CN113221788B - Method and device for extracting ridge culture characteristics of field blocks - Google Patents
Method and device for extracting ridge culture characteristics of field blocks Download PDFInfo
- Publication number
- CN113221788B CN113221788B CN202110557228.0A CN202110557228A CN113221788B CN 113221788 B CN113221788 B CN 113221788B CN 202110557228 A CN202110557228 A CN 202110557228A CN 113221788 B CN113221788 B CN 113221788B
- Authority
- CN
- China
- Prior art keywords
- image
- ridge
- field
- remote sensing
- line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
Abstract
The invention relates to a field block ridge culture feature extraction method and device, which are characterized in that a full-color remote sensing image is obtained, Laplace calculation and binarization are carried out on the full-color remote sensing image to obtain a ridge line image, ridge lines in the ridge line image are thinned by adopting an image thinning method to obtain a line central line, the line central line is cut into line segments, the line segments are screened according to a preset curvature threshold value to obtain line segments, an optical remote sensing satellite image is obtained, a field block image is obtained by identification from the optical remote sensing satellite image, and the ridge length, the ridge distance and the ridge direction corresponding to the field block in the field block image are obtained according to the line segments and the field block image, so that the auxiliary judgment of the crop type of the field block is facilitated, and the crop type identification precision is improved.
Description
Technical Field
The invention relates to the technical field of agricultural cultivation, in particular to a method and a device for extracting field block ridge culture characteristics.
Background
The field tillage characteristic is a basic characteristic for judging the planting type and the crop rotation mode, and is also an auxiliary characteristic for knowing the growth vigor and the yield of crops. For the description of the field tillage characteristics, a Gray-level Co-occurrence Matrix (GLCM for short) statistical texture characteristic is usually adopted, and the characteristic is originally developed for the application of the land cover and represents the disorder degree of the texture.
However, structural textural features are lacking to describe knowledge that is closely related to crop type, such as ridge culture, direction, spacing, etc.
Disclosure of Invention
Based on this, the present invention aims to provide a method and an apparatus for extracting features of field ridge culture, which have the advantage of improving the accuracy of crop type discrimination.
In order to achieve the above object, a first aspect of the present invention provides a field ridge feature extraction method, including:
acquiring a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image;
thinning ridge lines in the ridge line image by adopting an image thinning method to obtain line center lines;
cutting the line central line into line segments, and screening the line segments according to a preset curvature threshold value to obtain straight line segments;
acquiring an optical remote sensing satellite image, and identifying and acquiring a field image from the optical remote sensing satellite image;
and calculating the ridge length, the ridge distance and the ridge direction corresponding to the field in the field image according to the straight line segment and the field image.
Further, the step of obtaining a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image comprises: acquiring a panchromatic remote sensing image, and performing Laplace calculation on the panchromatic remote sensing image by adopting a Laplace filtering template; comparing the pixel value of the panchromatic remote sensing image after the Laplace calculation with a threshold value 0, and carrying out binarization processing on the panchromatic remote sensing image to obtain a ridge line image; when the pixel value is greater than the threshold 0, the pixel value is set to 1, and when the pixel value is less than the threshold 0, the pixel value is set to 0.
Further, the step of truncating the line center line into line segments and screening the line segments according to a preset curvature threshold to obtain straight line segments comprises: calculating the connectivity of each pixel in the line central line, deleting the pixels of which the connectivity is greater than or equal to a first preset threshold value, and cutting the line central line into N line segments; acquiring the line segments with the pixel number larger than a second preset threshold; calculating the curvature of the line segment, and determining the line segment with the curvature larger than a preset curvature threshold value as a straight line segment; and calculating the curvature of the line segment according to the ratio of the number of pixels of the line segment to the number of pixels corresponding to a straight line formed by connecting two end points of the line segment.
Further, the step of acquiring an optical remote sensing satellite image and identifying and acquiring a field image from the optical remote sensing satellite image comprises: acquiring an optical remote sensing satellite image, and inputting the optical remote sensing satellite image into a trained field block recognition model to obtain a field block image; wherein training the field recognition model comprises: sketching a field block boundary of the optical remote sensing satellite sample image to obtain field block sample data; and inputting the optical remote sensing satellite sample image as input and the field sample data as output into an FCIS deep learning network for training and learning to obtain a field recognition model.
Further, the step of calculating the ridge length, the ridge distance and the ridge direction corresponding to the field in the field image according to the straight line segment and the field image comprises: acquiring a first image layer corresponding to the straight line segment and a second image layer corresponding to the field image, and matching the first image layer and the second image layer according to longitude and latitude to determine the straight line segment corresponding to the field in the field image; calculating the pixel number of each straight line segment to obtain the ridge length of each straight line segment, and taking the maximum ridge length in the ridge lengths as the ridge length of the field block; calculating the distance between each straight line segment and the left and right adjacent line segments, and taking the median of the distance as the ridge distance of the field; and calculating the angle of each straight line segment, and taking the median of the angle as the ridge direction of the field block.
A second aspect of the present invention provides a field ridge culture feature extraction device, including:
the line image unit is used for acquiring a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image;
the thinning unit is used for thinning the ridge lines in the ridge line image by adopting an image thinning method to obtain line center lines;
the screening unit is used for cutting the line center line into line segments and screening the line segments according to a preset curvature threshold value to obtain straight line segments;
the field identification unit is used for acquiring an optical remote sensing satellite image and identifying and acquiring a field image from the optical remote sensing satellite image;
and the calculation unit is used for calculating the ridge length, the ridge distance and the ridge direction corresponding to the field in the field image according to the straight line segment and the field image.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a field block ridge culture feature extraction method and device, which are characterized in that a full-color remote sensing image is obtained, Laplace calculation and binarization are carried out on the full-color remote sensing image to obtain a ridge line image, ridge lines in the ridge line image are thinned by adopting an image thinning method to obtain a line central line, the line central line is cut into line segments, the line segments are screened according to a preset curvature threshold value to obtain line segments, an optical remote sensing satellite image is obtained, a field block image is obtained by identification from the optical remote sensing satellite image, and the ridge length, the ridge distance and the ridge direction corresponding to the field block in the field block image are obtained according to the line segments and the field block image, so that the auxiliary judgment of the crop type of the field block is facilitated, and the crop type identification precision is improved.
Drawings
FIG. 1 is a schematic flow chart of the ridge culture feature extraction method of the present invention;
FIG. 2 is a schematic flow chart of S10 in the method for extracting ridge culture characteristics of field blocks according to the present invention;
FIG. 3 is a schematic flow chart of S30 in the method for extracting ridge culture characteristics of field blocks according to the present invention;
FIG. 4 is a schematic flow chart of S41 in the method for extracting ridge culture characteristics of field blocks according to the present invention;
FIG. 5 is a schematic flow chart of S50 in the method for extracting ridge culture characteristics of field blocks according to the present invention;
FIG. 6 is a block diagram of the block ridge culture feature extraction apparatus of the present invention;
fig. 7 is a block diagram of a line image unit 70 of the ridge culture feature extraction device of the present invention;
fig. 8 is a block diagram of the truncating and screening unit 90 of the ridge culture feature extraction device of the field blocks of the present invention;
fig. 9 is a block diagram of the input unit 102 of the ridge culture feature extraction device of the present invention;
fig. 10 is a block diagram of the computation unit 110 of the ridge culture feature extraction device of the present invention.
Detailed Description
For a better understanding and practice, the invention is described in detail below with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a method for extracting ridge culture characteristics of field blocks, including the following steps:
and S10, acquiring a panchromatic remote sensing image, and performing Laplace calculation and binarization on the panchromatic remote sensing image to obtain a ridge line image.
Full-color bands, typically a single band around 0.5 to 0.75 microns, i.e., the visible band from green onward, are used. The panchromatic remote sensing image is an image pickup of a panchromatic waveband in ground object radiation, and is a single waveband, so that a gray level picture is displayed on the image. In the embodiment of the application, a panchromatic remote sensing image is obtained, and Laplace calculation and binarization are performed on the panchromatic remote sensing image to obtain a preprocessed image, namely a ridge line image.
