Embedded System-Based Sticky Paper Trap with Deep Learning-Based Insect-Counting Algorithm

József Sütő

doi:10.3390/electronics10151754

Abstract

Flying insect detection, identification, and counting are the key components of agricultural pest management. Insect identification is also one of the most challenging tasks in agricultural image processing. With the aid of machine vision and machine learning, traditional (manual) identification and counting can be automated. To achieve this goal, a particular data acquisition device and an accurate insect recognition algorithm (model) is necessary. In this work, we propose a new embedded system-based insect trap with an OpenMV Cam H7 microcontroller board, which can be used anywhere in the field without any restrictions (AC power supply, WIFI coverage, human interaction, etc.). In addition, we also propose a deep learning-based insect-counting method where we offer solutions for problems such as the “lack of data” and “false insect detection”. By means of the proposed trap and insect-counting method, spraying (pest swarming) could then be accurately scheduled.

Highlights

In order to achieve high crop yields, farmers use insecticides at scheduled times [1]
We proposed a new embedded system-based insect trap with an OpenMV Cam H7 microcontroller board, which can be used anywhere in the field without the restrictions of previous traps
To overcome the problem of having a lack of data seen in previous articles, we generated our own dataset and used a pre-trained deep model for the classification

Summary

Introduction

In order to achieve high crop yields, farmers use insecticides at scheduled times [1]. Insect monitoring and the evaluation of insect density in the field are very important tasks in agriculture because they make the forecast of insect invasion possible Such a forecast has significant environmental and economic effects because farmers can apply insecticides at the right time to defend their crops. Counting pests on pheromone trap images is a very slow, labor-intensive, and expensive process. It requires a skilled person capable of distinguishing insect species. An automated pest management system would be a great improvement as it would solve additional problems such as the lack of agricultural experts In such a system, an automated insect recognition method would be necessary in order to minimize human interaction

Methods

Results

Discussion

Conclusion