Image-based epidemic wood detection and positioning method and system

Abstract

The invention provides an image-based epidemic wood detection and positioning method and system. According to a multi-scale candidate region fusion network designed by the invention, the RPN(S-RPN) of a small target is detected to obtain a feature map through ResNet18, the RPN(M-RPN) of a medium target is detected to obtain a feature map through ResNet32, and the RPN(L-RPN) of a large target is detected to obtain a feature map through ResNet50. Non-maximum suppression (NMS) is then employed to reduce redundant candidate regions. The network can solve the problem that a current target detection network model based on a deep neural network is difficult to deal with multi-scale pine forest epidemic wood detection. According to the invention, the three-dimensional positioning geometric model of the epidemic wood is constructed according to the camera imaging principle of the holder monitoring tower and digital terrain elevation data (DEM), and the three-dimensional positioning of the epidemic wood outbreak point can be accurately realized. The method and the system are simple and rapid, and can improve the monitoring and early warning capability of the pine forest epidemic wood.

Description

technical field [0001] The invention belongs to the technical field of image detection and positioning, and in particular relates to a pest tree detection network based on deep learning-based fusion of multi-scale candidate regions and a three-dimensional positioning geometric model of the pest tree. Background technique [0002] Pine wood nematode is one of the most harmful pine forest diseases in my country's forestry industry, and is called the "cancer" of pine trees. It only takes 3-5 years from the infection of a pine tree to the death of the whole pine forest, and the infected pine trees are called diseased wood. Therefore, detection of infested trees is the most important thing in pine forest protection. The most distinguishing feature of infected pine trees from normal healthy pine trees is that the entire canopy needles of the trees are yellowish-brown or reddish-brown, whereas the canopy needles of normal healthy pine trees are green. At present, the detection me...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose an image-based method for detecting and locating diseased trees in the image of the cloud platform observation tower, which is used to solve the problem that the current target detection network model based on deep neural networks is difficult to deal with multi-scale pine forest diseases. Wood detection, and the problem of not being able to accurately obtain the geographic location of the diseased wood

Images