N表示图像序列中的图像的总量；u,v分别表示中药材三维解析模型中的每一点的行坐标和列坐标；u’,v’分别表示图像序列中第i张图像中包含的具有相同尺寸的点的行坐标和列坐标。For each point (u, v) in the three-dimensional analytical model of Chinese medicinal materials, according to the light intensity, light direction and gray level of the point (u', v') with the same size contained in the i-th image in the image sequence The value is corrected to refine the texture map; i=1, 2, 3, ..., N; N represents the total number of images in the image sequence; u, v represent the row coordinates of each point in the three-dimensional analytical model of Chinese medicinal materials, respectively and column coordinates; u', v' represent the row and column coordinates of the point with the same size contained in the ith image in the image sequence, respectively.
本发明实施例的中药材三维数字模型的制作方法，能够还原中药材的纹理、形态等鉴定特征，获得高精度的中药材三维数字标本模型，模型贴图清晰、完整性高、还原度高，实现了中药材的三维数字化；本发明可以重建出中药材的纹理、形态等鉴定特征，本发明能够构建中药材切片或中药饮片的三维数字标本和中药材的三维数字标本，可实现中药材局部(如切片、饮片)和整体的三维数字化。The method for making a three-dimensional digital model of a traditional Chinese medicinal material according to the embodiment of the present invention can restore the identification features such as texture and shape of the traditional Chinese medicinal material, and obtain a high-precision three-dimensional digital specimen model of the traditional Chinese medicinal material. The three-dimensional digitization of Chinese medicinal materials is realized; the present invention can reconstruct the texture, shape and other identification features of Chinese medicinal materials, and the present invention can construct three-dimensional digital specimens of Chinese medicinal materials slices or Chinese medicinal decoction pieces and three-dimensional digital specimens of traditional Chinese medicinal materials, and can realize partial ( Such as slices, decoction pieces) and the overall 3D digitization.
应当理解，前述构思以及在下面更加详细地描述的额外构思的所有组合只要在这样的构思不相互矛盾的情况下都可以被视为本公开的发明主题的一部分。另外，所要求保护的主题的所有组合都被视为本公开的发明主题的一部分。It is to be understood that all combinations of the foregoing concepts, as well as additional concepts described in greater detail below, are considered to be part of the inventive subject matter of the present disclosure to the extent that such concepts are not contradictory. Additionally, all combinations of the claimed subject matter are considered to be part of the inventive subject matter of this disclosure.
结合附图从下面的描述中可以更加全面地理解本发明教导的前述和其他方面、实施例和特征。本发明的其他附加方面例如示例性实施方式的特征和/或有益效果将在下面的描述中显见，或通过根据本发明教导的具体实施方式的实践中得知。The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description when taken in conjunction with the accompanying drawings. Other additional aspects of the invention, such as features and/or benefits of the exemplary embodiments, will be apparent from the description below, or learned by practice of specific embodiments in accordance with the teachings of this invention.
附图说明Description of drawings
附图不意在按比例绘制。在附图中，在各个图中示出的每个相同或近似相同的组成部分可以用相同的标号表示。为了清晰起见，在每个图中，并非每个组成部分均被标记。现在，将通过例子并参考附图来描述本发明的各个方面的实施例，其中：The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in every figure. Embodiments of various aspects of the present invention will now be described by way of example and with reference to the accompanying drawings, wherein:
图1是本发明实施例的中药材三维数字模型的制作方法的流程图。FIG. 1 is a flowchart of a method for making a three-dimensional digital model of a Chinese medicinal material according to an embodiment of the present invention.
图2是根据本发明实施例的中药材三维数字模型的制作方法的一个示例流程图。FIG. 2 is an exemplary flowchart of a method for making a three-dimensional digital model of a Chinese medicinal material according to an embodiment of the present invention.
图3-1、3-2是拍摄的黄芪片中药材以及通过本发明实施例的方法制作的三维数字标本的图片示例。Figures 3-1 and 3-2 are examples of pictures of Chinese medicinal materials of Astragalus tablets and three-dimensional digital specimens produced by the method of the embodiment of the present invention.
图4-1、4-2是拍摄的人参中药材以及通过本发明实施例的方法制作的三维数字标本的图片示例。Figures 4-1 and 4-2 are examples of pictures of Chinese medicinal materials of ginseng and three-dimensional digital specimens produced by the method of the embodiment of the present invention.
具体实施方式Detailed ways
为了更了解本发明的技术内容，特举具体实施例并配合所附图式说明如下。In order to better understand the technical content of the present invention, specific embodiments are given and described below in conjunction with the accompanying drawings.
