49 results about "Micrograph" patented technology

The invention provides an ore recognition method and system based on a large model algorithm, and is applied to the technical field of computer vision and data processing, and the method comprises the steps: obtaining an ore microscopic image, and generating a feature point location candidate set through a convolutional neural network model; calculating a spectral value and a morphological parameter of each feature point location to determine a multi-dimensional data point; clustering the multi-dimensional data points to obtain a feature point cluster; and when the spatial distance between the adjacent feature point position clusters is smaller than a first preset threshold value and the spectral similarity is higher than a second preset threshold value, combining the adjacent feature point position clusters into the same mineral region. The segmentation precision and the type identification accuracy of the complex symbiotic mineral area can be obviously improved.

The present application relates to microlens three-dimensional imaging technical field, specifically relates to micrograph array method and three-dimensional dynamic imaging material with micrograph, aims at solving the problem that microlens three-dimensional imaging cannot simultaneously meet fixed visual angle stable imaging and variable visual angle dynamic response.The technical scheme main points are: including transparent substrate layer;Micrograph structure layer;Microlens structure layer, microlens structure layer is located on the side of transparent substrate layer away from micrograph structure layer;Vacuum coating layer, vacuum coating layer is arranged on the surface of microlens structure layer, microlens structure layer is matched with micrograph structure layer, so that three-dimensional dynamic imaging material presents integrated micrograph three-dimensional dynamic stereoscopic image that changes with observation angle in the visual angle range, micrograph structure of different visual angle frame rate is fused according to the viewing angle segmentation, and a kind of three-dimensional dynamic imaging film material is made, and the image gradually changes when micrograph is seen through vacuum coating layer reflection formed when viewing visual angle or deflecting material changes.

Provided herein is a method of imaging and analyzing plant microspores. The method comprises obtaining one or more two-dimensional micrographs of a microspore population, wherein the microspore population has not been labeled; sending the one or more two-dimensional micrographs to an image processing system; segmenting, via the image processing system, the one or more two-dimensional micrographs to distinguish individual microspores, thereby creating a segmented population of microspores; obtaining one or more z-stacks of the population of segmented microspores or the population of microspores to be segmented; and using one or more of pixel intensity information and intensity signal processing within the one or more z-stacks of segmented cells to find nuclear edges and nuclear geometry and determining whether cells are uninucleate or binucleate based on the nuclear edges and nuclear geometry. Also provided herein is a method of imaging and analyzing plant microspores comprising determining microspore viability.

The utility model provides a kind of microscope external optical fiber lighting device, it is related to microscope lighting device technical field, including optical fiber, the one end of optical fiber corresponds with light source, and the other end extends to the light path entrance of the illumination system of microscope;The illumination system is sequentially provided with variable aperture diaphragm, condensing system and objective along light path propagation direction, and the aperture size of variable aperture diaphragm is changed to control the numerical aperture of illumination;The light emitted by light source is transmitted to variable aperture diaphragm by optical fiber, and clear micrograph is formed by condensing system, objective.The utility model has the beneficial effect of not only breaking through the constraint of traditional built-in illumination in physical space, and heat source can be completely isolated in microscope outside, avoid the occupation of light source to microscope internal space and the influence of cooling fan vibration on microscopic observation, in addition, different microscopes can be matched by different light sources, with flexible adaptive expansibility, can be widely applied to various microscopic observation systems.

The present invention belongs to the field of composite material detection technology, and specifically discloses a method and system for quantitatively evaluating the orientation of composite material filling particles, which includes S1, obtaining a cross-sectional micrograph of the composite material to be tested and converting it into a grayscale image; S2, processing the grayscale image to obtain a binary image; S3, traversing all pixel blocks in the binary image, obtaining the image longitudinal continuity SSV and the image transverse continuity SSH, and then obtaining the vertical orientation ORI; S4, rotating the binary image multiple times according to a preset angle, repeating steps S2 and S3 each time the image is rotated, and obtaining a curve of the vertical orientation varying with the rotation angle, wherein the maximum value of the vertical orientation in the curve and its corresponding rotation angle are the image orientation and orientation angle. The present invention does not require other complex equipment and instruments, and is not affected by the crystal structure of the filler particles; at the same time, the present invention performs full-image processing on the binary information, and also has excellent evaluation effects on complex filler particle information.

