43 results about "Tree canopy" patented technology

The invention relates to a method for extracting point cloud data of tree height based on a space three-dimensional template, which comprises the following steps: classifying original point cloud data to generate a digital elevation model (DEM), a digital surface model (DSM) and a tree canopy model (CHM) of a domain to be extracted; respectively computing the related coefficients of subdomains ofthe CHM and the template by introducing the space three-dimensional template; then marking the subdomains meeting requirements by utilizing a threshold distinguishing method; and finally performing adjacent superposition analysis in the marked candidate domain and merging the adjacent domains with over-high overlapping degree to generate tree height information. When ensuring the extracting precision of tree heights, the method effectively solves the problem of tree height judgment ambiguity of adjacent domains so that the extracted tree height information has higher quality and larger practical value, wherein the problem of tree height judgment ambiguity is caused by the phenomenon of tree superposition.

The invention discloses a method for measuring the tree canopy height through satellite remote sensing, wherein the method has the advantages of being wide in coverage range, short in data obtaining period and low in data obtaining cost. The method is realized through the following calculation scheme: extracting target forest regions in different directions from an overlapping region in acquired satellite three-dimensional relative image data by adopting a deep neural network model and an object-oriented remote sensing image classification technology; obtaining satellite stereo relative image data and a ground control point file of a target forest area according to the satellite stereo relative image data and the target forest area extracted by superposing an RPB file and the ground control point file; calculating canopy structure characteristic parameters of the target forest area, and generating a digital surface model DSM and a digital elevation model DEM of the target forest area; and in combination with image calculation of raster data, subtracting the obtained digital elevation model from the obtained digital surface model to obtain a canopy height DHM, and calculating the tree canopy height from a three-dimensional angle to realize measurement of the tree canopy height.

Systems and methods for property feature detection and extraction using digital images. The image sources could include aerial imagery, satellite imagery, ground-based imagery, imagery taken from unmanned aerial vehicles (UAVs), mobile device imagery, etc. The detected geometric property features could include tree canopy, pools and other bodies of water, concrete flatwork, landscaping classifications (gravel, grass, concrete, asphalt, etc.), trampolines, property structural features (structures, buildings, pergolas, gazebos, terraces, retaining walls, and fences), and sports courts. The system can automatically extract these features from images and can then project them into world coordinates relative to a known surface in world coordinates (e.g., from a digital terrain model).

The invention discloses a picking point detection model construction method and a picking point positioning method based on machine vision, and belongs to the technical field of artificial intelligence detection. Collecting a multi-angle RGB image and a depth image of a picking target, generating a first point cloud set of the picking target through visual angle mapping, and cooperating with the first point cloud set as a training set; a three-dimensional reconstruction method is adopted to reconstruct a second point cloud set of a picking body tree canopy in real time to cooperate as a verification set and a test set, and a picking point detection model Alpha-VoteNet is trained. During picking, a third point cloud set of a picking object tree canopy layer is subjected to three-dimensional reconstruction in real time by adopting a three-dimensional reconstruction method, picking area marking frames of all picking targets of the picking object tree canopy layer are obtained by utilizing the picking point detection model, and the bottom surface center is calculated according to the top points of the bottom surfaces of the marking frames to obtain the three-dimensional coordinates of the picking points. The method overcomes the problems that in a traditional method, three-dimensional information of a picking target cannot be obtained, and the recognition precision is not high due to obstacle shielding, and has the advantages of being high in detection precision, high in recognition speed, high in generalization ability and the like.

The invention discloses a method for quickly extracting the height of tree obstacles in power grid transmission lines, which adopts sub-meter high-resolution optical stereo image and RPC files, combined with high-precision control points collected on the ground, and extracts high-precision surface DSM and points through processing. Cloud data; on this basis, the filter classification technology is used to classify the point cloud data, and the ground points in the classification result are corrected by integrating the features of the ground objects in the satellite image, and the DEM is generated by interpolation; finally, the spatial difference based on DSM and DEM is established. The model's tree canopy height extraction method is used to verify the accuracy of the method through field collection points. The invention solves the problem that the height extraction precision is out of control due to DEM in the original spatial difference calculation model.

The invention discloses a tree crown extraction method based on multi-source remote sensing of UAV, which includes obtaining visible light image and laser radar point cloud, taking the visible light orthophoto image of UAV and laser radar point cloud as data sources, adopting watershed segmentation and oriented The method of combining object multi-scale segmentation to extract single tree canopy information under different canopy density. First, the object-oriented multi-scale segmentation method is used to extract the canopy and non-canopy areas of UAV visible light images, and the canopy distribution range is extracted using the canopy area as a mask; Based on the CHM constructed by UAV lidar, the watershed segmentation method is used to obtain the preliminary segmentation results of the single tree canopy; based on the canopy boundary of the watershed segmentation, and the brightness value of the visible light orthophoto image is characterized, the canopy area of ​​the visible light orthophoto image is binary. Subdivided to obtain the optimized single tree canopy boundary information, which greatly increases the accuracy of remote sensing canopy extraction.

The invention relates to the field of laser radar data processing, discloses a DBI tree vertex detection method for a canopy height model of airborne laser radar point cloud data, and invents a method for detecting a tree vertex based on a DBI control foreground pixel of a canopy height canopy model , including: 1. Standardize the marker pixel generation conditions of the canopy height model; 2. Use the gray level change characteristics of the height canopy model to filter and eliminate pseudo foreground pixels; 3. Introduce the similarity judgment factor DBI to filter the tree canopy with DBI-K vertex. The invention provides a tree top position detection method for a canopy height model of airborne lidar point cloud data, which can effectively solve the threshold dependence problem of the traditional window detection tree top method and improve the recognition accuracy of the tree top.

The invention discloses a method for obtaining forest canopy parameters of artificial forests by using two-dimensional images, which is suitable for the investigation of artificial forest resources and belongs to the technical field of photogrammetry in forestry applications. The key technical points include: 1. Extracting the tree point cloud model Useful information; 2. Use a specific method to extract information about the relative relationship between trees; 3. Use a specific estimation method to calculate the tree canopy parameters in the target area. The key problems to be solved include: 1. Only need to take pictures of the forest land to obtain the sample plot information, reduce the workload of the ground survey and improve the survey efficiency; 2. Solve the problem that close-range photogrammetry cannot obtain high-density forest environment forest canopy due to the occlusion of forest branches and leaves information problem. The invention is suitable for plantations in high-density forest environments, and it is the first time to use the method of combining photogrammetry technology and crown width estimation method to extract single-tree canopy information of high-density forest environment plantation forests. The results can provide basic data basis for rapid and accurate investigation of plantation forest resources.

The invention discloses a secondary inarching cultivation method for controlling passion fruit stem base rot. The method comprises the following steps of (1) selection of an anti-stem base rot virus-free grafting seedling, (2) selection of an inarching variety, (3) soil preparation and ground under tree canopy arrangement, (4) planting, (5) vine guiding, pinching and topping, (6) primary inarching, (7) secondary inarching and (8) cutting of a stem base rot occurrence part. Influences of the passion fruit stem base rot can be controlled; three anti-stem base rot main branches provide nutrientsfor each fruit-bearing plant, the plant growth is promoted significantly, the yield is improved, if one branch is infected with the stem base rot, the stem of the infected part needs to be timely cutoff, the planted plant can be ensured not to die due to infection of the stem base rot within one year, the purposes of high quality, high efficiency and high yield of passion fruits can be realized,and the practical significance in production is achieved.

The embodiment of the invention discloses a tree canopy spray collecting and processing system and method. The system comprises a spraying monitoring device, at least two fog drop collecting devices and a server; the spraying monitoring device is configured to send a fog drop collecting starting instruction to the fog drop collecting devices if it is monitored that spraying operation equipment reaches a target spraying area of a target tree canopy to start spraying operation; the fog drop collecting devices are configured to respond to the fog drop collecting starting instruction and effectively collect fog drops at the corresponding canopy part, and shooting to obtain a target fog drop image under the condition that fog drop collecting is stopped; and the server is configured to analyze the deposition coverage rate and the deposition density of the fog drops in the target fog drop image, determine the effective fog drop collecting duration of the fog drop collecting devices at the corresponding canopy part, and determine the unit time deposition coverage rate and the unit time deposition density of the corresponding canopy part for evaluating the spraying effect of the target treecanopy. By the adoption of the scheme, effective fog drops can be collected at all canopy positions, and inaccurate measurement and analysis caused by too many collected fog drops are avoided.

The invention discloses a profiling variable sprayer based on plant canopy phenotype and an adaptive control method. The profiling variable sprayer includes a profiling mechanism module A, a profiling mechanism module B, a lifting and sliding module, a profiling bracket module, Liquid supply system, electrical box and detection module; on the basis of dividing the function and mechanism of the profiling sprayer, it adopts a modular and lightweight design. The whole machine has the characteristics of compact structure, lightness and flexibility, easy manufacture and easy maintenance. . The adaptive control method of the profiling variable sprayer automatically collects the ultrasonic echo signal and processes it to obtain the characteristic information of the tree canopy, and then calculates the air volume according to the canopy volume and canopy density phenotype characteristic parameters, The spray volume is calculated by the leaf area density phenotypic characteristic parameters; the profiling control amount calculated by the ultrasonic distance measurement information drives the profiling mechanism to profil, and realizes the automatic profiling of the crown contour, with high profiling accuracy, accurate and efficient spraying, Save medicine and environmental protection, easy to operate.

The invention discloses a street tree target recognition method based on vehicle-mounted 2D LiDAR point cloud data, which includes using the vehicle-mounted 2D LiDAR system to obtain street data, saving it as a point cloud file, marking the point cloud file with tree crowns and non-tree crown categories; constructing a variable scale Grid, extract the three-dimensional spherical domain from the variable-scale grid; calculate the point cloud feature vector in the three-dimensional spherical domain, and obtain the point cloud feature vector set; the SVM algorithm learns the tree crown point cloud classifier from the point cloud feature vector set; according to the tree crown point cloud The classifier performs online recognition on the point cloud frame to obtain the spray prescription map. The invention adopts the vehicle-mounted 2D LiDAR with small data redundancy and easy online processing to acquire urban street data, and provides accurate spray basis for target application. Establish a point cloud sequence spatial index structure that preserves data accuracy, improves neighborhood retrieval efficiency, and meets online processing requirements to achieve accurate identification of street tree targets.

