100results about How to "Effective segmentation" patented technology

The invention discloses an intelligent inspection system and method for a photovoltaic module of a photovoltaic power generation system, and relates to the technical field of intelligent inspection. The method comprises the steps that a high-definition camera and an infrared imager carried by an unmanned aerial vehicle automatically inspect a photovoltaic panel, real-time operation data are acquired and transmitted to a preprocessing module, the preprocessing module optimizes an image through denoising, graying, contrast enhancement and geometric correction, a Unet algorithm is utilized to accurately segment strings and assemblies in the image, and the photovoltaic panel is obtained. And a defect identification module analyzes the processed image, calculates the average temperature, compares the average temperature with a threshold value, identifies hot spot defects, and finally outputs specific information, fault types and maintenance suggestions of a defective component by the system, and generates an inspection report. According to the technical scheme, a fault source can be quickly positioned, and cloud storage and an AI analysis tool are used for managing data and predicting potential maintenance requirements.

The invention belongs to the technical field of medical image processing, and particularly relates to an ultrasonic mammary gland lesion area automatic segmentation method based on multi-scale hybrid convolution, which comprises the step of carrying out image segmentation by adopting U2-Net as a network framework, and is characterized in that: at a jump connection part in each residual block, a multi-scale hybrid convolution algorithm is adopted to carry out image segmentation; the feature maps are processed through a multi-scale attention gate and then jump connection with the corresponding feature maps is carried out. According to the method, a hybrid convolution block and a multi-scale attention gate are fused with a U2-Net network structure, and unimportant features are further inhibited and valuable features are enhanced through three different receptive field convolution blocks, namely point-by-point convolution, common convolution and expansion convolution, so that a breast cancer lesion area is quickly and effectively segmented; secondly, a hybrid convolution block is connected to the last convolution layer of the encoder of the residual block and is used for extracting global context information of the small U shape; the method can improve the recognition and segmentation efficiency of the lesion area of the ultrasonic image, reduces the workload of a doctor, and assists the clinical diagnosis.

The invention discloses an SEM image granularity analysis method and system based on a visual large model, and relates to the technical field of image recognition. The method comprises the steps that an SEM image is acquired, automatic scale identification is performed on the SEM image, and the actual length and the pixel length of the scale and the actual length of each pixel are obtained; an SEM image set containing particle features is collected, particles in the SEM image set are labeled, and an SEM image training set and an SEM image test set are constructed based on a labeling result; constructing a visual large model, and performing fine tuning on the visual large model by using the SEM image training set; reasoning the SEM image test set by using the fine-tuned visual large model to obtain a particle segmentation result; attributes of each segmented particle are calculated based on the actual length of each pixel for particle size analysis. According to the scheme, the efficiency, precision and consistency of particle analysis are improved, and the efficient and accurate analysis requirements of microscopic particles are met.

The invention provides a large model segmentation and deployment method and device based on cloud edge end collaboration and a medium, and the method comprises the steps: giving a pre-trained large model as input, and evaluating the performance and resource consumption conditions of the large model during cloud edge end deployment; determining a segmentation point of the large model according to a performance evaluation result and a resource consumption condition; decomposing the large model into a plurality of sub-models according to the segmentation points; deploying each sub-model at a cloud end and an edge end; and establishing a cloud side end cooperative work mechanism, so that the sub-models cooperatively work at the cloud side and the edge side. The method can reduce the demand for computing resources and improve the response speed while ensuring the model performance, and is suitable for the demand in the field of intelligent manufacturing.

The invention discloses a remote sensing data monitoring method and system based on distributed computing, and belongs to the technical field of remote sensing data processing. The method comprises the following steps of performing comprehensive analysis according to environment interference data and remote sensing equipment performance data to obtain an environment interference degree evaluation value of each target area and a remote sensing equipment performance evaluation value of each target area, and performing adjustment feedback according to the remote sensing equipment performance evaluation value of each target area; performing comprehensive analysis according to the environment interference degree evaluation value of each target area and the remote sensing equipment performance evaluation value of each target area to obtain a remote sensing data accuracy evaluation value of each target area, and screening according to the remote sensing data accuracy evaluation value of each target area to obtain each effective comparison area; and processing the feature image data of each effective comparison region to obtain an abnormal change index of the to-be-detected region, and performing early warning feedback according to the abnormal change index of the to-be-detected region.

The invention discloses a water absorption rate detection method and system for a macromolecular water absorption material, and relates to the technical field of macromolecular water absorption materials.The water absorption rate detection method for the macromolecular water absorption material comprises the following steps that the initial weight of a to-be-tested macromolecular water absorption material sample is weighed; dropwise adding test liquid to the surface of the polymer water-absorbing material sample by using a burette; obtaining test feature data; factors influencing the water absorption performance of the polymer water absorption material are obtained; predicting key factors influencing the water absorption rate of the polymer water absorption material; and judging whether the water absorbing material meets a preset performance standard or not. Conditions of each experiment can be ensured to be consistent, so that repeatability and reliability of results are improved, the water absorption process can be accurately controlled by dropwise adding test liquid through a burette, the water absorption behavior of the material can be better understood, key factors are adjusted through an optimization algorithm, then the performance of the material can be improved, and the test accuracy is improved. And the preset performance standard is met.

The invention discloses a lateral bending spine model rapid restoration method based on semantic feature optimization, and relates to the technical field of medical image analysis and biomedical engineering, and the method comprises the steps: carrying out the segmentation recognition operation of a pre-collected medical image through the morphological feature and geometric analysis technology, obtaining the three-dimensional semantic information of the spine structure based on the segmentation recognition result; extracting morphological parameters on the basis of the three-dimensional semantic information of the spine structure, and exploring an association relationship among the morphological parameters by using a correlation analysis technology; and generating a spinal semantic parameter combination based on the association relationship and the morphological parameters, and searching an optimal spinal semantic parameter combination by using an optimization algorithm to repair the scoliosis spinal model. According to the method, based on a global optimization strategy of an optimization algorithm, a reasonable evaluation function is designed to quantify the performance of each parameter combination on a scoliosis model repair task, and the optimal semantic parameter combination weight is automatically found, so that effective adjustment of scoliosis spine model parameters is guided.

The invention relates to a semantic segmentation method and device for a remote sensing image, electronic equipment and a storage medium. Comprising the steps that an obtained remote sensing image data set is preprocessed, feature extraction and fusion are conducted on the preprocessed remote sensing image data set through a preset model framework, the fused features of each pixel are obtained, and the preset model framework comprises a Mamba2 encoder and a CNN decoder; and classifying the fused features of each pixel by using a preset classification network, and determining the semantic category of the remote sensing image corresponding to each pixel. Therefore, by combining the advanced Mamba2 model and the convolutional neural network CNN, remote sensing images of various scales and complexities can be effectively processed and segmented, and the problems that in the prior art, when large-size images are processed, the calculation cost is too high, and when complex dependency relationships are processed, the calculation cost is limited by the deficiency of training data and the generalization ability are solved. Therefore, more accurate ground feature classification and semantic information are provided.

The invention relates to the technical field of scene recognition, in particular to a real-time semantic segmentation method and system for a multi-shape pyramid in a traffic scene, and the method comprises the steps: initializing an original to-be-segmented image, and obtaining an initial segmentation feature map; channel splicing is carried out on the initial segmentation feature map and an output feature map obtained after first downsampling is carried out on the original to-be-segmented image, downsampling is carried out after a fused feature map is obtained, detail information is extracted, and a detail information feature map is obtained; down-sampling the detail information feature map and extracting semantic information; the method comprises the following steps: extracting transverse and longitudinal receptive field information of a semantic information feature map, performing expansion convolution to obtain feature maps of different receptive fields, obtaining a multi-scale and multi-shape context information feature map, performing up-sampling, performing point-by-point convolution dimension reduction on a detail information feature map, and performing channel splicing on the two features to obtain a fusion feature map containing semantic and detail information. And adjusting the channel number into a prediction category number, and normalizing to obtain a prediction segmentation result. The method has good segmentation precision and reasoning speed.

The invention discloses a polyp segmentation method and system based on edge detection and multi-scale fusion, and the method comprises the steps: obtaining polyp image data, carrying out the feature extraction of the polyp image data through a backbone network, and obtaining primary features; performing convolution connection on the primary features to obtain polyp global multi-scale feature information; edge information of the polyp global multi-scale feature information is further extracted and fused, and polyp edge local feature information and final polyp edge global feature information are obtained; and performing feature interaction and fusion on the final polyp edge global feature information to obtain interaction features and a final polyp feature map. According to the method, edge detection and feature extraction are combined, and the polyp segmentation precision is improved through feature fusion and enhancement; a space and channel attention mechanism is adopted to strengthen features, details are supplemented during up-sampling, and the recognition and segmentation accuracy is improved. Compared with a baseline model, the method has higher segmentation accuracy and generalization, and has important clinical application value.

The invention provides a remote sensing image photovoltaic identification method based on an SAM model, and relates to the technical field of computer vision, and the method comprises the steps: collecting and preprocessing a multi-resolution photovoltaic segmentation data set, and enabling the data set to be used for adjusting an image encoder; constructing an SAM segmentation network comprising the image encoder, the prompt encoder and the mask decoder, and inputting interactive click information and the data set into the SAM segmentation network; adding a segmentation output layer into the mask decoder to generate a preliminary segmentation result; and feeding the preliminary segmentation result back to an interaction module for interactive adjustment, and decoding again after adjustment to generate a final segmentation result.

The invention relates to the technical field of power equipment monitoring, provides a transformer substation portal frame drainage wire safety intelligent monitoring method based on a point cloud semantic segmentation technology, designs a point cloud semantic segmentation model with higher precision, and introduces an attention mechanism and a feature extraction module into a traditional coding-decoding structure. The extraction of different network layer features is enhanced, and the requirement of a substation scene for segmentation precision is better met. By calculating the Hausdorff distance, the system can monitor the position and the state of the drainage wire in time and give an alarm. Meanwhile, Gaussian filtering and RANSAC methods are adopted, the adaptability of the system to environmental noise and complex terrains is improved, the model is optimized through multi-feature extraction and an attention mechanism, efficient segmentation of complex scenes is achieved, and the generalization ability of the model is enhanced.

The invention discloses a photomask defect identification method, and belongs to the technical field of photomask detection. The method comprises the following steps: S1, acquiring a surface defect image; s2, preprocessing the surface defect image based on the imaging characteristics of the defect; s3, performing filtering algorithm correction on the preprocessed surface defect image; s4, performing colorimetric analysis on the surface defect image corrected by the filtering algorithm to obtain a particle region; s5, on the basis of morphological characteristics of defects, segmenting the particle region to obtain defect islands; and S6, performing morphological analysis on the defective island to obtain a defect type and a defect size. The photomask defect identification method provided by the invention combines morphological and colorimetric analysis methods to automatically process the surface defect image and extract the defect island, improves the accuracy of defect identification, gets rid of the constraint of the difference of different equipment imaging systems, realizes automatic batch processing, and saves the manpower and material investment in the mass production link.

The utility model belongs to the technical field of garbage treatment, and particularly relates to a garbage incineration homogenizing device which comprises an incinerator. The top of the incinerator is fixedly connected with a feeding hopper; the side wall of the feeding hopper is fixedly connected with a second motor; the output end of the second motor is fixedly connected with a second rotating rod; the output end of the second motor penetrates through the side wall of the feeding hopper; the middle part of the second rotating rod is fixedly connected with a blade; the side wall of the incinerator is fixedly connected with a smoke outlet pipe; a first motor is fixedly connected to the side wall of the first motor; the output end of the first motor penetrates through the side wall of the incinerator; the output end of the first motor is fixedly connected with a first rotating rod; a plurality of groups of blades are fixedly connected to the middle part of the first rotating rod; the blades are equidistantly distributed in the middle of the first rotating rod; two fixing rods are fixedly connected to the side wall of the first motor; the garbage incinerator can effectively reduce non-uniform accumulation of garbage in the incinerator, effectively reduces a large amount of harmful gas and particulate matters released by insufficient garbage combustion, and effectively improves the garbage combustion efficiency.

The invention provides an agaricus bisporus picking sequence planning method based on YOLOv8-USD instance segmentation and 3D point cloud, and the method comprises the steps: processing color image information through employing a YOLOv8n-USD instance segmentation algorithm, and generating a segmentation mask corresponding to label information; performing registration on the generated segmentation mask and the corresponding depth image to generate a 3D point cloud; morphological processing is conducted on the color images, the circle center and the radius of each mushroom are calculated, whether picking is needed or not is judged according to the radius, and picking and sorting of the agaricus bisporus are achieved; mapping the circle center and radius information of the mushrooms into the 3D point cloud through the depth image, performing rapid neighbor search by using a KD tree, and segmenting the mushrooms in each mushroom cluster; a global picking sequence planning strategy and a local picking sequence planning strategy are provided, the global picking sequence strategy is used for carrying out classification and priority ranking on different types of agaricus bisporus clusters, and the local picking sequence strategy is used for planning the picking sequence of each mushroom in the mushroom clusters. The picking sequence can be adjusted according to the overlapping degree of agaricus bisporus.

The utility model relates to the technical field of automatic cutting equipment, in particular to novel automatic cutting equipment for building wall materials, which comprises a frame body, Y-axis guide rails are fixedly arranged on two sides of the frame body, a movable inverted U-shaped seat is arranged outside the frame body, and a Y-axis slider is arranged on the inner side of the inverted U-shaped seat. A Y-axis motor is fixedly arranged on the inverted U-shaped seat; z-axis guide rails and tool rests in sliding fit with the Z-axis guide rails are arranged on two side plates of the inverted-U-shaped base respectively, and the tool rests are driven by a Z-axis motor to move along the Z-axis guide rails; a forming tool used for machining a V-shaped groove along a to-be-cut line of a wall material and a cutting tool used for cutting the wall material along the V-shaped groove are arranged on the portions, on the two sides of the inverted-U-shaped base, of the tool rest correspondingly, and the cutting center of the forming tool and the cutting center of the cutting tool are located on the same straight line. The cutting device is convenient and practical, low in operation difficulty and high in automation degree, meets the requirement of a continuous production process, firstly forms and then cuts, and greatly improves the cutting yield.

The invention provides an electrical equipment ultraviolet imaging automatic detection method and system based on a diffusion model, and belongs to the technical field of electric power measurement on-line monitoring, and the method comprises the steps: collecting ultraviolet imaging image data in various normal scenes for to-be-detected electrical equipment, and taking the ultraviolet imaging image data as an original input image x; adding noise to the original input image x to obtain a noise-added image, incorporating the noise-added image into a training set, and training a diffusion model by using the training set; an ultraviolet image of electrical equipment to be detected is collected, the diffusion model obtained through training is input, the total time step length T of denoising and reconstruction of the diffusion model is dynamically set, the ultraviolet image to be detected is denoised and reconstructed, and a reconstructed and denoised image is generated; and carrying out abnormal region segmentation and positioning on the obtained reconstructed and denoised image by using a U-Net network, and identifying an external insulation discharge abnormal region of the electrical equipment. The combination of the U-Net structure and the dynamic diffusion model enables the method to be more efficient and accurate in the detection and segmentation of the abnormal region.

The invention provides a multi-scale local contrast attention infrared small target detection method, and the method comprises the steps: inputting an initial infrared image input feature into a preset backbone network for feature coding, and inputting at least one obtained down-sampling infrared image feature and an integrated infrared image feature into a feature enhancement module, the feature enhancement module comprises a plurality of attention modules which are respectively used for processing each down-sampling infrared image feature and processing integrated infrared image features, and the attention modules are used for performing multi-scale channel attention feature and multi-scale space attention feature fusion on input features to obtain attention features; inputting the attention features of different scales output by each attention module into a feature fusion module to obtain fused infrared image features; and inputting the fused infrared image features into a prediction module to obtain a result image of infrared small target segmentation in the infrared image. According to the method, the infrared small target can be effectively segmented.

The utility model relates to a special sampling tool for a freeze-drying preparation laboratory, which belongs to the technical field of sampling of freeze-drying preparations in laboratories, and adopts the technical scheme that the special sampling tool for the freeze-drying preparation laboratory comprises a rod-shaped spoon handle, a spoon head arranged at one end of the rod-shaped spoon handle, a first spoon tail and a second spoon tail, the first spoon tail and the second spoon tail are installed at the other end of the rod-shaped spoon handle in a replaceable mode, the first spoon tail comprises a cylindrical pressing head, the second spoon tail comprises a connecting curved bar, a longitudinally-arranged blade is arranged at the other end of the connecting curved bar, and due to the arrangement of the cylindrical pressing head, freeze-drying preparations with different loading quantities and different distribution can be effectively ground; and processing the powder into uniform powder.

The invention discloses a novel multi-scale Top-Hat infrared small target detection method. The method comprises the following steps: step S110, obtaining an original image matrix containing an infrared small target; step S120, respectively constructing annular structure elements under different scales; step S130, performing multi-scale Top-Hat transformation on the original image matrix by using the constructed annular structure elements to generate enhanced target matrixes under multiple scales; step S140, calculating a non-zero mean value of each enhanced target matrix, and screening out the enhanced target matrix with the maximum mean value; s150, in the enhanced target matrix with the maximum mean value, extracting a part higher than the mean value as a suspected target area; and S160, calculating each connected domain in the suspected target area, and screening the connected domains according to a preset small target pixel number, thereby realizing target detection. According to the technical scheme, compared with structural elements of a single scale, targets of different sizes can be more effectively captured.

A heart image segmentation method based on a state space model and dual-path multi-scale enhancement relates to the technical field of image segmentation, and a state space attention module integrates an SSM state space model branch, a convolution branch and an attention branch, so that the recognition precision of the model on each structure boundary in a heart image is improved, and the recognition accuracy of the model is improved. And the unique advantage of processing the long-distance dependence while keeping the calculation complexity of linear correlation on the input size of the SSM is utilized, so that the heart image can be segmented more accurately and efficiently. The dual-path multi-scale enhancement module can enhance the recognition capability of the model for different scale structures through a channel aggregation branch and a multi-scale fusion branch, improves the adaptability and robustness of the model, and facilitates the processing of heart structures of various sizes. Therefore, the method can better adapt to the complexity and diversity of heart image segmentation, a segmentation result with higher precision and efficiency is provided, and the accuracy of heart disease diagnosis and treatment planning is remarkably improved.

The invention discloses a grinding wheel abrasive particle image segmentation method based on machine vision, and the method comprises the following steps: inputting a grinding wheel abrasive particle sample image, converting the sample image into a gray-scale image, and carrying out the preprocessing of image enhancement on the gray-scale image; performing operation on the gray-scale map line by line and column by column by two-dimensional Gaussian filtering; calculating the partial derivative of each pixel point in the x-axis direction and the y-axis direction, and obtaining the gradient of each pixel point according to the partial derivative; performing non-maximum suppression according to the gradient; acquiring a global threshold by using an Otsu algorithm, and introducing the global threshold into an adaptive function for setting image segmentation by high and low thresholds of a Canny algorithm to realize edge connection; lag compensation is carried out, and middle edges are accepted and rejected; and morphologically processing irrelevant areas and filling holes. According to the method, edge connection is achieved by setting the adaptive function of image threshold segmentation, the problem of noise interference caused by the grinding wheel base body area and the environment is effectively solved, the boundary of the grinding wheel abrasive particle image contour can be accurately divided, and good flexibility is achieved.

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 invention discloses a cable number identification system based on machine vision, and the system comprises a line character obtaining module which is used for obtaining a line character image of a cable number; the depth feature map extraction module is used for extracting a depth feature map of the cable number from the line character image; the feature vector module and the linear mapping module are used for inputting the high-dimensional feature vector of the line character image into a full-connection layer, and the full-connection layer performs linear mapping on the high-dimensional feature vector to obtain one-dimensional vector representation of the line character image; the classification module is used for enabling the one-dimensional vector representation of the line character image to pass through a softmax function to generate a classification label of each line character image; and the cable number generation module is used for acquiring the classification labels of all the line character images, and generating cable numbers according to the sequential combination of the characters in the images. According to the method, the significant feature map with higher expressive power is generated through the significance score, and finally the cable number is generated, so that the identification of the cable number is complete.

The invention discloses an image segmentation model training method and device of a stained image, equipment and a medium, and relates to the technical field of computers, and the method comprises the steps: obtaining a plurality of pairs of HE stained images and mIHC stained images of the same tissue slice, carrying out the pixel-by-pixel image registration processing of each pair of stained images, and obtaining a plurality of HE stained images and mIHC stained images; obtaining target mask information of the HE dyeing image according to an image registration result, and constructing an image training set based on the HE dyeing image and the corresponding target mask information; training a U-Net network structure by using the image training set so that each convolutional layer of the U-Net network extracts each slice image feature of the HE dyeing image and constructs a feature map, and converting the feature map into initial probability prediction results of different cell class groups; and according to the initial probability prediction result and the target mask information, performing back propagation updating on network parameters of the U-Net network to update the U-Net network, and obtaining the updated U-Net network as a preset image segmentation model.

The long-service-life hub bearing unit with the outer dust cover comprises a flange plate, a steering knuckle and a driving shaft, the steering knuckle is connected to one side of the flange plate, wheel bolts are arranged at the two ends of the flange plate, a locking nut is arranged in the center of the flange plate, the driving shaft is arranged on one side of the locking nut, and the driving shaft is arranged on the other side of the locking nut. A hub bearing is arranged on one side of the driving shaft, roller paths are arranged on the two sides of the steering knuckle, rollers are arranged in the roller paths, a clamping spring is connected to one side of each roller path, a static magnetic ring is arranged on one side of the steering knuckle, and a dustproof cover is connected to one side of each static magnetic ring. A sealing ring is arranged at the joint of the dustproof cover and the static magnetic ring, the dustproof cover is fixed through magnetic force, meanwhile, the operation stability and safety of the hub unit are improved through the sealing ring arranged at the joint, internal air pressure balance can be kept through the design of the sawtooth-shaped surface of the dustproof cover and vent holes, invasion of dust and pollutants can be effectively prevented, and the service life of the hub unit is prolonged. The service life is prolonged.

The invention discloses a highway construction quality detection method. The method comprises the following steps: obtaining a road image subjected to image preprocessing, and determining each initial recess sub-region in the road image; according to the gray value of each pixel point in each concave sub-region and the number of the pixel points, determining a morphological characteristic value of the concave sub-region; according to the morphological characteristic value, judging whether a first concave sub-region exists or not; if yes, determining a spatial distribution characteristic value of the first concave sub-region according to the form characteristic value and the position of each pixel point; according to the spatial distribution characteristic value, judging whether a second concave sub-region exists or not; if yes, the number of pixel points of the second sunken sub-area is obtained, and whether sunken parts are caused by overweight of the vehicle or not is judged; and removing the second concave sub-region to form a third concave sub-region, and analyzing the distribution characteristics of the third concave sub-region. The method can effectively identify and eliminate the road surface sunken area caused by overweight of the vehicle, and improves the accuracy and practicability of the detection result.

The invention relates to the field of path planning, in particular to a cross-floor memory path reconstruction method, device and equipment and a vehicle. According to the embodiment of the invention, the original path points are firstly densified, so that the path is finer, the actual road condition can be better reflected, and the inaccuracy caused by the sparse path is avoided. And secondly, the point cloud sets are matched and divided in combination with the lane line point cloud data, so that the lane line information of the vehicle at different positions can be more accurately determined, a more accurate data basis is provided for path reconstruction, and meanwhile, deviation caused by sensor errors or obstacle avoidance and the like can be corrected. And finally, path reconstruction is performed based on lane line segmentation, and a non-optimal part in an original memory path can be removed, so that the driving path quality and the parking effect during actual parking are improved, and the problems of poor experience, low efficiency, parking failure and the like caused by the non-optimal memory path are avoided.

The invention discloses a workpiece image motion distortion correction and accurate segmentation method based on a directed potential energy field. The method comprises the following steps: 1) obtaining a target original image to be segmented; 2) analyzing the periodic change of the distorted image, and establishing a fluctuation model; 3) image adaptive preprocessing: removing an image background and reducing interference; 4) carrying out self-adaptive initialization on geometric contour vertexes based on directed potential energy field integrated clustering; according to the method, the real-time motion distortion correction, the self-adaptive correction and the dynamic segmentation of the image obtained by the linear array camera are realized, so that the problems of motion distortion, self-adaptive correction and dynamic cutting of the metal plate in image processing are solved.

The present invention belongs to the field of MPI image segmentation, and specifically relates to an MPI image segmentation method, system and device based on local maximum point threshold expansion, and aims to solve the problem. The present invention includes: based on the threshold segmentation method, distinguishing the foreground signal and the background signal of the MPI image to be segmented; clustering the foreground MPI image signal by the K-means clustering method to obtain clusters of different categories; selecting the neighborhood pixels of the set area of ​​each cluster center, and obtaining the location of the local maximum signal intensity point of each cluster; based on the location of the local maximum signal intensity point of each cluster, combined with the signal intensity corresponding to each location, performing a threshold expansion operation to obtain the expansion area of ​​each cluster; merging the expansion areas of each cluster to obtain the segmentation result of the MPI image to be segmented. The present invention realizes a robust and accurate MPI image segmentation based on the threshold expansion method of the local maximum point, and reduces the influence of the selective bias existing in the MPI image segmentation.