In an alternative embodiment, referring to fig. 2, the step S10 includes steps S11-S12, which are as follows:
and S11, acquiring a full-color remote sensing image, and performing Laplace calculation on the full-color remote sensing image by adopting a Laplace filtering template.
The Laplace algorithm is a linear second-order differential algorithm, namely, the sum of the values of 4 adjacent pixels at the upper part, the lower part, the left part and the right part is added, and four times of the value of the pixel is subtracted to be used as the brightness value of the pixel. In the embodiment of the application, a Laplace filtering template is adopted to perform Laplace calculation on the panchromatic remote sensing image, wherein the Laplace filtering template is as follows:
s12, comparing the pixel value of the panchromatic remote sensing image after Laplace calculation with a threshold value 0, and performing binarization processing on the panchromatic remote sensing image to obtain a ridge line image; when the pixel value is greater than the threshold 0, the pixel value is set to 1, and when the pixel value is less than the threshold 0, the pixel value is set to 0.
The image binarization is a process of setting the gray value of a pixel point on an image to be 0 or 255, namely, the whole image presents an obvious black and white effect. In the embodiment of the application, the pixel value in the panchromatic remote sensing image after the Laplace calculation is compared with the threshold value 0, the pixel value is a positive value, the pixel value is reset to 1, the pixel value is a negative value, the pixel value is reset to 0, and the binarized image is the ridge line image.
And S20, thinning the ridge lines in the ridge line image by adopting an image thinning method to obtain line center lines.
Image Thinning (Image Thinning), which refers to an operation of skeletonization (Image Skeletonizing) of a binary Image, is to remove some points from an original Image through layer-by-layer peeling, but still maintain the original shape until a skeleton of the Image is obtained, which can be understood as a central axis of the Image. In the embodiment of the application, the ridge lines in the ridge line image are thinned by adopting an image thinning method to obtain line center lines. The image thinning method can be an image thinning algorithm in MATLAB and SKINMAGE, and pixels of which the line edges meet certain conditions are repeatedly deleted from ridge lines in the ridge line image, so that a line center line with a single pixel width is finally obtained.
And S30, cutting the line central line into line segments, and screening the line segments according to a preset curvature threshold value to obtain straight line segments.
In the embodiment of the application, the obtained line center line has an intersection part, the line center line with the intersection point is cut into line segments, the line segments comprise partial curved line segments, and the line segments are screened according to a preset curvature threshold value to obtain straight line segments.
In an alternative embodiment, referring to fig. 3, the step S30 includes steps S31-S33, which are as follows:
s31, calculating the connectivity of each pixel in the line central line, and deleting the pixels of which the connectivity is greater than or equal to a first preset threshold value to cut the line central line into N line segments;
s32, acquiring the line segments with the pixel number larger than a second preset threshold value;
s33, calculating the curvature of the line segment, and determining the line segment with the curvature larger than a preset curvature threshold value as a straight line segment; and calculating the curvature of the line segment according to the ratio of the number of pixels of the line segment to the number of pixels corresponding to a straight line formed by connecting two end points of the line segment.
In the embodiment of the present application, the connectivity of each pixel in the line center line is calculated, that is, whether 8 fields corresponding to a certain pixel include more than 2 pixels is calculated, pixels with connectivity greater than or equal to 3 are deleted, and the line center line is cut into a plurality of line segments. And reserving the line segments with the pixel number of more than or equal to 20, calculating the curvature of the reserved line segments, and determining the line segments with the curvature of more than or equal to 0.99 as straight line segments. Wherein, the curvature calculation formula is as follows:
nl is the number of pixels of the line segment, n is the number of pixels corresponding to a straight line formed by connecting two end points of the line segment, and b is the curvature of the line segment.
S40, obtaining an optical remote sensing satellite image, and identifying and obtaining a field image from the optical remote sensing satellite image.