在本公开中参照附图来描述本发明的各方面，附图中示出了许多说明的实施例。本公开的实施例不必定意在包括本发明的所有方面。应当理解，上面介绍的多种构思和实施例，以及下面更加详细地描述的那些构思和实施方式可以以很多方式中任意一种来实施，这是因为本发明所公开的构思和实施例并不限于任何实施方式。另外，本发明实施例公开的一些方面可以单独使用，或者与本发明实施例公开的其他方面的任何适当组合来使用。Aspects of the invention are described in this disclosure with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to include all aspects of the invention. It should be understood that the various concepts and embodiments described above, as well as those described in greater detail below, can be implemented in any of a number of ways, as the concepts and embodiments disclosed herein do not limited to any implementation. In addition, some aspects disclosed in the embodiments of the present invention may be used alone or in any suitable combination with other aspects disclosed in the embodiments of the present invention.
结合图1所示的示例性实施例的中药材三维数字模型的制作方法，包括以下步骤：In conjunction with the manufacturing method of the three-dimensional digital model of Chinese medicinal materials of the exemplary embodiment shown in FIG. 1, the following steps are included:
获取在同一高度和背景下拍摄的中药材的多角度图像，构造图像序列；Obtain multi-angle images of Chinese herbal medicines taken at the same height and background, and construct image sequences;
基于图像序列进行三维重建，获得中药材的点云空间模型；Perform 3D reconstruction based on image sequences to obtain the point cloud space model of Chinese herbal medicines;
对该中药材的点云空间模型进行网格化，生成网格，获得该中药材的初始模型；Gridding the point cloud space model of the Chinese medicinal material to generate a grid to obtain the initial model of the Chinese medicinal material;
在该初始模型的基础上，生成纹理，获得有纹理的中药材的三维解析模型；On the basis of the initial model, a texture is generated to obtain a three-dimensional analytical model of the textured Chinese herbal medicine;
基于该图像序列，根据中药材特性，计算该中药材的三维解析模型的综合质量因子；以及Based on the image sequence, according to the properties of the Chinese medicinal material, calculate the comprehensive quality factor of the three-dimensional analytical model of the Chinese medicinal material; and
基于综合质量因子与预设要求的比较，获得最终的中药材的三维解析模型，并输出。Based on the comparison of the comprehensive quality factor and the preset requirements, the final three-dimensional analytical model of the Chinese herbal medicine is obtained and output.
可选地，该基于图像序列进行三维重建包括：Optionally, the three-dimensional reconstruction based on the image sequence includes:
对该图像序列中的多角度图像提取特征值，进行稀疏重建；Extract feature values from the multi-angle images in the image sequence, and perform sparse reconstruction;
基于该图像序列以及稀疏重建获得的拍摄参数，进行稠密重建，获得中药材的点云空间模型。Based on the image sequence and the shooting parameters obtained by sparse reconstruction, dense reconstruction is performed to obtain the point cloud space model of Chinese medicinal materials.
其中，拍摄参数包括相机内参和相机外参。The shooting parameters include camera internal parameters and camera external parameters.
相机内参用以描述相机的固有属性，包括焦距和主点位置。相机外参用以描述相机在世界空间的坐标关系，包括旋转关系与平移关系。Camera intrinsics describe the intrinsic properties of the camera, including focal length and principal point position. The camera external parameters are used to describe the coordinate relationship of the camera in the world space, including the rotation relationship and translation relationship.
可选地，对该多角度图像提取特征值，进行稀疏重建，包括：Optionally, extract feature values from the multi-angle image and perform sparse reconstruction, including:
基于该多角度图像提取特征值，进行该多角度图像之间的特征点匹配，确定该多角度图像之间的匹配关系，从而获得极几何图和拍摄参数，完成稀疏重建。Extract feature values based on the multi-angle images, perform feature point matching between the multi-angle images, and determine the matching relationship between the multi-angle images, thereby obtaining a polar geometry map and shooting parameters, and completing sparse reconstruction.
其中，极几何图的每个顶点表示一幅图像；每一条边连接了公共可见区域的两张图像，每一条边表达了对应两个图像对之间的极几何关系。Among them, each vertex of the polar geometric graph represents an image; each edge connects two images in the common visible area, and each edge expresses the polar geometric relationship between the corresponding two image pairs.