The invention relates to a method for generating a three-dimensional structure of a porous material through image data processing and application of the method, and belongs to intersection of three-dimensional reconstruction, deep learning and material characterization. The problems that when the prior art is directly used for constructing the three-dimensional structure of the material, computing resource consumption is large, functions are split, simulation adaptability is poor, data dependence is high, and angle features are limited are solved. The invention relates to a method for generating a porous material three-dimensional structure through image data processing, which comprises the following steps of: 1, screening two-dimensional images to obtain images conforming to a signal-to-noise ratio and a contrast ratio as input microscopic image samples, and preprocessing the input microscopic images to generate a training sample set; 2, constructing a model of the algorithm; 3, training models, including discriminator model training and generator model training; and 4, inputting a sample image into the trained model, and obtaining a corresponding three-dimensional voxel structure based on the output three-dimensional probability voxel structure.

The application discloses a method for judging the sensitivity of a titanium alloy to sustained fatigue, which comprises the following steps: step (1): a mathematical model formula y=kx+b is simulated by using titanium alloy forging samples with different feather organization contents and a sustained fatigue sensitivity coefficient; in the formula, x represents the feather organization content, and y represents the sustained fatigue sensitivity coefficient; and step (2): the feather organization content of a titanium alloy forging sample to be detected is substituted into the mathematical model formula to obtain the result; the feather organization is a flaky cluster alpha phase in a net basket organization observed by a metallographic microscope, and the content of the feather organization is a ratio of an area of the feather organization to a total area of the net basket organization in a metallographic micrograph taken. The judging method can quantitatively judge the sustained fatigue sensitivity coefficient of the titanium alloy forging sample according to the microstructure characteristics of the titanium alloy forging sample in engineering, and the sustained fatigue value of the material does not need to be measured when mass-produced forgings or a process window is adjusted.

A method of automated control of a microscope in cryogenic electron microscopy (cryo-EM), wherein the microscope is configured to collect high-magnification micrographs of particles suspended in vitreous ice. Such particles are found in grid squares, and a square contains holes from which high-magnification micrographs are imaged. The method is carried out during an active data collection session, leveraging a pipeline that comprises a set of models. The pipeline evaluates a set of collection locations to determine whether to continue collection at a current grid / square or instead at a new grid / square. The evaluation is based on a set of one or more quality scores derived from one or more pretrained models and machine learning-based active learning. Based on the determination, control information is provided to automatically control the microscope to move to a next target for data collection.

The present invention belongs to the field of cartridge case manufacturing, specifically to a method for predicting cartridge case material properties from SEM images based on explicit Fast Fourier Transform (FFT). The method comprises the following steps: S1, sampling the cartridge case material before stamping to produce an SEM sample; S2, taking a micrograph of the sample and performing electron backscatter diffraction analysis to obtain an SEM image and EBSD data; S3, creating a three-dimensional voxel map of a representative volume unit from the SEM image, and using the EBSD data to generate Euler angle information for each three-dimensional pixel; S4, calculating stress and strain data for the three-dimensional voxel map of the representative volume unit using an explicit Fast Fourier Transform (FFT) method; and S5, using a tensor sparse symbolic regression method to learn the stress and strain data generated in step S4 to generate a constitutive model for the cartridge case material. The present invention can accurately predict the large deformation plastic mechanical properties of the cartridge case material from the results of the microstructural analysis of the cartridge case material.

The invention discloses a patterned wafer microstructure data acquisition system in a bright field environment and a data set construction method. The system comprises an adjustable light source assembly, a microscope imaging assembly and a three-axis electric displacement platform, the data set construction comprises the following steps: fixing a patterned wafer sample on a three-axis electric displacement platform, and setting illumination brightness through an adjustable light source assembly to obtain uniform bright field illumination; the three-axis electric displacement platform is controlled to move along a preset path, automatic focusing scanning in the Z-axis direction is executed at each point position, and an optimal focal plane image is generated; the collected images are classified after being cut and preprocessed, and normal samples and abnormal samples are distinguished; for the abnormal sample, generating a defect mask file through interactive labeling; and integrating images and masks to form a structured data set which can be directly used for algorithm research. According to the method, microscopic images with high depth of field and high contrast can be obtained, and a high-quality anomaly detection data set directly used for deep learning model training is constructed.

