Weed detection method based on deep neural network
A technology of deep neural network and detection method, which is applied in the field of weed detection, can solve problems such as difficult to accurately and efficiently identify the position and outline area of weeds, and achieve good recognition effect
Active Publication Date: 2021-10-22
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF5 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006]
Aiming at the above-mentioned deficiencies in the prior art, a weed detection method based on a deep neural network provided by the present invention solves the problem that it is difficult to accurately and efficiently identify the position and outline area of weeds in the prior art
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0041] The specific embodiments of the present invention are described below so that those skilled in the art can understand the present invention, but it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, as long as various changes Within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are included in the protection list.
[0042] Such as figure 1 As shown, the weed detection method based on deep neural network includes the following steps:
[0043] S1. Obtain the original image, perform instance segmentation and labeling of the crops in the original image to obtain a training image, and construct an image training set;
[0044] S2. Constructing an initial neural network model;
[0045]S3. Train the initial neural network model through the i...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a weed detection method based on a deep neural network, and the method comprises the steps: dividing pixel points belonging to crops on an image through employing instance segmentation, and then carrying out the binaryzation of the image according to the feature that the features of green plants can be enhanced more effectively through ultra-green features, and obtaining the pixel point information of all green plants in the image; subtracting the pixel point of the crop from the pixel points of all the green plants to obtain the pixel points of the green plants which are not the crops; and finally, removing the pixels with a small number of pixel points in some pixel point groups, wherein a pixel communication region in the remaining pixel points is a weed; obtaining the position information and contour information of the weed, so a technical support is provided for precision agriculture.
Description
technical field [0001] The invention relates to the field of weed detection, in particular to a weed detection method based on a deep neural network. Background technique [0002] In traditional agricultural operations, there are usually two methods to remove weeds: one is to identify weeds by human eyes, and then manually pull out or eradicate weeds; the other is to spray herbicides directly on the farmland. The traditional method not only wastes a lot of manpower, but also sprays a lot of pesticides, which will cause serious environmental pollution. With the introduction of the concept of precision agriculture, the model of weed positioning + robotic weeding has been widely used. [0003] In the prior art, the method of directly identifying weeds is usually used. Although weeds can be effectively detected, only the types of weeds included in the training set can be detected, and the types of weeds not included in the training set cannot be detected. However, there are ma...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00
CPCG06N3/08G06N3/045Y02A40/10
Inventor 朱嘉静兰荻刘勇国张云李巧勤杨尚明
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA