12 results about "Otsu's method" patented technology

The invention discloses a tidal flat automatic extraction method fused with an irregular hexagonal grid. The method comprises the following steps: step 1, constructing a time sequence remote sensing data set; step 2, establishing a precision verification sample library; step 3, screening spectral indexes; 4, synthesizing an extreme tide level image; 5, generating an irregular hexagonal grid; step 6, carrying out adaptive threshold segmentation; step 7, secondary elimination of confused ground features; step 8, tidal flat range extraction and precision verification; according to the method, based on a time sequence remote sensing image set, pixels at all positions are screened through a specific spectral index maximum value, a new extreme tide level image is formed through combination, spatial heterogeneity of a complex coastal zone is adapted through an irregular hexagonal grid and a bottom-to-top hierarchical merging strategy, local self-adaptive threshold segmentation is achieved in cooperation with a local Otsu algorithm, and a new extreme tide level image is obtained. The optimal grid scale is automatically determined by means of an elbow method, and the confused ground features are secondarily eliminated through the time sequence standard deviation, so that the problem of miscarriage of the confused ground features is effectively solved, and high-precision full-automatic extraction of the tidal flat is realized.

A method for automatically monitoring dynamic changes of water blooms based on remote sensing images comprises the following steps: step 1, acquiring remote sensing reflectivity data of a multi-source remote sensing satellite in a screened time sequence range, calculating a spectral index, automatically determining an extraction threshold through pixel gradient statistics and an Otsu algorithm, and accurately identifying a water bloom area; and 2, calculating the water bloom area based on the multi-temporal data, researching the seasonal change, the long-term trend and the spatial distribution rule of the water bloom area in combination with a time sequence analysis method, and evaluating the accuracy of water bloom inversion by comparing the actually measured chlorophyll-a concentration with a water bloom inversion result. The application can monitor the change of the algal bloom area in real time.

A method for monitoring the grassland vegetation coverage based on continuous photography, comprising: determining an observation quadrat and continuously taking multiple vegetation photos of the vegetation within the observation quadrat throughout the entire growing season; cropping and geometrically correcting the multiple vegetation photos; obtaining the vegetation pixel ratio corresponding to the vegetation photos and comparing the vegetation pixel ratio with a preset second threshold; when the vegetation pixel ratio is greater than the second threshold, calculating the vegetation coverage corresponding to the vegetation photos by using the Otsu algorithm; otherwise, using the vegetation pixel ratio as the vegetation coverage; obtaining a dynamic change curve of the vegetation coverage based on the vegetation coverage corresponding to each vegetation photo. The present invention effectively improves the monitoring accuracy and efficiency of the dynamic change characteristics of the grassland vegetation coverage by combining continuous photography and the Otsu algorithm.

The present application relates to high-speed railway subgrade construction quality detection technical field, disclose a kind of railway subgrade filler grading identification method and system based on airborne unmanned vision perception, comprising: through the image of filler field collected by airborne unmanned aerial vehicle, carry out gray conversion, size reduction etc. in turn pretreatment;Adopt Otsu's method to carry out self-adapting binaryzation, exert morphological open operation, erosion, the sequence operation of inflation, to accurately segment and optimize filler particle profile;Profile is extracted using Canny algorithm, with the long diagonal line of particle circumscribed rectangle as equivalent particle diameter;Introduce shape correction factor to correct the error of sphere volume hypothesis, calculate particle mass based on the preset filler density-particle size relationship model;Finally, statistical generation grading curve.The present application realizes the automatic, fast, non-contact identification of railway subgrade filler grading, improves detection efficiency and accuracy, and provides technical support for intelligent quality control of subgrade construction.

The invention relates to the technical field of an over-fire area extraction method, in particular to an over-fire area extraction method fusing time sequence spectral features and a multi-source fire product, which comprises the following steps of: S1, acquiring Sentinel-2 data, preprocessing and constructing a median image; s2, constructing a sample library of the overfire area and a sample library of the non-overfire area; s3, combining the training sample and the median image to screen and identify a spectral feature set of a fire passing area, and extracting an initial fire passing area; s4, the initial fire passing area is corrected by combining the multi-source fire point product and the enhanced combustion index, and a precise fire passing area is obtained; s5, merging the MCD64A1 product and the precise fire passing area, creating an envelope rectangle, and performing threshold segmentation on the month-by-month median image of the optimal spectral index by adopting an Otsu algorithm to obtain final fire passing area data; and S6, evaluating the precision of the fire passing area by using the verification sample, and drawing. Based on Sentinel-2 images and multi-source fire products, the invention provides a method for quickly identifying a large-scale and long-time-sequence fine fire passing area.

The invention provides a lane line identification method and device fusing artificial fish and a particle swarm algorithm, and electronic equipment, and belongs to the technical field of road identification. The method comprises the steps that an Otsu algorithm is adopted to fuse artificial fish and a particle swarm algorithm, each pixel value of a road image corresponds to a particle position, each particle corresponds to the artificial fish for foraging optimization, and according to the difference value between the inter-class variance value of the next particle position and the inter-class variance value of the current particle position, the current particle position is determined according to the difference value between the inter-class variance value of the next particle position and the inter-class variance value of the current particle position. And adjusting the visual field range of the artificial fish, searching an optimal segmentation threshold value based on the adjusted visual field range of the artificial fish, and segmenting lane lines in the road image according to the optimal segmentation threshold value by adopting an Otsu algorithm. According to the method, the artificial fish and the particle swarm algorithm are fused, all pixel values do not need to be traversed, a large amount of operand is saved, the operation time is shortened, and meanwhile the defect that a traditional particle swarm algorithm is caught in a local optimal solution in premature convergence is overcome.

The invention relates to a cross-scene concrete surface crack unsupervised identification method based on transfer learning, and the method comprises the steps: collecting concrete surface crack images in different scenes, and constructing a data set; binarizing the image based on an Otsu algorithm, removing noise pixels, and constructing a data set with a label and a data set without a label; embedding a residual structure in the cyclic generative adversarial network, constructing a feature migration model, and inputting a labeled data set and an unlabeled data set for adversarial training to generate a crack image with consistent feature distribution; an SE attention mechanism is embedded in the U-Net, a semantic segmentation model is constructed, the original image and the generated image are input for training, and the trained semantic segmentation model is obtained; and embedding the trained feature migration model and the semantic segmentation model into a sliding window, collecting a to-be-identified concrete surface image, and inputting and outputting crack edge and geometric information. The semantic segmentation precision of the model in different appearance feature differences can be improved.

This invention relates to outdoor iris image processing technology, providing a real-time comprehensive quality assessment method and an adaptive optical parameter adjustment system. The assessment method includes: locating the eye boundary based on grayscale projection and local normalization; generating dual thresholds using an improved Otsu algorithm and dynamic grayscale subspace search; calculating a sharpness index based on second-order gradient entropy change; locating the pupil through a reliable edge point fitting mechanism, detecting the outer circle of the iris using an improved integral-differential operator, and evaluating segmentation accuracy using the ratio of the local standard deviation of the edge to the standard deviation of the target region; fusing gradient fitting and grayscale distribution to achieve eyelid and eyelash detection; and finally outputting a quality score based on 18 weighted indicators. The optical adjustment system senses the environment and initializes optical parameters, dynamically adjusting parameters based on weighted voting from multiple frames of images, and achieving rapid convergence with environmental consistency verification. This invention improves the speed and stability of pupil localization, sharpness assessment, eyelash detection, and illumination adjustment in low-computing-power scenarios.

The application discloses a mural virtual restoration method and device, computer equipment and a readable storage medium, relates to the mural disease detection and image processing technical field, is suitable for large-area mural with various diseases, improves the restoration efficiency, and can overcome small noise. The method comprises the following steps: in response to a mural restoration instruction, a hyperspectral imager is used to collect mural images, the mural images are corrected, and target hyperspectral reflectivity images are obtained; a principal component analysis algorithm and a high-pass filter enhancement algorithm are used to enhance the target hyperspectral reflectivity images, an image operation formula is used to operate the enhanced images, and preprocessed images are obtained; a non-disease area mask is made based on a remote sensing image processing platform, a multi-scale bottom-hat transform algorithm is used to extract the images after the mask, disease information is obtained, the disease information is segmented by using an Otsu algorithm, and a disease binary graph is obtained; a fast marching algorithm is used to restore the denoised disease binary graph, and target restoration images are obtained.

The present invention provides a method for automatically identifying and detecting mixed coal vitrinite, comprising: step one, collecting multiple images of various single coals in an actual industrial environment and calibrating the types of coals. Step two, using global median filtering and local mean filtering to eliminate some noise points in the image. Step three, using the clustering-based Otsu algorithm to binarize the image and remove the background image. Step four, using texture features and grayscale histograms to remove inertin and exinin from the image. In step four, the texture features include contrast, correlation, energy and entropy. Step five, using the Random Patchwork algorithm to generate a mixed coal image data set for multiple single coal images. Step six, constructing an image semantic segmentation model. Step seven, inputting the data set into the network for iterative training; and performing detection after the training is completed. The present invention combines traditional feature extraction with an improved deep learning algorithm, and can quickly and accurately detect and identify the types and proportions of mixed coal vitrinite.

The Zn—Al—Mg-based hot-dip plated steel member includes: a steel member; and a hot-dip plating layer arranged on a surface of the steel member, wherein the hot-dip plating layer includes, as an average composition, in terms of mass %, 4 to 22% of Al, 1.0 to 10.0% of Mg, 0.0001 to 2% of Fe, and a balance including Zn and impurities; in binarized image data of image data obtained by imaging a surface of the hot-dip plating layer, the total area ratio of white pixels is 30% or more and 70% or less; and the binarized image data is obtained by binarizing the image data by Otsu's method.

This invention discloses a method for combating intermittent sampling interference based on Otsu's method for random orthogonal subpulses. The method includes: establishing intra-pulse and inter-pulse randomly coded orthogonal frequency modulated waveform models; encoding and orthogonally frequency modulating intra-pulse and inter-pulse subpulses to obtain compressed subpulses; obtaining the inter-class variance of the compressed subpulses according to a multi-level Otsu algorithm; obtaining an optimal threshold based on the inter-class variance of the compressed subpulses; and determining the difference between the target signal and three types of intermittent sampling interference signals based on the optimal threshold. This invention can improve the accuracy of identifying and distinguishing the target signal and the three types of intermittent sampling forwarding interference signals, effectively combating intermittent sampling forwarding interference.