In an alternative embodiment, the step S40 includes step S41, which is as follows:
s41, obtaining an optical remote sensing satellite image, inputting the optical remote sensing satellite image into a trained field block recognition model, and obtaining a field block image.
In the embodiment of the application, the optical remote sensing satellite image acquired from Google Earth is input into a trained field recognition model, and all fields in the optical remote sensing satellite image are recognized, so that the field image is obtained.
In an alternative embodiment, referring to fig. 4, the steps S41 including S412 to S414 include:
s412, sketching a field block boundary of the optical remote sensing satellite sample image to obtain field block sample data;
and S414, inputting the optical remote sensing satellite sample image as input, and inputting the field block sample data as output into an FCIS deep learning network for training and learning to obtain a field block identification model.
In the embodiment of the application, a large number of optical remote sensing satellite sample images are collected in advance, then field block boundaries in the sample images are manually outlined one by one aiming at each optical remote sensing satellite sample image to obtain field block sample data, the optical remote sensing satellite sample images are used as input, the field block sample data are used as output and input to a full volume example perception Semantic Segmentation (FCIS) deep learning network for training and learning, and therefore a field block identification model is obtained.
S50, calculating ridge length, ridge distance and ridge direction corresponding to the field in the field image according to the straight line segment and the field image.
In the embodiment of the application, the straight line segments are classified into each field block in the field block image according to the field block image so as to calculate the ridge length, the ridge distance and the ridge direction corresponding to the field block in the field block image.
In an alternative embodiment, referring to fig. 5, the step S50 includes steps S51 to S54, which are as follows:
s51, acquiring a first image layer corresponding to the straight line segment and a second image layer corresponding to the field image, and matching the first image layer and the second image layer according to longitude and latitude to determine the straight line segment corresponding to the field in the field image;
s52, calculating the number of pixels of each straight line segment to obtain the ridge length of each straight line segment, and taking the maximum ridge length in the ridge lengths as the ridge length of the field block;
s53, calculating the distance between each straight line segment and the left and right adjacent line segments, and taking the median of the distance as the ridge distance of the field block;
s54, calculating the angle of each straight line segment, and taking the median of the angle as the ridge direction of the field block.
In this embodiment of the present application, a first layer corresponding to the linear segment and a second layer corresponding to the field image are matched according to longitude and latitude, that is, the linear segment is classified into each field in the field image.
And calculating the number of pixels contained in each straight line segment, and multiplying the number of pixels by the width of each pixel to obtain the ridge length. Each field block comprises a plurality of straight line segments, ridge lengths corresponding to the straight line segments are compared with each other, and the largest ridge length is determined to be the ridge length of the field block. And calculating the distance between each straight line segment and the left and right adjacent straight line segments, sequencing the distances from small to large, and taking the median of the distances as the ridge distance of the field block. When the angle of each straight line segment is calculated, a plane coordinate system is set, for example, a field for cultivating crops in the longitudinal direction is taken as an example, an included angle between each straight line segment and a vertical line in the plane coordinate system is calculated, the angles of the included angle are sorted from small to large, and the median of the angles is taken as the ridge direction of the field.
By applying the embodiment of the invention, the full-color remote sensing image is obtained, the Laplace calculation and the binarization are carried out on the full-color remote sensing image to obtain the ridge line image, the ridge lines in the ridge line image are thinned by adopting an image thinning method to obtain the line center line, the line center line is cut into line segments, the line segments are screened according to the preset curvature threshold value to obtain straight line segments, the optical remote sensing satellite image is obtained, the field block image is obtained by identifying from the optical remote sensing satellite image, and the ridge length, the ridge distance and the ridge direction corresponding to the field block in the field block image are obtained according to the straight line segments and the field block image, so that the crop type of the field block is conveniently and auxiliarily judged, and the crop type identification precision is improved.
