482 results about "Edge extraction" patented technology

The invention discloses a weldment welding seam automatic detection method and device based on machine vision, and relates to the technical field of machine vision intelligent detection. The weldment welding seam automatic detection method and device based on machine vision comprises the steps that S1, surface images and forming feature data of a weldment are collected and preprocessed to construct a standardized image feature data set; s2, the boundary clearness of the weld joint is evaluated by combining the edge strength and the contour coherence, and the main contour extraction range is dynamically adjusted; s3, analyzing abnormal focusing characteristics of the candidate area, and adjusting a defect labeling range and a detection priority; and S4, integrating the boundary definition and the abnormal focusing features, analyzing the structure abnormality, and dynamically controlling and verifying a resource allocation strategy. The problems that in the weldment detection process, obvious light reflection and texture blurring phenomena exist in a heat affected area at a weld joint, a traditional image enhancement and edge extraction algorithm is difficult to stably recognize microdefects, and the credibility of a detection result is reduced are solved.

The invention provides a remote sensing coastline automatic extraction method and system based on residual space pyramid segmentation and two-dimensional attention, and relates to the technical field of space analysis. The method comprises the steps of high-resolution remote sensing image acquisition and preprocessing, sea-land segmentation network reasoning, probability graph thresholding and edge extraction and vectorization processing. According to the method, the segmentation precision is improved through multi-scale feature aggregation and attention enhancement, coastline vector data with geographic coordinates are generated in combination with edge detection and topological repair, and the method is suitable for spatial analysis and coastline monitoring.

The invention discloses a circuit breaker image edge detection method fusing spatial constraint fuzzy clustering and lightweight network optimization, and the method carries out the local adaptive threshold calculation through combining spatial constraint FCM and Otsu algorithms, and optimizes the edge detection process of a Canny operator. According to the method, fuzzy classification is carried out on a circuit breaker image by adopting spatial constraint FCM to obtain a strong marginal probability graph; the circuit breaker image is subjected to block processing through local adaptive threshold calculation, a global threshold is generated through integration, and then the global threshold is input into a Canny operator for accurate edge extraction. In order to further improve the detection effect, a lightweight neural network PiDiNet is used to correct a Canny output image. According to the method, the edge detection precision of the circuit breaker image can be effectively improved, and the method is suitable for edge extraction tasks in high-noise and complex background environments.

The invention discloses a color printed matter defect automatic detection system and method based on machine vision, and relates to the technical field of computer vision, and the method comprises the steps: converting a color image into a gray level image, generating a boundary enhanced gray level image through a Canny edge detection algorithm, calculating the offset of edge pixels through sub-pixel recovery, and obtaining a color printed matter defect detection result. Extracting an edge pixel proportion based on the sub-region division table, performing amplification tracking on edge dense sub-regions, and analyzing RGB color deviation; through a multi-scale Gaussian pyramid decomposition and edge refinement technology, analyzing multi-scale features of the boundary enhanced grayscale image and optimizing edge precision; and classifying edge defects and color defects of each sub-region by utilizing a defect threshold value, calculating a comprehensive risk value, increasing a risk level according to a boundary condition, and generating a detection report. According to the method, the Canny edge detection algorithm is combined with sub-pixel recovery, so that the edge extraction precision and the positioning capability of color printed matter defect detection are improved.

The invention discloses an online AOI detection system based on an industrial intelligent sensor, and the system comprises an image collection and preprocessing module which is used for collecting and preprocessing image data of an industrial product; the mask auto-encoder modeling module is used for constructing a mask auto-encoder model; the structure parameter optimization module is used for optimizing the mask auto-encoder model; the image reconstruction and difference extraction module is used for generating a reconstructed image, extracting an image difference region and determining a candidate defect region; the defect identification module is used for carrying out edge extraction and aggregation analysis, identifying a final defect area and acquiring spatial position information; the defect classification and labeling module is used for extracting defect area features and generating corresponding classification labels and grade labels; and the control response module is used for generating a control instruction and issuing the control instruction to the production line control device. According to the method, the high-precision automatic identification and real-time classification processing of the surface defects of the industrial product are realized by fusing the mask auto-encoder and the Tiancattle herd optimization algorithm.

The invention discloses a geological map-oriented breakpoint repairing and closed curve reconstruction method and system and a medium. The method comprises the following steps of: preprocessing an original geological map to obtain a grayscale image; carrying out binarization and morphological closed operation processing to obtain a grayscale image of the edge of the smooth contour curve; performing pixel skeletonization to obtain a pixel-level skeleton diagram; performing secondary edge extraction, connected domain marking and endpoint statistics on the pixel-level skeleton diagram, and screening out a non-closed curve and an endpoint set; carrying out nearest neighbor retrieval by adopting KD-Tree to obtain a candidate pairing set of each end point; performing priority pairing in combination with the spatial distance and the direction smoothness; and generating a transition line segment according to a pairing result to supplement the fracture part of the non-closed curve. According to the method, non-closed breakpoints generated by scanning or drawing errors in the geological map can be automatically restored, and a closed curve with topological integrity and a smooth boundary is generated.

The invention relates to the technical field of measurement and automation, in particular to a non-contact wire diameter identification method and system, and the method comprises the steps: enabling a CCD camera and a laser range finder to aim at a remote cable identification target, and obtaining a cable image and the spatial position information of the cable image; inputting the cable image into a trained cable type identification model for feature extraction and classification processing to obtain a cable type, and determining an edge detection preset parameter corresponding to the cable type; edge extraction and contour segmentation are carried out based on the cable image and edge detection preset parameters, and the pixel length of the cable diameter is obtained according to geometric characteristic estimation of the segmented contour; based on the spatial position information of the cable and the pixel length of the diameter of the cable, performing scale conversion from an image to a physical world to obtain the actual length of the diameter of the cable; and based on the actual length of the cable diameter and the cable type, searching a matching item in a preset cable model database to obtain a cable model identification result.

The invention discloses a fan blade clearance intelligent monitoring method based on multi-modal fusion. A laser radar and a visual sensor are adopted to collect point cloud data and image data of the fan blade; environment parameters are collected, validity judgment is conducted on the point cloud data and the image data of the fan blade according to the environment parameters, and radar monitoring data and visual monitoring data are obtained; modeling the fan blade according to the radar monitoring data to obtain a three-dimensional model of the fan blade, and further obtaining a radar measurement value of the blade clearance according to the three-dimensional model of the fan blade; performing image segmentation, edge extraction and depth estimation on the visual monitoring data in sequence to obtain a visual measurement value of the blade clearance; and the radar measurement value and the vision measurement value of the blade clearance are combined, and a multi-modal fusion algorithm is adopted for processing to obtain the fusion clearance distance of the fan blade. The accuracy and real-time performance of fan blade clearance monitoring are remarkably improved, blade collision accidents can be effectively prevented, and reliable guarantee is provided for safe operation of a wind turbine generator.

A video generation framework that is controllable, unsupervised and based on events (CUBE) includes an event camera, which captures changes in light intensity at each pixel of a scene asynchronously and generates event camera data. A text-to-image diffusion model that is conditioned on textual descriptions integrates the event camera data to control video synthesis. Further, an edge extraction module translates event data into a format usable by the text-to-image diffusion model, whereby the diffusion model synthesizes detailed and contextually accurate videos based on textual prompts. Further, an improved system (CUBE Plus) includes a content frame identification module which selectively identifies and uses only the most information-rich event segments of the event camera data to drive cross-frame attention, and an event driven attention mechanism that allows the framework to focus on event-dense moments.

The invention relates to the technical field of machine vision and product defect visual prediction, in particular to a product defect visual prediction method and device based on X rays. According to the method, edge extraction and entropy-based analysis are utilized to enhance defect positioning, meanwhile, the challenges of image noise and non-uniform intensity change are solved, and in order to improve robustness, a self-adaptive threshold strategy combined with DBSCAN clustering is adopted to distinguish defects and noise. Therefore, the problems that in the prior art, the ratio of product defects is low, the visual prediction efficiency of edge blurring is low, and the accuracy is not high are solved. Compared with the prior art, noise reduction, edge detection and crack identification are enhanced. By integrating X-ray image processing, the method improves accuracy, reduces false positive and improves detection efficiency.

The invention discloses a defect edge extraction method and device based on a terahertz image, terminal equipment and a storage medium, and belongs to the technical field of defect detection, and the method comprises the steps: obtaining an original terahertz image; performing enhancement processing on the original terahertz image to obtain an enhanced terahertz image; segmenting the enhanced terahertz image into a plurality of target areas through a mean shift algorithm; for each target area, through a maximum between-class variance method, calculating segmentation threshold values of a foreground and a background in the target area; according to the segmentation threshold, segmenting a foreground region from the target region as a defect region; and inputting the defect area into an active contour model, so that the active contour model generates a defect edge curve according to the defect area. The problems that in the prior art, a defect area needs to be specified manually in advance, and the defect edge extraction error is large can be solved.

The invention relates to the technical field of image data processing, in particular to an automobile part quality visual inspection method and system, and the method comprises the steps: obtaining a plurality of original images of a target object; preprocessing each original image to obtain a plurality of target images; performing edge extraction on each target image to obtain all edges of each target image; performing crack growth characteristic screening on each edge to confirm a crack corresponding to each target image, wherein the crack growth characteristics comprise curvature change and gradient change; respectively evaluating each crack to obtain a feature score of each crack; color marking is conducted on the image of the position where each crack is located according to the feature score of the crack, a marked image containing a color mark is obtained, accurate recognition is conducted by comprehensively considering the curvature and gradient change characteristics of crack growth, the recognition efficiency is improved through feature scoring and marking, quality evaluation deviation caused by edge misjudgment is reduced, and the quality evaluation accuracy is improved. And the product quality and safety are ensured.

The invention relates to a pavement disease detection method, system and device based on a hybrid architecture, and a storage medium. The method comprises the following steps: obtaining pavement image data; the method comprises the steps that data are input into a re-parameterized feature extraction network, the network is based on an HGNetV2 architecture, a RCHGBlock module is formed by embedding a RepConv structure into HGBlock, and a plurality of modules are cascaded and stacked to construct a four-stage progressive feature pyramid structure; based on a high-level semantic layer of a feature pyramid, integrating a space edge perception enhanced attention mechanism, enhancing disease edge features through a dual-path complementary processing framework of an edge extraction path and a standard convolution path, and performing multi-scale feature fusion by using an improved C3K2-SEAM module as a feature fusion unit; and performing end-to-end disease detection based on the fusion features, and outputting disease types, positions and confidence information. Compared with the prior art, the method has the advantage that the disease detection precision and robustness in a complex scene are remarkably improved.

The invention discloses a machine vision-based automatic identification and calculation method for pore change of a rock section under an SEM (scanning electron microscope), and the method comprises the steps: obtaining original image data of the rock section through a scanning electron microscope, transmitting the original image data to a machine vision SEM pore intelligent quantification cloud platform, and dividing the original image data into a plurality of image sub-regions which are not overlapped with each other; calling a special threshold inversion model to analyze the gray information of each sub-region, and determining a dual-phase pore segmentation threshold; carrying out edge extraction on the sub-regions after threshold processing by adopting a pore edge gradient enhancement algorithm; performing topological structure analysis on the intermediate image data by using a biphase pore topological inversion model, constructing and optimizing a pore space topological network, and eliminating false pore structure data; and according to the optimized pore space topology network, integrating calculation results of all the sub-regions to obtain a pore change identification calculation result of the whole rock section. The method improves the pore identification precision and processing efficiency, and adapts to different types of rock samples.

The invention discloses a road traffic detection control method based on a vehicle-mounted camera module, and belongs to the technical field of traffic detection, and the method specifically comprises the steps: collecting road image data in real time through the vehicle-mounted camera module; preprocessing the acquired road image data, and extracting a road area; performing edge extraction on the preprocessed road image data by using an edge detection algorithm to obtain road edge information; according to the road edge information, detecting straight-line segments in the road through Hough transform, and calculating an included angle between adjacent straight-line segments to determine the line type of the road; calculating the distance between the vehicle and the road boundary and the offset of the vehicle relative to the road center line according to the line type of the road in combination with the speed and position information of the vehicle; according to the distance between the vehicle and the road boundary and the offset of the vehicle relative to the road center line, the driving direction and speed of the vehicle are adjusted through a preset control strategy; according to the invention, real-time safety detection of the vehicle in the driving state is realized.

The invention provides a remote sensing image building extraction method fusing semantic and edge features, and relates to the technical field of remote sensing image processing. On the basis of a semantic segmentation network HRNet, PSABlock is designed to reconstruct a backbone network, the backbone network serves as an image semantic information extraction branch, input features are decomposed into spatial components and channel components, spatial relations and channel dependency are captured respectively, and effective fusion of global and local features is achieved. According to the method, a module fusing multilayer edge features is designed as an edge feature extraction branch, an edge feature map of a building is obtained by generating an edge prediction result and performing weighted fusion, and an edge extraction effect is optimized through explicit edge loss. In addition, the invention provides a'main body-edge-fusion 'joint loss function. Compared with a semantic segmentation model of single feature extraction, the method has the advantages that the precision of the building extracted from the test data set is higher, the edge is more complete, and the requirements of practical application are met.

The invention relates to the technical field of medical image processing, in particular to an ultrasonic image segmentation method based on edge guidance, and the method comprises the steps: inputting an ultrasonic image into an edge extraction branch and an image encoder; outputting an edge mask image corresponding to the image through the edge extraction branch; performing noise suppression and topological correction on the edge mask image to generate a closed edge constrained by an anatomical structure; generating prompt box information based on the closed edge, and inputting the prompt box information to a prompt encoder; fusing the image prompt features output by the prompt encoder, the image features extracted by the image encoder and the edge mask information; the fusion features are input into a decoding module, a final segmentation result is obtained, the synergistic effect of edge information and visual prompt is fully utilized, the perception ability of the model for the anatomical structure in the ultrasonic image is improved, and the accuracy and robustness of segmentation are remarkably improved. Therefore, the problems of fuzzy boundary, inaccurate prompt, weak structure identification capability and the like in related technologies are solved.

The invention discloses a low-light image enhancement method and system based on edge extraction and feature fusion, and aims to improve the image quality under a low-light condition, and the method comprises the steps: differential convolution edge extraction, feature fusion and image enhancement, effective extraction of image edge information through a differential convolution kernel, and combination of global and local features through a feature fusion module. The image enhancement module utilizes a deep learning network to improve image brightness and suppress noise, the method is suitable for embedded equipment with a high-performance computing environment and limited resources, the performance is excellent on a plurality of data sets through experimental verification, the image definition and details are remarkably improved, and the method is suitable for the fields of security monitoring, automatic driving and the like and has a wide application prospect.

The invention discloses a semantic segmentation method based on phase contour constraint, and belongs to the technical field of computer vision. The method comprises the following steps of: acquiring a two-dimensional texture image and a deformed stripe image of a scene to be detected in the same view field by using a calibrated camera-projector system; resolving the deformed fringe image through a multi-step phase shift and Gray code technology to obtain absolute phase distribution, and calculating a gradient field; according to a preset sudden change condition, extracting a physical edge contour reflecting the depth change of the object from the gradient field; and mapping the physical edge contour into the two-dimensional texture image based on a pixel coordinate corresponding relation to accurately define a semantic object region, endowing a corresponding category label, and finally generating a semantic label mask image. According to the method, the uncertainty of the visual texture edge is corrected by utilizing the physical phase information, the problem of edge extraction in a complex illumination scene is effectively solved, automatic generation of semantic segmentation data is realized, and the training precision and robustness of a semantic segmentation model are improved.

The invention discloses a quartz surface defect detection method based on visual identification, and the method comprises the following steps: S1, carrying out the image collection of a quartz surface through a high-resolution camera, and carrying out the preprocessing of the collected image; s2, carrying out edge extraction on the image by adopting a Canny edge detection algorithm; s3, generating more defect images by using a hybrid generation model based on a variational auto-encoder and an energy guide mechanism; s4, combining the generated defect image and the real image, and performing defect classification through an image segmentation network in combination with a self-attention mechanism; s5, jointly training the improved variational auto-encoder and the image segmentation network; and S6, feeding back the category of the defect and the edge information of the defect to a production line control system in real time. According to the method, a high-resolution image acquisition technology, a variational auto-encoder and an image segmentation network are combined, and automatic detection and accurate classification of quartz surface defects are realized through fusion of deep learning and a traditional image processing method.

The invention relates to a fine-grained target real-time image segmentation method and system based on a dynamic state modeling network, and belongs to the technical field of intelligent image processing. The method comprises the following steps: extracting multi-scale detail features by using a lightweight backbone network; through a dual-scale two-dimensional selective scanning module, the features are divided into a thin branch and a thick branch, and local scanning and global scanning are executed respectively; a dynamic cross-scale feature selection and aggregation module is adopted, redundancy is suppressed through reweighting and statistical filtering, and key target responses are highlighted; at a decoding end, local details and global semantics are fused through jump connection and an edge extractor; and finally, introducing a form-guided pseudo label hierarchical supervision strategy, and improving the structure learning ability of the model by using a coarse-to-fine morphological prior. According to the method, the segmentation precision, the boundary integrity and the tiny target recall rate of the fine-grained target under the scenes of ore separation, industrial defect detection, pavement crack recognition and the like are remarkably improved.

The invention relates to the technical field of video target tracking, in particular to a camera automatic target tracking method and system, and the method comprises the steps: obtaining video image data of a target region through a camera, and carrying out the edge extraction and region division; according to the edge distribution condition in each region in the single-frame video image and the gray level distribution difference in the region, obtaining an illumination concentration degree; suspected shadow areas in the single-frame video image are extracted, and target compensation coefficients of all the areas are obtained; the target compensation coefficient of each region is corrected in combination with the difference degree between the gray level fluctuation conditions of each region in the adjacent frames of video images, and the target existence coefficient of each region in the single frame of video image is obtained in combination with the illumination concentration degree of each region, so that the region containing the target in the single frame of image is judged, and target tracking is further realized. The target tracking precision of the camera can be improved.

The invention discloses a fan blade defect detection method based on direction constraint edge extraction and texture discrimination, and the method comprises the steps: image enhancement processing: carrying out the brightness and contrast enhancement processing of an input original fan blade image, and obtaining an enhanced image; fan blade positioning: analyzing the main direction of the image by using a structure tensor, and positioning a fan blade area in combination with a direction constraint edge extraction method; and defect detection and extraction: performing morphological operation, connected domain analysis and texture fusion discrimination on the suspected region in the positioning result, and extracting a final defect region. The method is suitable for automatic identification of surface defects such as cracks and corrosion in an offshore wind power inspection image; the method has the advantages of high robustness, low false detection rate and adaptability to complex sea and sky backgrounds.

The invention discloses a road width automatic calculation and quality verification method for road network data, relates to the technical field of road network data processing, and is used for solving the problem of low multi-source data fusion efficiency in an existing method. The method comprises the following steps: acquiring multi-source road network data, pre-standardizing the multi-source road network data, unifying coordinate system mapping and labeling metadata to generate an original data set with a label; performing noise adaptive filtering and shielding detection, identifying an interference region by using U-Net semantic segmentation, and outputting a purified data set by applying GAN local completion; constructing a multi-modal network of an attention mechanism to calculate a dynamic fusion weight, and performing parallel integration to form a fusion feature vector; a width parameter is calculated through Canny optimization edge extraction and DBSCAN clustering, and a topological structure is embedded; checking a result by using a multi-scale consistency index, and performing backtracking optimization if the result exceeds a threshold value; and outputting a result and iteratively updating parameters to form a closed-loop mechanism. The method improves fusion efficiency and calculation precision, is suitable for traffic navigation and urban planning, and supports real-time dynamic application.

The invention belongs to the technical field of computer vision, and discloses a semantic edge guided three-dimensional scene reconstruction method based on an SDF diffusion model. The objective of the invention is to solve the problems of high bandwidth occupation of input data, poor structural consistency and insufficient reconstruction quality in traditional three-dimensional reconstruction. A point fine segmentation and gradient semantic edge extraction method is provided, edge information with consistent semantics and rich geometric details is extracted, and lightweight scene representation retaining key geometric information is realized. On the basis, a three-dimensional reconstruction framework of staged diffusion is proposed: firstly, a global occupancy field is generated to determine an object space range; then executing high-resolution SDF diffusion in the local area to recover a fine geometric surface; and finally, carrying out space alignment and fusion to ensure the consistency of a reconstruction result under a global coordinate system. According to the method, the geometric accuracy, semantic consistency and resource overhead of reconstruction are effectively balanced, and a novel solution is provided for high-quality three-dimensional scene reconstruction under a low-bandwidth condition.

The invention discloses a network optimization InSAR large gradient deformation phase unwrapping method based on edge detection, and the method specifically comprises the steps: 1, carrying out the differential interference processing of an original image, and obtaining a differential interference pattern; 2, Gaussian filtering is carried out on the differential interferogram, and the gradient magnitude and the gradient direction of each pixel in the image after Gaussian filtering are calculated; a gradient magnitude image is obtained; step 3, performing non-maximum suppression on the gradient magnitude image, and completing preliminary extraction of edges; 4, setting double thresholds to further screen edge points, and finally completing edge extraction; 5, generating a coherence graph based on the differential interferogram, extracting coherence points, and generating an unwrapping network based on the extracted coherence points and the edges extracted in the step 4; and step 6, phase unwrapping is carried out based on the unwrapping network. The method has certain universality for monitoring application scenes with large-gradient characteristic deformation, such as landslide, mining and the like, and can provide important technical support for geological disaster prevention and control.

The invention discloses a multi-mode human body recognition system and method based on photoelectric metasurface and radar fusion, and belongs to the technical field of optoelectronics and radar perception. According to the system, optical edge enhancement is realized through a phase programmable metasurface, a continuous wave radar and a frequency modulation continuous wave laser radar are combined to obtain a micro-Doppler spectrum and a three-dimensional point cloud, and a near-infrared reflection spectrum is collected at the same time. According to the algorithm, a cross-modal Transform-GAT architecture is adopted to fuse multi-source features, edge extraction is optimized through a self-adaptive kernel function, and low-power-consumption edge reasoning is achieved through a Mach-Zehnder interference device or a ReRAM chip. Experiments show that the recognition accuracy of the method in complex scenes such as multi-target shielding and low light is remarkably superior to that of a single-mode scheme, the method has the edge deployment capability with delay smaller than or equal to 25 ms and power consumption smaller than or equal to 4 W, and an efficient solution is provided for multi-scene human body dynamic recognition and health monitoring.

The invention relates to a fire area inversion method based on airborne dual-spectrum detection and depth estimation, and belongs to the field of unmanned aerial vehicle detection. The method comprises the following steps: acquiring a multi-dimensional data set of a dual-spectrum image, a temperature image, an unmanned aerial vehicle attitude and the like; constructing a multi-modal space collaborative perception segmentation network, combining temperature change characteristics with temperature space distribution characteristics of flames to generate temperature region distribution characteristics, and coupling absolute temperature and pixel information to enhance flame weak edge extraction; designing a temperature-guided space structure loss function TSSLoss, and combining gradient change consistency constraint and temperature weight constraint on a segmentation loss function for network training; and according to the unmanned aerial vehicle pose, the target depth and the fire area segmentation pixel area, an early fire area is derived in combination with an airspace transmission inversion formula, and an actual fire area is calculated. According to the method, multi-source information is effectively combined for physical constraint, and more accurate fire detection segmentation and fire area calculation can be realized.

The invention relates to a reconstructed image anti-aliasing processing method based on an adaptive sampling rate, belongs to the technical field of CT (Computed Tomography) reconstructed image processing, and solves the problem that the existing anti-aliasing processing method for a CT reconstructed image is poor in effect. The anti-aliasing processing method for the reconstructed image comprises the following steps: preprocessing the reconstructed image to obtain a to-be-processed image; performing edge extraction on the to-be-processed image based on a preset gradient threshold and a preset Sobel operator, and segmenting the to-be-processed image into a to-be-replaced image region and a reserved image region; determining an adaptive sampling rate based on the horizontal direction edge pixel connection number and the vertical direction edge pixel connection number of all sawtooth vertex pixels in the to-be-processed image; determining a pixel value of each pixel in the to-be-replaced image region based on the adaptive sampling rate and the to-be-processed image, and obtaining a replaced image region; and combining the replaced image area and the reserved image area to obtain an image after anti-aliasing processing. And the anti-aliasing effect of the CT reconstructed image is improved.

A ferrite structure edge extraction method based on a microstructural phase diagram of steel, comprising: performing edge detection on a metallographic image of a low-carbon steel cold rolled sheet; closing grain boundaries; removing redundant boundaries; detecting grain parts of unconnected parts and grain unclosed parts in a labeled image on the basis of a network topology structural diagram, performing grain boundary edge determination, and when the vertex of each unclosed part grain remains unchanged, continuously traversing different grain boundary arcs connected to the unclosed part grain so as to determine whether the unclosed part grain satisfies the characteristics of a completely closed grain, thereby achieving grain reconstruction; and then performing reconstruction detection to finally complete extraction and reconstruction of ferrite structure edges. Thus, ferrite grains of a microstructural phase diagram of steel are accurately reconstructed, two grains connected by a narrow area are effectively segmented, accurate ferrite grains of a low-carbon steel cold rolled sheet can be extracted, and ferrite grain size grading of the low-carbon steel cold rolled sheet is assisted on the basis of the extracted accurate ferrite grains of the low-carbon steel cold rolled sheet.