A method for crop target feature recognition

Abstract

The present invention relates to the field of agricultural operation technology, specifically a method for identifying crop target features, comprising the following steps: capturing images of crops; processing images, including noise reduction, grayscale processing and image edge detection, extracting crop veins and stem and leaf trunks; vectorizing crop veins and stem and leaf trunks; finding an end point based on the vector, and generating crop top coordinates based on the end point. This patent is based on the growth characteristics of the plant itself, and by identifying the distribution of the plant trunk and stems and leaves, the growth position of the plant top is inferred. The logic identifies the plant trunk and stems and leaves in the image and converts them into a vector graph. By connecting the vector graph, the starting point is obtained, which is the plant top. Compared with the traditional direct target recognition of the plant top, this logic is less affected by image quality, and takes into account the structural characteristics of the plant itself, which can effectively improve the recognition rate.

Description

Technical Field

[0001] The invention relates to the technical field of agricultural operations, and in particular to a method for identifying crop target features. Background Art

[0002] Some crops (such as cotton) have a growth advantage at the top. In order to inhibit the growth of the main trunk and increase the growth rate of branches and trunks, these crops need to be toppled. Taking cotton as an example, in the existing mechanical cotton topping technology, target detection and other methods are usually used to identify the cotton tops. However, the target detection method is greatly affected by the quality of the photo, and the cotton tops are small and very similar to the background color. The accuracy of the traditional target detection method is low, far less than that of manual detection. Therefore, a new crop target feature recognition method is needed. Summary of the invention

[0003] In order to overcome the defects existing in the prior art, the purpose of the present invention is to provide a crop target feature recognition method for stem and leaf plants, which is less affected by image quality and takes into account the structural characteristics of the plant itself, and can effectively improve the recognition rate.

[0004] The object of the present invention is achieved through the following technical solutions:

[0005] A method for identifying crop target features, wherein the crop is a stem-leaf plant and the target feature is the top of the stem and leaf of the crop, comprises the following steps:

[0006] B1. Use visual sensors to collect images of crops;

[0007] B2, performing Gaussian filtering to reduce noise on the image, performing background threshold segmentation on the image, and extracting a relatively complete crop image;

[0008] B3, performing dimensionality reduction processing on the acquired color crop image, converting the image into a grayscale image, retaining the shape distribution information of the crop trunk and stems and leaves in the original image, performing Canny operator edge detection, and obtaining a binary image of the crop edge and the stem and leaf trunk edge;

[0009] S4, processing the acquired binary image: performing straight line detection on the binary image, wherein the curved image is the leaf edge, and the remaining image is the stem and leaf trunk, extracting the straight line image as C1, and the remaining image as C2, and the pixel values ​​in the images C1 and C2 are the edge information of the binary image;

[0010] S5. Let C1(x,y) and C2(x,y) be the pixel points at the coordinate positions in the straight line image C1 and the residual image C2, and judge C1(x,y) and C2(x,y). If C1(x,y)=C2(x,y), it means that (x,y) is the leaf tip coordinate. If C1(x,y)≠C2(x,y), it means that (x,y) is not the leaf tip coordinate. Save all leaf tip coordinates {(x,y)}.

[0011] S6. Convert the straight line image C1 into a vector, with the starting point being the leaf tip coordinate {(x,y)}, obtain the intersection coordinates {(x',y')} of all vectors, draw a circle with i pixels as the radius, and use each pixel in C1 as the center of the circle for judgment. If there are intersection coordinates (x',y') in the circle and there are more than n intersections, it is considered that the intersections are clustered, and calculate the average coordinates of all intersections in the circle (Σx / n,Σy / n). The coordinates are the center coordinates of the top of the stem and leaf of the crop, that is, the position coordinates of the target feature; wherein i and n are both natural numbers greater than 0.

[0012] The present application provides a crop target feature recognition method. Compared with the traditional target feature detection method, the present application is based on the growth characteristics of stem-leaf plants (such as cotton) themselves. By identifying the main trunk and stem and leaf distribution of the plant, the growth position of the top of the plant is inferred. The logic identifies the main trunk and stem and leaves of the plant in the image and converts them into vector graphs. By connecting the vector graphs, the starting point is obtained, which is the top of the plant. Compared with the traditional direct recognition of target features, this logic is less affected by image quality, and takes into account the structural characteristics of the plant itself, which can effectively improve the recognition rate.

[0013] The concept, specific structure and effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 1 is a schematic diagram of the process of the cotton top identification method in the embodiment. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0016] Example:

[0017] Take cotton top identification as an example. Figure 1 As shown, this embodiment provides a cotton top identification method (logic), comprising the following steps:

[0018] S1. Use a binocular camera or a depth camera (visual sensor) to collect images of cotton in the working area from a downward perspective.

