8894 results about "Image manipulation" patented technology

The invention provides a microcrack precise segmentation method integrating feature fusion and convolution attention, and belongs to the field of image processing. According to the method, a crack segmentation network based on an encoder-decoder architecture is constructed, a convolution block attention module is introduced at an encoder end, background noise is adaptively suppressed and obvious characteristics of cracks are enhanced through a channel and space dual attention mechanism, and the method is suitable for the adaptive segmentation of the cracks on the premise of almost not increasing the calculation overhead. The sensitivity of the model to microcracks is improved; a feature fusion module is introduced at a decoder end, and cooperation of low-layer details and high-layer semantics is realized through cross-layer fusion, so that a semantic gap is effectively bridged, detail loss caused by traditional convolution stacking is avoided, and continuity and a complete topological structure of a long and narrow crack are ensured. According to the method, through collaborative optimization of multi-scale feature extraction and an attention mechanism, accurate capture of the saliency features of the crack and effective suppression of complex background interference are realized, and the detection sensitivity and overall segmentation consistency of the micro-crack are remarkably improved.

The invention belongs to the technical field of computer vision and artificial intelligence, belongs to the cross technical field of target detection, deep learning and image processing, and particularly relates to a YOLOv8 algorithm improvement method based on an unmanned aerial vehicle aerial image small target detection model, which comprises the following steps of: introducing a user-defined feature enhancement module into a YOLOv8 backbone network, a neck part and a detection head part; the self-defined feature enhancement module comprises a context guide self-adaptive fusion module introduced into a backbone network so as to replace part of traditional convolution operation; a space edge sensing feature up-sampling module and a space sensing enhanced convolution module are adopted in the neck fusion network; a fine-grained dynamic pruning detection head is introduced into a detection head detection network. According to the method, the performance of the model in a small target detection scene is effectively enhanced, and the accuracy, robustness and real-time response capability of a detection system are remarkably improved.

The invention discloses an alimentary canal pathological diagnosis visual language large model construction method based on reinforcement learning and application thereof, and belongs to the technical field of medical image processing. The method comprises the following steps: firstly, extracting pathological information through layout analysis and adaptive threshold processing, and recombining the pathological information into a structured data set containing an inference chain; secondly, a visual encoder and a multi-branch classifier are used for extracting features and confidence coefficients, and dynamic structured cue words are generated; and finally, inputting the image and the cue word into a multi-modal large model, carrying out supervised fine-tuning hot start, and carrying out reinforcement learning training by adopting a group relative strategy optimization algorithm in cooperation with a composite reward function containing format, semantics and diagnosis dimensions. The problems that a general model is prone to generating illusion in a pathological scene and lacks reasoning logic are solved, and the accuracy and logicality of pathological report generation are remarkably improved.

The invention relates to the technical field of medical image processing, in particular to an engineered heart tissue pulsation trajectory automatic analysis method based on an image recognition algorithm, which comprises the following steps: S1, multi-modal image fusion: performing space-time registration and feature fusion on acquired multi-modal heart images to generate a fused image sequence; s2, cardiac muscle tissue segmentation: outputting a cardiac muscle tissue segmentation result with a timestamp; s3, motion track modeling: generating three-dimensional track point cloud data in a pulsation period; s4, feature parameter extraction: performing spatial-temporal feature analysis on the track point cloud data, and extracting multi-dimensional motion parameters; and S5, heterogeneity atlas generation: generating a cardiac pulse heterogeneity atlas according to the multi-dimensional motion parameters. According to the method, automatic analysis of the cardiac pulse track and generation of the heterogeneity atlas based on the multi-modal image and space-time modeling are realized, and the precision and the intelligent level of cardiac motion anomaly recognition are remarkably improved.

