173 results about "Digital reconstruction" patented technology

This document generally describes methods and systems for generating digital reconstructions of tissue from humans or other species. The method can, for example, include receiving, at a computing system, image data of a tissue sample, where one or more sections of the tissue sample are stained with hematoxylin and eosin (H&E), registering the image data to generate registered image data, identifying tissue subtypes based on application of a machine learning model to the registered image data, annotating the identified tissue subtypes to generate annotated image data, and determining a digital volume of the tissue sample in three dimensional (3D) space based on the annotated image data. The disclosed technology can provide for single-cell analysis and other analysis of tissue samples, such as early detection of cancer in human tissue samples.

The invention discloses a cultural heritage digital reconstruction system based on deep learning, and the system comprises a data collection module, a data preprocessing module, a feature description module, a modeling module, and a model optimization module. The feature description module is used for acquiring data processed by the data preprocessing module, feature extraction is performed on point cloud data through a point cloud processing network in deep learning, the data preprocessing module is used for data preprocessing, the data preprocessing comprises correction, denoising, color correction and cutting, and the modeling module is used for modeling the point cloud data according to the extracted features. Three-dimensional reconstruction of the cultural heritage is carried out, and a model optimization module is used for smoothing a generated cultural heritage model, filling the deficiency and eliminating noise; the cultural heritage digital reconstruction method has the advantages that the digital system has rich details and sense of reality, people can conveniently visit and learn, and the efficiency and precision of cultural heritage digital reconstruction are improved.

The invention discloses a tunnel face digital reconstruction method based on multi-source data fusion, and the method comprises the steps: collecting the multi-source data, such as geological, terrain and landform exploration data, while-drilling real-time data, laser scanning point cloud data and unmanned aerial vehicle shot image infrared data, and carrying out the preprocessing and fusion of the multi-source data; according to the method, deep integration of multi-source data fusion and a digital technology is realized, on the basis of real-time collection of multi-source data, through combination of geometric modeling and mechanical analysis, a crack expansion model is constructed, a tunnel face digital reconstruction technology is constructed, and the shape and stress state of the tunnel face are visually and dynamically displayed, so that the tunnel face is reconstructed. And accurate reconstruction and dynamic tracking of the tunnel face are realized.

The invention discloses a digital human rendering method based on Gaussian splashing and multi-scale characteristic field distillation, and belongs to the field of three-dimensional human body digital reconstruction. According to the method, feature extraction is carried out through a Vision Transformer encoder, based on an SMPL model, through a cross-modal parameter estimation module and dynamic human body modeling of three-dimensional Gaussian splashing and semantic feature rendering, three-dimensional Gaussian is projected to a two-dimensional image plane to calculate a covariance matrix and color mixing, and after a rendered color image and an initial feature field image are output, a three-dimensional image is obtained. A student feature map is obtained through a convolution acceleration module, a teacher feature map is obtained after feature extraction is carried out through a two-dimensional basic model, the constructed model is trained, and optimization training is completed through comprehensive total loss function calculation; according to the method, the problems of fuzzy semantics, detail missing, low rendering efficiency and inaccurate human body-scene separation in the existing method are effectively solved, so that more efficient, fine and robust monocular or multi-view human body three-dimensional reconstruction is realized.

The invention discloses a rear-mounted holographic focusing schlieren imaging method. The method comprises the following steps: after a laser emits a laser beam, the laser beam is divided into a main light source and holographic reference light; a main light source sequentially passes through a multi-stage beam expander, a soft light screen, a Fresnel lens and a source grating to obtain light beams with alternate brightness and darkness, and the light beams irradiate a flow field; the light beam penetrating through the flow field is focused by the focusing lens, an image of the source grating is formed at the knife edge grating through the focusing lens, and the light beam penetrating through the flow field is modulated by the knife edge grating, so that an image of the section of the flow field is formed behind the knife edge grating; the light beam modulated by the knife edge grid passes through the field lens and is recorded together with the holographic reference light to obtain a holographic focusing schlieren map; performing digital reconstruction to obtain three-dimensional flow field position information; and calculating the three-dimensional density gradient and the density value of the flow field based on the imaging principle of focusing schlieren. The invention further discloses a rear-mounted holographic focusing schlieren imaging system. According to the method and the system, the ultra-high-speed flow field can be measured, and the requirement of an ultra-high-speed aircraft test on fine quantitative measurement of the three-dimensional flow field is met.