[0019] S2, performing Gaussian filtering on the image to remove noise interference, improve image quality, perform background threshold segmentation on the cotton image, and extract a relatively complete cotton plant image;

[0020] S3, performing dimensionality reduction processing on the obtained colored cotton image, converting the image into a grayscale image, retaining the main trunk and stem and leaf shape distribution information of the cotton in the original image, performing edge detection using the Canny operator, obtaining the cotton edge and stem and leaf main trunk image, achieving effective separation of the cotton and background image, and retaining only the cotton edge and stem and leaf main trunk information;

[0021] S4, processing the obtained cotton edge binary image, according to the growth characteristics of cotton, the edge of cotton leaves is relatively curved, so the curved image should be the leaf edge, and the remaining image is the stem and leaf trunk, and performing straight line detection on the cotton edge binary image, extracting the straight line image as C1, and the remaining image as C2. The pixel values ​​in images C1 and C2 are the edge information of the cotton edge binary image;

[0022] S5. Let C1(x,y) and C2(x,y) be the pixel points at the coordinate positions in the straight line image C1 and the residual image C2, and judge C1(x,y) and C2(x,y). If C1(x,y)=C2(x,y), it means that (x,y) is the leaf tip coordinate. If C1(x,y)≠C2(x,y), it means that (x,y) is not the leaf tip coordinate. Save all leaf tip coordinates {(x,y)}.

[0023] S6. Convert the straight line image C1 into a vector, with the starting point being the leaf tip coordinate {(x,y)}, obtain the intersection coordinates {(x',y')} of all vectors, draw a circle with 5 pixels as the radius, and use each pixel in C1 as the center of the circle for judgment. If there are intersection coordinates (x',y') in the circle and there are more than n intersections, it is considered that the intersections are clustered, and calculate the average coordinates of all intersections in the circle (Σx / n,Σy / n), which are the center coordinates of the cotton top; n is a natural number greater than 0, and this parameter can be set.

[0024] The cotton top recognition method provided in this embodiment first collects images of the cotton in the working area; then processes the image, including noise reduction, grayscale processing and image edge detection, extracts cotton veins and stem and leaf trunks; then vectorizes the cotton veins and stem and leaf trunks; finds the end point according to the vector, and generates the cotton top coordinates according to the end point. This application is based on the growth characteristics of cotton itself, and infers the growth position of the cotton top by identifying the distribution of the cotton trunk and stem and leaves. This logic identifies the cotton trunk and stem and leaves in the image and converts them into a vector graph. By connecting the vector graph, the starting point is obtained, which is the cotton top. Compared with the traditional direct target recognition of cotton tops, this logic is less affected by image quality, and takes into account the structural characteristics of the plant itself, which can effectively improve the recognition rate.

[0025] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other.

Claims

1. A crop target feature recognition method, characterized in that: The crop is a plant with stems and leaves, and the target feature is the top of the stems and leaves of the crop, comprising the following steps: S1. Use a visual sensor to collect images of crops in the working area from a downward perspective; S2, performing Gaussian filtering to reduce noise on the image, performing background threshold segmentation on the image, and extracting a relatively complete crop image; S3, performing dimensionality reduction processing on the acquired color crop image, converting the image into a grayscale image, retaining the shape distribution information of the crop trunk and stems and leaves in the original image, performing Canny operator edge detection, and obtaining a binary image of the crop edge and the stem and leaf trunk edge; S4, processing the acquired binary image: performing straight line detection on the binary image, wherein the curved image is the leaf edge, and the remaining image is the stem and leaf trunk, extracting the straight line image as C1, and the remaining image as C2, and the pixel values ​​in the images C1 and C2 are the edge information of the binary image; S5. Let C1(x,y) and C2(x,y) be the pixel points at the coordinate positions in the straight line image C1 and the residual image C2, and judge C1(x,y) and C2(x,y). If C1(x,y)=C2(x,y), it means that (x,y) is the leaf tip coordinate. If C1(x,y)≠C2(x,y), it means that (x,y) is not the leaf tip coordinate. Save all leaf tip coordinates {(x,y)}. S6. Convert the straight line image C1 into a vector, with the starting point being the leaf tip coordinate {(x,y)}, obtain the intersection coordinates {(x',y')} of all vectors, draw a circle with i pixels as the radius, and use each pixel in C1 as the center of the circle for judgment. If there are intersection coordinates (x',y') in the circle and there are more than n intersections, it is considered that the intersections are clustered, and calculate the average coordinates of all intersections in the circle (Σx / n,Σy / n). The coordinates are the center coordinates of the top of the stem and leaf of the crop, that is, the position coordinates of the target feature; wherein i and n are both natural numbers greater than 0.

Images