24results about How to "Accurate segmentation" patented technology

This invention discloses a 3D scene segmentation and interaction method based on 3D Gaussian, comprising: Step 1, instance discovery: inputting 3D Gaussian scene data, dividing the 3D Gaussian points into structurally coherent instance-level Gaussian groups to obtain refined instances; Step 2, instance and scene semantic assignment: selecting representative viewpoint images of refined instances and inputting them into a visual-language model to generate semantic description labels for the instances; filtering instance pairs in the scene through spatial geometric relationships, and clarifying the spatial relationship description of instance pairs through a large model, constructing a static scene graph that integrates geometric proximity relationships and instance semantic relationships, forming a structured description of all instances; Step 3, natural language-driven instance localization and interaction: receiving user input commands, combining the instance semantic description, the static scene graph, and the real-time viewpoint direction relationship during the query to perform multi-dimensional matching, determining the target instance ID, and executing interactive operations.

The utility model belongs to medical apparatus and instruments, concretely relates to a kind of pathological auxiliary material taking tool of endoscopic mucosal resection specimen, including total connecting plate and comb plate;The side of total connecting plate is connected comb plate;Comb plate is composed of several parallel straight boards, the width of each straight board is less than target material taking width;The distance between the upper edge of the n-1th straight board to the upper / lower edge of the nth straight board is equal to the distance between the upper / lower edge of the nth straight board to the upper / lower edge of the n+1th straight board, and the distance between the upper / lower edge of the n-1th straight board to the upper / lower edge of the nth straight board is equal to target material taking width, wherein, n is natural number greater than 1 and less than total number N.Compared with prior art, the utility model solves the problem that it is difficult to control the cutting size and easy to cause specimen contamination when cutting specimen into mucosal strips in prior art.The tool of the present scheme can effectively assist cutting positioning and control the cutting size.

The application discloses a complex specification multi-modal understanding method and system based on YOLO two-stage segmentation, comprising the following steps: performing pre-processing on an input specification image; inputting the pre-processed image into a coarse-grained detection stage of an improved YOLOv8 network to generate a coarse-grained candidate region set; inputting the coarse-grained candidate region set into a fine-grained segmentation network based on an improved YOLOv8 feature extraction structure; constructing a layout hierarchical relationship structure based on an accurate layout segmentation result; according to the layout hierarchical relationship structure, performing multi-modal feature extraction on the accurate layout segmentation result, and calculating a matching weight; inputting a graphic-text semantic alignment result and the layout hierarchical relationship structure into a logical structure modeling module together to construct a layout element graph; and generating a knowledge graph semantic output result according to a structured logical reasoning result. The application realizes automatic structured understanding of a complex specification by adopting YOLO two-stage segmentation and multi-modal semantic alignment.

The application discloses a kind of for the method, device, equipment and medium of tab abnormality detection in battery cell production process, comprising: the tab area image collected in battery cell production process is segmented to generate tab mask;Based on the tab mask extraction multi-scale pixel-level feature vector, and the multi-scale pixel-level feature vector is input into unsupervised probability GMM model to identify pixel-level abnormal point;The pixel-level abnormal point is analyzed to obtain candidate abnormal area, and the parameter index of the candidate abnormal area is obtained;Determine whether the parameter index meets the set tab abnormality judgment rule, if meet, it is judged as tab abnormality, and tab abnormality result is output.Therefore, the application can be widely applied to the detection of tab folding, crease, fracture, misplacement and other abnormalities on battery cell production line.

The present application belongs to the technical field of ground laser radar forestry data processing, and discloses a broad-leaved forest single tree segmentation method and system based on branch information guidance, taking ground-based laser radar broad-leaved forest point cloud as a processing object, using RANSAC cylindrical fitting to combine the growth characteristics of the tree trunk to detect the tree trunk; starting from the top of the tree trunk in the low vegetation area, the tree branches are extracted by segmenting and growing the branches and combining the thickness changes of the branch segments; starting from the end of the branch, the tree crown leaf point cloud is segmented by layer-by-layer growth. The present application has stronger trunk detection capability under the conditions of lush low vegetation and complex terrain; on the other hand, the present application can accurately segment the tree crown when the large and small crowns are intertwined and the multiple crowns are closely surrounded. The algorithm is efficient, simple and easy to use, and has great practical significance for improving the semantic understanding ability of forest scenes, assisting forest resource investigation, vegetation ecological research, and satellite remote sensing product calibration and verification.

This invention discloses a non-contact tire deformation recognition method based on a fine-tuned large-scale visual model. The method includes calibrating a monocular high-speed camera, fine-tuning the parameters of the mask decoder of the large-scale visual model using a tire image dataset, generating a point coordinate cue sequence based on pixel calculation using OpenCV, inputting the cue sequence into the fine-tuned large-scale model for segmentation, further processing the pixel matrix using OpenCV to generate a point cue sequence, and post-processing using iterative geometric fitting and region growing algorithms to obtain the mechanical deformation parameters of the target sample tire. This invention can achieve accurate and rapid tire deformation recognition, reaching pixel-level precise segmentation. It solves the problems of large errors, limited measurement environments, nonlinear image distortion caused by cameras, weak generalization ability, and high training costs in existing computer vision-based vehicle tire recognition methods.

The application discloses a method for realizing dynamic visualization of non-heritage pictures, and belongs to the cross field of computer vision and digital cultural heritage protection. The method comprises the following steps: collecting non-heritage pictures and text instructions containing dynamic requirements and respectively preprocessing; completing semantic labeling and image segmentation based on a hierarchical labeling system combined with a ResBlock-CBAM enhanced U-Net; detecting key points according to non-heritage object types; introducing a cultural constraint mechanism, selecting a GANs model to fuse features to generate dynamic data; previewing and iteratively optimizing, and the method can also integrate style picture features. The method can improve labeling and segmentation accuracy, guarantee the cultural authenticity of dynamic effects, improve generation efficiency, adapt to various non-heritage scenes, and optimize user experience, and is suitable for non-heritage digital inheritance.

This invention discloses a polyp segmentation method based on edge guidance and deep-supervised PVT. For colonoscopy images containing polyps, this method utilizes a constructed polyp segmentation network to output the location and shape of the polyps. The polyp segmentation network uses PVT-v2 as the backbone network for feature extraction and also includes a context fusion decoding branch, a boundary extraction branch, a boundary awareness module, and a cascaded feature fusion branch. The context fusion decoding branch acts as an image decoder, receiving the semantic feature map output by the PVT-v2 backbone network. The boundary extraction branch uses a thinning module to enrich the content of the detail feature map, obtaining edge prediction results. The boundary awareness module fuses the thinned features with the output feature map of the context fusion decoding to obtain a boundary-aware prediction result. The cascaded feature fusion branch fuses features at different scales layer by layer in a deep-to-shallow order, outputting the final polyp segmentation prediction result P5.

The application provides an endoscope image enhancement method and device, electronic equipment and a storage medium. The method comprises the following steps: extracting a plurality of image features from an acquired endoscope image, fusing the image features, and performing segmentation processing on the fused feature image through a double-threshold method to determine a blood vessel region and a mucous membrane region in the image. In different image processing modes, the blood vessel region and the mucous membrane region in the endoscope are accurately segmented to obtain a target image. In this way, the contrast of the blood vessel region and the mucous membrane region in the endoscope image can be improved, and in different image processing modes, the image region corresponding to the image processing mode in the image can be enhanced to obtain an enhanced target image, improve the overall image clarity, reduce a large number of data processing steps, ensure the accuracy of image enhancement processing, and improve the image processing efficiency.

This invention discloses a passive visual weld seam tracking method based on deep learning semantic segmentation, comprising the following steps: establishing a vision system and acquiring weld seam images during PAW welding without the aid of external light; segmenting the weld seam images based on semantic segmentation to extract the arc light and weld seam components; labeling connected components in the segmented images to eliminate missegmented parts; determining the welding torch position based on the geometric features of the arc light and performing straight line fitting on the weld seam to calculate the offset between the two; and filtering the current result based on the previously calculated offset. The beneficial effects of this invention are: by constructing a BiseNetV2 network, accurate segmentation of the arc light and weld seam and accurate positioning of the welding torch are achieved; at the same time, the filtering method can obtain higher accuracy to meet the accuracy requirements of real-time weld seam tracking.

The present application relates to a kind of methods for training target segmentation model, comprising: obtaining training set, training set includes multiple frames of continuous original image, each frame original image is annotated with the real label of target;The original image in training set is input into convolutional neural network, obtains the feature map of multiple scales of each frame original image;First feature map of current frame original image and second feature map of previous frame original image are input into gated recurrent unit network, to obtain fusion feature map;Multiple scales of the feature map of current frame original image and fusion feature map are input into feature fusion decoding network, obtain the prediction label of target;Loss between real label and prediction label is calculated based on loss function, and target segmentation model is iteratively trained according to loss, until target segmentation model converges, and the trained target segmentation model is obtained.The target segmentation model obtained can accurately segment target, reduce the interference of complex background and the generation of artifact.

The application is a kind of multi-organ segmentation method based on parallel deep U-shaped network and probability density map, and the steps are as follows: step one: self-supervised pre-training of deep convolutional instance segmentation network encoder based on contrast learning on large-scale unlabeled image dataset; step two: based on organ probability density map and pre-training network weight, multi-organ preliminary segmentation of input image is carried out by using multi-scale parallel deep U-shaped instance segmentation network; step three: the preliminary segmentation result is improved by using multi-scale adaptive fusion module, and the fine multi-organ segmentation of the input image is obtained. The application can make full use of large-scale unlabeled data which is difficult to use in traditional supervised learning, and at the same time, the probability density map guide and multi-scale adaptive fusion module are introduced for multi-organ segmentation task, which ensures the accuracy and reliability of multi-organ segmentation result. The application can be closely combined with clinical diagnosis, radiotherapy and chemotherapy plan making and other related fields, and has broad market prospect and application value.

The application discloses a kind of based on U-Net model's complex building classification extraction method, including with steps:Step1.Getting unmanned aerial vehicle image data;Step2.Make building classification dataset.Step3.Using classification building dataset respectively trains U-Net model, obtains building extraction model;Step4.Enhance dataset;Step5.Using enhanced dataset, U-Net model is fine-tuned training, obtains fine-tuned building extraction model;Step6.Model test: the building image of prediction dataset is input to step Step5 obtains fine-tuned building extraction model, and classification is identified to building.The application can be under the condition that building dataset is limited, to complex building type is effectively identified extraction, the building contour after segmentation is clear, building is complete, can exist in the form of single house, effectively improve the extraction accuracy of complex building.

The invention provides an image data processing method and device and a storage medium. According to the method, the medical image registration problem of the whole segment of the spine is decomposed into segmentation, pose transformation and splicing processing of a plurality of vertebral segments, so that the calculation complexity and the technical overhead of direct registration of the whole segment are remarkably reduced. According to the method, accurate segmentation of vertebral segments is realized by utilizing respective characteristics of a first modal medical image and a second modal medical image on bone imaging and soft tissue imaging; a pose transformation information set is determined based on a segmentation result, calculation of a transformation matrix is further optimized, and high efficiency and accuracy of the registration process are ensured; finally, the complete registration image is generated by performing weighted fusion splicing on the transformed segmented data set, so that the defect of overweight calculation burden caused by overall processing in the related technology is overcome, and the method has the characteristics of high real-time performance and wide applicability, and is particularly suitable for an application scene of real-time registration in an operation.

The application relates to the technical field of image segmentation, and particularly discloses a focused ultrasound micro-damage image segmentation method based on a composite energy field guided Snake model. In view of the problems of single external energy construction, insufficient anti-interference capability, poor time sequence stability and the like of the existing focused ultrasound micro-damage image segmentation method in processing weak boundary, strong noise and dynamic expansion ultrasound images, the external energy item based on regional texture features, boundary structure and dynamic information is integrated into a composite energy field, multi-source joint driving is applied to the contour evolution of a Snake active contour segmentation model from two dimensions of spatial structure and time evolution, the interference of false critical points and the phenomena of contour drift, collapse and border crossing are significantly inhibited in weak boundary, strong noise and damage dynamic expansion ultrasound images, stable and accurate segmentation of the damage region in a complex ultrasound background is realized, and good time sequence consistency is maintained.

The application relates to the technical field of welding automation, and provides a corrugated plate welding seam extraction method and system which are self-adaptive and anti-deformation, the method comprising the following steps: acquiring a three-dimensional point cloud of a corrugated plate surface; segmenting two adjacent geometric surfaces to be welded; selecting a reference surface and a projection surface; calculating the normal distance of each point of the projection surface to the reference surface, screening effective projection points reflecting the real welding seam boundary based on a set threshold value; projecting the effective projection points to the reference surface and performing fitting to obtain a welding seam track. The application automatically excludes local deformation interference through a threshold screening mechanism, directly extracts the welding seam from the real point cloud with high precision, and overcomes the defect that the prior art relies on an ideal geometric model and causes a large error.

A method for segmenting and 3D visualization of chest CT images, belonging to the field of medical image processing and computer-aided diagnosis technology, comprises the following steps: S1, importing and standardizing preprocessing of medical image data; S2, segmenting of multiple tissue images of the thoracic cavity based on a hybrid strategy; S3, dual-mode 3D geometric reconstruction; and S4, interactive 3D visualization and rendering. Through these steps, this invention directly achieves accurate segmentation of multiple tissues and dual-mode 3D reconstruction of surfaces and volumes without relying on high-end hardware or complex manual operations, thus improving image analysis efficiency.

The application provides an eye fundus image lesion segmentation method and device, computer equipment and medium, and belongs to the technical field of medical image processing. The method improves the generalization ability of the model through data preprocessing, uses an eye fundus image lesion segmentation model containing an encoder, a global-local attention module, a multi-scale feature capturing module and a decoder, and realizes accurate segmentation of four kinds of sugar net disease lesions in the eye fundus image. The encoder adopts a three-branch structure to extract multi-scale features, the global-local attention module fuses global and local attention features, the multi-scale feature capturing module extracts multi-scale lesion features through different convolution kernels, and the decoder generates the final segmentation result through channel splicing and up-sampling. The application adopts the above-mentioned eye fundus image lesion segmentation method, device, computer equipment and medium, can simultaneously segment four kinds of sugar net disease lesions in the eye fundus image, has fewer calculation parameters, and has good accuracy and robustness.

This invention belongs to the field of medical detection, specifically relating to a method, system, and device for multi-organ segmentation in medical images using frequency domain perception. The invention designs a U-shaped network with PVT modules as the initial backbone. The encoder in the U-shaped network uses a four-layer PVT module for downsampling, and the decoder uses a four-layer PE module for upsampling. In addition to the four-layer PVT module branches in the encoder, a feature fusion branch consisting of a five-layer TAF module is added. The TAF module is used to extract frequency domain features and fuse them with spatial features. A three-layer SRA module is added to the decoder. The SRA module adds relative position encoding to the upsampling results between adjacent PE modules, and then uses two layers of attention to calculate refined features. Finally, the newly designed network model is trained using multi-organ medical image samples to obtain a high-precision multi-organ segmentation model. This invention solves the problem of insufficient accuracy of existing image segmentation techniques in multi-organ segmentation scenarios.