The invention discloses a body-in-white welding seam recognition and automatic welding method based on a machine vision technology, particularly relates to the technical field of computer vision and image processing, and is used for solving the problem of welding seam track recognition accuracy caused by insufficient processing capability of an existing three-dimensional vision recognition method on incomplete and uncertain point cloud data. Through the steps of multi-view point cloud acquisition and registration, probabilistic confidence evaluation, region growth of track continuity constraint, multi-track fusion optimization and the like, accurate identification of a body-in-white welding seam track under a complex working condition is realized; firstly, multi-view point cloud data are obtained, probabilistic registration is carried out to generate a confidence evaluation result, then candidate tracks are generated based on confidence weighting and semantic constraint, finally, an optimal track is generated through intelligent optimization and converted into a welding instruction which can be executed by a robot, and the accuracy and robustness of weld joint recognition are effectively improved.

The invention relates to the technical field of medical image processing, in particular to an artificial intelligence assisted intraoperative imaging method, which comprises the following steps: S1, preprocessing a multi-modal medical image, segmenting and recognizing an anatomical structure by a deep learning model according to the preprocessed image, measuring anatomical parameters based on a segmentation and recognition result, and generating an operation planning path by artificial intelligence according to the anatomical parameters; s2, collecting a C-shaped arm perspective image stream in real time, dynamically tracking space coordinates of a surgical instrument, comparing the position of the instrument with a surgical planned path, calculating offset, and when the offset is greater than an offset threshold, outputting correction guidance through an AR superposition layer; and S3, monitoring an image quality index in real time, dynamically adjusting exposure parameters through a reinforcement learning model, and when a metal implant is detected, switching a dual-energy-spectrum mode and executing an artifact suppression algorithm. According to the method, preoperative precise planning and intraoperative assistance are realized through artificial intelligence, the problems of poor image quality and high radiation risk are solved through technical optimization, and the method has important clinical application value.

The invention discloses an industrial part defect sample accurate generation method based on a conditional diffusion model, and belongs to the field of image processing and artificial intelligence. The method forms a closed-loop cooperative system by constructing four deep coupling modules of physical constraint noise scheduling, multi-scale feature coupling, double-domain feedback optimization and adaptive weight adjustment; a defect physical forming mechanism is converted into a dynamic noise scheduling strategy, deep interaction between condition information and a feature map is established at multiple levels of a diffusion network, quality closed-loop optimization is achieved through dual evaluation of a pixel domain and a frequency domain, and training weight is dynamically adjusted according to defect scarcity. And multi-scale accurate control is realized, a quality guarantee closed loop is established, the problem of data imbalance is effectively solved, and the performance of an industrial defect detection model is remarkably improved.

The invention discloses a short video network public opinion information identification method based on an image processing technology, and relates to the technical field of artificial intelligence and image processing, and the method comprises the following steps: S001, through obtaining image frames, audio tracks and time sequence information of a short video, constructing a multi-modal fusion model, extracting continuous image frames with suspicious identity features, and carrying out the recognition of the short video network public opinion information; generating a forgery risk area distribution map; and S002, performing semantic consistency verification according to the counterfeit risk region distribution map, and extracting space and time anomaly features existing among facial micro-expressions, pronunciation actions and background semantics in the image frame. According to the method, a multi-modal model is constructed by fusing image, audio and time information, fine abnormal features, traceability forgery starting points and propagation paths in a deep forgery video are identified, and an identification strategy and a public opinion response mechanism are dynamically adjusted, so that accurate identification, adaptive processing and closed-loop control of short video public opinion risks are realized; and the identification accuracy and the treatment efficiency are improved.

The invention relates to the technical field of image compensation, in particular to a reflection compensation method based on image processing, and provides the following scheme: acquiring an environment panoramic image by using a panoramic camera to establish a scene global coordinate system, and acquiring a multi-view original image in combination with a camera array arranged in the circumferential direction of an object; determining a rotation symmetry axis according to the multi-view contour features, reconstructing a three-dimensional geometric body, and calculating normal distribution and curvature change; generating a prediction image based on diffuse reflection and specular reflection hypothesis in the surface expansion domain through virtual visual angle disturbance, identifying a reflection region according to color and gradient consistency, and generating a reflection mask; and performing texture reconstruction on the reflective area by using geometric registration and multi-view compensation, and finally performing splicing and fusion to obtain a non-reflective high-fidelity panoramic image. The method does not need to change the field illumination condition, and can achieve the precise recognition and compensation of the complex curved surface reflection in the cultural relic in-situ collection environment.