The embodiment of the application relates to a kind of processing method and device for optimizing electron micrograph quality, the method comprises: one complete model pre-training DDPM model is recorded as first pre-training model;And by the way of fusing multiple LoRA adapters on first pre-training model, obtain first fusion model;Collect multiple spherical aberration electron micrograph to form corresponding first training atlas;The parameter set L of first fusion model is fine-tuned based on first training atlas;After fine-tuning, corresponding electron micrograph quality optimization processing is carried out to any ordinary transmission electron micrograph based on the second denoising autoencoder of first fusion model, and corresponding optimized electron micrograph is obtained.Through the application, the image quality of ordinary transmission electron micrograph can be optimized to reach the quality level of spherical aberration electron micrograph, and the model fine-tuning efficiency can be improved, and the model use cost is reduced.

The application discloses a kind of over-fire cable protection multi-material path micrograph generation methods based on Mamba, it is related to bridge cable protection material microstructure intelligent characterization field after fire.The method obtains the micrograph of different materials in cable protection system under different over-fire temperatures and different electron microscope magnifications, and constructs material micrograph sample library;Classify, differentially pretreat and conditionally label material micrograph, form conditional sample set;Construct micro-morphology generation model based on Mamba, and use latent space condition generation training strategy and multi-path parameter optimization method to train;The micro-morphology generation model that completes verification is encapsulated as lightweight plug-in, and target micrograph generation and output are realized.The application can realize the controllable generation of multi-material micrograph under small sample condition, and improve the characterization ability of material specific micro-morphology feature.

The invention provides a semiconductor defect detection method and device based on a microscopic atlas, and the method comprises the steps: selecting a spectrum band of infrared multi-spectrum-band microscopic imaging according to the type of a detected semiconductor material, carrying out the scanning imaging of a sample in a first preset view field and a first preset resolution ratio through an electric objective table, and splicing the scanning images into a complete microscopic image; analyzing the complete microscopic image, and positioning a suspected pollutant defect micro-area; performing infrared multi-spectrum microscopic imaging on the suspected pollutant defect micro-area in a second predetermined view field and a second predetermined resolution, identifying the defect type through an image processing algorithm, and if the suspected pollutant defect micro-area is determined to be a pollutant defect micro-area, switching to an infrared wide-spectrum microscopic spectrum measurement light path; fourier transform is carried out on the infrared spectrum light source, spectrum detection is carried out on the pollutant defect micro-area according to the Fourier transform result, collected spectrum data is compared with data in a preset database, and pollutant components and distribution are determined. The method provides guarantee for semiconductor quality control.

The present application relates to the technical field of agricultural product traceability, and discloses a kind of agricultural product pre-sale traceability method and system based on plant micrograph features, and macro image is collected during pre-sale period, and biological intrinsic coordinate system is established by extracting natural intersection node, and high-precision microscope camera is driven to collect reference micrograph, and reference micrograph is converted into gray scale scalar field function, and reference topological persistent graph is generated by extracting topological generator, and after coding, it is bound with pre-sale order in traceability cloud database by combining digital signature, and during review period, natural intersection node is extracted again to reconstruct coordinate system and calculate growth expansion coefficient to compensate deformation, and real-time topological persistent graph is calculated to construct cost objective function by extracting topological persistent graph, and according to space negative gradient closed loop servo micro-motion scanning and locking best deformation alignment point, verification pass signal is output under the condition that minimum value topological distance is lower than legal homology threshold and tamper-proofing check passes, and the present application realizes traceless right confirmation of living agricultural product features.

The invention belongs to the technical field of computer image processing, and relates to a metal material phase content analysis method and system, a storage medium and equipment, and the method comprises the following steps: collecting a color metallographic micrograph of a sample of a metal material, and constructing a data set of a metal material microstructure by using the color metallographic micrograph; using a label labeling tool to label the data set of the microscopic structure of the metal material; the basic operation module is used for constructing a metal material phase content analysis model based on an environment and an interface provided by a deep learning framework; constructing a model based on an improved U-shaped convolution / self-attention codec algorithm; training a model based on an improved U-shaped convolution / self-attention codec algorithm by using a data set marked by a metal material microstructure; and calculating the phase content of the metal material. The method solves the problem that a traditional image processing method is not easy to analyze the phase content formed by a complex structure.

The invention discloses a marking method and system for a compact sandstone microscopic image, and the method comprises the steps: carrying out the edge enhancement of particles in the compact sandstone microscopic image through an image processing technology, and obtaining a first image; the sandstone microscopic image segmentation model is used to carry out particle pre-labeling processing on the first image to obtain a second image, and the second image comprises at least one piece of editable labeling information formed according to the particle identification result; and modifying the particle recognition result of the second image according to the editable annotation information. According to the method, the labeling efficiency can be remarkably improved, and the labeling fineness and accuracy are improved.

The present application relates to the field of material science and engineering, in particular to a kind of porous material pore structure fast identification and three-dimensional model generation method.The present application is first to the two-dimensional micrograph of porous medium phase decoupling and feature extraction, obtains topological parameter set;Then through parameter space mapping establishes statistical feature matching model;Then generate geometric topological element in the feature space that satisfies the condition;Finally, the final continuous phase space domain is constructed using spatial domain topological mapping algorithm, for the digital reconstruction and performance simulation of porous material structure.The present application can effectively realize the identification and three-dimensional model construction of porous material pore structure, and the advantages of the model construction of the present application are verified through numerical simulation, which provides strong technical support for the research and application of porous materials.

The application discloses a kind of based on biological visual characteristic micrograph optimization method, it is related to microorganism microscopic imaging technical field, comprising: S1, according to the characteristics of input image, the parameter of retinal model is adaptively adjusted;S2, using retinal model to carry out multilevel central vision channel and peripheral vision channel processing to input image;S3, the output data of central vision channel and peripheral vision channel are fused, and optimization image is generated.The application simulates the working principle of human visual system, especially the space-time filtering characteristics of retina, realizes the efficient optimization processing of microorganism microscopic image, solves the problems of low contrast, noise suppression not thoroughly, overlapping area separation difficulty of existing microorganism microscopic image, realizes the high-fidelity imaging of dense microorganism colony.

The invention relates to the technical field of image segmentation, in particular to a steel microscopic image ferrite and pearlite grain boundary segmentation method. The invention provides a segmentation method based on deep learning, and aims to solve the problems of insufficient grain boundary segmentation precision, serious noise interference and fuzzy boundary in a steel microscopic image in an existing image segmentation method. The method comprises three main steps of image acquisition and preprocessing, U-Net model construction and training, and post-processing of a segmentation result. Specifically, the image quality is improved and the training data set is expanded through the image preprocessing step and in combination with the technologies of targeted denoising, contrast enhancement, data enhancement and the like. In the model training stage, composite loss of cross entropy and a Dice loss function is adopted, and the problem of imbalance between the grain boundary and the background is solved. And in the post-processing stage, fine grain boundary deletion and noise residue are repaired by using morphological operation and an eight-neighborhood completion algorithm, and the integrity and connectivity of the grain boundary are ensured.

The invention discloses a pathological section full-automatic quality detection method and system based on an OCT technology, and relates to the technical field of pathological detection. The system comprises an OCT imaging module, a sample bearing and positioning module, a data processing module and a quality evaluation module, according to the method, automatic flattening and positioning of a pathological section are achieved through a sample bearing and positioning module, three-dimensional scanning is conducted on the pathological section through an OCT imaging module to obtain an original interference signal, signal noise reduction, wave number resampling, Fourier transform and other processing are completed through a data processing module, and a three-dimensional microscopic image is constructed. And finally, the quality evaluation module automatically identifies slice defects, quantifies defect levels and outputs a quality detection result. According to the invention, full-process automation of pathological section quality detection is realized, manual intervention is not needed, the detection precision can reach a microscopic scale, the problems of low efficiency, high omission ratio and dependence on experience judgment in traditional manual sampling inspection are effectively solved, and an objective and efficient technical scheme is provided for pathological section quality control.

This invention discloses a method and system for dislocation defect identification based on image recognition and deep learning networks, belonging to the fields of image recognition and machine learning. The method first calibrates the dimensions of a high-resolution electron micrograph, identifies atomic coordinates using a deep learning object detection model, and calculates the local atomic environment vector for each atom. Second, using an atomic model dataset covering various materials, dislocation types, and deformation states, a deep learning network is trained capable of predicting the distance between atoms and dislocations based on the local environment vector. Finally, the image is meshed, and the neighboring atomic information of each mesh node is extracted. Combined with the dislocation distance information output by the trained deep learning network, unsupervised clustering is used to identify the dislocation location. This method achieves automated and high-precision identification of dislocation defects in face-centered cubic materials, while also enabling dynamic characterization of crystal defect deformation behavior.

A method for identifying and characterizing surface defects on an object by means of electronic processing is described. The method comprises the following steps: acquiring, by means of a microscope, at least one microphotograph or digital image of an object or a part of the object to be confirmed for the defect; then providing the acquired at least one microphotograph or digital image to an algorithm trained by means of artificial intelligence and / or machine learning technology; a defect is then identified by a trained algorithm and the identified defect is ascertained by means of a first mask, and a first processed image is provided as an output of the trained algorithm, in which the identified defect is ascertained by the first mask. The method also provides for processing the first processed digital image by means of filtering and morphological closing operations for highlighting the edges of the defects in a cleaner and clearer manner and by means of segmentation operations adapted to highlight the individual defects visually confirmed in the image, a second processed image and a second processed mask adapted to confirm the defect and highlight and characterize the contour and shape of the defect in an improved manner are obtained. The method further comprises the step of applying a computer vision technique / algorithm to the aforementioned second processed image to determine characterization information for each of the detected defects. For each of the identified defects, this characterization information comprises at least the following information: a defect category, a position of the defect relative to a reference coordinate system associated with the image, at least one geometric / dimensional parameter of each defect. The at least one trained algorithm is trained in a preliminary training step performed on training digital images, each of which is marked by means of marking a known defect and filtered to produce a mask corresponding as much as possible to the marked defect. A system for detecting and characterizing defects on a micrograph of an object is also described, which system is capable of performing the above method.

A method comprising: estimating (106) a first quantity indicative of a porosity rate of a sample from a tomogram representing the sample; polishing (108) a surface of the sample using initial polishing parameters; estimating (114) a second quantity indicative of the porosity rate of the sample from a micrograph showing the surface of the sample after polishing; from the initial polishing parameters, determining (116) adjusted polishing parameters making it possible to reduce a difference between the porosity rate indicated by the first quantity and the porosity rate indicated by the second quantity; polishing (208) a surface of another sample in the same material using the adjusted polishing parameters; estimating (214) a quantity indicative of the porosity rate of the other sample from a micrograph showing the surface of the other sample after its polishing.Figure for abstract: Fig. 2.

The invention relates to the technical field of micro-lens three-dimensional imaging, in particular to a micro image-text array method and a three-dimensional dynamic imaging material with micro images and texts, and aims to solve the problem that micro-lens three-dimensional imaging cannot meet the requirements of fixed-view-angle stable imaging and variable-view-angle dynamic response at the same time. According to the technical scheme, the film is characterized by comprising a transparent base material layer; a micro image-text structure layer; the micro lens structure layer is located on the side, away from the micro image-text structure layer, of the transparent base material layer; the vacuum coating layer is arranged on the surface of the micro lens structure, and the micro lens structure layer is matched with the micro image-text structure layer, so that the three-dimensional dynamic imaging material presents a three-dimensional dynamic stereoscopic image which is integrated with micro images and texts and changes along with the observation angle within the viewing angle range, and the micro image-text structures with different viewing angle frame rates are segmented and fused according to the viewing angle. According to the three-dimensional dynamic imaging thin film material, the gradual change of the image formed by reflecting the micro image-text through the vacuum coating layer can be seen by converting the viewing angle or deflecting the material.

The invention provides a shale permeability analysis method considering multi-media distribution based on an electron microscope image, and relates to the technical field of oil and gas development, and the method comprises the following steps: based on a shale organic matter electron microscope image and a shale inorganic matter electron microscope image, carrying out statistics on pore sizes and distribution rules of organic matters and inorganic matters to obtain a statistical result; considering the micro-scale effect, and respectively calculating the gas permeability in the organic matter and the gas permeability in the inorganic matter according to the statistical result; performing micro-crack identification on the shale multi-medium electron microscope image, and calculating the gas permeability in the micro-crack through the opening degree of the micro-crack; and substituting the gas permeability in the organic matter, the gas permeability in the inorganic matter and the gas permeability in the microfractures into the shale core scale model to solve a shale gas flow mathematical model, and calculating to obtain the shale permeability. According to the method, the microscale effect in shale multiple media is considered, and the problem that the gas microcosmic flow mechanism cannot be comprehensively considered in shale pressure pulse experiment permeability calculation is solved.

The invention discloses an activity evaluation method and system for the whole stem cell culture process, and belongs to the technical field of stem cell culture.The technical scheme includes the steps that stem cell microscopic images are periodically collected through a microscopic imaging system, and the stem cell microscopic images covering the convergence degree interval of 0-100% are obtained and preprocessed; carrying out weak supervision segmentation on the preprocessed microscopic image to obtain a morphological segmentation image; performing recursive multilayer decomposition on the preprocessed microscopic image through two-dimensional discrete wavelet transform to obtain a texture feature map; inputting the segmentation image and the texture feature image into a convolutional neural network for convolution calculation, and then performing flattening to obtain a morphological feature vector and a texture feature vector; splicing the morphological feature vector and the texture feature vector to obtain a fused feature vector; and inputting the fusion feature vector into an evaluation network to output an activity score. The method and the system have the beneficial effect that the activity evaluation method and the system for the whole stem cell culture process are provided.

The application discloses a kind of by image processing software calculation metal material EBSD grain size method and device, it is related to the measurement technical field of physics. Including: by electron backscattering diffraction device, the metal material to be measured is carried out electron backscattering diffraction EBSD analysis, and analysis data is obtained;By the supporting software of electron backscattering diffraction device, analysis data is handled, and orientation imaging micrograph is exported;According to orientation imaging micrograph, the numerical value A of measurement target Area is obtained m And the numerical value P of measurement target Perimeter m ;According to numerical value P m And numerical value A m , the average equivalent area effective grain size of metal material to be measured is calculated. When the effective grain size of metal material is obtained using the application, a relatively large EBSD step can be set, the time of detection is reduced under the premise of ensuring the accuracy of the results, and the average grain size obtained is more accurate than other methods.

The invention discloses a two-dimensional bar code decoding method and device, equipment and a storage medium, and relates to the technical field of digital image processing, and the two-dimensional bar code decoding method comprises the steps: obtaining an original code image corresponding to a to-be-decoded two-dimensional bar code, and carrying out the image positioning and bar code correction of the original code image through a target detection model, obtaining a correction image corresponding to the to-be-decoded two-dimensional bar code; performing boundary prediction on the corrected image through a codeword segmentation network to obtain a codeword boundary probability graph, and performing codeword construction based on the corrected image and the codeword boundary probability graph to obtain a plurality of codeword micrographs; performing codeword normalization based on each codeword micrograph and a codeword micrograph normalization model to obtain a plurality of binary micrographs; and assembling and decoding the binarized micrographs to obtain a decoding result corresponding to the two-dimensional bar code to be decoded. The method can stably work in difficult scenes such as low definition, strong distortion and complex background, and the decoding success rate is improved.

The present disclosure relates to an automatic measurement method for electron micrograph images obtained in a semiconductor process. The method comprises: obtaining an electron micrograph image to be analyzed as a first image; obtaining a region selected by a user in the first image as a first region, and obtaining information of a manual measurement operation performed by the user in the first region; obtaining a value of a specified image feature of the first region, and selecting one from a plurality of preset region matching algorithms as a first algorithm based on the specified image feature value; obtaining a value of a specified edge feature related to the manual measurement operation information, and selecting one from a plurality of preset edge recognition algorithms as a second algorithm based on the specified edge feature value; and recognizing at least one second region from the first image by using the first algorithm, recognizing an edge to be measured in the second region by using the second algorithm, and performing automatic measurement by using the edge.