The invention belongs to the technical field of customer service, and particularly relates to an intelligent customer service practical training system and method, and the system comprises a business scene digital reconstruction layer which is used for constructing a business training scene to form a training scene library, managing the business training scene library, and reconstructing a training scene according to scene configuration data; the AI simulation interaction core layer is used for receiving and recognizing voice and text input of students and generating behavior strategies of AI partner training according to input content on the basis of a current training scene; the training execution layer is used for generating scene configuration data according to a student selection instruction and sending the scene configuration data to the business scene digital reconstruction layer, and the scene configuration data content comprises a training scene, training difficulty and role setting; the training difficulty is dynamically adjusted according to the real-time performance of the trainee; and the quantitative evaluation layer is used for performing multi-dimensional analysis on the training data and generating a quantitative evaluation result and an improvement suggestion. The problems that existing customer service training is low in efficiency and high in cost are solved.

The invention discloses a 3D GS cultural relic digital reconstruction method and system based on a block chain, and the method comprises the steps: collecting the RGB image data and depth perception data of a cultural relic, eliminating the influence of different shooting conditions through an illumination separation processing technology, and building a standardized image data set; recognizing a surface area suitable for reconstruction based on image analysis, determining feature point distribution by using kernel density estimation, and generating initial three-dimensional representation through Gaussian ellipsoid fitting; performing gradient calculation and feature extraction on the depth data, and combining with Gaussian representation to form a geometric constraint mechanism; self-adaptive encryption based on visual importance is executed for a sparse region, and a layered rendering effect is achieved through opacity parameter adjustment; the rendering characteristics and the conversion relation of different view angles are analyzed, key observation points are determined through stability analysis, and a smooth multi-view-angle display sequence is constructed; and integrating multi-view rendering information to generate volumetric representation, and completing right confirmation of the high-quality three-dimensional digital model through digital signature.

The invention discloses a traffic accident scene digital reconstruction method based on ground-air cooperation. The method comprises the following steps: acquiring image data of an accident scene by using a five-point calibration method and a three-direction camera shooting method; performing denoising, enhancement and geometric distortion correction on the image data of the accident scene to obtain preprocessed image data; key feature points are extracted from the preprocessed image data, and feature point matching is carried out between different images to obtain a feature point matching result; and reconstructing an accident scene digital reconstruction model according to a feature point matching result and the preprocessed image data. According to the invention, the precision of a measurement result can be ensured, and the safety of traffic accident scene investigation work and the comprehensiveness of investigation data are ensured.

The invention relates to the technical field of case analysis, and discloses a multi-case collision method based on file-related personnel archives, which comprises the following steps of: integrating multi-source heterogeneous data such as file-related personnel, time tracks, geographic positions and the like, constructing a three-dimensional digital twinborn body of a case through a hierarchical feature extraction and cross-modal alignment technology, and establishing a three-dimensional digital twinborn body of the case; causal association between case elements is discovered based on data driving and domain knowledge double channels, after an initial causal atlas is constructed, illegal logic connection is automatically filtered through a reality constraint library, and a trusted causal network conforming to law enforcement specifications is generated. According to the method, a constraint enhanced deduction framework is constructed, physical logic compliance guarantee of an anti-fact scene is realized on the basis of keeping case total-factor digital reconstruction and dynamic association mining capability, and a reconnaissance strategy scheme conforming to real law enforcement constraints and having tactical innovation is generated; and the performability and decision credibility of a deduction result in a complex actual combat environment are remarkably improved.

The invention relates to the technical field of factory visualization, in particular to an intelligent factory production visualization method and system based on digital twinning, and aims to automatically construct a three-dimensional space model of factory production line layout and equipment structure by fusing laser point cloud and image data and replace manual labeling to realize digital reconstruction of a physical workshop. A deep learning model is utilized to analyze product paths and state changes, exposed nodes are identified, associated equipment is reasoned, intelligent division of the equipment is completed, and manual modeling work is reduced. The behavior rhythm characteristics of the operation parameters are extracted and compared with the historical modes of the equipment sub-models, the rhythm similarity matrix is automatically constructed, the parameter binding relation is determined, and tedious manual configuration is avoided. The scheme has good flexibility and self-adaptive ability, the model can be quickly updated after process change or equipment replacement, it is guaranteed that the digital twin system is synchronous with the actual factory state, and efficient deployment and visual updating are achieved.

The invention discloses a layered rock mass preparation method and device based on a digital image, and the method comprises the following specific steps: S1, carrying out the intelligent recognition and parameter extraction of the stratified rock mass bedding of a target layer rock sample, obtaining the bedding structure feature information of the sample, carrying out the high-precision digital three-dimensional reconstruction of the bedding structure, and obtaining the high-precision digital three-dimensional reconstruction of the stratified rock mass; a corresponding digital reconstruction model is obtained; and S2, according to the digital reconstruction model and bedding structure feature information of the sample, preparing the layered rock-like test piece based on a 3D printing technology. According to the invention, the preparation process of the layered rock mass test piece is simplified, the sample printing efficiency is improved, the sample difference is reduced, the consistency and surface flatness of the test piece are greatly improved, the discreteness among the samples is reduced, and the rock mass characteristics under the real geological condition are better simulated; meanwhile, the period of stratified rock mass research and engineering application is effectively shortened, and the reliability and practicability of a simulation experiment are improved.

The invention discloses a transformer substation three-dimensional modeling method based on laser point cloud and visible light fusion, and the method comprises the steps: carrying out the scanning and shooting of a transformer substation through employing an unmanned plane, carrying a laser radar and a visible light camera, and obtaining laser point cloud data and visible light image data correspondingly; acquiring laser point cloud data and visible light image data; the obtained data is preprocessed; constructing a three-dimensional curved surface model based on the preprocessed data; executing texture mapping of the three-dimensional curved surface model to generate a transformer substation three-dimensional model; according to the three-dimensional modeling method for the transformer substation, high-precision, high-efficiency and high-reality digital reconstruction of the transformer substation is realized by combining an advanced data acquisition technology and a processing algorithm. According to the method, the modeling efficiency and precision are improved, powerful technical support is provided for intelligent management and optimization of the transformer substation, the problems that in the prior art, the modeling workload is large, the modeling efficiency is low are solved, and the modeling efficiency is improved.

The invention provides a parallel battlefield environment construction method and system based on digital twinning. The method comprises the steps that a laser echo set reflected by the earth surface is obtained; aggregating the intensity characteristic of each laser echo in the laser echo set and the time difference between the departure time and the arrival time of each laser echo, and generating point cloud data; performing space coordinate association on a three-dimensional elevation model constructed based on the point cloud data and a preset battlefield terrain material feature map to generate a battlefield comprehensive terrain entity grid; performing spatial feature analysis on the battlefield comprehensive terrain entity grid to generate a dynamic threat distribution representation; performing spatio-temporal evolution deduction on the dynamic threat distribution representation to generate enemy unit maneuvering path prediction data; and inputting the enemy unit maneuvering path prediction data into a preset digital twinborn battlefield environment to drive the preset digital twinborn battlefield environment to dynamically update. According to the invention, accurate digital reconstruction of the complex battlefield terrain is realized, and a real-time and high-confidence battlefield situation deduction platform is provided for command decision.

A method including acquiring a hand movement video obtained by capturing images during a process in which a target user rotates a wrist joint from a first angle to a second angle while maintaining a hand posture with five fingers naturally extended, so that the hand movement video comprises a plurality of image content frames captured during a rotation process of the hand from a front / side view to a side / front view; acquiring curvature information of the fingers and / or nails from the plurality of image content frames acquired during the rotation process; and creating a 3D hand model of the target user based on the hand movement video, and optimizing key point parameters of the fingers and / or nails in the 3D hand model based on the curvature information of the fingers and / or nails. The present disclosure enables digital reconstruction of the hand and improve the reconstruction accuracy of the hand.

The invention discloses an AR digital reconstruction and visual navigation system and method for an agricultural cultural heritage scene. The system comprises a digital modeling module, a scene recognition module, an AR reconstruction module, a path navigation module and an interactive navigation module. According to the invention, by fusing laser point cloud, unmanned aerial vehicle images and historical images, agricultural cultural heritage scenes such as rice terraces, mulberry-based fish ponds, waterwheel irrigation systems and the like with extremely high local characteristics are reconstructed, historical scenes in which natural or human factors disappear are recovered, and AR live-action is superposed with virtual farming behaviors, traditional festival activities and clothing styles and features, so that the real-scene and real-scene virtual farming is realized. A user can experience thousand years of agricultural culture personally on the scene, the visiting experience that the user cannot understand and cannot see completely in the past is changed, an exclusive route is planned for the user, various interaction modes such as voice, images and gestures are supported, and the method is suitable for users of different age groups, language backgrounds and cultural cognition levels.