The invention discloses an air-assisted orchard spraying device, which comprises a plurality of spray heads, a blower and a liquid medicine tank; all the spray heads are arranged in an orderly manner; each spray head has an air chamber and Atomizer; the air chamber and the atomizer are respectively connected to the blower and the liquid medicine tank, and a liquid delivery pump is connected between the liquid medicine tank and the atomizer; the atomizer The atomizer is placed inside the air chamber, and the direction of the mist outlet of the atomizer is consistent with the direction of the air outlet of the air chamber. The air-assisted orchard spraying device of the present invention is provided with an air-assisted spray head and a blower, so that the airflow provided by the blower can engulf the medicine mist and spray through the spray head, so that the medicine mist can be sprayed to a higher crown sprayer. The powerful airflow of the nozzle blows the mist to all parts of the plant canopy. The airflow helps to increase the range of the orchard sprayer, assists the mist to penetrate the dense canopy, and promotes the turning of the leaves to increase the attachment rate of the leaves.

The invention belongs to the technical field of monitoring systems, and discloses an air-ground integrated mangrove monitoring system and a control method, which obtain high-resolution aerial images of statistical target areas through unmanned aerial vehicles or other aerial photography means, and analyze the structural distribution of mangroves; Collect hydrological and meteorological environmental factors to conduct meteorological and hydrological detection in the area; according to the data detected in step 1, open the image with high-resolution image tree crown extraction software for processing, and compare it with the original image to determine the number of mangroves; detect according to step 1 Combined with image spectrum and texture features, the pest status of mangroves is estimated; statistical system sampling principle is used to count the species of mangroves in a single area, and according to the number of detected animal species, the overall mangroves The type of action in the woods is evaluated to derive the type of animal. The invention adopts remote sensing information technology to monitor mangroves and transmit information, thereby improving the quality and efficiency of monitoring.

The invention relates to a method for estimating the height of a single tree in a forest by combining LiDAR point clouds and synchronous remote sensing images: obtaining low-density airborne LiDAR point cloud data and synchronous high-resolution aerial remote sensing images in forest areas. By mosaicing and orthorectifying the synchronous aerial remote sensing images of the target area, the digital orthophoto map DOM of the target area is obtained; the single tree canopy position, canopy boundary information and single tree canopy information are obtained from the digital orthophoto map; The airborne LiDAR point cloud data is processed, combined with the obtained single tree canopy information to obtain the point cloud data corresponding to the single tree canopy; based on the above results, a three-dimensional tree height model is constructed; according to the number of laser points in the single tree canopy is greater than 1 , equal to 1 and no laser point to optimize and reconstruct the three-dimensional tree model, so as to calculate the height value of the corresponding forest canopy vertex. Raise the tree height estimation level to the single tree level, and effectively improve the accuracy of forest stand single tree height estimation.

The invention discloses a method for precisely extracting tree canopy temperature based on infrared thermal images. The method comprises the following steps: detecting and obtaining infrared thermal image data of a tree to be detected; extracting all pixel point temperature values in the infrared thermal image data; according to the pixel point temperature values, converting infrared thermal images from color images into grayscale images; according to canopy structure features, determining area range of an irregular tree canopy through corresponding area shapes; and within the area range of the tree canopy, and according to gray values in the grayscale images, extracting corresponding number of pixel points, and obtaining distribution state of the tree canopy temperature. The method can reduce interference of redundant data, especially interference of soil background, and the extracted pixel points can accurately reflect the tree canopy temperature. Therefore, the method for preciselyextracting tree canopy temperature based on the infrared thermal images can accurately extract the tree canopy temperature.

The invention discloses small landscape tree crown dust removing equipment. The small landscape tree crown dust removing equipment comprises a dust removing box. A dust removing cavity with a downwardopening is formed in the dust removing box. An annular wrapping airbag is fixedly arranged on the inner wall of the dust removing cavity. The left side of the wrapping airbag is higher than the lowerside of the wrapping airbag. An annular cavity is formed in the dust removing box and communicates with a dust collector through a communicating pipe. The dust collector is activated to inflate the wrapping airbag so as to wrap a trunk. By means of the small landscape tree crown dust removing equipment, a tree crown of a landscape tree can be wrapped. Air can be blown to the crown to separate dust accumulated on leaves from the leaves, and the dust separated from the leaves is absorbed by a dust suction mechanism, so that the dust on the crown can be cleaned. When the dust is large in quantity and difficult to clean, after dust cleaning is completed by the small landscape tree crown dust removing equipment, the crown can be washed with a small amount of water, and the water is guided intoa flowerpot so as to be utilized and prevented from sputtering all around.