36 results about "Local scale" patented technology

A method and multi-scale attention system for the detection of objects and temporal change regions by a spatiotemporal attention operator of an image sequence, which linearly aggregates temporal change filter with spatial saliency filter and includes an extractor of salient maxima, which selects consecutive salient maxima of the spatiotemporal operator to produce the locations of the objects of interest and the centers of temporal change regions. The concept of spatial local scale introduced into the system and method for its determination allows for a scale-adaptive integration of the temporal change with spatial saliency and effective detection of different in size and location objects of interest. Can be used for object and spatiotemporal change detection in monitoring pollution, natural disasters, weather conditions and environmental changes based on satellite remote sensing imagery from various sensors.

A spectral clustering method for automatically determining the number of clusters based on a neighboring point method comprises the steps of 1) normalizing all dimensions of a data set; 2) calculating an interval sparse distance matrix by a neighboring point method and defining the matrix as local scale parameters of distance mean values of the neighboring points to obtain a whole sparse similarity matrix; 3) determining the local density of each data point and the minimum distance to other points with a higher local density by calling a CCFD method, and obtaining the number of singular points generated by the fitting outside a confidence interval; 4) calculating a degree matrix D and a Laplacian matrix L according to a formula and extracting an eigenvector group by eigen decomposition of L; 5) outputting clustering results; and 6) selecting and outputting the clustering result with the optimal number of neighboring points corresponding to the maximum Fitness function value. According to the invention, the local scale parameter of each data point can be estimated according to data distribution, the number of clustering centers is automatically determined, and the parameter adaptation of the number of neighboring points is realized.

The invention relates to a structure multiscale health monitoring device and a structural multiscale health monitoring method on the basis of a novel integrated type composite sensor array. The device comprises a computer, a broadband light source with a plurality of usable wavelengths, a multiscale demodulating device, a composite-material flexible plate and a novel integrated type composite sensor module, wherein the novel integrated type composite sensor is arranged on the composite-material flexible plate. The method applies the novel integrated type composite sensor, simultaneously measures the modal frequency of the structure and the strain distribution near the sensor and seeks a multiscale damage distribution form by population evolution and individual operation of selection, cross and variation and the like of a hierarchical genetic algorithm. The structural multiscale health monitoring device and the structural multiscale health monitoring method have the advantages of more sensing parameters, small volume of a sensing head and high system integration level and the like, and has the beneficial effects that the problems of data fusion and noise test and the like on the same structure can be effectively solved and health monitoring for the structure of the composite material from the whole and local scales can be realized.

A system and method using a text extraction application for identifying words with multiple orientations from an image are described. The text extraction application receives an input image, generates progressively blurred images, detects blobs in the blurred images, outputs ellipses over the blobs, detects a word in the input image, orients and normalizes a first version of the word, generates an inverted version of the word, performs OCR on the first version and the inverted version of the word, generates confidence scores for the first version and the inverted version of the word and outputs text associated with the word.

The invention relates to an industrial remote metering system, which is mainly used for industrial enterprise logistics metering, and belongs to the technical field of the industrial enterprise logistics metering. In a technical scheme, the system comprises a local scale house part, a metering hall part and a server part; and the three parts exchange data through a metering private network. The invention has the advantages of integrating the audio and video information with a metering service, and integrating with the systems such as a logistics system, an integrated circuit (IC) access card and the like, so that a plurality of technologies of the metering system, an entering/leaving-factory logistics system, an audio-video system, a radio frequency IC card, and a door access control system and a vehicle management system are completely integrated once and for all; the horizon of the metering staff is expanded by the system by using the technical means such as a radio frequency card technology, a network communication technology, an image acquiring and processing technology and the like, so that the metering staff can see metering vehicles and goods from all directions; the scale staff are isolated from the supplies or transporters, therefore the cheating cases are reduced, and the working staff are protected; each worker can supervise as many scale surfaces as possible; and the metering gap time is fully used, and the metering efficiency is improved.

The invention discloses a local scale parameter, entropy and cosine similarity-based spectral feature selection method. A Gaussian kernel function is adopted as a similarity measurement method, defined local scale parameters which are of features and based on local standard deviation of the features are used as kernel function parameters, and problems that uniform scale parameters in calculating feature affinity matrices cannot reflect data distribution information, and local scale parameters are impacted by off-group points are overcome; entropy and cosine similarity sorting is respectively adopted to measure feature importance degrees, a suitable feature subset can be quickly selected; and technical support is provided for data analysis of diseases such as tumours, and the method has important biomedical significance.

The invention discloses a local quick updating method of a mixed voxel geological body model. Firstly, the method builds an overall mixed voxel geological body model, which is an IHG-TIN model (Irregular Hexahedron Grid and Triangle Irregular Network). When the mixed voxel geological body model needs to be updated because of a local new bore hole or other local exploration data, the method only needs to build a local mixed voxel geological body model on the same local scale. A new voxel is used to replace a voxel at the same position of the original model locally, then a new layer model is extracted from the updated mixed voxel geological body model, and the topological relation and geological properties of the layer model are updated, so that quick local dynamic updating is achieved. The original model does not need to be rebuilt from the beginning, and thus the dynamic updating efficiency and model precision of geological bodies can be improved.

