8342 results about "Image segmentation" patented technology

An exemplary image segmentation method and system that employs, in a deep neural network, (i) an attention gate that fuses features with skip connections and (ii) a graph or multi-scale convolutional attention module that enhances the long-range and local context. The segmented region or image data derived in part from the segmented region can be subsequently employed for diagnosis, controls, planning, assessment, or analysis of various diseases. The image data can be medical images (e.g., from medical instruments), sensor images or videos (e.g., from smart phones, cameras, satellites, etc.), as well as 3D objects or volume.

The invention discloses a Mama-based spectrum dynamic fusion and double-attention enhancement medical image segmentation method, which comprises the following steps of: firstly, constructing a Mama integrated spectrum domain and attention pyramid module, fusing spectrum dynamic characteristics and a self-attention pooling mechanism, and performing frequency domain information compensation and local characteristic enhancement to obtain a spectrum dynamic fusion image; the spatial correlation loss caused by image blocking processing is relieved; secondly, designing a layered enhanced U-shaped architecture, deploying an MISAP module in a shallow layer of an encoder to capture multi-scale global context features, introducing a bipolar routing attention mechanism in a deep layer, and dynamically allocating sparse attention weights to focus a key pathological region; according to the method, the segmentation precision of complex edge textures and tiny lesions in medical images can be remarkably improved, and the Dice coefficient in breast tumor, polyp and abdominal organ segmentation tasks is averagely improved by 6.5%.

SOLUTION: A semi-supervised medical image segmentation method based on a visual language model includes the steps of: obtaining a medical image; inputting an unlabeled image and a text description into a visual language model, and obtaining a text-guided mask based on obtained dense image embedding and text embedding; inputting a labeled image into a student model, and calculating supervised loss by using obtained labeled image prediction; respectively inputting the unlabeled image into the student model and a teacher model to obtain unlabeled image prediction and a pseudo label, merging the text-guided mask with the pseudo label, and calculating semi-supervised loss by using the merged pseudo label and unlabeled image prediction; and performing medical image segmentation by using a trained student model on the basis of the supervised loss and the semi-supervised loss.EFFECT: A target segmentation region can be accurately identified by using advantages of text descriptions.SELECTED DRAWING: Figure 1

The invention relates to the technical field of welding processing, and discloses a dissimilar metal welding joint temperature field optimization control method and system. Real-time temperature field scanning is conducted on the dissimilar metal welded joint, image segmentation processing is conducted on a temperature field image obtained through scanning, temperature data of a weld joint area, a heat affected area and a base metal area are extracted, and whether the temperature gradient exceeds a preset threshold value or not is judged; aiming at an area with non-uniform distribution of heat input quantity, analyzing the performance of the welding joint under the combined action of thermal stress and mechanical stress, obtaining the difference between residual stress distribution and plastic deformation capacity, and analyzing a stress concentration area by combining with the yield strength of the material; and a mapping relation between the welding process parameters and temperature field distribution is established, and a welding process parameter adjustment prediction model is obtained through historical welding process parameters and corresponding temperature field data training in combination with heat flux density distribution. The problems of uneven heat input, stress concentration and the like in dissimilar metal welding are effectively solved, and the welding quality and efficiency are improved.

The invention provides a crop whole growth cycle identification method and system based on deep learning, and the method comprises the steps: obtaining a multispectral image sequence of a target crop in a continuous time period through an image collection device, and carrying out the standardized illumination adjustment processing of each image frame in the multispectral image sequence, obtaining a standard illumination image set corresponding to the multispectral image sequence; carrying out crop region segmentation processing on each image, extracting a local feature region related to a target crop, generating a standardized crop image set containing the local feature region, inputting the standardized crop image set into a pre-trained multi-task deep learning model, extracting combined features through parallel convolution branches, and carrying out image segmentation processing on the combined features; and executing cross-stage correlation analysis in the full connection layer, outputting a multi-task classification result corresponding to the target crop growth cycle, and generating a stage identification report corresponding to the target crop full growth cycle. According to the invention, the crop whole growth cycle identification precision and the agricultural management efficiency can be improved.

The invention discloses a manipulator grabbing method based on deep learning target detection and image segmentation, and relates to the technical field of artificial intelligence and robotics.The manipulator grabbing method comprises the following steps that a scene image to be processed is collected, the image quality is improved through the multi-light-source fusion image enhancement technology, and recognition errors caused by uneven illumination are reduced; and inputting the enhanced image to a pre-trained deep learning model, executing a target detection task, and outputting an initial bounding box and a category label of the target object. According to the method, through multi-light-source image enhancement and high-precision image segmentation, the accuracy of target recognition and contour extraction is remarkably improved, and the capture failure rate caused by image misjudgment is reduced. And meanwhile, geometric consistency verification and a multi-factor grabbing scoring mechanism are introduced, dynamic screening and collision pre-detection are conducted on the paths, the grabbing stability and safety of the mechanical arm in the complex environment are effectively guaranteed, and the intelligence and robustness of the whole system are remarkably improved.

The invention provides a machine vision-based heavy-load large holding tool driving wheel multi-mode damage detection method, and relates to the technical field of automatic detection, and the method comprises the steps: collecting the image, vibration and temperature data of a driving wheel, extracting the visual features of the image through a multi-scale pyramid network, and obtaining the multi-mode damage of the driving wheel; inputting a cross-modal feature fusion network in combination with eccentricity and runout features of the vibration signals and stress concentration features of the temperature data to obtain fusion features; inputting the fusion features into a multi-task detection module to identify the damage, and measuring the damage area, depth, crack growth rate and stress concentration degree based on image segmentation, a photometric stereo method, vibration signal analysis and temperature gradient analysis; and finally, in combination with the equipment operation condition and the damage evolution law, an improved evidence theory is adopted to fuse the evaluation indexes, and an evaluation report including the damage state, the development trend and the maintenance suggestion is generated.

The invention discloses a close planting farmland growth vigor assessment method and system based on image processing, and relates to the field of agricultural information, and the method comprises the following steps: S1, multi-source data collection and preprocessing; s2, improving image segmentation, and extracting crop features; and S3, multi-dimensional growth vigor evaluation. According to the method, the field block level, the plant level and the whole growth period are covered through multi-source data collection, a generative adversarial network is used for repairing and shielding the plant image and restoring complete form information, the segmentation problem in a close planting scene is solved, the accuracy of close planting crop image analysis is improved, accurate registration of multi-modal data is achieved by means of feature point matching, and the accuracy of close planting crop image analysis is improved. The graph neural network optimizes image segmentation, effectively distinguishes overlapped leaves and stalks, deeply fuses multi-modal features and dynamically selects a fusion strategy, improves feature distinguishability, constructs a dynamic adaptive evaluation model, improves generalization ability and evaluation precision, identifies and intervenes abnormities in real time, and improves crop anti-risk ability and yield prediction accuracy.

The invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based image segmentation and dynamic target recognition method, which comprises the following steps of: accurately positioning a candidate region through multi-modal space-time fusion and dynamic confidence coefficient screening; strengthening spatial-temporal feature expression in a layering manner through a multi-level feature decoupler, and generating a multi-dimensional feature enhanced spatial-temporal candidate region; through a deformable segmentation network, a deformation convolution kernel and edge motion matching loss are combined, joint optimization of a geometric boundary and motion continuity is realized, and the segmentation robustness of a flexible target is improved; through optical flow back propagation dynamic correction and confidence coefficient propagation, high-precision segmentation masks with consistent time and space are output; and through a target trajectory re-identification and completion mechanism driven by a graph attention network, and in combination with optical flow deformation prediction, stable tracking in a shielding scene is realized.

A three-dimensional modeling method and apparatus, which relate to the technical field of modeling, and aim to implement an automated three-dimensional scene modeling task, thereby reducing the modeling duration and the modeling cost. The embodiments of the present application involve: after three-dimensional point cloud data is reconstructed on the basis of collected images, on the basis of the three-dimensional point cloud data, segmenting a scene to be modeled, so as to obtain a point cloud feature of a target object; then obtaining a mask feature map of the target object on the basis of the images; and registering the point cloud feature and mask feature map of the target object in a three-dimensional space, so as to obtain geometric structure parameters of the target object. Instead of manually selecting points from a cloud point feature, more accurate geometric structure parameters of the target object are obtained by means of registering a point cloud feature of a target object that is obtained by means of point cloud segmentation and a mask feature map of the target object that is obtained by means of image segmentation, without the need for manual intervention, such that the automation rate of modeling operations is increased, thereby reducing the modeling duration and the labor cost consumed by modeling.

The invention provides a robust multi-modal image segmentation method and system based on instance perception query, and the method comprises the steps: obtaining original image data and text description data, and generating training data through semantic analysis and causal graph construction; extracting image features and text features by adopting a multi-layer coding network, and establishing a feature dependency relationship through causal reasoning; generating instance clustering features based on attention calculation, and obtaining fusion features in combination with a multi-scale fusion strategy; and calculating query confidence and generating candidate segmentation masks, and outputting a final segmentation result through post-processing optimization. According to the method, by introducing causal reasoning and instance awareness query mechanisms, the cross-modal feature alignment precision and the robustness of the segmentation result are improved, and the method has better segmentation performance and generalization ability in a complex scene.

The invention discloses a Transform-CNN medical image segmentation method and system based on multi-scale fusion semantic enhancement, and relates to the technical field of image segmentation, and the method comprises the steps: collecting medical image data, and constructing a medical image data set; constructing a medical image segmentation network model; training the medical image segmentation network model through the medical image data set; and inputting real-time acquired data into the trained medical image segmentation network model to obtain a medical image segmentation result. According to the invention, double encoders, LGDA modules and the like are adopted to capture multi-scale global features of kidney tumors and enhance representation, and the LGDA module is used for adapting to size and form changes of the kidney tumors; the MLCF module is used for supplementing information of the main encoder; the PSE module captures multi-scale global semantic information and integrates local context information, local features are fused through the SAM module, and kidney tumor positioning in an endoscope image is achieved.

According to the medical image segmentation method and device based on spatial perception and frequency domain information, the precision and robustness of medical image segmentation are effectively improved by combining frequency domain information guidance and a multi-head state spatial perception technology. Frequency domain transformation is performed on a medical image, high-frequency and low-frequency components of the image are separated by using a multi-scale decomposition technology, and low-frequency features are extracted to obtain global information. By introducing a learnable noise filtering mechanism, noise and irrelevant background information in a frequency domain are suppressed, so that the model can be focused on a lesion area more accurately. A multi-head perception visual state space module is designed on a bottleneck layer, lesion features of different scales are captured through a multi-scale adaptive feature fusion mechanism, and the capability of segmenting small-size lesions and complex structures is enhanced. A context focusing attention mechanism is introduced into jump connection, fusion of global information and local details is further enhanced, and the accuracy of a segmentation result is ensured; and finally, recovering a high-resolution segmented image through a decoder.

The invention discloses a self-prompting semantic segmentation method and device based on SAM and a storage medium, and belongs to the technical field of image segmentation. The method comprises the following steps: acquiring a target image containing a target object; inputting the target image into a semantic segmentation model based on SAM to obtain a segmentation result of the target image; the segmentation result comprises the position and the type of the defect of the target object; wherein the semantic segmentation model comprises an image encoder, a target detector, a prompt encoder and a mask decoder; the target detector is used for positioning the position area of the defect of the target object on the target image according to the image features output by the image encoder, and outputting a defect detection frame; the prompt encoder is used for encoding the defect detection frame and generating position prompt characteristics; and the mask decoder is used for fusing the image features and the position prompt features and generating a segmentation result of the target image. According to the invention, end-to-end industrial defect detection can be realized, and the efficiency of industrial defect detection is improved.

The invention belongs to the technical field of iron stick yam image recognition detection, and discloses an iron stick yam intelligent system and method based on image recognition, and the method comprises the steps: dynamically adjusting the polarization direction to generate a polarization suppression image, and generating a space mapping relation; based on the polarization suppression image, segmenting the main body area of the iron stick yam, filling and optimizing the hole edge, generating a high-precision contour mask, further extracting the texture difference between wrinkles and cracks, generating a direction sensitive characteristic pattern, and evaluating morphological defects; constructing a comprehensive feature vector, training a Gaussian mixture model to generate reference distribution, calculating a quality deviation index, and judging a risk level; constructing a double-branch feature vector, outputting a preliminary comprehensive quality score through a complementary aggregation model, introducing a planting density-curvature physical model, obtaining a final quality score, dynamically adjusting a quality judgment threshold, dividing quality grades, and triggering a response strategy according to confidence grade to form a closed loop; and comprehensiveness and accuracy of Chinese yam quality evaluation are improved.

The present invention relates to the technical field of agricultural information, and specifically to a plant factory information automatic acquisition method and system based on unmanned aerial vehicle machine vision. The method comprises the following steps: using a camera mounted on an unmanned aerial vehicle to capture an image, and using a deep learning and image segmentation technology to generate a plant region segmentation image. According to the present invention, the unmanned aerial vehicle is used to carry the camera to capture a plant image, and efficient and accurate plant region segmentation is realized by using the deep learning technology. A plant health assessment report is generated by means of a computer vision technology, and a health problem is early warned. A plant health analysis report comprehensively considering environmental factors is generated on the basis of ground sensor data and time series analysis, and comprehensive and accurate health assessment is provided. A multi-dimensional plant health monitoring model provides support for all-round health monitoring. A plant disease nutrition analysis report is generated on the basis of a machine learning classification algorithm, a problem is accurately recognized, and a fertilization and pesticide-spraying strategy is formulated on the basis of a decision support system, thereby optimizing resource utilization, and improving the yield and quality.

The invention provides a division method and system for a forest pest control area, and relates to the technical field of data processing, and the method comprises the steps: building a forest pest control database through collecting pest data and environment data of a control area; and training a pest risk prediction model based on historical data to generate a risk distribution map. And then, in combination with a GIS technology and a deep learning image segmentation algorithm, intelligent division is performed on the target area, and it is ensured that the unmanned aerial vehicle efficiently covers the prevention and control unit. According to the method, the optimal spraying route can be planned by adopting a path optimization algorithm based on the flight capability of the unmanned aerial vehicle, the pesticide carrying capacity and the environmental conditions, spraying parameters are dynamically adjusted by monitoring the wind speed, obstacle information and the like in real time, precise pesticide application is ensured, the pest control efficiency can be improved, repeated spraying and spraying blind areas are reduced, the pesticide use amount is reduced, and the pesticide application cost is reduced. Intelligent and precise forest pest control is achieved, and the method is suitable for large-scale unmanned aerial vehicle autonomous operation scenes.

The invention relates to the technical field of medical image processing, and provides a medical image segmentation method based on wavelet enhancement and multi-scale feature fusion. According to the method, a CNN-Transform double-branch coding structure is combined, a multi-scale wavelet fusion module is provided, from the perspective of a frequency domain, Haar wavelet transform is adopted to extract an image high-frequency sub-band so as to enhance edge and texture detail expression, dynamic weighting is performed on different frequency band features through grouping convolution and a sub-band attention mechanism, and the discrimination capability is improved; meanwhile, a multi-scale cavity pyramid structure is fused in a spatial domain, and after cross attention dynamic fusion is introduced, a feature alignment mechanism of a wavelet domain and the spatial domain is established; and collaborative fusion of frequency domain and space domain features is realized. The method effectively improves the segmentation precision of the fuzzy boundary and the fine-grained structure under the complex background, has good universality and adaptability, and is suitable for various medical image segmentation tasks.

The invention relates to the technical field of medical image segmentation, in particular to a few-sample medical image segmentation method based on edge perception multi-prototype learning, and the method comprises the steps: inputting support and query images into a feature encoder, and extracting support and query feature maps of different sizes; inputting into a local attention fusion prototype generator to generate a support foreground prototype; processing the support mask through dynamic corrosion operation to generate an inner boundary prototype; generating a multi-foreground local prototype through a multi-layer perceptron; local and global information is optimized through multi-scale feature extraction, and a multi-scale prototype is obtained; fusing to obtain a multi-prototype foreground prototype; dynamic calculation weighting is carried out on the multi-prototype foreground prototype by using a double-stage prototype optimization network, and automatic calibration is carried out; then prediction is carried out through a prototype prediction module, and finally collaborative optimization is carried out through a loss calculation module; the method can effectively solve the problem of edge detail loss involved in the background technology.

The invention discloses a large-format image segmentation method and system fusing frequency domain perception and state space modeling. According to the method, three core parts including an encoder, a state space modeling module and a frequency domain perception feature fusion module are included, a feature extraction network based on a ResNet structure is adopted by the encoder, multi-scale feature extraction of an input image is achieved, and effective capture of deep semantic information is ensured. The state space module realizes adaptive optimization of image data through visual state space modeling, and constructs a hybrid architecture in combination with a wavelet multi-scale convolutional network, thereby effectively solving the limitation of the traditional CNN in the aspect of remote dependence capture. And the frequency domain perception feature fusion module effectively optimizes the feature reconstruction process through interactive fusion of multi-frequency domain information, so that efficient and fine pixel-level semantic segmentation is realized. The decoder can integrate a low-frequency global structure and high-frequency edge detail information, and the recovery effect of edges and textures in a high-resolution image is remarkably enhanced.

The invention discloses a tumor space occupying brain function partition nerve influence image segmentation method based on a deep learning model, and particularly relates to the technical field of artificial intelligence and medical cross. Comprising the following collaborative operation steps: multi-modal data fusion acquisition, tumor edge feature enhancement, anatomy-functional feature collaborative modeling, occupation deformation compensation, adversarial boundary optimization, dynamic loss regulation and control, cascade segmentation strategy, clinical interaction verification, and construction of a U-Net + + based improved three-dimensional feature fusion network. Utilizing a multi-scale cavity convolution module to extract edge gradient features of the tumor infiltration area in a layered manner; in order to solve the problem of multi-modal fusion and dynamic adaptation mechanism deficiency in the prior art, a three-dimensional cross-modal fusion network of a channel competition mechanism is constructed, multi-modal key features are dynamically screened through spatial pyramid pooling and soft attention weight, and a tumor volume self-adaptive dynamic loss function is combined, so that multi-modal fusion and dynamic adaptation are realized. The problems of feature confusion and poor generalization of a traditional method are solved.

The invention relates to the technical field of prostate MRI (Magnetic Resonance Imaging) image segmentation, in particular to a prostate MRI image segmentation method based on Mamb-Unet, which realizes effective fusion of features of different scales between an encoder and a decoder, and fuses multi-scale information on a space axis by using a space attention bridge segment to fuse fusion features of different levels from the encoder, so that the segmentation accuracy of the prostate MRI image is improved. And then global and local information is captured through 1D convolution and full connection in the channel attention bridge segment, so that a channel attention map is generated, channel information is better integrated, and multi-level different-scale feature fusion is realized.

The invention relates to a cerebrovascular three-dimensional evaluation method based on multi-scale feature fusion, and the method constructs a multi-scale feature fusion model which comprises a pilot segmentation network, a fusion segmentation network and a topology-style loss module. The pilot segmentation network is based on a dynamic Unet network, the response of the model to the complex boundary of the cerebrovascular image is enhanced, and the pathological detection performance is improved. The fusion segmentation network introduces a feature selection module, and effectively fuses the long-range dependence feature of the Transform layer and the local context information of the CNN layer, so that the model can accurately extract the cerebrovascular features under different scales. The topology-style loss module optimizes model parameters by combining a topology loss function and style self-consistency loss, restrains the topology continuity of a segmentation result, and effectively solves the problem that blood vessels are easy to break in a traditional segmentation method. The method effectively improves the cerebrovascular image segmentation precision.

The invention discloses a rock debris image segmentation method based on multi-scale feature enhancement and edge perception gating. The method comprises the following steps: covering 13 * 13 to 3 * 3 pixel receptive fields through four-stage parallel depth separable cavity convolution; the dynamic weight generation network generates a space attention weight to realize self-adaptive fusion of multi-branch features; a traditional edge operator kernel is initialized in a shallow network, and 5 * 5 learnable convolution is adopted in a deep layer to enhance edge continuity. Through channel grouping initialization and a random disturbance strategy, the direction sensitivity of a traditional operator is reserved. A mixed weight is generated through channel attention and space attention, and edge features are injected into a main path in a residual scaling mode. The attention weight dynamically adjusts the contribution degree of the edge and semantic features, and high-precision segmentation of a boundary region is ensured; the shallow edge features are optimized preferentially by combining the loss function, and the robustness of the model to complex textures is improved.

The invention discloses an ophthalmologic operation real-time navigation system and method based on dynamic visual field tracking, and relates to the technical field of medical systems. Comprising the following steps: matching operation data closest to a current operation from a database, and generating standard operation navigation of the current operation; the real-time image of the operation process is compared with the image of the standard operation navigation in real time, the step of the current operation is determined, and meanwhile subsequent image navigation of the operation of the patient is displayed; generating a spatial mapping matrix between an eyeball movement track and a target region by combining target region boundary data of preoperative OCT image segmentation and an intra-operative navigation coordinate system; the virtual navigation path of the surgical instrument is dynamically corrected according to the real-time eyeball movement track, so that the surgical instrument is aligned with the target area, the spatial relation between the instrument and the target area can be visually mastered through the overlapping display of the three-dimensional virtual path and the real-time image, unconscious displacement or physiological tremor of a patient in an operation is actively adapted, and the operation accuracy is improved. And the continuous accuracy of the navigation path is ensured.

The invention relates to the technical field of concrete member detection, and discloses a concrete member defect identification and labeling system, which comprises an image acquisition equipment matching module, a defect feature analysis module and a real-time labeling regulation and control module, and a defect classification priority judgment module capable of being additionally arranged. The image acquisition equipment matching module calculates and matches the optimal equipment through the adaptive characteristic value based on the image resolution, the equipment acquisition precision, the working distance and the illumination compensation parameter; the defect feature analysis module performs quantitative analysis on features such as textures, crack forms and hole distribution of zoning images by using algorithms such as multi-scale image segmentation and frequency domain transformation; the real-time labeling regulation and control module dynamically adjusts the labeling position according to the defect position offset, the size change rate and the illumination fluctuation parameters; and the defect classification priority judgment module divides defect grades according to crack width, hole density and the like. The system improves the automation level and accuracy of concrete member defect detection, and is suitable for constructional engineering member quality detection.

The invention provides a titanium alloy bar visual detection method based on image segmentation. The method comprises the following steps: carrying out pretreatment and feature enhancement operation on an acquired titanium alloy bar surface image; constructing an adaptive segmentation algorithm based on region growth to carry out defect region segmentation on the surface image of the titanium alloy bar; performing defect feature extraction on the segmented defect region; a defect classification model is constructed, and titanium alloy bar defect types are classified based on defect features; performing quantitative analysis on the defect severity through a defect severity quantitative model; establishing a defect causal chain model, and carrying out defect traceability analysis and causal derivation; after defect traceability analysis is completed, fusion analysis is carried out on the multi-scale features, and decision suggestions are given. According to the method, the surface defects of the titanium alloy bar can be efficiently and accurately recognized and classified, the accuracy and robustness of defect detection are improved, and the defects of a traditional detection method in the aspects of efficiency, recognition capability, quantitative analysis and the like are effectively overcome.

The invention discloses an adaptive medical image segmentation method during testing based on prototype alignment. The method comprises the following steps: acquiring target domain medical image data; obtaining a source domain model, wherein the source domain model comprises a feature extractor and a classifier; extracting a class prototype of each class by using a classifier, and initializing two models with the same structure as the source domain model by using a feature extractor and the classifier of the source domain model; the two initialization models are respectively used as a teacher model and a student model; the target domain image is input into a teacher model and a student model, the teacher model outputs a pseudo tag and an entropy graph of a target domain, the student model carries out supervised training according to the pseudo tag, and a target domain prototype output by the student model is aligned with a source domain prototype; the teacher model is updated through the index moving average value of the student model parameters, and finally the teacher model outputs a segmentation result. According to the method, the relation between the source domain and the target domain can be effectively established, and the cross-modal image segmentation precision is improved.

The invention discloses a rapid construction method and system for a three-dimensional space model of a multi-level building, and relates to the field of building model construction, and the method comprises the following operation steps: S1, constructing a dynamic demand-oriented detail level model; s2, automatic component modeling based on image segmentation; s3, performing multi-version difference judgment and model updating; and S4, cloud collaboration and load optimization. According to the method and the system for quickly constructing the three-dimensional space model of the multilevel building, the region is dynamically divided through the semantic weight, so that the rendering efficiency during model construction can be effectively improved, the memory can be effectively reduced during rendering, the visual fidelity can be optimized, key curvature characteristics can be forcibly reserved for a high semantic weight region, and the construction efficiency is improved. Based on this, deformation distortion caused by simplification can be avoided, a visible area in the model can be automatically identified, an invisible area can be degraded, only the contour line frame of the building is reserved, and the overall memory occupation can be greatly reduced.

Disclosed is a method for establishing a 3D medical image segmentation model based on masked modeling and application thereof includes: establishing a semi-supervised learning network, wherein a student network includes an encoding module for extracting latent features and a segmentation decoder that predicts segmentation results, a teacher network includes an encoding module and a segmentation decoder that are structurally consistent with the student network; training the semi-supervised learning network, wherein during training, two random masking operations are performed on each image, and the image is input to the two networks respectively; optimizing and updating the weight of the student network, and transferring the updated weight to the teacher network, wherein the training loss function includes prototype representation loss, which is used to characterize the difference between the prototypes extracted and generated by the two networks; the student network may further include a reconstruction decoder and an auxiliary segmentation decoder.