31 results about "Segmentation system" patented technology

This application relates to the field of artificial intelligence technology and discloses a semantic segmentation system and method for satellite remote sensing images. The system includes: an encoding module configured to extract multi-scale features of the target image; a feature fusion module configured to fuse the original feature maps at various scales through linear attention calculation to obtain a first fused feature, and then split it to generate multiple mixed feature maps with the same scale as the original feature maps; a collaborative decoding module configured to extract spatial domain features and frequency domain features from the original feature maps and the mixed feature maps of the same scale, and perform cross-attention calculation to iteratively fuse spatial domain information and frequency domain information to generate a first tensor; and an output module configured to process the first tensor to obtain the segmentation result of the target image. This system can improve the recognition accuracy of detailed textures and small targets in remote sensing images, and improve the accuracy of semantic segmentation of remote sensing images.

The application belongs to the technical field of computer vision and edge computing, and relates to a cloud-edge collaborative lightweight semantic segmentation method for an automatic driving scene, which comprises the following steps: inputting a target image containing road scene elements into a semantic segmentation model to output a pixel-level class probability map consistent with the size of the target image, and obtaining a semantic segmentation map through Softmax processing; the construction steps of the semantic segmentation model are as follows: firstly, constructing and initializing a cloud-edge collaborative segmentation system, then generating a candidate encoder of a student model and combining the candidate encoder with a decoder of the student model to obtain a candidate student model, performing short training and evaluation on the cloud, and selecting a target student model in combination with edge device constraints and filtering, and finally deploying the teacher model and the target student model to the cloud for full training and knowledge distillation optimization, and then deploying the target student model to an edge device. The application combines neural architecture search and knowledge distillation, can balance the performance of the edge device and the limitation of the computing resources, is high in accuracy and efficiency, and is suitable for the automatic driving environment perception demand.

Receiving, by a multi-tenant platform, a segment query, wherein the segment query includes static attributes and dynamic attributes, wherein the dynamic attributes include interaction data. Receiving, by the multi-tenant platform in response to the execution of the segment query, a segment result set, wherein the segment result set comprises a segment of a population of a tenant of the multi-tenant platform. Automatically synchronizing, by the multi-tenant platform in response to receiving the segment result set, the segment result set with one or more third-party systems. Triggering, in response to the automatic synchronization, the one or more third-party systems to perform one or more campaign actions of a campaign using the segment included in the segment result set. Storing the segment query. Receiving updated data for the static attributes and the dynamic attributes. Re-executing, either periodically or on-demand, the stored segment query using the updated data for the static attributes and the dynamic attributes.

This invention discloses an automatic vein segmentation system and method for ultrasound images. The system includes: an image acquisition unit for acquiring vein ultrasound images; a data processing unit for preprocessing the vein ultrasound images and performing pixel-by-pixel annotation on the preprocessed images to form ground truth labels, thereby obtaining a training sample set; and a segmentation inference unit for generating training configurations based on the training sample set using a self-configured network segmentation framework and performing K-fold cross-validation training to obtain multiple vein segmentation models. The preprocessed vein ultrasound image to be segmented is then input into the multiple trained vein segmentation models for inference and fusion, outputting a vein region segmentation mask. This invention can reduce training instability and inference fluctuations caused by differences in sample sizes, improve the model's adaptability to cross-sample differences, and enhance the stability and robustness of the segmentation results. It can accurately and automatically segment veins in complex ultrasound backgrounds.

This invention provides a rapid landslide area segmentation method based on high-resolution satellite remote sensing data and deep learning algorithms, belonging to the field of ground scene technology. The invention downsamples the original satellite remote sensing image to obtain a first image, extracts reflectance data from the red and near-infrared channels, calculates the reflectance difference and sum of reflectance values, and normalizes them to obtain a vegetation index to construct a vegetation layer. The first image and the vegetation layer are superimposed and stitched together, and feature enhancement is performed to obtain a vegetation feature map. This map is input into a convolutional neural network to output a landslide probability map. A two-dimensional coordinate set is constructed and cropped to obtain high-resolution test patches. The spatial gradient magnitude of the local vegetation index is calculated and weighted by a preset enhancement coefficient to generate edge-enhanced patches. The edge-enhanced patches are input into a semantic segmentation network to output a local landslide map, which is then backfilled into the initial image to obtain a landslide area segmentation map. This invention constructs a cascaded segmentation system to achieve efficient and accurate segmentation of landslide areas.

