Detection method of Spodoptera frugiperda based on staged depth inpainting images
A step-by-step technology of Spodoptera frugiperda, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as the difficulty of accurate detection of Spodoptera frugiperda in corn, and achieve the effect of improving image detection and recognition capabilities
Active Publication Date: 2021-12-10
HEFEI INSTITUTES OF PHYSICAL SCIENCE - CHINESE ACAD OF SCI
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The purpose of the present invention is to solve the defect that it is difficult to accurately detect Spodoptera frugiperda with different ages in the prior art, and provide a detection method for Spodoptera frugiperda based on staged depth repair images to solve the above problems
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
[0083] In order to have a further understanding and understanding of the structural features of the present invention and the achieved effects, the preferred embodiments and accompanying drawings are used for a detailed description, as follows:
[0084] like figure 1 As shown, the method for detecting Spodoptera frugiperda based on staged depth repair images of the present invention comprises the following steps:
[0085] The first step is the collection of training samples. Collect images of Spodoptera frugiperda and its corresponding hazard shape and feces size as training data. The focus of the image is on the part of Spodoptera frugiperda at the position of the heart leaf of corn, and the size of all training images is normalized to 64 ×64 pixels.
[0086] In the second step, the training samples are preprocessed. Construct and train a staged depth restoration model, and use staged depth restoration technology to repair the missing part of the image of Spodoptera mesoph...
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 relates to a method for detecting Spodoptera frugiperda based on staged deep image restoration and improved Double-DQN technology, which solves the defect that Spodoptera frugiperda of different ages is difficult to accurately detect compared with the prior art. The invention comprises the following steps: collection of training samples; preprocessing of the training samples; construction and training of image detection models of Spodoptera frugiperda of different ages; collection and preprocessing of images of Spodoptera frugiperda to be detected and corresponding context information ; markers of specific positions in maize images of different instars of fall armyworm. The present invention realizes phased restoration of large-area missing images by constructing the edge map model of different ages of Spodoptera frugiperda and the image depth restoration model of Spodoptera frugiperda, and on this basis, trains Spodoptera frugiperda of different ages and the corresponding context The information image detection model has improved the image detection and recognition capabilities of Spodoptera frugiperda in different ages.
Description
technical field [0001] The invention relates to the technical field of image target detection, in particular to a method for detecting Spodoptera mesophila based on staged depth restoration images. Background technique [0002] In my country, the fall armyworm was first discovered in Yunnan in January 2019, and it has expanded to 21 provinces in less than 7 months. It is a major migratory pest with strong reproductive ability, and it is the first invasion, with almost no natural enemies. It is estimated that the maize seedling stage can generally reduce the yield by 10%-25%, and seriously endangering the field can cause the destruction of the seed and the harvest. It has been included in the national monitoring object of major plant diseases and insect pests. [0003] How to accurately detect and identify fall armyworm of different ages is the primary task of monitoring and control. Because fall armyworms often hide in the corn heart leaves to eat them, it is more difficul...
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 Patents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/25G06F18/24G06F18/214
Inventor 贾秀芳李伟王儒敬谢成军黄河张洁周满胡海瀛
Owner HEFEI INSTITUTES OF PHYSICAL SCIENCE - CHINESE ACAD OF SCI