The invention discloses a cross-modal eye fundus image generation method and system based on a generative adversarial network, relates to the technical field of medical image processing, and constructs an eye fundus focus perception and edge consistency generative adversarial network by taking a cyclic consistency generative adversarial network as a baseline. The core of the method is that a lesion perception mixed attention module is embedded in a bottleneck layer of a generator so as to strengthen the extraction capability of fine features of a lesion area; an edge information extraction module is designed, and key edge features are accurately extracted in combination with Roberts edge detection, wavelet transform and non-local mean denoising; and a joint loss function containing edge consistency loss is constructed, and the semantic consistency of a focus structure during cross-modal generation is ensured by minimizing the feature difference between the source image and the generated image. According to the method, the problems of disordered content, inconsistent structure and unstable training of the generated image in the prior art are effectively solved, and the simulation degree and clinical availability of the generated image are remarkably improved.

The invention relates to the technical field of medical image processing, in particular to a CT image intelligent analysis system for pneumonia auxiliary screening. The method comprises the following steps: firstly, preprocessing a chest CT image and detecting a candidate focus area; secondly, extracting a topological feature, a deep convolution feature and a texture statistical feature based on a persistent coherence theory from each candidate focus, and performing feature fusion through a multi-head self-attention mechanism to generate a unified focus representation vector; mapping the lesion characterization vectors to a pre-constructed radiology knowledge graph, adopting a graph neural network for reasoning, and outputting the pneumonia suspected probability and lesion classification of each lesion; and finally, performing fusion and uncertainty quantification on the analysis results of the plurality of focuses by adopting an evidence theory, and generating a comprehensive screening report. According to the method, complex-form lesions are effectively identified through topological features, accurate identification of lesion types is realized through knowledge graph reasoning, and diagnosis uncertainty quantification is provided through an evidence theory.

The invention relates to the technical field of artificial intelligence and remote sensing image processing, and particularly provides a hyperspectral image and laser radar data classification method based on a dynamic fusion network. The method comprises the following steps: preprocessing acquired multi-modal data, and constructing multi-scale input; a dual-scale local attention module is designed, and context information of different scales is fused in a self-adaptive weighted mode through gating soft pooling; a dynamic down-sampling feature enhancement module is designed, the down-sampling rate is dynamically adjusted according to the complexity of the feature map, and deep multi-scale interaction is carried out based on a Mama backbone; constructing a directional interactive attention module, extracting features in horizontal, vertical and diagonal directions through directional gating convolution, and capturing an anisotropic structure of a linear ground feature; through the design of a double-path classifier, fusing shallow space details and deep semantic information; and the model is trained, optimized and reasoned to obtain data classification, and the method improves the classification precision and the calculation efficiency.

The invention discloses a vehicle queuing length dynamic estimation method and system based on unmanned aerial vehicle cruise, belongs to the technical field of traffic state monitoring, and is used for solving the technical problems of how to realize accurate identification and tracking of queuing vehicles and panoramic scene construction and solving the problem of vehicle position deviation caused by time difference under dynamic traffic flow. And the accuracy and stability of vehicle queuing length estimation are improved. The method comprises the following steps: acquiring cruise video data of a congested road section through an inspection unmanned aerial vehicle; performing image processing on the cruise video data, and extracting vehicle characteristic parameters of the congested road section; constructing a lane-level queuing vehicle discrimination model based on two indexes of vehicle speed and linear vehicle distance; acquiring an ordered queuing vehicle set of the congested road section through the model; determining the initial queuing length of each lane according to the ordered queuing vehicle set; and dynamically correcting the initial queuing length based on a space-time compensation correction algorithm to obtain the real-time queuing length of each lane in the congested road section.