The application discloses a coronary artery and pericoronary fat segmentation system based on self-supervised learning, and belongs to the technical field of medical image analysis.The application solves the problem that an existing supervised learning model needs to depend on a large-scale labeled data set.The CCTA image is acquired through coronary artery CT angiography technology, a coronary artery segmentation model is pre-trained on unannotated CCTA images in a self-supervised learning mode, rich feature representation is learned by the model, then the coronary artery segmentation model is fine-tuned on annotated data, and thus the dependence on a large-scale labeled data set is reduced.The pericoronary fat segmentation algorithm adopts a traditional image processing algorithm, does not need to be annotated, and saves network training cost.The application constructs an efficient segmentation framework, and solves the high-cost problem in existing coronary artery and pericoronary fat segmentation work.The method can be applied to coronary artery and pericoronary fat segmentation.

The application relates to the technical field of medical image processing, and particularly discloses a cavity image real-time lesion recognition and boundary segmentation system. A cross-sectional image sequence in a cavity is acquired through an interventional imaging catheter, and after shared features are extracted by using a deep network, pixel-level tissue boundary segmentation and overall lesion property recognition are synchronously performed. The core lies in that a dynamic gradient coordination mechanism is introduced, gradient direction conflicts between the segmentation and recognition double tasks are detected and resolved in real time during the training process, and adaptive projection transformation based on the task convergence state is used to fuse the gradients, so that the shared network can learn balanced features which are optimal for both tasks; the method realizes end-to-end real-time processing from image input to synchronous output of the segmentation and recognition results, and significantly improves the synergy, accuracy of lesion analysis and real-time auxiliary efficiency of clinical operation navigation.

The application provides a kind of semi-supervised semantic segmentation system, method, equipment and storage medium, belong to the image segmentation technical field based on computer vision;The application proposes a frequency auxiliary consistency disturbance method FreqMatch, constructs joint disturbance space IFFAP by image feature disturbance flow IFPS and frequency domain auxiliary disturbance flow FAPS, dynamically balances disturbance intensity and direction to optimize quality-efficiency trade-off;At the same time, introduce multi-feature channel class memory mechanism (FCCM), by integrating the multi-channel information of original features, enhanced features and disturbance features, alleviate the class imbalance problem and suppress pseudo-label noise. This method significantly improves the segmentation accuracy and model robustness in the limited labeling scenario, and achieves the current optimal performance on Pascal VOC and COCO datasets.

The application discloses a kind of based on multi-source spatial resolution's remote sensing building target segmentation measurement method, belong to remote sensing image processing technical field.Processing remote sensing image to be obtained, and receive the model type selected by user, weight path and optional manual input resolution;Execute spatial resolution analysis;Call building target segmentation model to infer image, output building candidate binary mask or probability graph, and obtain building area mask according to threshold strategy;Building area mask is executed target level measurement analysis;Based on multi-source spatial reference information, execute real area measurement;According to target level analysis result, generate superimposed rendering graph, black background color mask, statistical table and structured export file, realize the closed-loop output of segmentation, measurement, review and export.The present application is used to solve the problems that real area output is not reliable in existing segmentation system, target level measurement results are inconsistent, and no geographic reference image misoutput real area.

This invention discloses a polyp segmentation system based on multi-scale feature fusion and edge awareness, belonging to the field of polyp segmentation technology. The system first extracts features from intestinal polyp images at multiple scales, from deep to shallow, forming feature maps with at least four levels of resolution, laying the foundation for subsequent multi-level representation. Then, multi-scale feature refinement is performed on the higher-resolution feature maps, introducing a multi-receptive-field structure to enhance the expression of local structure and texture details, while retaining the lowest-resolution features as the starting point for fusion. Furthermore, cross-scale feature fusion is performed in ascending order of resolution, achieving progressive supplementation of semantic features at different scales. Finally, edge aggregation processing using unified channels, unified sizes, and fixed Gaussian convolution kernels strengthens the consistency of structural boundary region representation. The overall scheme has the advantages of clear structural hierarchy, full feature utilization, and strong ability to express polyp edge details.

This application relates to the field of medical image processing technology, and discloses a fully automated lung nodule segmentation system and method based on ultra-high resolution target scanning CT. The system includes a data preprocessing module, a 3D target detection module, an adaptive resampling module, a 3D fine segmentation module, and a post-processing and feature extraction module. The detection module outputs a 3D bounding box to provide nodule in-plane size information for adaptive resampling; the resampling module maintains the Z-axis unchanged while adaptively downsampling the XY axes; the segmentation module adopts a dual-cascaded feature pyramid network architecture to achieve two-stage optimization of coarse and fine segmentation; the post-processing module generates a binarized segmentation mask and calculates quantitative features. This application effectively mitigates the impact of some volume effects on the segmentation accuracy of subsolid nodule boundaries.

This invention relates to the field of image data processing technology, specifically to an OCT-based image choroid segmentation system and method. The scheme performs discrete wavelet transform on the input image to separate low-frequency and high-frequency components; extracts low-frequency global context features and high-frequency boundary direction features through a dual-branch network; in the frequency domain cross-attention module, features are aligned using high-frequency boundary direction features as query keys and low-frequency global context features as keys and values ​​to suppress speckle noise interference and output fused features; the decoder receives the fused features and outputs a choroid segmentation mask. This invention solves the problem of the contradiction between speckle noise suppression and boundary preservation in existing spatial domain processing, filtering out noise while retaining weak high-frequency boundary information, thus improving the accuracy of choroid segmentation boundary localization.

This invention belongs to the field of medical image processing and deep learning, and provides a deep learning-based system and method for fat segmentation of facial and maxillofacial CT images. The system includes: an image acquisition module, an image preprocessing module, and a segmentation network module. The segmentation network module includes a facial and maxillofacial fat segmentation network and an error calculation module. The facial and maxillofacial fat segmentation network includes an encoding module, a decoding module, a bridging module, and cross-layer skip connections. The encoding module includes four sequentially connected encoding convolutional modules, which downsample layer by layer through discrete wavelet transform. The decoding module includes four sequentially connected decoding convolutional modules, which upsample layer by layer through inverse wavelet transform. The samples are then fused with the output features of each encoding convolutional module and input into the next level decoding convolutional module. The cross-layer skip connections include a residual attention module. This application can significantly improve the accuracy and clinical applicability of fat segmentation, providing technical support and algorithmic foundation for automated quantitative analysis of facial fat.

Systems and techniques are provided for processing data (e.g., image data). For instance, according to some aspects of the disclosure, a method may include receiving, at a transformer of a machine learning system, learnable queries, keys, and values obtained from a feature map of a segmentation model of the machine learning system. The method may further include learning, via the transformer, a mapping between an unsupervised output and a supervised output of the segmentation model based on the feature map.

Systems and methods for performing video object segmentation are provided. In an example, video data representing a sequence of image frames and video data representing an object mask can be received at a video object segmentation server. Image features can be generated based on a first image frame of the sequence of image frames, image features can be generated based on a second image frame of the sequence of image frames; and object features can be generated based on the object mask. A transformation matrix can be computed based on the image features of the first image frame and the image features of the second image frame; the transformation matrix can be applied to the object features, resulting in transformed object features. A predicted object mask associated with the second image frame can be obtained by decoding the transformed object features.

This invention relates to the field of image processing technology, and more particularly to an RGBD image instance segmentation system based on adaptive fusion and boundary enhancement. The system includes a data preprocessing module, a dual-stream feature extraction backbone network, an adaptive cross-modal fusion module, a multi-scale feature fusion network, a selective boundary aggregation module, and an instance segmentation prediction head. Specifically, the data preprocessing module encodes the RGB and depth images into a dual-modal input; the dual-stream feature extraction backbone network extracts RGB and depth features in parallel; the adaptive cross-modal fusion module employs a separation and aggregation gating mechanism to suppress noise and adaptively fuse the dual-modal features at multiple levels; the multi-scale feature fusion network aggregates the fused multi-level features; the selective boundary aggregation module extracts detailed information from shallow features and selectively injects high-level semantic features to refine the boundaries; and the instance segmentation prediction head finally generates the instance segmentation result. This invention significantly improves instance segmentation accuracy in complex indoor scenes, especially at small targets and blurred boundaries, through adaptive fusion and boundary refinement.