可选地，该基于该图像序列以及该拍摄参数，进行稠密重建，包括：Optionally, the dense reconstruction is performed based on the image sequence and the shooting parameters, including:
以该图像序列为基础，按顺序为其中的每一个图像选择领域图像，构造立体图像对；Based on the image sequence, select a domain image for each image in sequence, and construct a stereo image pair;
计算每一个图像的深度图；Calculate the depth map of each image;
基于该立体图像对的领域图像将该每一个图像的深度图进行融合，完成稠密重建。Based on the domain image of the stereo image pair, the depth map of each image is fused to complete the dense reconstruction.
在可选的示例中，可线性融合函数进行深度图的融合。In an optional example, a linear fusion function can be used to perform fusion of depth maps.
可选地，该计算每一个图像的深度图，包括：Optionally, the calculation of the depth map of each image includes:
根据该相机内参以及该极几何图的极线计算每一个图像的每一个像素上的深度值；Calculate the depth value on each pixel of each image according to the camera intrinsic parameter and the epipolar line of the polar geometry map;
根据该相机外参，确定每一个像素的前后关系，并根据该前后关系，舍弃后方被遮挡的像素并保留前方未被遮挡的像素，获得每一个图像的深度图。According to the camera extrinsic parameters, determine the front-to-back relationship of each pixel, and according to the front-to-back relationship, discard the occluded pixels in the back and keep the unoccluded pixels in the front, and obtain the depth map of each image.
可选地，该计算中药材的三维解析模型的综合质量因子，包括：Optionally, the comprehensive quality factor of the calculated three-dimensional analytical model of Chinese medicinal materials includes:
以下述方式获得中药材三维解析模型的综合质量因子F(x)：The comprehensive quality factor F(x) of the three-dimensional analytical model of Chinese medicinal materials is obtained in the following way:
F(x)＝αf1(x)+βf2(x)+γf3(x)F(x)=αf1 (x)+βf2 (x)+γf3 (x)
其中，f1(x)、f2(x)、f3(x)分别表示模型差异度、模型完整性和贴图清晰度；α、β、γ分别表示模型差异度、模型完整性和贴图清晰度的权重值，α、β、γ∈[0,1]，并且α+β+γ＝1。Among them, f1 (x), f2 (x), f3 (x) represent the model difference, model integrity and map clarity, respectively; α, β, γ represent the model difference, model integrity and map clarity, respectively degree weight value, α, β, γ∈[0,1], and α+β+γ=1.
可选地，该模型差异度、模型完整性和贴图清晰度的权重值被设置成根据中药材的纹理、形态的复杂程度和重建还原的效果动态调整。例如，权重值根据中药材的纹理、形态的复杂程度和重建还原的效果而定，对于形态较为复杂的中药材，模型完整性为主要影响因素，模型完整性权重值β∈[0.5,1)，如威灵仙中药材；对于纹理比较复杂的中药材，贴图的清晰度为主要影响因素，贴图清晰度权重值γ∈[0.5,1)，如羌活中药材；对于纹理和形态均比较复杂的中药材，模型完整性和贴图清晰度的权重值β、γ∈(0,0.4]，如黄连(味莲)中药材。Optionally, the weight values of the model difference degree, model integrity and map clarity are set to be dynamically adjusted according to the texture of the Chinese medicinal material, the complexity of the shape and the effect of reconstruction and restoration. For example, the weight value is determined according to the texture of the Chinese medicinal material, the complexity of the shape and the effect of reconstruction and restoration. For the Chinese medicinal material with complex shape, the model integrity is the main influencing factor, and the model integrity weight value β∈[0.5,1) , such as Weilingxian Chinese medicinal materials; for Chinese medicinal materials with complex textures, the clarity of the map is the main factor, and the weight of the map clarity is γ∈[0.5,1), such as Qianghuo Chinese medicinal materials; for the texture and shape are more complex The weight values of model integrity and map clarity are β, γ∈(0, 0.4], such as Coptis chinensis (flavor lotus) Chinese herbal medicine.
可选地，该基于综合质量因子与预设要求的比较，获得最终的中药材的三维解析模型并输出，包括：Optionally, based on the comparison of the comprehensive quality factor and the preset requirements, the final three-dimensional analytical model of the Chinese medicinal material is obtained and output, including:
基于综合质量因子达到预设要求，则输出该中药材的三维解析模型；否则，修正该中药材的三维解析模型，直到综合质量因子达到预设要求；Based on the comprehensive quality factor meeting the preset requirements, output the three-dimensional analytical model of the Chinese medicinal material; otherwise, correct the three-dimensional analytical model of the traditional Chinese medicinal material until the comprehensive quality factor meets the preset requirements;
在达到预设修正结束条件时，如果综合质量因子仍未达到预设要求，则选择综合质量因子评估最高值对应的模型作为最终输出的中药材三维解析模型。When the preset correction end condition is reached, if the comprehensive quality factor still does not meet the preset requirements, the model corresponding to the highest value of the comprehensive quality factor evaluation is selected as the final output three-dimensional analytical model of traditional Chinese medicinal materials.
可选地，该修正该中药材的三维解析模型，包括：Optionally, the correction of the three-dimensional analytical model of the Chinese medicinal material includes:
对于中药材三维解析模型中的每一点(u,v)，根据该图像序列中第i张图像中包含的具有相同尺寸的点(u’,v’)的光照强度、光照方向和灰度值进行修正，细化纹理图贴；i＝1，2，3，…， N；N表示图像序列中的图像的总量；u,v分别表示中药材的三维解析模型中的每一点的行坐标和列坐标；u’,v’分别表示图像序列中第i张图像中包含的具有相同尺寸的点的行坐标和列坐标。For each point (u, v) in the three-dimensional analytical model of Chinese medicinal materials, according to the light intensity, light direction and gray value of the point (u', v') with the same size contained in the i-th image in the image sequence Make corrections and refine texture maps; i=1, 2, 3, ..., N; N represents the total number of images in the image sequence; u, v represent the row coordinates of each point in the three-dimensional analytical model of Chinese medicinal materials, respectively and column coordinates; u', v' represent the row and column coordinates of the point with the same size contained in the ith image in the image sequence, respectively.
优选的实施例中，在每一次修正中药材三维解析模型之后，以修正后的中药材三维解析模型为评估对象，继续评估综合质量因子，如果综合质量因子达到预定要求，则输出修正的模型，否则判断当前的图像序列编号i是否等于N，如果否，则置i＝i+1，并继续进行模型修正，如果i等于N，则选择综合质量因子评估最高值对应的模型作为最终输出的中药材三维解析模型。In a preferred embodiment, after each correction of the three-dimensional analytical model of Chinese medicinal materials, the corrected three-dimensional analytical model of traditional Chinese medicinal materials is used as the evaluation object, and the comprehensive quality factor is continuously evaluated. If the comprehensive quality factor meets the predetermined requirements, the corrected model is output. Otherwise, judge whether the current image sequence number i is equal to N, if not, set i=i+1, and continue to perform model correction, if i is equal to N, select the model corresponding to the highest value of the comprehensive quality factor evaluation as the final output of traditional Chinese medicine 3D analytical model of the material.
以植物中药材为例，更加具体的介绍前述中药材三维解析模型的制作过程。Taking plant Chinese medicinal materials as an example, the production process of the aforementioned three-dimensional analytical model of Chinese medicinal materials is introduced in more detail.
结合图2的示例流程的中药材三维数字模型的制作方法，包括以下步骤：The method for making a three-dimensional digital model of a traditional Chinese medicinal material in conjunction with the example process shown in FIG. 2 includes the following steps:
步骤1、获取在同一高度和背景下拍摄的中药材的多角度图像，构造图像序列；Step 1. Obtain multi-angle images of Chinese medicinal materials taken at the same height and background, and construct an image sequence;
步骤2、对多角度图像提取特征值，进行稀疏重建，获得极几何图和拍摄参数；Step 2. Extract feature values from the multi-angle image, perform sparse reconstruction, and obtain a polar geometry map and shooting parameters;
步骤3、基于图像序列以及拍摄参数，进行稠密重建，获得中药材的点云空间模型；Step 3. Based on the image sequence and shooting parameters, dense reconstruction is performed to obtain the point cloud space model of the Chinese herbal medicine;
步骤4、对中药材的点云空间模型进行网格化，生成网格，获得中药材的初始模型；Step 4. Gridding the point cloud space model of the Chinese medicinal material to generate a grid to obtain the initial model of the Chinese medicinal material;
步骤5、在初始模型的基础上，生成纹理，获得有纹理的中药材的三维解析模型；Step 5. On the basis of the initial model, a texture is generated to obtain a three-dimensional analytical model of the textured Chinese herbal medicine;
步骤6、基于图像序列，结合中药材特性，计算中药材的三维解析模型的综合质量因子，包括模型差异度、模型完整性和贴图清晰度，如果综合质量因子达到预定的要求，则输出模型，否则进入步骤7；Step 6: Calculate the comprehensive quality factor of the three-dimensional analytical model of the Chinese medicinal material based on the image sequence and the characteristics of the Chinese medicinal material, including the model difference, model integrity and map clarity. If the comprehensive quality factor meets the predetermined requirements, the model is output. Otherwise, go to step 7;
步骤7、对于中药材三维解析模型中的每一点(u,v)，根据图像序列中第i张图像中包含的具有相同尺寸的点(u’,v’)的光照强度、光照方向和灰度值进行修正，细化纹理图贴；i＝1， 2,3,…，N；Step 7. For each point (u, v) in the three-dimensional analytical model of Chinese medicinal materials, according to the light intensity, light direction and gray level of the point (u', v') with the same size contained in the ith image in the image sequence Correct the degree value and refine the texture map; i=1, 2, 3, ..., N;
步骤8、以修正后的中药材三维解析模型为评估对象，如果综合质量因子达到预定的要求，则输出模型，否则判断i是否等于N，如果否，则置i＝i+1，并继续执行步骤7进行模型修正，如果i等于N，则选择综合质量因子评估最高值对应的模型作为最终输出的中药材三维解析模型。Step 8. Take the revised three-dimensional analytical model of traditional Chinese medicinal materials as the evaluation object. If the comprehensive quality factor meets the predetermined requirements, output the model, otherwise judge whether i is equal to N, if not, set i=i+1, and continue to execute In step 7, model correction is performed. If i is equal to N, the model corresponding to the highest value of the comprehensive quality factor evaluation is selected as the final output three-dimensional analytical model of Chinese medicinal materials.