The invention belongs to the technical field of gearboxes, and particularly discloses a gearbox abnormal state monitoring method and system based on an infrared imaging change rate, and the method comprises the steps: collecting the surface infrared thermal imaging of a gearbox in real time, visualizing the gradient distribution condition of an infrared temperature field, and determining the position information of a calibration point on the gearbox; a temperature sensor is arranged at a calibration point on the gear box to obtain a temperature signal of the gear box; calculating a difference value matrix of the gearbox temperature signal and the infrared thermal imaging temperature, judging whether the maximum value of the difference value matrix exceeds a preset first alarm threshold value or not, and preliminarily judging whether the state of the tested piece is abnormal or not; and reconstructing the temperature field gradient image, comparing the maximum temperature value in the reconstructed temperature field gradient image with a preset second alarm threshold value, and re-determining whether the state of the measured piece is abnormal or not. By adopting the technical scheme, digital reconstruction and visual modeling of the temperature field are carried out, gradient zone imaging of the temperature field of the gearbox is monitored and analyzed, and the abnormal state of the gearbox is detected.

The invention discloses a portable X-ray back scattering security check system based on AI image enhancement. The system belongs to the technical field of radiation imaging, and comprises a sensing unit used for sensing physical situation information in an imaging process so as to generate time-sequenced physical prior data; the prediction unit is used for predicting a future scanning behavior based on the physical prior data to generate a feed-forward control instruction and prediction confidence; the regulation and control unit is used for adaptively regulating and controlling scanning parameters based on the feedforward control instruction and recording execution information including regulation and control residual errors; and the AI reconstruction processor is used for performing enhanced reconstruction on the original X-ray image based on the physical priori data, the prediction confidence and the execution information. By constructing an intelligent closed loop of perception, prediction, regulation and control and reconstruction, physical uncertainty information in the imaging process is quantized and used for guiding AI reconstruction, so that the problem of information fault between physical acquisition and digital reconstruction is solved, and the quality and reliability of a final image are improved.

The invention discloses a three-dimensional point cloud-based power distribution network digital reconstruction method, which comprises the following steps of: automatically acquiring high-precision point cloud and image data of a power distribution network site through an unmanned aerial vehicle, and eliminating noise points, simplifying data volume and improving data quality in combination with a gross error elimination algorithm and a voxel filtering technology; a pyramid type hierarchical storage and octree indexing technology is utilized to realize regional management and rapid space retrieval of massive point clouds; and performing automatic semantic classification on the point cloud based on a PointNet deep learning algorithm, performing high-precision alignment on a classification result and a real scene in combination with a homonymy matching method, and constructing a real and restored power distribution network field three-dimensional working condition model. According to the method, the problems of low field investigation efficiency and incomplete information of a traditional power distribution network are solved, digital and automatic investigation of non-power-cut operation of the power distribution network is realized, transformation from operation and maintenance of the power distribution network to scientific and technical transformation is promoted, manpower and material resources are saved, and power supply reliability and operation and maintenance management efficiency are remarkably improved.

A system is provided for digitally reconstructing a structure of a region of interest for an object. The system comprises an imaging device configured to obtain a sequence of images of the region of interest and a processor. The processor is configured to analyze system data which at least coveys information about the imaging device position relative to the region of interest. Quasi-periodic behaviors are extracted from the system data, redundancy is removed in the sequence of images and / or system data using the identified quasi-periodic behaviors, and the structure of the region of interest is reconstructed using the sequence of images and system data.

Disclosed is a computer-implemented method of determining imaging control data for controlling medical imaging for localizing the position of an anatomical body part. The disclosed method encompasses generation of synthesized medical images from a tomographic planning image at an image contrast which allows to appropriately recognize a representation of soft tissue in the synthesized medical images. An imaging parameter, for example the energy, of a (virtual) beam of imaging radiation is adjusted to achieve such a contrast. The determined value of the imaging parameter can then be used later for image-based tracking of the anatomical body part during a medical procedure such as radiation treatment. For example, too high energy values should be avoided because they would result in a comparably low contrast of image constituents representing soft tissue because only a small fraction of the energy would be absorbed by the soft tissue.

The invention discloses an existing bridge high-precision three-dimensional reconstruction method based on laser point cloud data, and relates to the technical field of bridge engineering digital reconstruction. Multi-source laser point cloud cooperatively collects detail point cloud of key components of an existing bridge, point cloud of a bridge floor and an upper structure and point cloud of a shielding part; multi-source point cloud coarse registration is completed based on a target ball of a preset specification, and then fine registration is carried out by adopting an improved iterative nearest point algorithm of integrated normal vector constraint; and adopting a layered denoising strategy to carry out statistical filtering denoising on the flat area, and carrying out adaptive radius filtering on the vulnerable area to reserve cracks and peel off disease points. Through collaborative design of multi-source acquisition, precise processing, intelligent segmentation, parameterized modeling and hierarchical verification, data integrity improvement, consideration of point cloud processing precision and disease retention, semantic segmentation and instance identification precision optimization, BIM model engineering value improvement and achievement adaptability enhancement are realized.