This invention provides a method, device, and medium for intelligent decision-making in meteorological auxiliary operations. First, Doppler radar data is acquired. Then, based on the Doppler radar data, the spatial distribution information of cloud clusters is determined. Next, based on the spatial distribution information and a clustering algorithm, the cloud clusters are adaptively segmented to obtain independent cloud cluster units. The neighborhood radius and core point threshold of the clustering algorithm are adaptively adjusted according to the spatial distribution information of the cloud clusters. Then, the cloud trajectory of the independent cloud cluster units is analyzed, and the current cloud cluster position is determined based on the cloud trajectory and Doppler radar data. Finally, based on the current cloud cluster position, auxiliary operation decision information is determined. This invention achieves intelligent identification of cloud spatial distribution through Doppler radar data and adaptively and dynamically adjusts the neighborhood radius and core point threshold of the clustering algorithm according to the actual spatial characteristics of the cloud clusters, achieving accurate segmentation of cloud clusters at different scales such as hail nuclei and rainbands, improving the accuracy of cloud cluster identification, and ensuring the effectiveness of operations.

The application relates to a kind of based on deep learning's regional feature classification method and system of distribution network planning, belong to artificial intelligence technical field.The method includes the following steps: using unmanned aerial vehicle to collect distribution network planning area orthophoto and label to form data set;The main network of DeeplabV3+ is replaced by MobileNetV3, the global pooling operation in ASPP module is replaced by NSCT, and EMA is introduced at the end of the encoder to obtain an improved model;The model is trained based on the data set, and then the orthophoto of the region to be classified is input into the trained model to obtain a pixel-by-pixel feature classification prediction result map, wherein the improved model includes an encoder and a decoder, each part uses a specific structure and operation, such as NSCT for feature decomposition and fusion, EMA for multi-scale attention processing, and the decoder performs upsampling, splicing and other operations to output the final result, solving the limitations of manual visual interpretation of orthophoto.

The present invention discloses a camouflaged target segmentation method and system based on light intensity and polarization cues. The method includes: obtaining a test picture and preprocessing the test picture to obtain a polarization information map of the test picture, where the polarization information map includes an intensity map, a degree of polarization map, and a polarization angle map; inputting the polarization information map of the test picture into a pre-trained camouflaged target segmentation network model, and successively performing multi-layer feature extraction processing, compression processing, multi-layer feature fusion processing, multi-branch search processing, and three-modal fusion processing on each polarization information map to obtain a mask image of the camouflaged target. By introducing polarization information including light intensity and polarization cues and using a two-stage fusion method of multi-layer fusion and multi-branch search, the problem of camouflaged target segmentation using light intensity and spectral polarization as cues is solved, helping the polarization vision system to accurately find camouflaged targets highly similar to the current scene.

This application discloses a data augmentation method and apparatus for a multi-task model, a training method and apparatus for a multi-task model, an equipment, and a computer program product. The method includes: acquiring raw point cloud data, including the 3D bounding box of a target object and its corresponding semantic label data; dividing the 3D bounding box of the target object into multiple parts based on the raw point cloud data using a differentiated partitioning strategy, obtaining the 3D bounding boxes of multiple parts corresponding to the target object and the point cloud data contained in each part, thereby constructing a mapping relationship between the point cloud data in each part and the semantic label data; and enhancing the point cloud data contained in each part based on the mapping relationship and a preset multi-dimensional point cloud enhancement strategy. This application, through an innovative data augmentation method, ensures that the augmented data maintains accurate semantic correspondence in the multi-task model, improves the quality and richness of the training data for the multi-task model, and enhances the model's adaptability to different scenarios.

This invention discloses a 3D mesh segmentation method based on boundary awareness and contrastive learning, relating to the field of 3D segmentation in computer graphics and 3D vision. The method includes the following steps: data acquisition, labeling the segmented regions of the constructed 3D mesh, assigning a single integer label to each segmented region, labeling each edge of the 3D mesh with a category, forming a dataset from several labeled 3D meshes, and dividing the entire dataset into a training set and a test set; boundary determination and training a deep learning network using a contrastive learning method; model segmentation, obtaining the trained model, inputting other 3D meshes into the model, and finally outputting the mesh segmentation result. This method can improve the quality of mesh boundary segmentation results, making the boundary regions more accurate.