Referring to fig. 6, an embodiment of the present invention provides a device 60 for extracting ridge culture characteristics of field blocks, including:
the line image unit 70 is used for acquiring a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image;
a thinning unit 80, configured to thin ridge lines in the ridge line image by using an image thinning method to obtain line center lines;
the truncation and screening unit 90 is used for truncating the line central line into a line segment and screening the line segment according to a preset curvature threshold value to obtain a straight line segment;
the field block identification unit 100 is used for acquiring an optical remote sensing satellite image and identifying and acquiring a field block image from the optical remote sensing satellite image;
and the calculating unit 110 is configured to calculate, according to the straight line segment and the field image, a ridge length, a ridge distance, and a ridge direction corresponding to a field in the field image.
Optionally, referring to fig. 7, the line image unit 70 specifically includes:
the Laplace calculation unit 72 is used for acquiring a panchromatic remote sensing image and performing Laplace calculation on the panchromatic remote sensing image by adopting a Laplace filtering template;
a binarization unit 74, configured to compare a pixel value of the panchromatic remote sensing image after Laplace calculation with a threshold value 0, and perform binarization processing on the panchromatic remote sensing image to obtain a ridge line image; when the pixel value is greater than the threshold 0, the pixel value is set to 1, and when the pixel value is less than the threshold 0, the pixel value is set to 0.
Optionally, referring to fig. 8, the truncating and screening unit 90 specifically includes:
a line segment unit 92, configured to calculate a connectivity of each pixel in the line centerline, delete the pixels whose connectivity is greater than or equal to a first predetermined threshold, and truncate the line centerline into N line segments;
an obtaining unit 94, configured to obtain the line segment with the number of pixels greater than a second predetermined threshold;
a straight line segment unit 96, configured to calculate curvature of the line segment, and determine a line segment with curvature greater than a preset curvature threshold as a straight line segment; and calculating the curvature of the line segment according to the ratio of the number of pixels of the line segment to the number of pixels corresponding to a straight line formed by connecting two end points of the line segment.
Optionally, the identifying field block unit 100 specifically includes:
the input unit 102 is used for acquiring an optical remote sensing satellite image, inputting the optical remote sensing satellite image into a trained field block recognition model, and acquiring a field block image;
optionally, referring to fig. 9, the input unit 102 specifically includes:
the delineating unit 1022 is configured to delineate a field boundary of the optical remote sensing satellite sample image to obtain field sample data;
and a training learning unit 1024, configured to input the optical remote sensing satellite sample image as input and the field sample data as output to an FCIS deep learning network for training and learning, so as to obtain a field recognition model.
Optionally, referring to fig. 10, the calculating unit 110 specifically includes:
a matching unit 112, configured to obtain a first layer corresponding to the screened line segment and a second layer corresponding to the field image, match the first layer and the second layer according to longitude and latitude, and determine the straight line segment corresponding to the field in the field image;
the ridge length unit 114 is configured to calculate the number of pixels of each line segment to obtain the ridge length of each line segment, and use the maximum ridge length in the ridge lengths as the ridge length of the field block;
a ridge distance unit 116, configured to calculate a distance between each line segment and a left-right adjacent line segment, and use a median of the distances as a ridge distance of the field;
a ridge unit 118, configured to calculate an angle of each line segment, and use a median of the angles as a ridge direction of the field.
By applying the embodiment of the invention, the full-color remote sensing image is obtained, the Laplace calculation and the binarization are carried out on the full-color remote sensing image to obtain the ridge line image, the ridge lines in the ridge line image are thinned by adopting an image thinning method to obtain the line center line, the line center line is cut into line segments, the line segments are screened according to the preset curvature threshold value to obtain straight line segments, the optical remote sensing satellite image is obtained, the field block image is obtained by identifying from the optical remote sensing satellite image, and the ridge length, the ridge distance and the ridge direction corresponding to the field block in the field block image are obtained according to the straight line segments and the field block image, so that the crop type of the field block is conveniently and auxiliarily judged, and the crop type identification precision is improved.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, to those skilled in the art, changes and modifications may be made without departing from the spirit of the present invention, and it is intended that the present invention encompass such changes and modifications.