其中，在步骤1中，包括拍摄准备和采集照片。Among them, in step 1, including shooting preparation and collecting photos.
拍摄准备是指，开启柔光灯进行多点式布补光，优选为三点均匀分布式补光，在均匀的光照条件下，将中药材样品垂直固定于转盘中心台上，中药标本保持竖直。Shooting preparation refers to turning on the soft light for multi-point cloth fill light, preferably three-point evenly distributed fill light, and under uniform lighting conditions, the Chinese medicinal material sample is vertically fixed on the center table of the turntable, and the traditional Chinese medicine sample is kept upright. straight.
然后，使用三脚架稳定拍摄相机，拍摄采集照片。Then, use a tripod to stabilize the camera and take the capture photo.
拍摄过程中，尤其选择相机在微距的模式下，对中药材进行分段堆叠拍摄。During the shooting process, the camera is especially selected to shoot the Chinese herbal medicines in sections and stacks in the macro mode.
在可选的实施例中，还包括拍摄标准色卡的操作，以利于在后续图像处理之前，通过色卡校色。In an optional embodiment, the operation of photographing a standard color card is also included, so as to facilitate color correction through the color card before subsequent image processing.
在步骤2中，基于多角度图像提取特征值，进行图像之间的特征点匹配，确定图像之间的匹配关系，确定出极几何图，从而可获得稀疏点云。In step 2, feature values are extracted based on the multi-angle images, feature point matching between the images is performed, the matching relationship between the images is determined, and a polar geometric map is determined, thereby obtaining a sparse point cloud.
在步骤2中，通过稀疏重建，可获得拍摄参数中的相机内参与相机外参。In step 2, through sparse reconstruction, the in-camera participating camera extrinsic parameters in the shooting parameters can be obtained.
在可选的实施例中，可基于开源库工具，例如OpenMVG库(open Multiple ViewGeometry)，通过SIFT特征提取和匹配，进行稀疏重建。In an optional embodiment, sparse reconstruction can be performed through SIFT feature extraction and matching based on open source library tools, such as OpenMVG library (open Multiple ViewGeometry).
在步骤3中，基于深度图融合的方式进行稠密重建，建立中药材的点云空间模型，包括：In step 3, dense reconstruction is performed based on depth map fusion, and a point cloud space model of Chinese medicinal materials is established, including:
以图像序列为基础，按顺序为其中的每一个图像选择领域图像，构造立体图像对；Based on the image sequence, select a domain image for each image in sequence, and construct a stereo image pair;
基于图像一致性约束的原则，计算每一个图像的深度图；Calculate the depth map of each image based on the principle of image consistency constraints;
采用addWeighted线性融合函数，基于领域图像对每一个图像的深度图进行融合，完成稠密重建。The addWeighted linear fusion function is used to fuse the depth map of each image based on the domain image to complete the dense reconstruction.
其中，在步骤4中，计算每一个图像的深度图，包括：Among them, in step 4, the depth map of each image is calculated, including:
根据相机内参以及极几何图的极线(Epipolar line)计算每一个图像的每一个像素上的深度值；Calculate the depth value on each pixel of each image according to the camera internal parameters and the epipolar line of the polar geometry map;
根据相机外参，确定每一个像素的前后关系，并根据前后关系，舍弃后方被遮挡的像素并保留前方未被遮挡的像素，获得每一个图像的深度图。According to the external parameters of the camera, determine the front-back relationship of each pixel, and according to the front-back relationship, discard the occluded pixels in the back and keep the unoccluded pixels in the front, and obtain the depth map of each image.