The invention discloses an edge-cloud collaborative civil aviation equipment intelligent detection platform, which relates to the technical field of intelligent detection and comprises an edge computing layer, a cloud computing layer, a collaborative scheduling layer, a data collaborative channel and an image processing layer. According to the method, a dynamic edge-cloud task scheduling mechanism is adopted, a traditional static task allocation mode is broken through, millisecond-level elastic task migration is achieved through real-time network state monitoring and computing power perception, incremental federated learning collaborative modeling and a gradient aggregation mechanism based on differential privacy are adopted, collaborative optimization of a multi-edge-node model is supported, and the task migration efficiency is improved. Multi-modal sensor fusion and lightweight anomaly detection are fused, the composite fault detection precision is improved, adaptive sampling and an edge knowledge graph are adopted, the sampling rate is dynamically adjusted based on the equipment risk level, offline rapid diagnosis can be achieved in combination with a graph database, and the problems that data redundancy is caused by fixed sampling, and the detection accuracy is high are solved. And the traditional system loses the diagnosis capability when the network is interrupted.

The invention provides an automatic plant identification method and system based on a neural network model, and relates to the technical field of intelligent agriculture, and the method comprises the steps: inputting a composite feature map into a compressed and optimized lightweight convolutional neural network model, and outputting a plant type identification result and a confidence value in real time; dynamically binding a plant type identification result with a geographic information system (GIS) coordinate, generating an endangered plant space distribution thermodynamic diagram and triggering a real-time alarm; images are automatically collected and marked based on newly-added plant sample positions marked in the spatial distribution thermodynamic diagram, and weight parameters of the lightweight convolutional neural network model are periodically updated. The plant identification and protection capability is comprehensively improved through the links of data acquisition diversification and standardization, image processing high efficiency and scientificity, feature extraction and identification accuracy, endangered plant protection and dynamic monitoring, model continuous optimization and adaptability and the like.

The invention discloses a thermal printing image processing method, device and equipment and a medium, and relates to the field of thermal printing. The method comprises the following steps: acquiring an original image to be printed, and converting the original image into a grayscale image; calculating a gradient difference value of the grayscale image, and generating a threshold distribution map; generating a heat accumulation risk map based on an average gray value in a preset neighborhood window taking each pixel point of the gray image as a center; generating a preliminary binary image according to the threshold distribution map; extracting image features of the preliminary binarization image, and classifying the original image to obtain an image classification result; calling a target processing parameter set corresponding to the image classification result from a plurality of preset processing parameter sets according to the image classification result; performing optimization processing on the preliminary binary image according to the target processing parameter set and the heat accumulation risk map to obtain an optimized binary image; and the final printing image adaptive to the resolution of the target printer is generated, so that the printing definition is improved.

The invention discloses a visible light, infrared and IQ signal fusion individual identification method based on cross-modal cross attention, and belongs to the technical field of artificial intelligence and multi-modal image processing. Aiming at the problems of insufficient multi-modal heterogeneous feature fusion capability, unbalanced modal semantic expression and unstable classification precision in the prior art, a visible light image, an infrared image and an original IQ signal are acquired, and after preprocessing, a convolutional neural network is combined with a space attention module to extract image features; using a convolutional hybrid network to extract signal spectrum features; three groups of modal pairs are constructed by adopting a cross-modal bidirectional cross attention mechanism to carry out bidirectional semantic interaction and feature fusion; and finally, inputting the fusion features into a classifier to obtain an identification result. According to the method, deep semantic fusion can be realized, modal quality changes can be dynamically adapted, and the precision and robustness of target recognition in a complex environment are improved.