Claims (8)
1. A field block ridge culture feature extraction method is characterized by comprising the following steps:
acquiring a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image;
thinning ridge lines in the ridge line image by adopting an image thinning method to obtain line center lines;
cutting the line central line into line segments, and screening the line segments according to a preset curvature threshold value to obtain straight line segments; calculating the connectivity of each pixel in the line central line, and deleting the pixels of which the connectivity is greater than or equal to a first preset threshold value so as to cut off the line central line into N line segments; acquiring the line segments with the pixel number larger than a second preset threshold; calculating the curvature of the line segment, and determining the line segment with the curvature larger than a preset curvature threshold value as a straight line segment; calculating the curvature of the line segment according to the ratio of the number of pixels of the line segment to the number of pixels corresponding to a straight line formed by connecting two end points of the line segment;
acquiring an optical remote sensing satellite image, and identifying and acquiring a field image from the optical remote sensing satellite image;
and calculating the ridge length, the ridge distance and the ridge direction corresponding to the field in the field image according to the straight line segment and the field image.
2. The method for extracting ridge culture characteristics of farmland blocks according to claim 1, wherein the step of obtaining a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image comprises the following steps of:
acquiring a panchromatic remote sensing image, and performing Laplace calculation on the panchromatic remote sensing image by adopting a Laplace filtering template;
comparing the pixel value of the panchromatic remote sensing image after the Laplace calculation with a threshold value 0, and carrying out binarization processing on the panchromatic remote sensing image to obtain a ridge line image; when the pixel value is greater than the threshold 0, the pixel value is set to 1, and when the pixel value is less than the threshold 0, the pixel value is set to 0.
3. The method for extracting ridge culture features in farmland blocks as claimed in claim 1, wherein the step of obtaining an optical remote sensing satellite image and identifying and obtaining the farmland block image from the optical remote sensing satellite image comprises:
acquiring an optical remote sensing satellite image, and inputting the optical remote sensing satellite image into a trained field block recognition model to obtain a field block image;
wherein training the field recognition model comprises:
sketching a field block boundary of the optical remote sensing satellite sample image to obtain field block sample data;
and inputting the optical remote sensing satellite sample image as input and the field sample data as output into an FCIS deep learning network for training and learning to obtain a field recognition model.
4. The method for extracting features of ridge culture of field blocks according to claim 1, wherein the step of calculating the ridge length, ridge distance and ridge direction corresponding to the field block in the field block image according to the straight line segment and the field block image comprises:
acquiring a first image layer corresponding to the straight line segment and a second image layer corresponding to the field image, and matching the first image layer and the second image layer according to longitude and latitude to determine the straight line segment corresponding to the field in the field image;
calculating the pixel number of each straight line segment to obtain the ridge length of each straight line segment, and taking the maximum ridge length in the ridge lengths as the ridge length of the field block;
calculating the distance between each straight line segment and the left and right adjacent line segments, and taking the median of the distance as the ridge distance of the field;
and calculating the angle of each straight line segment, and taking the median of the angle as the ridge direction of the field block.
5. A field block ridge culture feature extraction device is characterized by comprising:
the line image unit is used for acquiring a full-color remote sensing image, and performing Laplace calculation and binarization on the full-color remote sensing image to obtain a ridge line image;
the thinning unit is used for thinning the ridge lines in the ridge line image by adopting an image thinning method to obtain line center lines;
the truncation and screening unit is used for truncating the line center line into a line segment and screening the line segment according to a preset curvature threshold value to obtain a straight line segment; calculating the connectivity of each pixel in the line central line, deleting the pixels of which the connectivity is greater than or equal to a first preset threshold value, and cutting the line central line into N line segments; acquiring the line segments with the pixel number larger than a second preset threshold; calculating the curvature of the line segment, and determining the line segment with the curvature larger than a preset curvature threshold value as a straight line segment; calculating the curvature of the line segment according to the ratio of the number of pixels of the line segment to the number of pixels corresponding to a straight line formed by connecting two end points of the line segment;
the field identification unit is used for acquiring an optical remote sensing satellite image and identifying and acquiring a field image from the optical remote sensing satellite image;
and the calculation unit is used for calculating the ridge length, the ridge distance and the ridge direction corresponding to the field in the field image according to the straight line segment and the field image.