在步骤5中，可采用现有的纹理模型，进行纹理生成，例如基于GAN网络或者VGG-19卷积神经网络的纹理生成模型而获得。In step 5, an existing texture model can be used for texture generation, for example, a texture generation model based on a GAN network or a VGG-19 convolutional neural network can be obtained.
在计算中药材的三维解析模型的综合质量因子时，可根据重建的中药材的特性和还原效果确定三个权重值α、β、γ的大小。例如，权重值根据中药材的纹理、形态的复杂程度和重建还原的效果而定，对于形态较为复杂的中药材，模型完整性为主要影响因素，模型完整性权重值β∈[0.5,1)，如威灵仙中药材；对于纹理比较复杂的中药材，贴图的清晰度为主要影响因素，贴图清晰度权重值γ∈[0.5,1)，如羌活中药材；对于纹理和形态均比较复杂的中药材，模型完整性和贴图清晰度的权重值β、γ∈(0,0.4]，如黄连(味莲)中药材。When calculating the comprehensive quality factor of the three-dimensional analytical model of Chinese medicinal materials, the three weight values α, β and γ can be determined according to the characteristics of the reconstructed Chinese medicinal materials and the reduction effect. For example, the weight value is determined according to the texture of the Chinese medicinal material, the complexity of the shape and the effect of reconstruction and restoration. For the Chinese medicinal material with complex shape, the model integrity is the main influencing factor, and the model integrity weight value β∈[0.5,1) , such as Weilingxian Chinese medicinal materials; for Chinese medicinal materials with complex textures, the clarity of the map is the main factor, and the weight of the map clarity is γ∈[0.5,1), such as Qianghuo Chinese medicinal materials; for the texture and shape are more complex The weight values of model integrity and map clarity are β, γ∈(0, 0.4], such as Coptis chinensis (flavor lotus) Chinese herbal medicine.
在本发明的实施例中，如果达不到预定的三维重建的图像质量要求，可根据不同的应用预设质量要求的具体限定，例如中药材三维解析模型的综合质量因子F(x)未达到预设的阈值，或者至少两个质量因子未达到预设值的要求，则需要对获得三维解析模型进行协同修复，修正纹理贴图质量，以提高中药材表面纹理表征。In the embodiment of the present invention, if the predetermined image quality requirements for 3D reconstruction cannot be met, the specific limitations of the quality requirements can be preset according to different applications. For example, the comprehensive quality factor F(x) of the 3D analytical model of Chinese medicinal materials does not meet If the preset thresholds, or at least two quality factors do not meet the requirements of the preset values, it is necessary to perform collaborative repair on the obtained 3D analytical model and correct the quality of texture maps to improve the surface texture representation of Chinese herbal medicines.
在本发明的实施例中，通过UV拆分，结合光照强度、光照方向和灰度值进行修正，提高表面纹理贴图的细节表征和图像重建质量。In the embodiment of the present invention, through UV splitting, correction is carried out in combination with illumination intensity, illumination direction and grayscale value, so as to improve the detailed representation of the surface texture map and the quality of image reconstruction.
作为可选的实施例中，针对中药材三维解析模型中的每一点(u,v)，根据图像序列中第i 张图像中包含的具有相同尺寸的点(u’,v’)的光照强度、光照方向和灰度值进行修正，细化纹理图贴；i＝1，2,3,…，N；在本发明的实施例中，从图像序列中的第1张图开始，以其所包含的具有相同尺寸的点(u’,v’)的光照强度、光照方向和灰度值进行修正，在后续的迭代锡主义很过程中，顺序的执行其他的图像进行修正。As an optional embodiment, for each point (u, v) in the three-dimensional analytical model of Chinese medicinal materials, according to the light intensity of the point (u', v') with the same size contained in the ith image in the image sequence , the illumination direction and gray value are corrected, and the texture map is refined; i=1, 2, 3, ..., N; in the embodiment of the present invention, starting from the first picture in the image sequence, the The illumination intensity, illumination direction and gray value of the included points (u', v') with the same size are corrected, and in the subsequent iterative process, other images are sequentially performed for correction.
作为可选的方式，图像尺寸优选统一为4096*2160。As an optional method, the image size is preferably unified to 4096*2160.
结合图3-1、3-2以及4-1、4-2所示，分别表示拍摄的黄芪片中药材的照片以及通过本发明实施例的方法制作的三维数字标本的图片示例，通过图片的对比可见，通过本发明实施例的方法可以构建出中药材切片的三维数字标本和中药材的三维数字标本，实现中药材局部和整体的三维数字化，并且贴图清晰、完整性高、还原度高。3-1, 3-2, and 4-1, 4-2 respectively represent the pictures of the Chinese medicinal materials of Astragalus tablets taken and the picture examples of the three-dimensional digital specimens produced by the method of the embodiment of the present invention. It can be seen from the comparison that the three-dimensional digital specimen of sliced Chinese medicinal materials and the three-dimensional digital specimen of Chinese medicinal materials can be constructed by the method of the embodiment of the present invention, so as to realize the three-dimensional digitization of the part and the whole of the traditional Chinese medicinal material, and the mapping is clear, the integrity is high, and the reduction degree is high.
在另外的实施例中，还可以对图像序列中的每一个图像进行按照其分段的位置进行标记，在进行纹理贴图修正时，根据中药材三维解析模型中的点(u,v)的位置，根据其在不同序列的图像中的光照强度进行强弱排序，选择光照强度最大的进行优先修正。In another embodiment, each image in the image sequence may be marked according to its segmented position, and when the texture map is corrected, according to the position of the point (u, v) in the three-dimensional analytical model of the Chinese medicinal material , according to its light intensity in different sequences of images, the intensity is sorted, and the one with the largest light intensity is selected for priority correction.
然后，以修正后的中药材三维解析模型为评估对象，如果综合质量因子达到预定的要求，则输出模型，否则继续判断i是否等于N，如果否，则置i＝i+1，并继续执行步骤7，如果i 等于N，则输出综合质量因子评估最高值对应的模型。Then, take the revised three-dimensional analytical model of Chinese medicinal materials as the evaluation object, if the comprehensive quality factor meets the predetermined requirements, output the model, otherwise continue to judge whether i is equal to N, if not, set i=i+1, and continue to execute Step 7, if i is equal to N, output the model corresponding to the highest value of the comprehensive quality factor evaluation.
根据本发明公开的实施例，还提出一种中药材三维数字标本的制作系统，包括：According to the embodiments disclosed in the present invention, a system for making three-dimensional digital specimens of Chinese medicinal materials is also proposed, including:
一个或多个处理器；one or more processors;
存储器，存储可被操作的指令，指令在被一个或多个处理器执行时实现前述实施例的中药材三维数字模型的制作方法。The memory stores operable instructions, and the instructions, when executed by one or more processors, implement the method for making a three-dimensional digital model of Chinese medicinal materials in the foregoing embodiments.
根据本发明公开的实施例，还提出一种存储计算机程序的计算机可读取介质，计算机程序包括能被一个或多个计算机执行的指令，指令在被一个或多个计算机执行时实现前述实施例的中药材三维数字模型的制作方法。According to the disclosed embodiments of the present invention, there is also provided a computer-readable medium storing a computer program, the computer program comprising instructions executable by one or more computers, the instructions implementing the aforementioned embodiments when executed by one or more computers The production method of three-dimensional digital model of traditional Chinese medicinal materials.
虽然本发明已以较佳实施例揭露如上，然其并非用以限定本发明。本发明所属技术领域中具有通常知识者，在不脱离本发明的精神和范围内，当可作各种的更动与润饰。因此，本发明的保护范围当视权利要求书所界定者为准。Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those skilled in the art to which the present invention pertains can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be determined according to the claims.
Claims (10)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210619124.2ACN114998549A (en) | 2022-06-01 | 2022-06-01 | Method for making three-dimensional digital model of traditional Chinese medicinal materials and computer readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210619124.2ACN114998549A (en) | 2022-06-01 | 2022-06-01 | Method for making three-dimensional digital model of traditional Chinese medicinal materials and computer readable medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114998549Atrue CN114998549A (en) | 2022-09-02 |
Family
ID=83031748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210619124.2APendingCN114998549A (en) | 2022-06-01 | 2022-06-01 | Method for making three-dimensional digital model of traditional Chinese medicinal materials and computer readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114998549A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116704152A (en)* | 2022-12-09 | 2023-09-05 | 荣耀终端有限公司 | Image processing method and electronic device |
| CN120374870A (en)* | 2025-06-25 | 2025-07-25 | 苍南县求是中医药创新研究院 | Vision-based traditional Chinese medicine defect detection method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105868677A (en)* | 2015-01-19 | 2016-08-17 | 阿里巴巴集团控股有限公司 | A live face detection method and device |
| WO2017026839A1 (en)* | 2015-08-12 | 2017-02-16 | 트라이큐빅스 인크. | 3d face model obtaining method and device using portable camera |
| CN113345063A (en)* | 2021-08-05 | 2021-09-03 | 南京万生华态科技有限公司 | PBR three-dimensional reconstruction method, system and computer storage medium based on deep learning |
| CN114037802A (en)* | 2021-11-24 | 2022-02-11 | Oppo广东移动通信有限公司 | Three-dimensional face model reconstruction method and device, storage medium and computer equipment |
| CN114549729A (en)* | 2022-03-25 | 2022-05-27 | 北京奇艺世纪科技有限公司 | Method and device for generating texture map, electronic device and storage medium |
- 2022
- 2022-06-01CNCN202210619124.2Apatent/CN114998549A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105868677A (en)* | 2015-01-19 | 2016-08-17 | 阿里巴巴集团控股有限公司 | A live face detection method and device |
| WO2017026839A1 (en)* | 2015-08-12 | 2017-02-16 | 트라이큐빅스 인크. | 3d face model obtaining method and device using portable camera |
| CN113345063A (en)* | 2021-08-05 | 2021-09-03 | 南京万生华态科技有限公司 | PBR three-dimensional reconstruction method, system and computer storage medium based on deep learning |
| CN114037802A (en)* | 2021-11-24 | 2022-02-11 | Oppo广东移动通信有限公司 | Three-dimensional face model reconstruction method and device, storage medium and computer equipment |
| CN114549729A (en)* | 2022-03-25 | 2022-05-27 | 北京奇艺世纪科技有限公司 | Method and device for generating texture map, electronic device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 3D视觉工坊: "基于图像的三维模型重建——稠密点云重建", Retrieved from the Internet <URL:https://blog.csdn.net/yong_qi2015/article/details/105963647>* |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116704152A (en)* | 2022-12-09 | 2023-09-05 | 荣耀终端有限公司 | Image processing method and electronic device |
| CN116704152B (en)* | 2022-12-09 | 2024-04-19 | 荣耀终端有限公司 | Image processing method and electronic device |
| CN120374870A (en)* | 2025-06-25 | 2025-07-25 | 苍南县求是中医药创新研究院 | Vision-based traditional Chinese medicine defect detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110363858B (en) | Three-dimensional face reconstruction method and system | |
| CN109255831B (en) | A method for single-view face 3D reconstruction and texture generation based on multi-task learning | |
| Weber | Another link between archaeology and anthropology: virtual anthropology | |
| CN108537191B (en) | A three-dimensional face recognition method based on structured light camera | |
| US10217293B2 (en) | Depth camera-based human-body model acquisition method and network virtual fitting system | |
| WO2021238673A1 (en) | Tooth scanning method and apparatus, computer device, and computer-readable storage medium | |
| WO2015149302A1 (en) | Method for rebuilding tree model on the basis of point cloud and data driving | |
| Tian et al. | Efficient computer-aided design of dental inlay restoration: a deep adversarial framework | |
| CN110222628A (en) | A kind of face restorative procedure based on production confrontation network | |
| DE102019106666A1 (en) | Dental CAD automation using deep learning | |
| CN114998549A (en) | Method for making three-dimensional digital model of traditional Chinese medicinal materials and computer readable medium | |
| CN110363750B (en) | An automatic extraction method of root canal morphology based on multimodal data fusion | |
| CN102222357B (en) | Foot Shape 3D Surface Reconstruction Method Based on Image Segmentation and Mesh Subdivision | |
| CN105160346B (en) | A kind of greasy recognition methods of curdy fur on tongue based on texture and distribution characteristics | |
| CN111583128B (en) | Face image highlight removal method based on deep learning and photorealistic rendering | |
| CN103871100A (en) | Method for rebuilding tree model based on point cloud and data driving | |
| CN107221029A (en) | A 3D Image Reconstruction Method | |
| Buzi et al. | Measuring the shape: performance evaluation of a photogrammetry improvement applied to the Neanderthal skull Saccopastore 1 | |
| CN116402943A (en) | Indoor 3D reconstruction method and device based on signed distance field | |
| CN110047075A (en) | A kind of CT image partition method based on confrontation network | |
| CN111986216A (en) | RSG liver CT image interactive segmentation algorithm based on neural network improvement | |
| Zhang et al. | Data-driven flower petal modeling with botany priors | |
| CN108986210A (en) | The method and apparatus of 3 D scene rebuilding | |
| CN116740703A (en) | Wheat phenotype parameter change rate estimation method and device based on point cloud information | |
| Gunz | Computed tools for paleoneurology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |