51 results about "Thinning algorithm" patented technology

The invention discloses a mobile robot path planning method and system, and belongs to the technical field of robots, the method comprises global path planning and local path planning; wherein the global path planning comprises the steps of enabling a path to be constantly away from an obstacle by improving an A * algorithm and introducing a bilateral obstacle detection algorithm and an offset obstacle avoidance method so as to avoid collision with the obstacle during turning, and performing optimization and smoothing processing on the path by combining a thinning algorithm and a middle point outward expansion method so as to generate a global optimal path; the local path planning comprises the steps of improving a TEB algorithm, introducing a sensing window and an evaluation function, proposing an acceleration and jerk double-constraint dynamic obstacle avoidance strategy, and ensuring stable movement of the robot in the obstacle avoidance process. According to the scheme, by combining the advantages of global path planning and local path planning, the path planning capability and the transportation efficiency of the robot in a complex and changeable environment are improved.

The invention relates to a pavement crack image skeleton extraction method based on an improved Zhang-Suen refinement algorithm. The method comprises the following steps: S1, carrying out binarization processing on a pavement crack image; s2, carrying out boundary expansion on the crack binary image; s3, carrying out run-length coding on the crack binary image in the horizontal direction, and carrying out compression storage on continuous non-zero pixel segments in each row; s4, using an improved Zhang-Suen refinement algorithm to process the crack binary image processed in the step S3; s5, removing boundary extension; s6, performing simplification processing on the preliminarily refined crack skeleton image by using four predefined 3 * 3 L-shaped structure templates; s7, removing burrs in the image; and S8, outputting a crack skeleton image. The technical problems that according to an existing skeleton extraction method, a topological structure is prone to being damaged, the calculation efficiency is low, the extraction result cannot be completely a skeleton with the single pixel width, and burrs exist in the skeleton extraction result are solved.

The invention relates to an engine stator part small target crack length measurement method based on a deep convolutional neural network, and the method comprises the steps: carrying out the labeling of an initial fatigue crack image obtained through carrying out a fatigue crack propagation experiment, converting the initial fatigue crack image into a PNG label, and carrying out the expansion of the original image and the label through a data intensifier, constructing a VOC (volatile organic compound) format small target crack data set; with the help of the strong feature extraction capability and fine distinguishing capability of a deep convolutional neural network, a semantic segmentation network for small target crack segmentation is constructed based on a full convolutional network, a U-shaped architecture and jump connection, and training is performed to obtain a trained model weight for high-quality crack segmentation; and based on the segmented image, extracting a crack skeleton with a pixel width by adopting preprocessing means such as expansion and a refinement algorithm, counting the number of pixel points in different arrangement modes based on the crack skeleton, calculating to obtain a crack pixel length, and performing scale conversion to obtain an actual crack length. According to the invention, high-precision measurement of the crack length of a large batch of small targets is realized.

Described herein is NLS using a bounded linear initial search space and a fixed grid with pre-calculated variables. Specifically, first and second signals with unknown first and second elevation angles, respectively, are received that have been reflected by an object, with the second signal also having been reflected off the ground. A line of second angles is then established as a function of first angles, a sensor height, and a range to the object. The first angles being bound by a function of the sensor height and the range and a function of the sensor height, the range, and the maximum height. A search algorithm is then used to search for an initial elevation angle pair along the line. The initial elevation angle pair may then be fed into a refinement algorithm (e.g., non-linear least squares) to determine the elevation angles associated with the first and second signals.

The present application belongs to the technical field of construction engineering equipment, and specifically relates to a ring area steel mesh hole positioning test device and method based on 3D vision. Mainly includes the following contents: the 3D sensor is driven to scan the steel mesh through a rotating mechanism, and a steel mesh point cloud image is obtained; the steel mesh image is subjected to binaryzation processing, facilitating subsequent image analysis; the steel mesh image is skeletonized by using a 3D refinement algorithm, and a steel mesh skeleton is extracted; the neighborhood integral algorithm is used to find a steel intersection point set; the least square sphere fitting algorithm is used to fit and position a steel binding point; and the Varignon parallelogram method is used to position a steel mesh center hole. The method reduces the consumption of manpower, improves the construction efficiency, and helps to realize the automation and intelligentization of steel mesh area concrete construction.