The invention discloses an architectural drawing management method and system based on multi-modal AI recognition, and relates to the technical field of data management, and the method comprises the steps: extracting drawing frame information of a drawing file through a preset image recognition large model when the drawing file is received; based on a preset field mapping relation, converting the drawing frame information into database fields and storing the database fields into a preset drawing database; when a query instruction is received, the query instruction is converted into a library query statement, and retrieval is performed in a preset drawing database based on the library query statement; and determining a target drawing file according to the retrieval result, and sending the target drawing file to a permission user corresponding to the query instruction. According to the method, three core technologies of dynamic drawing frame positioning, natural language processing and an intelligent workflow engine are fused, so that full-process digital reconstruction from drawing analysis and structured storage to intelligent retrieval and multi-user collaboration is realized, and the building drawing management efficiency is improved.

The invention relates to the technical field of three-dimensional digital reconstruction, in particular to an archaeological site reconstruction method based on image point cloud cross-modal alignment, and the method comprises the steps: obtaining an optical image of an archaeological site and archaeological site point cloud data; obtaining a point cloud distribution characterization value based on the archaeological site point cloud data, determining a thinning mode of the archaeological site point cloud according to the point cloud distribution characterization value, and carrying out thinning processing based on the thinning mode; performing cross-modal alignment on the thinned archaeological site point cloud data and the optical image of the archaeological site; determining the qualification of the alignment effect according to the cross-modal alignment rate, and determining an adjustment strategy when the alignment effect is unqualified according to the mismatch rate of the key region when the alignment effect is unqualified; and fusing the archaeological site optical image with the qualified alignment effect with the archaeological site point cloud data, and executing dense point cloud reconstruction, three-dimensional grid generation and texture mapping to complete three-dimensional reconstruction of the archaeological site. According to the method, the cross-modal alignment efficiency of archaeological site reconstruction is improved.

The invention relates to the technical field of agricultural breeding phenotype detection, in particular to a small-particle-size seed microscopic three-dimensional reconstruction device and a size feature extraction method.The small-particle-size seed microscopic three-dimensional reconstruction device comprises a rack and an image acquisition mechanism and further comprises a horizontal rotating mechanism, a light supplementing mechanism and a main control system; the image collecting mechanism, the horizontal rotating mechanism and the light supplementing mechanism are all arranged in the rack, the horizontal rotating mechanism is installed on the lower portion of the rack, to-be-detected seeds are placed on the horizontal rotating mechanism, the light supplementing mechanism and the image collecting mechanism are installed on the upper portion of the rack and conduct light supplementing and image collecting on the to-be-detected seeds, and the main control system is arranged on the side of the rack. And the image acquisition mechanism, the horizontal rotating mechanism and the light supplementing mechanism are electrically connected with the main control system. According to the method, high-quality three-dimensional reconstruction and size feature accurate extraction of the small-particle-size seeds can be realized, so that three-dimensional digital reconstruction and size feature detection of the small-particle-size seeds are completed, the detection precision and efficiency are improved, and the development of the breeding industry is promoted.

The present invention relates to a digital testing method and computer equipment for a nuclear power plant simulator. The method comprises the following steps: obtaining key system elements by deconstructing the simulator's software configuration and calculation results, and performing digital modeling based on the key system elements to obtain a test object; obtaining a test mode by decomposing the simulator's test capabilities according to different quality attributes and performing digital reconstruction; constructing different digital test scenarios based on the test object and the test mode, and implementing the digital scenarios into automated test tools through programming to build a digital test system; obtaining control data, and performing test operations on the simulator using the digital test system based on the control data, and outputting difference results. Efficient, accurate, comprehensive, and maintainable testing of the nuclear power plant simulator is achieved. The efficiency, accuracy, and comprehensiveness of the test are collectively improved, and the maintainability and scalability of the test system are enhanced.

The invention discloses an environment situation scanning three-dimensional reconstruction method based on infrared and SLAM, and belongs to the technical field of computer vision and three-dimensional reconstruction. According to the method, thermal radiation information and environment image information of a target environment are obtained through infrared sensing and visual imaging respectively, after feature extraction and matching processing are conducted on the two kinds of information, synchronous positioning and map construction are completed, and environment pose data and environment map data are generated; a sensing uncertainty probability model is established, after multi-source data denoising is completed through Bayesian filtering, feature level fusion is carried out to obtain fusion environment data, and finally a three-dimensional reconstruction model of a target environment is constructed based on the fusion environment data. According to the method, the full-process closed loop of environment situation scanning and three-dimensional reconstruction is realized, the application limitation of single sensing information is made up, the comprehensiveness of environment sensing data and the scene adaptability of three-dimensional reconstruction are improved, and the digital reconstruction requirements in various complex environments can be met.

The invention provides a construction method of a virtual electromagnetic environment model and a test device. The construction method comprises the following steps: acquiring multi-dimensional test data of a mobile carrier in a target scene; taking a timestamp of the baseband signal data as a reference, performing synchronous fusion on the positioning data and the attitude data, and generating a synchronous data set; performing parameter estimation on the synchronous data set, and separating a relative channel parameter of the multipath component in a mobile carrier coordinate system; converting the relative channel parameters into absolute space parameters under a geographic coordinate system based on the instantaneous space state, and determining physical scattering source characteristics corresponding to the multipath component based on the relative channel parameters and the absolute space parameters; and generating a virtual electromagnetic environment model based on the absolute space parameters and the physical scattering source characteristics. In the mode, a structured virtual electromagnetic environment model containing absolute space attributes can be generated, and then high-fidelity digital reconstruction and reproduction of a complex dynamic electromagnetic environment in a real scene are realized.

The invention belongs to the field of geological monitoring, and discloses a loess macropore seepage observation device and method based on a loess digital reconstruction body, and the loess macropore seepage observation device comprises the loess digital reconstruction body, a coloring agent container and a water discharging bottle; the bottom of the loess digital reconstruction body is provided with a permeable stone layer, and the top is provided with a hermetically connected top cover; a plurality of image acquisition devices are arranged around the loess digital reconstruction body; one sides of the drainage bottle and the permeable stone layer are communicated with the bottom of the coloring agent container; the first end of the three-way pipe is communicated with the interior of the top cover, the second end of the three-way pipe is provided with a pressure meter, and the third end of the three-way pipe is communicated with the atmosphere. The loess digital reconstruction body duplicates the macropore structure of the loess sample, the water flow migration process in the macropores of the loess digital reconstruction body can be rapidly observed, the water-driven gas process in the macropores of the loess digital reconstruction body can be rapidly determined, the working efficiency is high, the device is flexible and detachable, and the cost is low. The problem that the three-dimensional migration process of the water flow in the unsaturated loess macropores is difficult to accurately capture at present can be effectively solved.

The invention discloses a quantitative analysis method for ardealite leachate-carbonate rock fracture corrosion effect, and belongs to the technical field of material analysis. The quantitative analysis method for the ardealite leachate-carbonate rock fracture corrosion effect comprises the following steps: preparing a carbonate sample; preparing ardealite leachate; mounting the sample in a seepage device in a sealing manner, pouring the leachate into the seepage device, and carrying out a seepage test; performing CT scanning on the sample before and after corrosion to obtain a two-dimensional section slice image of the sample, and preprocessing the slice image based on a gray value and threshold segmentation method; and performing three-dimensional digital reconstruction on the preprocessed CT slice data, and calculating pore structure parameters, fracture structure parameters, fractal dimensions and permeability of the sample to complete quantitative analysis. By adopting the quantitative analysis method for the ardealite leachate-carbonate rock fracture corrosion effect, the problem of low analysis accuracy of the ardealite leachate corrosion effect by an existing analysis method can be solved.

The invention is suitable for the technical field of medical images, and provides a two-dimensional projection image generation method, computer equipment and a computer program product. The method comprises the following steps: determining a first projection area corresponding to first three-dimensional voxel data of a target patient and a second projection area corresponding to second three-dimensional voxel data of a target medical device through a CPU (Central Processing Unit), and determining a first number of thread blocks to be started according to performance parameters of a GPU (Graphics Processing Unit); starting a first number of thread blocks through the GPU, respectively executing digital reconstruction projection operation on the first three-dimensional voxel data and the second three-dimensional voxel data in respective corresponding projection areas in parallel through the first number of thread blocks, and fusing a first projection image and a second projection image which are respectively obtained into a target projection image; therefore, the application range of the method can be expanded while the overall generation efficiency of the target projection image is improved.