The invention belongs to the technical field of medical image processing, and provides a heart image segmentation method fusing multi-receptive-field convolution and distracting to solve the technical problem that construction of Transform variants with global modeling capability and controllable calculation complexity is still an urgent breakthrough in the current medical image segmentation field. Comprising the following steps: S1, preprocessing training data to obtain processed training data; s2, constructing a multi-modal heart image segmentation model; s3, training a multi-modal heart image segmentation model through the processed training data to obtain a completely trained segmentation model; and S4, segmenting the to-be-detected heart image by using the completely trained segmentation model to obtain a segmentation result. According to the method, the convolutional neural network and the global self-attention mechanism are deeply fused, so that the precision of heart multi-modal image segmentation is remarkably improved, and the adaptability of the model to a complex anatomical structure is enhanced.

The invention discloses a PDF (Portable Document Format) drawing recognition and information structured extraction method. The method comprises the following steps: generating a high-resolution bitmap through image preprocessing; positioning and classifying a text region, a table region and a symbol region in the drawing based on a target detection model of transfer learning; hough transform is combined with SIFT feature matching to identify engineering symbols, and sub-pixel positioning is realized through an RANSAC algorithm; after the oblique text is corrected through affine transformation, the content is extracted through OCR; reconstructing a table structure based on OPTICS clustering and projection analysis; constructing an RDF knowledge graph according to a coordinate association rule; and using U-Net difference to detect and position an omission area and complementing the omission area. According to the method, deep learning and image processing technologies are fused, the problems of low rotating text recognition rate, table structure loss and semantic association deficiency in a traditional method are solved, through lightweight model compression and TensorRT acceleration, the analysis accuracy is remarkably superior to that of the traditional method, and the method can be widely applied to the fields of constructional engineering, petrochemical engineering and the like and has wide application prospects. And the drawing information processing efficiency and the data integrity are improved.

The invention provides an intelligent paper marking system based on a large language model. The intelligent paper marking system comprises an examinee test paper character recognition module used for carrying out image processing and content recognition on scanned or shot student answer sheets or answer sheets; the subject knowledge base is used for performing systematic arrangement and representation modeling on multi-subject teaching contents; the subject knowledge retrieval module is used for performing semantic analysis and matching on test paper questions and examinee answering contents to obtain subject knowledge related to the test questions and a scoring basis; a scoring template generator module; a large language model scoring module; and the comment correction module is used for optimizing and adjusting the preliminary comments generated by the large language model and outputting final comments with more pertinence and teaching guidance significance. The technical scheme can be widely applied to automatic evaluation scenes of subjective questions in the education field.

The invention discloses a remote sensing video segmentation method and segmentation system based on text guidance, belongs to the crossing field of remote sensing image processing and computer vision, and relates to a remote sensing video segmentation method and segmentation system. The invention aims to solve the problems that the existing remote sensing video segmentation technology is poor in flexibility, cannot interact with natural languages, is insufficient in generalization ability for new categories or complex targets, and cannot meet the requirement of quickly and accurately extracting semantic information in a dynamic remote sensing scene. The method comprises the following steps: 1, acquiring a video frame sequence, and acquiring a key frame based on the video frame sequence; 2, obtaining an initial segmentation mask; 3, obtaining an optimized mask; 4, calculating a minimum bounding rectangle of the optimized mask, obtaining a bounding box of the minimum bounding rectangle, and obtaining an expanded bounding box; and 5, inputting the expanded bounding box and the video frame sequence obtained in the step 1 into an improved SAM2 video segmentation model, and outputting a frame-by-frame segmentation result of the region of interest by the improved SAM2 video segmentation model.

The invention discloses a lightweight display interface rendering optimization system, which relates to the technical field of image processing, and comprises an acquisition module, an interface analysis module, a behavior analysis module and an attention analysis and rendering module, dynamically calculating an effective view radius by utilizing a visual tunneling effect; the interface analysis module performs character string fuzzy matching in combination with the context input by the user, identifies the search intention of the user and improves the weight of a corresponding component; the attention analysis module performs multi-modal weighted fusion on the physiological fixation data and the sparse content saliency thermodynamic diagram; according to the method, by recognizing the semantic intention and the physiological fixation point of the user, on the premise that core visual experience continuity is guaranteed, GPU load and video memory bandwidth occupation are remarkably reduced, and balance of high-performance display and low-power-consumption operation is achieved.

The invention belongs to the field of image processing, and provides a gynecological tumor image processing method and system based on AI multi-modal image analysis, and the method comprises the steps: 1, obtaining an original image of a patient, and obtaining a structure mask and an image frame sequence after period alignment and structure normalization based on the original image; step 2, obtaining a focus mask sequence after structure limitation based on the image frame sequence; step 3, respectively acquiring a modal structure semantic tensor of each image in the image frame sequence, and acquiring a fused semantic feature tensor based on the modal structure semantic tensor; 4, obtaining a final focus mask based on the fused semantic feature tensor and the structure mask; and step 5, obtaining a response visualization graph based on the focus mask. The method is clear in technical structure, coherent in task chain and independent in model interface, has real deployment and continuous evolution capabilities, and is particularly suitable for gynecological image AI auxiliary system scenes under periodic driving.

The invention relates to the technical field of medical evaluation, and discloses an Alzheimer's disease early cognition evaluation system and method based on combination of traditional Chinese medicine and western medicine. The system comprises a tongue image acquisition module, a pulse condition fluctuation signal acquisition module, a western medicine information acquisition module, a data preprocessing module, a feature extraction module, a data fusion module, a comprehensive data set establishment module and an evaluation decision module, wherein the tongue image acquisition module acquires tongue images, pulse condition fluctuation signals, inquiry text data and inquiry sound information; and the western medicine information acquisition module performs multi-modal fusion on feature data of traditional Chinese medicine and western medicine to construct a comprehensive data set. According to the system, multi-modal data is collected by integrating a traditional Chinese medicine four-diagnosis module and a western medicine detection module, the multi-modal data is processed and fused through algorithms including attention mechanism image processing, signal analysis and the like, and then an evaluation model is trained through integrated learning, transfer learning and the like. The method comprises the steps of equipment deployment, data acquisition and processing, model training, evaluation application and the like, multi-dimensional accurate evaluation is realized by combining optimization of a generative adversarial network, fuzzy logic and the like, and technical support is provided for early diagnosis.

The invention belongs to the technical field of image processing, and discloses a remote digital image analysis cooperation system. The virtual diagnosis module is used for collecting a continuous pathological section sequence and carrying out space registration to obtain a registration section; the model building module is used for performing three-dimensional voxel reconstruction on the registration slices to obtain a three-dimensional voxel reconstruction model, and performing semantic enhancement to obtain a three-dimensional pathological voxel model; the collaborative interaction module is used for generating virtual avatars of G experts based on the VR technology, labeling the three-dimensional pathological voxel model and obtaining space anchor point labeling data of the experts; the graph driving module is used for constructing a structured semantic tree according to the space anchor point annotation data, detecting annotation conflicts and obtaining an annotation conflict detection result; the conflict resolution module is used for performing conflict resolution on the marking conflict detection result to obtain a consensus suggestion; the collaborative efficiency and conclusion reliability of remote diagnosis are improved, and a systematic solution is provided for precise diagnosis of complex pathological cases.

The invention discloses an image recognition and correction system based on an optical model and environment perception, and relates to the technical field of image processing, and the system comprises an environment perception module which is used for collecting multi-medium refractive indexes, interface three-dimensional forms and initial image data through a refractive index sensor, a three-dimensional laser scanner and a visual camera; an environment perception data set is formed through dynamic correction and multi-source fusion; the optical model is constructed on the basis of environment sensing data, the light propagation path is calculated in combination with the interference compensation sub-model, and the interference compensation sub-model can quantify the interference influence by analyzing related data acquired by the environment sensing module for interference factors such as impurities and temperature gradient in a medium, so that the interference compensation precision is improved. The compensation term is generated and fused into the light propagation equation, so that the light propagation path is corrected, the calculation accuracy of the light propagation path is ensured, a firm and reliable data support is provided for subsequent image recognition and correction, and the precision and quality of image recognition and correction are improved.

The invention relates to the technical field of medical image processing and computer vision, in particular to a medical image segmentation method and system based on deep learning, the method is based on a U-shaped encoder-decoder architecture, a DSAB module is introduced into an encoder, and context perception of a directional anatomical structure is enhanced through complementary directional space shift and CSA mechanism weighting; an MGCF module is designed in a decoder, and a parallel multi-scale convolution path and an AGCA mechanism are combined, so that multi-level features are efficiently fused to recover boundary details. Meanwhile, links of data preprocessing, Transform structure details, segmentation result post-processing and the like are supplemented, the model performance is improved through a mixed loss function and an optimization training strategy, and the method has remarkable advantages in segmentation precision and boundary definition and provides powerful support for clinical auxiliary diagnosis.

The invention discloses a high-energy-efficiency logarithm floating point multiplier based on truncation error compensation. The system comprises a compensation parameter modeling unit and a floating point multiplier module. The compensation parameter modeling unit obtains an optimal compensation value k through linear fitting and parameter optimization, and realizes regional self-adaptive compensation in combination with block modeling so as to reduce truncation errors; the floating point multiplier module comprises a special value detection module, a symbol exclusive or module, an exponential addition module and a mantissa truncation compensation module. In a logarithm domain, multiplication is replaced by addition, and compensation logic is introduced to correct errors. The multiplier supports various floating point precisions and configurable truncation bit widths, provides three representative structures to give consideration to both the precision and the energy efficiency, and is suitable for high-energy-efficiency calculation scenes such as neural networks and image processing.

The invention belongs to the technical field of image processing, tracking and recognition, and relates to a multi-mode re-recognition method based on semantic-style decoupling distillation. The method depends on a multi-modal re-identification model which comprises a multi-modal feature extractor comprising a teacher branch module and a student branch module, a decoupling distillation module and a hierarchical self-supervised learning module, and comprises the following steps: constructing a mixed multi-modal feature extractor sharing a shallow layer and an independent deep layer to extract mixed features; performing dual supervision of semantic distillation and style distillation, modeling modal-invariant semantic information and modal-specific style information, and realizing effective decoupling of a feature space; a hierarchical self-supervised learning space is constructed, and in combination with intra-modal and cross-modal comparative learning, images under local damage and style disturbance conditions are scrambled; according to the method, recognition performance and reasoning efficiency are both considered, semantic features and modal specificity styles are effectively separated, semantic consistency, feature robustness and network learning efficiency are cooperatively improved, and modal specificity is also reserved.

The invention discloses a method for high-precision measurement of the diameter of an inhibition zone of a culture medium. The method comprises the following steps: S1, image acquisition and pretreatment: acquiring a culture medium image and carrying out standardized pretreatment on the acquired image; s2, semantic segmentation of the inhibition zone: performing network training on the preprocessed petri dish image based on deep learning; and S3, accurate edge detection: extracting an accurate boundary of the inhibition zone based on a semantic segmentation result, and adopting a sub-pixel-level edge detection algorithm. And S4, scale identification and calibration: automatically detecting and positioning the measurement scale in the image by adopting an intelligent image identification algorithm, and integrating a calibration precision verification mechanism to realize real-time automatic calibration of measurement data. And S5, diameter calculation and result output: based on the results of S3 and S4, calculating the equivalent diameter of the inhibition zone by adopting a geometric fitting algorithm, and generating a standardized measurement report. According to the method, deep learning and image processing technologies are fused, high-precision automatic measurement of the diameter of the inhibition zone is realized, and the method has strong environmental adaptability.