The invention provides a fingerprint image processing method which can be applied to the technical field of artificial intelligence. The fingerprint image processing method comprises the following steps: carrying out segmentation processing on an input fingerprint image by adopting a Gaussian mixture model based on an expectation maximization clustering algorithm to obtain a segmented fingerprint image; enhancing the segmented fingerprint image by using a direction and frequency adaptive Gabor filter to obtain an enhanced fingerprint image, the direction of the Gabor filter being determined by a fingerprint local ridge direction, and the frequency being determined by a local ridge frequency; performing binarization processing on the enhanced fingerprint image by adopting a local threshold method to obtain a binarized fingerprint image; and processing the binarized fingerprint image by adopting a sequential refinement algorithm, deleting boundary pixels until a skeleton with a single-pixel width is reserved, and obtaining a processed fingerprint image. The invention further provides a fingerprint image processing device and equipment, a storage medium and a program product.

The application is suitable for the technical field of image processing, and provides a battery glue line skeleton extraction method based on an improved ZS thinning algorithm, which comprises the following steps: preprocessing a collected battery glue line image to obtain a binary image; performing improved ZS thinning on the binary image, dividing the image into odd and even subdomains for batch processing, and introducing a constraint condition to iteratively delete redundant pixels to obtain a preliminary skeleton; using a T-shaped 4-neighborhood deletion template to remove T-shaped redundant pixels and realize complete single-pixelization; through 8-chain code tracking, identifying and removing burr branches caused by glue overflow and uneven glue coating, while protecting nodes to obtain a final glue line skeleton. The application solves the problems of skeleton redundancy, over-thinning and burr in existing thinning algorithms, realizes high-precision, single-pixel and burr-free glue line skeleton extraction, and can meet the online detection requirements of battery assembly lines, and has good industrial applicability.

The invention provides a sub-pixel interpolation method and device for binocular stereoscopic vision. The method comprises the following steps of: calculating an aggregation cost value of each pixel in a left eye image and a right eye image which are acquired at the same time in a preset parallax range; determining the optimal aggregation cost value of each pixel; determining an integer-level parallax value of each pixel in the left-eye image and the right-eye image according to the optimal aggregation cost value of each pixel; performing sub-pixel interpolation according to the optimal aggregation cost value of each pixel to obtain a decimal-level parallax value of each pixel; and determining a final parallax value of each pixel according to the integer-level parallax value and the decimal-level parallax value of each pixel. According to the method, the sub-pixel interpolation is carried out on the pixels, so that the distribution condition of the SGM algorithm in a spatial plane under a binocular visual angle can be remarkably refined, the finally obtained parallax value is enabled to present a smoother and more accurate characteristic, and the performance of a binocular vision system in an application scene is improved.

The application discloses a segmental straightening method and terminal equipment based on computer vision. The method comprises the following steps: installing a camera sensor at a position between a rack and a straightening roll in a horizontal direction; performing joint calibration of all cameras to facilitate determination of relative positions of two-dimensional information; recording two-dimensional image information of a workpiece to be straightened and the straightening roll by the camera, and simultaneously performing model training preparation; performing semantic segmentation processing on the workpiece by an image recognition hub, and obtaining a bending curve by a distance transformation method or a skeleton thinning algorithm; according to bending information of the workpiece, performing discrete curvature calculation on the workpiece by a point taking method or a spline curve fitting method, and then performing segmental processing on the workpiece according to a point detection method and a curvature threshold method; and according to material thickness, yield strength and other parameters, determining optimal roll spacing and a press-down amount of a current segment, and performing real-time full-automatic closed-loop control on a steel straightening machine by real-time control on a press-down system and a variable roll spacing system of the straightening machine.

The application provides a scoliosis detection device, system and computer readable storage medium, the scoliosis detection device comprises a processor, the processor is configured to implement the following steps: acquiring a full body image of a user performing a preset action by using an image acquisition device, the preset action comprising a standing action; adopting a thinning algorithm to perform thinning line extraction on the full body image of the standing action to obtain a thinning line of a first type of preset joint of the user, the first type of preset joint comprising a spine joint, a shoulder joint and a hip joint; based on the thinning line of the first type of preset joint, obtaining posture information of the user, and calculating the similarity between the posture information of the user and standard posture information; when the similarity is less than a preset similarity threshold, generating a prompt information and sending it to a user device, the prompt information is used to indicate that the user has scoliosis. The application does not need manual screening, the accuracy of detection is higher, and can meet the large amount of scoliosis screening work.

The invention provides a plane calibration method for cross-structured light, which is used for solving the technical problem that when the cross-structured light is used as a light source for shooting, laser stripes contained in a shot image belong to two laser planes, so that a matching error is generated when straight line fitting is carried out in a laser plane calibration process. According to the plane calibration method for the cross-structured light, the laser line image is divided into the laser stripe and the background, then the laser stripe center line is extracted from each divided laser line image through the image refinement algorithm, and accurate extraction of the laser stripe center line is achieved; and then a classification criterion is constructed based on a principal component analysis method of K nearest neighbor and normal vector consistency, so that whether the laser point on the center line of the laser stripe belongs to the first laser plane or the second laser plane can be quickly and accurately judged, and the problem of point cloud attribution ambiguity of the biplanar laser line is successfully solved. And accurate solving and calibration of cross laser plane parameters are realized.

The invention relates to the technical field of engineering geology and rock mass mechanics, and discloses a rock mass fracture trace skeleton extraction and grouping quantification method and system. Comprising the steps of performing denoising and size standardization preprocessing on an excavation face image; using an improved DeepLab V3 + model to carry out semantic segmentation on the fracture trace to obtain a binary image; extracting a single-pixel width skeleton through a refinement algorithm based on the binary image; constructing a two-dimensional grouping strategy containing inclination angle similarity and spatial proximity to cluster skeleton traces; and calculating the length, the inclination angle and the density of the grouping result, and performing statistical analysis. The system comprises an image acquisition and preprocessing module, a crack segmentation module, a skeleton extraction module, a trace grouping module, a quantitative analysis module and the like. According to the scheme, automatic extraction, grouping and quantitative characterization of fracture traces are achieved, and the method is suitable for rock mass structure investigation and engineering analysis.

The application provides a cross structured light plane calibration method, which is used to solve the technical problem that when cross structured light is used as a light source for shooting, the laser stripe contained in the shot image belongs to two laser planes, which causes matching error in the process of laser plane calibration. The cross structured light plane calibration method of the application first divides the laser line image into two parts of laser stripe and background, then extracts the laser stripe center line on each segmented laser line image by using an image thinning algorithm, so that the accurate extraction of the laser stripe center line is realized; and then based on the principal component analysis method of K nearest neighbors and the normal vector consistency, a classification criterion is constructed, which can quickly and accurately determine whether the laser point on the laser stripe center line belongs to the first laser plane or the second laser plane, successfully solves the point cloud belonging ambiguity problem of the double-plane laser line, and realizes the accurate solution and calibration of the cross laser plane parameters.

The invention discloses an inspection processing detection method based on a power distribution network. The method comprises the following steps: acquiring original image data of the surface of a suspension clamp fitting by using a high-resolution camera; preprocessing the original image data and segmenting the original image data in a target identification processing mode to obtain a hardware fitting groove area image; a Zhang-Suen refinement algorithm is used for carrying out skeleton extraction on a crack communication region in the hardware fitting groove region image by determining a pixel point deletion mode through local communication analysis, and a pixel-level crack topological structure diagram is obtained; and based on the crack topological structure diagram, classifying the crack topological structure diagram into transverse cracks or longitudinal cracks through a direction distribution histogram and curvature continuity analysis, and through local connectivity analysis in a Zhang-Suen refinement algorithm process, avoiding fracture or distortion of a crack skeleton.

This invention discloses a machine vision-based method for counting half-smooth tongue sole fry, comprising: acquiring half-smooth tongue sole fry images and binarizing them; acquiring all connected component parameters of the binarized image and constructing corresponding connected component images; processing the connected component images using a skeleton thinning algorithm to obtain fry skeleton images; classifying the pixels in the fry skeleton images; calculating the number of half-smooth tongue sole fry in the fry skeleton images using a crosspoint counting method; obtaining the average standard area of ​​the fry; applying a compensation algorithm to connected component images exceeding a set ratio coefficient; and summing the fry counts in all compensated connected component images to obtain the total number of fry. This invention uses crosspoint counting and a compensation algorithm to obtain the number of fry. The judgment method is simple and fast, non-contact with the fry, and enables automatic detection by machines instead of manual labor; it is applicable to all complex situations such as low or high adhesion and overlap of fry, and has the advantages of fast calculation speed, good detection stability, and high counting accuracy.

The invention discloses a pathological section diagnosis area marking and archiving method, which comprises the following steps: acquiring multi-temporal pathological section image data of a target tissue area, the multi-temporal pathological section image data comprising section images of the target tissue area collected at different time points; based on the multi-temporal pathological section image data, constructing a cable pipeline topology network in the target organization area through an image registration algorithm and a topology refinement algorithm; for the pathological section image data of each time phase, identifying an abnormal diagnosis area in the pathological section image data through an image identification algorithm; calculating geometric topological relation data between the spatial position of the abnormal diagnosis area identified in each time phase and the nearest branch node in the cable pipeline topological network; and based on the geometric topological relation data, generating fault source coordinate data in one-to-one correspondence with the abnormal diagnosis areas, and storing the fault source coordinate data into a diagnosis area coordinate database.

The application discloses a cable bending radius calculation method based on mask center line algebraic reconstruction, which comprises the following steps: an RGB-D dual-mode semantic segmentation network is used to acquire a cable mask, and a depth image is used to generate a cable space point cloud; the mask is subjected to an image thinning algorithm to obtain a cable plane feature center line, and the cable plane feature center line track is tracked, and the cable space point cloud is indexed according to the track pixel coordinates to obtain a cable space feature center line; the cable space feature center line is non-uniformly sampled to obtain a space sampling point, and the space sampling point is subjected to algebraic reconstruction to obtain a cable space feature curve; the cable space feature curve is regarded as a particle space motion track, and the space bending radius of the cable space feature curve is solved by using a particle space motion law.

This invention provides a geological disaster monitoring method and system based on UAV vision and target collaboration, belonging to the field of disaster monitoring technology. The method includes: acquiring geological images at different times; performing terrain modeling based on the geological images and target locations to obtain a digital surface model; identifying vegetated and bare areas based on the geological images; constructing a Delaunay triangulation based on targets in the vegetated areas to calculate vegetation displacement; calculating bare land displacement based on the digital surface model; fusing vegetation and bare land displacements using thin-plate spline functions to obtain global displacement information; detecting cracks using an improved Canny-Zernike operator based on the digital surface model and identifying crack skeleton lines using a morphological thinning algorithm; and modeling displacement gradient tensors based on crack skeleton lines and global displacement information to construct a crack opening distribution map and obtain monitoring results. This invention creatively combines physical noise reduction, intelligent fusion, and precise analysis to achieve high-precision positioning in harsh environments.

This application discloses a method and a main control chip for improving the accuracy of straight lines in a map. The main control chip includes a binarization unit, an image thinning unit, and a line detection unit. The robot pre-constructs a map by collecting environmental information. The method includes: Step 1, binarizing the map to obtain a binary image; then performing Step 2; Step 2, extracting a skeleton image from the binary image using an image thinning algorithm; then performing Step 3; Step 3, extracting edge detection lines from the skeleton image using a line detection algorithm, denoising the extracted edge detection lines, and then fitting the denoised edge detection lines with straight lines to obtain fitted line segments.

The invention discloses an intelligent marine route extraction method based on multi-resolution grid density. The method comprises the steps of preprocessing AIS data and converting the AIS data into a continuous track segment set; abrupt coordinate change caused by the daily change line is eliminated through a coordinate system processing mechanism; generating a coarse grid by using an improved super-sampling rasterization algorithm, and constructing a multi-resolution density field in combination with high-density region identification and recursive multi-resolution refinement; enhancing density field features through adaptive power law transformation, multi-scale Gaussian decomposition and adaptive morphological processing; extracting a grid skeleton of the enhanced density field by using a GPU parallel Zhangsuen refinement algorithm; a simplified vector path is obtained through intelligent path tracking, coordinate system conversion and DouglasPeucker adaptive simplification, a path topological structure is constructed, and finally a continuous and accurate marine route network is generated. According to the method, the high-precision air route can be efficiently and automatically extracted, and the accuracy and the calculation efficiency of air route identification are remarkably improved.

The application relates to the technical field of computer-aided semen analysis, and discloses a sperm flagellum movement parameter calculation method and a computer readable storage medium, which comprises the following steps: based on a preset sampling frequency, a plurality of sample images are extracted from a flagellum movement video of sperm to be analyzed; for each sample image, the sample image is input into an improved U-Net network to obtain a binary flagellum segmentation mask, after a morphological thinning algorithm is used to extract a flagellum skeleton, a parameterized flagellum skeleton curve is obtained; a convolution kernel in the improved U-Net network is improved into a convolution kernel with a constrained deformable offset; based on the parameterized flagellum skeleton curve, the flagellum length, the flagellum tangent angle, the flagellum propagation speed, the flagellum curvature, the flagellum amplitude, the flagellum frequency, the flagellum wavelength, the flagellum bending amplitude and the flagellum bending index are calculated. Through the convolution kernel with the constrained deformable offset, the receptive field is increased, the segmentation continuity is ensured, and the movement parameters of the flagellum can be quantitatively calculated.

The invention discloses an inland river navigation water area environment classification method based on image processing, and the method comprises the steps: converting an electronic chart into an image, and carrying out the preprocessing of the image, and obtaining a binary image; an improved Zhang-Suen refinement algorithm is adopted to extract a riverway skeleton, and the number of end points and branch points of the riverway skeleton is calculated; meanwhile, extracting a boundary contour by using a Mohr neighborhood boundary tracking algorithm, and calculating the proportion of a water area in the contour in combination with a scanning line algorithm so as to judge whether an island exists or not; and integrating the topological characteristics (the number of end points and branch points) of the river channel and an island judgment result, automatically classifying the environments of the inland river navigation water area, and realizing the analysis of the inland river navigation water area. According to the method, automation and intelligence of inland water area environment recognition are realized, the classification precision is high, the reliability is high, and key environment sensing information can be provided for inland waterway analysis.

The invention discloses an expressway-oriented sensing equipment state prediction and life evaluation system, which comprises a multi-source sensing data acquisition module, a sensing equipment state prediction module, a service life evaluation module and a service life evaluation module, wherein the multi-source sensing data acquisition module is used for acquiring multi-source sensing data; the data preprocessing module is used for generating standardized multi-source sensing data; the equipment space marking module is used for forming equipment space distribution; the road section area division module is used for constructing a cross check structure; the equipment credibility scoring module is used for generating a credibility score; the visual degradation feature extraction module is used for generating visual degradation features; the health index calculation module is used for generating a health index; the prediction and evaluation module is used for calculating a residual life interval; and the evaluation result output module is used for outputting a prediction result. According to the method, the improved Zhang-Suen refinement algorithm and the multi-dimensional degradation hierarchical expression structure are combined, so that accurate state evaluation and life prediction of the highway sensing equipment are realized.

Embodiments of the present application provide an inland ship trajectory space-time visualization interaction method, system and electronic device, which acquires the real-time scale of the current map view and the ship identifier selected by the user, the query time range and the working condition data type, and determines the corresponding trajectory data and working condition data accordingly; the trajectory data is cleaned, and the stop window in the trajectory is identified based on the voyage segmentation processing, the trajectory on both sides of the stop window is cut into multiple independent navigation segments; the real-time scale is used to dynamically calculate the thinning threshold value, the trajectory thinning algorithm is used to adaptively thin the trajectory points in each independent navigation segment; finally, each navigation segment is drawn in different colors on the map view and the stop window is marked with a preset graphic, a double time axis slider is generated, the first time axis slider is used to display the time interval of each segment and stop window, and the second time axis slider is used to synchronously play back the corresponding trajectory points and working condition data along the time axis for the target navigation segment selected by the user.

The invention relates to a power transmission line icing monitoring method, equipment and a medium. The method comprises the following steps: acquiring original image information about a power transmission line and target meteorological characteristic data of the position of a tower; performing image segmentation on the original image information through an image segmentation algorithm; conducting wire skeleton extraction on the image segmentation information through a skeleton refinement algorithm; determining actual maximum icing thickness information of the power transmission line according to the image segmentation information, the skeleton image information and the pixel mapping distance; inputting the target meteorological characteristic data into a meteorological prediction model, and outputting an icing index of the power transmission line through the meteorological prediction model; and determining the icing risk index of the power transmission line according to the actual maximum icing thickness information, the icing index, the maximum icing thickness information that the electric wire can bear, the current temperature information of the power transmission line, the freezing point temperature information of water, the image weight, the meteorological weight and the temperature smoothing factor. And the real-time monitoring accuracy of the icing risk of the power transmission line is greatly improved.

This application relates to a method, apparatus, equipment, and medium for extracting three-dimensional phenotypic parameters from plant leaves. The method includes: using Poisson surface reconstruction technology to perform geometric domain smoothing on the pose-standardized point cloud of plant leaves; fusing the local point density and principal component curvature features of the plant leaf point cloud; combining the color texture information mapped from the plant's RGB image; and enhancing the vein signal through nonlinear power function transformation to construct candidate vein data. The candidate vein data is then binarized using the Otsu dynamic threshold segmentation algorithm to extract the target vein region of the target plant leaf. The vein topology skeleton within the target vein region is extracted using the Zhang-Suen thinning algorithm. Based on the vein topology skeleton, the leaf length and width of the target plant leaf are calculated, and the vein density per unit area of ​​the target plant leaf is determined according to the leaf area and the total length of the vein branches. This application allows for phenotypic parameter extraction without leaf harvesting, avoiding measurement errors caused by water loss and curling of leaves after detachment.

The present application relates to a kind of circular pointer type instrument pointer fitting and instrument reading identification method, belong to image recognition field.The present application simultaneously divides the pixel area of scale value and pointer in instrument panel using semantic segmentation method;Using the method of image classification to identify scale value, determine the rotation center of pointer;The skeleton thinning algorithm and least square method are used to the pointer pixel area generated by segmentation to fit the straight line equation of pointer, determine the direction of pointer;Using multiple scale values weighted calculation pointer reading.The present application can automatically fit the straight line equation of pointer, consider multiple scale values to calculate instrument reading, it is advantageous to improve the precision of pointer instrument reading identification.

The present invention discloses a method for evaluating text printing quality based on text mask structural similarity: a text mask-based dynamic perceptual text segmentation network is used to process printed text images and standard text images to obtain a multi-channel text fine segmentation feature map; an image refinement algorithm is used to extract text skeleton information from the multi-channel text fine segmentation feature map, and different weights are assigned to each layer of pixels based on the order in which pixels are removed during the extraction process to establish a text skeleton template; a structural similarity image quality evaluation algorithm is used to perform text printing quality detection on the standard text skeleton template and the printed text skeleton template, obtaining an average structural similarity value between the printed text skeleton template and the standard text skeleton template; and the text printing quality is determined based on the average structural similarity value. The present invention constructs a text segmentation network, and the output is a feature map containing text mask information, thereby improving the detection range and accuracy.

The invention discloses a self-adaptive welding method for a sheet bending angle welding seam. The method comprises the steps that a 2D image and a 3D depth map of a bending angle position are obtained through scanning; the trained YOLOv5n-seg network is used to output a bending cavity probability mask at the bending angle position; processing the bending cavity probability mask by using a BLOB model to obtain a binary mask; based on a Zhang-Suen refinement algorithm, skeleton extraction is carried out on the binarized mask to obtain a weld joint center line pixel point set; performing B spline curve fitting on the center line of the welding seam; determining a corresponding ROI frame according to the points of the B-spline curve, extracting the boundary of the binarized mask, obtaining the 2D coordinates of boundary representative points of the boundaries on the two sides of the points of the B-spline curve in the ROI frame, and obtaining the 3D coordinates of the boundary representative points based on the corresponding relationship between the 2D image and the 3D depth map; and constructing a 3D welding line path according to the midpoint coordinates of the 3D coordinates of the boundary representative points, and welding according to the 3D welding line path. The device can adapt to welding of bent fillet weld joints with holes.

The application discloses an OCTA-based image processing method, comprising the following steps: acquiring an OCTA angiogram; pre-processing the OCTA angiogram; performing image segmentation on the pre-processed OCTA angiogram; extracting a blood vessel skeleton from the image segmented OCTA angiogram; and performing parameter measurement based on the extraction result. The OCTA-based image processing method can accurately obtain the skeleton and boundary of the blood vessel by performing a thinning algorithm and branch point detection on the binary image, and can split the blood vessel tree into single-section branch blood vessels, so that the measurement process of each branch blood vessel is simplified and unified, and the efficiency and accuracy of the measurement are ensured.