Artificial intelligence tobacco insect identification system based on Internet of Things

Abstract

The invention discloses an artificial intelligence tobacco insect identification system based on Internet of Things. The system comprises a data platform and an application display platform, the data platform comprises an image acquisition device, an image transmission module, an image processing server and a tobacco insect identification module, and the image acquisition device is used for acquiring tobacco insect image data; tobacco insect image data are acquired through the image acquisition module, the image transmission module transmits the tobacco insect image data to the image processing server through the Internet of Things, after the image processing server processes and analyzes the tobacco insect image data, the tobacco insect identification module marks and identifies the tobacco insect number and the tobacco insect type, and after tobacco insect number data are acquired, the tobacco insect number data are transmitted to the application display platform. The application display platform comprises an insect situation reporting module and a statistical analysis module, and the insect situation reporting module is used for displaying insect situation information acquired in the insect situation image acquisition period to the application display platform; and the statistical analysis module is used for displaying the insect condition information in a certain period of time in the monitoring area on the application display platform and archiving or/and exporting the insect condition information.

Description

technical field [0001] The invention relates to the technical field of tobacco raw material storage, in particular to an artificial intelligence tobacco insect identification system based on the Internet of Things. Background technique [0002] my country is a big tobacco country. In 2019, my country's cigarette production increased slightly to 2.36 trillion sticks, a year-on-year increase of 1.22%. In 2020, various cigarette production bases resumed work one after another in February. As the national logistics system gradually returned to the right track, cigarettes, as a relatively rigid and addictive category, were less affected by the epidemic. The output of cigarettes from January to February was 518.74 billion sticks. A year-on-year increase of 4.4%. It is understood that in addition to being made into cigarettes, dry tobacco, pipe tobacco, cigars, etc. for human consumption, tobacco also has a variety of medical uses. [0003] my country's tobacco industry is devel...

AI Technical Summary

Problems solved by technology

This method not only consumes a lot of manpower, but also leads to untimely detection of insect infestation, resulting in heavy losses

The traditional way of monitoring tobacco insects involves a lot of human interference, and the original data is recorded by the insect worker. It is impossible to know whether the insect worker regularly checks the traps, and there is no record of information such as drug killing and insect nests or the discovery of tobacco insects. The time point and other information records are not timely, which will mislead the follow-up analysis of the pest situation

If you need to know the actual situation of the pest situation, there are few channels to obtain information and you need to combine the recorded information of multiple departments for manual integration, which requires a lot of work.

Images