6. The ridge culture feature extraction device of claim 5, wherein the obtaining unit comprises:
the Laplace calculation unit is used for acquiring a full-color remote sensing image and performing Laplace calculation on the full-color remote sensing image by adopting a Laplace filtering template;
the binarization unit is used for comparing the pixel value of the panchromatic remote sensing image after the Laplace calculation with a threshold value 0, and performing binarization processing on the panchromatic remote sensing image to obtain a ridge line image; when the pixel value is greater than the threshold 0, the pixel value is set to 1, and when the pixel value is less than the threshold 0, the pixel value is set to 0.
7. The ridge culture feature extraction device of claim 5, wherein the identification unit comprises:
the input unit is used for acquiring an optical remote sensing satellite image, inputting the optical remote sensing satellite image into a trained field block recognition model and acquiring a field block image;
wherein training the field recognition model comprises:
the delineating unit is used for delineating a field block boundary of the optical remote sensing satellite sample image to obtain field block sample data;
and the training learning unit is used for inputting the optical remote sensing satellite sample image and the field sample data into an FCIS deep learning network for training learning to obtain a field recognition model.
8. The ridge culture feature extraction device of claim 5, wherein the calculation unit comprises:
the matching unit is used for acquiring a first image layer corresponding to the screened line segment and a second image layer corresponding to the field image, matching the first image layer and the second image layer according to longitude and latitude, and determining the straight line segment corresponding to the field in the field image;
the ridge length unit is used for calculating the pixel number of each line segment to obtain the ridge length of each line segment, and the maximum ridge length in the ridge lengths is used as the ridge length of the field block;
the ridge distance unit is used for calculating the distance between each line segment and the left and right adjacent line segments, and taking the median of the distance as the ridge distance of the field block;
and the ridge direction unit is used for calculating the angle of each line segment and taking the median of the angle as the ridge direction of the field block.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110557228.0A CN113221788B (en) | 2021-05-21 | 2021-05-21 | Method and device for extracting ridge culture characteristics of field blocks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110557228.0A CN113221788B (en) | 2021-05-21 | 2021-05-21 | Method and device for extracting ridge culture characteristics of field blocks |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113221788A CN113221788A (en) | 2021-08-06 |
| CN113221788B true CN113221788B (en) | 2022-06-03 |
Family
ID=77093711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110557228.0A Active CN113221788B (en) | 2021-05-21 | 2021-05-21 | Method and device for extracting ridge culture characteristics of field blocks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113221788B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116630650B (en) * | 2023-07-20 | 2023-09-29 | 武汉理工大学 | Ridge width identification method and unmanned vehicle device for ridges |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6684100B1 (en) * | 2000-10-31 | 2004-01-27 | Cardiac Pacemakers, Inc. | Curvature based method for selecting features from an electrophysiologic signals for purpose of complex identification and classification |
| CN107480706A (en) * | 2017-07-24 | 2017-12-15 | 中国农业大学 | A kind of seed production corn field remote sensing recognition method and device |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103559493A (en) * | 2013-10-21 | 2014-02-05 | 中国农业大学 | Method for extracting linear ground objects |
| CN105740759B (en) * | 2016-01-15 | 2019-01-15 | 武汉珈和科技有限公司 | Semilate rice information decision tree classification approach based on feature extraction in multi-temporal data |
| CN106447510A (en) * | 2016-08-30 | 2017-02-22 | 水利部水土保持监测中心 | Terrace field information extracting method and terrace field information extracting system |
| CN110555804B (en) * | 2018-05-31 | 2022-04-15 | 清华大学 | Correction method and device for high-resolution remote sensing data, computer equipment and readable storage medium |