The application belongs to the technical field of image processing, and discloses a coal flow segmentation system for underground belt conveyor in coal mine based on light-weight ViT, which comprises a decoding module and a processing module; the decoding module comprises pulling RTSP video stream, Opencv decoding, image pressing into stack q, judging the length of message queue q, and popping out images; and the processing module comprises acquiring images, image stack q depth, image adjustment, image feature coding, decoding and outputting results. The PID-Former proposed in the application realizes better balance between speed and accuracy, the detection speed thereof on an RTX3090 GPU reaches 204.1FPS, the speed thereof on an embedded microprocessor JersonTX2 reaches 67.1FPS, the parameter quantity is only 1.934M, the accuracy reaches 96.99mIoU, and real-time coal flow segmentation can also be realized on an edge computing platform, thereby providing a basis for fault detection and speed control of the belt conveyor.

This invention discloses a medical image enhancement and segmentation system based on image processing, comprising: an image acquisition module for preprocessing standardized images; an enhanced image generation module for inputting the standardized images into a pre-trained image enhancement model and outputting high-contrast images; a mask image generation module for generating binary mask images of K slice images; a voxel coordinate labeling module for anchoring the voxel index coordinates of muscle tissue; a three-dimensional coordinate mapping module for mapping the voxel index coordinates of the muscle tissue to three-dimensional spatial coordinates of the muscle tissue; a voxel statistics module for counting the number of voxels occupied by the muscle tissue; and a muscle quantization module for calculating the total volume and estimated mass of the muscle tissue. This invention achieves the processing from raw MRI images to physically interpretable muscle quantization results by using the mean of the remaining standard contrast sequence as a contrast threshold to generate high-contrast realistic images.

The application discloses an OCT blood vessel layering intelligent segmentation system and method, and relates to the technical field of medical image processing.The OCT blood vessel imaging image is acquired and spliced with an initialized feedback correction map, a double-channel segmentation model is input to obtain a first probability map set; a first certainty map is calculated according to the set, and a first segmentation mask is generated; end point sets are obtained through skeletonization and recognition processing, and correction values are calculated based on the end points to update the feedback correction map; the updated feedback correction map is spliced with the first segmentation mask, and second forward propagation, mean value processing and binary processing are performed to obtain a second segmentation mask; whether the Dice similarity coefficient of the second segmentation mask and the first segmentation mask is greater than a preset threshold value is judged, if yes, the second segmentation mask is output, otherwise, the feedback correction map is updated according to the correction value and iteration is continued until convergence or the maximum iteration number is reached; and the application can improve the accuracy and topological coherence of the OCT blood vessel image segmentation result.

Receiving, by a multi-tenant platform, a segment query, wherein the segment query includes static attributes and dynamic attributes, wherein the dynamic attributes include interaction data. Executing, by the multi-tenant platform, the segment query. Receiving, by the multi-tenant platform in response to the execution of the segment query, a segment result set, wherein the segment result set comprises a segment of a population of a tenant of the multi-tenant platform. Automatically synchronizing, by the multi-tenant platform in response to receiving the segment result set, the segment result set with one or more third-party systems. Triggering, in response to the automatic synchronization, the one or more third-party systems to perform one or more campaign actions of a campaign using the segment included in the segment result set. Storing the segment query. Receiving updated data for the static attributes and the dynamic attributes. Re-executing, either periodically or on-demand, the stored segment query using the updated data for the static attributes and the dynamic attributes. Receiving, by the multi-tenant platform in response to the re-execution of the stored segment query, a second segment result set, wherein the second segment result set comprises a second segment of the population of the tenant of the multi-tenant platform. Automatically synchronizing, by the multi-tenant platform in response to receiving the second segment result set, the second segment result set with the one or more third-party systems. Triggering, in response to the automatic synchronization, the one or more third-party systems to perform one or more second campaign actions using the second segment included in the second segment result set.

This invention discloses a segmentation system and batch processing and quality assessment method for concrete CT images, aiming to solve the problems of low segmentation accuracy, poor batch processing efficiency, and lack of objective quality assessment for multiphase components in complex gray-scale overlapping scenes. The method includes: preprocessing using adaptive filtering based on local gradient statistics; constructing a multi-scale feature pyramid network containing encoder-decoder and skip connections to generate initial segmentation; introducing a channel-spatial dual attention mechanism to optimize microcracks and pore boundaries; achieving parallel task allocation under heterogeneous computing resources through a batch processing scheduling engine; establishing a multi-dimensional quality assessment system covering segmentation consistency, topological integrity, and structural fidelity, and feeding the assessment results back to the preprocessing and model parameter adjustment units to form a closed-loop optimization. This enables high-precision, high-efficiency, and fully automated intelligent segmentation and quality control of concrete CT images.

This invention provides a multi-agent-based document hierarchical segmentation system, comprising: a document input interface for receiving documents to be processed; and a set of parsing and segmentation agent modules, including multiple executing agent modules for parsing the documents to be processed and segmenting them according to the parsing results. The executing agent modules use a workflow definition to define the main execution process and logic. The multi-agent-based document hierarchical segmentation method includes receiving documents to be processed through the document input interface; verifying the initial text modules through the set of parsing and segmentation agent modules; if the verification fails, segmentation is performed again to obtain qualified initial text modules. This invention converts the format of the documents to be processed to a unified format by setting the document input interface to facilitate subsequent processing, and sets the parsing and segmentation agent modules accordingly.

A lightweight Mamba right atrial segmentation system based on Atlas prior and boundary awareness is presented, belonging to the field of medical image processing technology. This invention addresses the challenge of existing methods simultaneously achieving segmentation accuracy, lightweight models, and real-time performance. It utilizes a constructed right atrial probabilistic map (Atlas) to provide anatomical structure prior guidance for subsequent segmentation tasks, enhancing performance in LGE-MRI right atrial image segmentation. An attention-guided boundary enhancement module strengthens boundary feature representation at the feature level, while a dynamic weight scheduler optimizes boundary segmentation at the loss function level. This approach obtains complementary information from both region segmentation and boundary optimization, ensuring high image segmentation performance. Furthermore, the lightweight Mamba segmentation network significantly reduces the number of parameters and computational complexity while maintaining long-range dependency modeling capabilities, guaranteeing real-time segmentation. This method can be applied to right atrial image segmentation.

An incremental micro-segmentation system includes a shared network and a network control device. The network control device is configured to perform operations of: retrieving multiple key values from a network flow; adding the multiple key values to be a policy rule of a temporary policy group based on an interested attribute of the temporary policy group of a candidate policy group set; computing a group score of each temporary policy group according to a recommended factor; when determining that the group score is greater than a threshold, generating a recommendation set including the temporary policy groups; and deploying the recommendation set to an access control list to make the temporary policy groups be enforced.

The application discloses a cell instance panorama segmentation method based on a basic visual large model without training, which comprises the following steps: S1, outputting a segmentation mask and a bounding box; S2, extracting a backbone feature vector and a morphological feature vector; S3, selecting a cell image as a support sample from each to-be-classified category; S4, performing leave-one-out method on a backbone feature space and a morphological feature space k A confusion pair set is obtained by calculating a confusion rate of a neighbor classification calculation category pair; S5, a Fisher discriminant direction is calculated and decoupled by using Ledoit-Wolf analytical optimal covariance shrinkage for each confusion pair; S6, a multi-backbone weighted k A neighbor classifier outputs a category prediction result and a corresponding confidence. The application further discloses a cell instance panorama segmentation system based on a basic visual large model without training. The application has the characteristics of not requiring a large amount of training, being cross-center, accurately distinguishing visually similar cells and being easy to integrate, and can be widely applied to the technical field of medical image analysis.

The application discloses a medical image continuous passive domain adaptation segmentation system, and belongs to the technical field of medical image processing. The application solves the problems of limited domain adaptation capability and catastrophic forgetting of previous target domains of the existing continuous passive domain adaptation method. The application constructs a causal chain among an image domain D, an anatomical structure S, an image x and a segmentation result y. In the causal chain, the anatomical structure S and the image domain D jointly determine the image x, and the anatomical structure S and the image x jointly affect the segmentation result y. In the domain adaptation sub-chain, the influence of the anatomical structure S on the segmentation result y is adjusted to adapt to a new target domain, and the domain adaptation capability of the model is improved. In the knowledge reservation sub-chain, the influence of the image domain D on the image x and the influence of the image x on the segmentation result y are processed, so that the memory of old knowledge is reserved while new knowledge is learned, and the anti-forgetting capability of the model is improved. The method can be applied to the technical field of medical image segmentation.