Automatic paper change system based on computer vision effect and pest monitoring system

Abstract

The invention discloses an automatic paper change system based on a computer vision effect and a pest monitoring system. The automatic paper change system comprises laminating pest adhering paper, a paper change device, an image acquisition device and a processor, wherein the laminating pest adhering paper is arranged in the paper change device, and the processor controls the paper change device to change the pest adhering paper. The pest adhering paper is of a double-layer (including a working layer and a protecting layer) structure, the working layer comprises a basal layer, an adhesive layer and a solid attractant layer, attractants are immersed in the basal layer, the adhesive layer is provided with breathable micropores, and the protecting layer is pasted on the surface of the workinglayer. The paper change device changes the pest adhering paper by new pest adhering paper and recovers the used pest adhering paper in a biaxial linkage mode. The pest monitoring system counts specific target insects based on a deep neural network and simultaneously acquires environmental data. Pests can be effectively and continuously trapped for a long time, pest image data can be acquired, types and the number of the pests are recognized and counted by the aid of a deep neural network technology, and pest scale and trend are accurately monitored.

Description

technical field [0001] The invention relates to the field of pest monitoring and early warning, in particular to an automatic replacement system for sticky insect paper based on computer vision effects and a pest monitoring system based thereon. Background technique [0002] The monitoring and management of diseases and insect pests is an important production link in agricultural production. Deploying pest disaster monitoring and early warning systems is an important means for agricultural production management departments and large modern farms to monitor and manage pests. The pest disaster monitoring and early warning system mainly includes: pest data collection, pest data analysis, statistics and trend forecasting. [0003] Among them, the long-term and timely collection of pest data is the most basic link of the pest disaster monitoring and early warning system. Most of them use the method of "distributed fixed-point collection and centralized collection" to collect pe...

AI Technical Summary

Problems solved by technology

However, these traps are not suitable for configuring automatic attractant replacement devices or storing large quantities of attractant at the installation point, and the attractant must be manually replaced to achieve long-term monitoring

[0006] CN103598169A discloses a kind of fruit fly monitoring device and system based on the McPhail type trap. The liquid attractant is filled in the trapping bottle, and the camera is installed on the bottle mouth to collect the images of the pests captured by the liquid. The device must be manually replaced regularly. The attractant is used to keep the trapping bottle working effectively for a long time, and the fruit flies are trapped in the liquid, and the insects can move in the liquid or be immersed in the liquid, resulting in blurred images of the collected pests or changes in distinguishable features, which seriously affect the later stage The identification statistics of

It must be replaced manually, and the automatic replacement of the sticky board cannot be realized

[0010] In summary, the attractants, traps, replacement devices, trapping devices and monitoring methods of existing devices and systems cannot be organically coordinated to achieve long-term, effective, continuous, and maintenance-free collection of pest data and identification and statistics of pests

Most of the pests are caused by fast failure of attractants, poor trapping effect, manual replacement of traps, poor control of automatic replacement effect on replacement accuracy or trap effectiveness, inability to collect pest data effectively for a long time, and low accuracy of automatic pest classification The problem of inaccurate and insufficient statistical information

Images