The invention provides a method for stably detecting straight line/ circle features. First, setting a base point and an initial detection area and repositioning the detection area through template matching; adjusting the detection area by use of position supplementation and correction; calculating and obtaining estimated value of noise variance through extracting measuring directional data, calculating measuring line local noise scales within the scope of a projection line, and performing filtering to measuring line local scales; then calculating edge points of one-dimensional data; detecting noise points in the points through fitting inner points and noise point segmentation and screening; keeping the inner points on the straight line/ circle, and finally, fitting out a straight line/ circle by the detected inner points by use of the least square method. By use of the method, edge feature points can be stably extracted in a strong noise image and fitted into a straight line/ circle. Detection area and ROI area are set through template matching so that the straight line/ circle detection is carried out within the ROI area, and therefore, the method has good calculation real-time performance and will be of great importance to industrial machine vision measurement techniques.

A design space analyzer generates a parametric model associated with a design problem. The design space analyzer then discretizes various parameters associated with the model and generates a plurality of sample designs using different combinations of discretized parameters. The design space analyzer also computes one or more metrics for each sample design. In this fashion, the design space analyzer generates a coarse approximation of the design space associated with the design problem. The design space analyzer then evaluates portions of that approximation, at both global and local scales, to identify portions of the design space that meet certain feasibility criteria. Finally, the design space analyzer modifies the design space to facilitate more efficient exploration during optimization.

The invention discloses an artificial acclimation method of agrocybe prauox fayod, comprising the following steps: a. culturing strains; b. culturing protospecies; c. culturing cultivated species. The invention utilizes common culture technology resources to improve a purely wild agrocybe praecox, thus eliminating the shortages of smaller agrocybe praecox body, thinner pileus and bacterial context, thinner stipitipellis, dispersed growth and the like and achieves satisfied acclimation effect. By improvement, the growth circle of one time per year is changed into two times per year, which improves yield and establishes technical basis for popularization on a large scale. The method of the invention can effectively accelerate local scale planting to improve cultivation benefit, fully utilize wastes produced in agricultural production, such as corn cobs, boll hulls, crop straws and the like to serve as raw materials, and effectively lighten air pollution caused by fume due to burning crop straws.

The invention discloses a point cloud filtering method based on Snake model multi-primitive fusion, By adopting the filtering strategy of multi-primitive fusion (point primitive and object primitive),the defect of poor robustness of single primitive filtering method can be effectively avoided, and the feature information provided by different primitive can be fully utilized to improve the adaptability of filtering method to complex terrain region. The Snake model is mainly constructed. By setting the target energy function with the ground point adsorption force, the model can take into account the local scale information and global scale information of the point cloud, and fully tap the automation potential of the point cloud data itself. According to this model, the filtering threshold is expressed as a linear function of terrain slope change, and the adaptive update of filtering threshold can be realized, which avoids the defect of poor filtering precision of single threshold and excessive parameter setting of multiple thresholds.

The invention discloses a multiple athletes behavior detection method based on scale adaptation local spatiotemporal features, and the method comprises the following steps: S1, a normalized Laplace operand is adopted to estimate a local scale and establish a scale adaptation local spatiotemporal feature detection algorithm, and a Harris spatiotemporal interest point detection operand and the Laplace operand are combined to infer a Harris-Laplace spatiotemporal interest point detection operand; S2, the normalized Laplace operand is generalized to a spatiotemporal domain; S3, local spatiotemporal feature descriptors are generalized to three-dimensional soccer match video images; then before a K-means clustering algorithm is adopted to generate a spatiotemporal codebook, each spatiotemporal interest point is normalized to ensure that the each spatiotemporal interest point is invariable under scaling and shifting.

The invention belongs to the technical field of buildings, in particular to a spray dust remover for building construction sites. Aiming at the problems occurring when in use of the prior spray dust remover that the dust removal scope is small, only dust removing on the local scale is feasible, and the dust removal effect is poor, the invention has the following proposal that the spray dust remover comprises a base; the top part of the base is connected with support boxes which are equidistantly distributed through bolts; a same support baffle is connected between the two connected support boxes through the bolts; the support baffle and the top parts of the support boxes are connected with a cover plate through the bolts; a rectangular straight slot is formed in the bottom part of the cover plate; and mounting insert slots which are equidistantly distributed are formed in one side of the support baffle. In the embodiment of the invention, the upper part and the lower part of the spraydust remover work together to remove the dust, therefore the dust removal effect of the spray dust remover is obviously improved; and a plurality of booster pumps are utilized for obviously increasingthe spray scopes of upper atomizing nozzles and lower atomizing nozzles, and therefore the spray dust removal scope of the device is increased, and the spray dust removal effect of the device is further improved.

In the convolutional neural network (CNN) that uses the encoder-decoder structure for image segmentation, the multi-scale context aggregation module receives the final encoded feature map from the encoder, and sequentially aggregates the feature maps from the global scale to the local scale. Scale context to strengthen the semantic relationship of different scale contexts, thereby improving segmentation accuracy. Multi-scale context is obtained by calculating the hole convolution with different expansion rates on the feature map. In order to reduce the amount of calculation required for CNN, achannel-based feature selection (CFS) module is used in the decoder to merge the two input feature maps. Each feature map is processed by a global pooling layer, followed by a fully connected layer or a 1*1 convolutional layer to select highly active channels. Through subsequent multiplication by channel and addition by element, only the channels with high activation in the two feature maps are retained and enhanced in the merged feature map.

The invention discloses a positioning map based on an RFID tag matrix. RFID tags with the same size are placed on a square mat according to X intervals; each RFID tag has its own ID number, thereby forming a numbered matrix. The invention has the beneficial effects of: accurate positioning in a local area, low cost and simple technical solution. The map mat with RFID tag matrix is ​​also convenient to carry, and multiple map mats can be combined to increase the space area.

The invention provides a multi-scale ship enterprise scene recognition and extraction method combined with spatial distance constraint. The method comprises the following steps: obtaining the nationwide ship enterprise distribution condition and structural features, and making a high-resolution ship enterprise remote sensing data set; utilizing the remote sensing data set to train a ship enterprise remote sensing image recognition model based on the global scale and the local scale; performing multi-model and multi-scale detection on the to-be-predicted remote sensing image by adopting a shipenterprise remote sensing image recognition model; based on a multi-scale detection result, extracting a spatial distance between the ship enterprise objects of the remote sensing image to be predicted; and performing scene category comprehensive evaluation on the to-be-predicted remote sensing image in combination with global and local multi-scale detection results, a local object label combination mode and an object space distance constraint condition. The ship enterprise complex scene extraction method has the beneficial effects that ship enterprise complex scene extraction is effectively carried out, theoretical support can be provided for ship industry development and coastal environment monitoring in China, and a macroscopic decision basis can be further provided for government capacity adjustment and industrial layout for overseas ship enterprise extraction.

The embodiment of the invention discloses a map display method, an apparatus, equipment and a storage medium. The method comprises the following steps: acquiring a plurality of target addresses, and alocking range of each target address and a local scaling ratio of each locking range; obtaining a locking range of each target address and a local scaling ratio of each locking range; getting the overall scale of the map; a locking range of each target address being displayed at the local zoom scale, and a map area outside the locking range being displayed at the overall zoom scale. A technical problem that that exist map is inconvenient to be used due to a single zoom scale is sol, so that different regions are displayed by the map through a plurality of zoom scales.

The invention discloses a rural landscape zoning method and system and a storage medium, and the method comprises the steps: employing a multi-level division method, and constructing a multi-scale and multi-level zoning architecture from the territorial scale to the regional scale to the local scale; determining a plurality of target feature elements causing landscape similarity and heterogeneity under different scales based on previous zoning research experience; according to a preset classification index, classifying the plurality of target feature elements, and based on an obtained classification result, determining a rural landscape zoning index corresponding to each scale; and performing rural landscape regionalization based on the rural landscape regionalization indexes corresponding to the corresponding scales under each hierarchical regionalization architecture. By implementing the method, the zoning comprehensiveness can be improved.

The invention relates to identification, monitoring and early warning of scattered plaque spartina alterniflora in four seasons of the year, in particular to a spartina alterniflora identification and early warning method, which comprises the following steps of: 1, building an acquisition platform, and acquiring a spartina alterniflora image; 2, splicing and segmenting the image; 3, making a training set, a verification set and a test set; and 4, selecting a mainstream convolutional neural network, and training, verifying and testing the model. 5, judging the recognition accuracy, and selecting the convolutional neural network meeting the conditions; 6, data of the four periods are trained respectively, and spatial distribution of the spartina alterniflora is carried out; 7, constructing a hazard grade evaluation model; 8, obtaining the diffusion distance of the spartina alterniflora; 9, acquiring biomass crop parameters of the spartina alterniflora; and 10, generating a spartina alterniflora early warning visual heat map. According to the method, accurate identification of the spartina alterniflora under the local scale is realized, the early warning visual heat map is generated, and data information is provided for early monitoring and prevention of the spartina alterniflora.

The present invention provides a method and system for performing multi-zone association and scaling display on a K line graph or a USA (United States of America) line graph. The method comprises the following steps: (a) by using time axis as horizontal axis, displaying global line graph data of a K line graph or a USA line graph in a line graph display window; (b) receiving local scaling instruction and local scaling parameter by means of human-computer interaction interface; (c) performing isotonic transformation of time axis dimension on line graph data within a corresponding range on the K line graph or the USA line graph according to the local scaling instruction and the local scaling parameter, and obtaining isotonic line graph data with global data and local data integrated; and (d) replacing the global line graph data with the isotonic line graph data after isotonic transformation in the line graph display window and displaying the isotonic line graph data. According to the method and system for performing multi-zone association and scaling display on the K line graph or the USA (United States of America) line graph provided by the present invention, by performing single-dimensional, multi-zone and variable-proportion isotonic mapping transformation on the local data of the K line graph or the USA line graph, associated display of the global data and the local data can be implemented, and the problem that organic correlation between ''global and local'' line graph information is cut off by the display window can be well solved.