| CN109146889B (en) * | 2018-07-13 | 2021-11-19 | 洛阳中科龙网创新科技有限公司 | Farmland boundary extraction method based on high-resolution remote sensing image |
| CN109215038A (en) * | 2018-09-29 | 2019-01-15 | 中国资源卫星应用中心 | A kind of intelligent information retrieval method and system based on remote sensing image |
| CN109840553B (en) * | 2019-01-17 | 2023-05-16 | 苏州中科天启遥感科技有限公司 | Extraction method and system of cultivated land crop type, storage medium and electronic equipment |
| CN109872292A (en) * | 2019-02-22 | 2019-06-11 | 东华理工大学 | Method for quickly being handled graph image |
| CN110232389B (en) * | 2019-06-13 | 2022-11-11 | 内蒙古大学 | Stereoscopic vision navigation method based on invariance of green crop feature extraction |
| CN112487999A (en) * | 2020-12-02 | 2021-03-12 | 西安邮电大学 | Remote sensing image robust feature extraction method based on cycleGAN |
-
2021
- 2021-05-21 CN CN202110557228.0A patent/CN113221788B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6684100B1 (en) * | 2000-10-31 | 2004-01-27 | Cardiac Pacemakers, Inc. | Curvature based method for selecting features from an electrophysiologic signals for purpose of complex identification and classification |
| CN107480706A (en) * | 2017-07-24 | 2017-12-15 | 中国农业大学 | A kind of seed production corn field remote sensing recognition method and device |
Non-Patent Citations (2)
| Title |
|---|
| Interactive Multiscale Classification of High-Resolution Remote Sensing Images;Jefersson Alex dos Santos 等;《IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING》;20130831;第6卷(第4期);第2020-2034页 * |
| 顾及局部特性的CNN在遥感影像分类的应用;杨嘉树 等;《计算机工程与应用》;20181231;第54卷(第7期);第188-195页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113221788A (en) | 2021-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111986099B (en) | Tillage monitoring method and system based on convolutional neural network with residual error correction fused | |
| CN106651872B (en) | Pavement crack identification method and system based on Prewitt operator | |
| Malambo et al. | Automated detection and measurement of individual sorghum panicles using density-based clustering of terrestrial lidar data | |
| CN106295789B (en) | Crop seed counting method based on image processing | |
| CN112907587B (en) | High mountain forest line extraction method based on Otsu and edge detection algorithm of GEE | |
| CN110070571B (en) | Phyllostachys pubescens morphological parameter detection method based on depth camera | |
| CN113077486B (en) | Method and system for monitoring vegetation coverage rate in mountainous area | |
| CN111727457B (en) | Cotton crop row detection method and device based on computer vision and storage medium | |
| Grigillo et al. | Automated building extraction from IKONOS images in suburban areas | |
| CN115115627B (en) | Soil saline-alkali soil monitoring method based on data processing | |
| CN113221788B (en) | Method and device for extracting ridge culture characteristics of field blocks | |
| Soares et al. | Plantation Rows Identification by Means of Image Tiling and Hough Transform. | |
| CN108961295B (en) | Purple soil image segmentation and extraction method based on normal distribution H threshold | |
| CN110610438B (en) | Crop canopy petiole included angle calculation method and system | |
| CN110298366B (en) | Crop distribution extraction method and device | |
| CN115331100A (en) | Spatial distribution monitoring method and system for cultivated land planting attributes | |
| CN115115954A (en) | Intelligent identification method for pine nematode disease area color-changing standing trees based on unmanned aerial vehicle remote sensing | |
| CN113033386A (en) | High-resolution remote sensing image-based transmission line channel hidden danger identification method and system | |
| CN111738936A (en) | Image processing-based multi-plant rice spike length measuring method | |
| CN115731257A (en) | Leaf form information extraction method based on image | |
| CN114219815A (en) | High-resolution remote sensing image farmland extraction method based on multilevel semantic boundary segmentation | |
| CN113554675A (en) | Edible fungus yield estimation method based on unmanned aerial vehicle visible light remote sensing | |
| CN117557915B (en) | Crop variety identification method, device, electronic equipment and storage medium | |
| CN113421301B (en) | Method and system for positioning central area of field crop | |
| CN117274844B (en) | Rapid extraction method for field peanut seedling emergence condition by using unmanned aerial vehicle remote sensing image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |