Pest Manager: A Systematic Framework for Precise Pest Counting and Identification in Invisible Grain Pile Storage Environment

Chuanyang Ma,Jiangtao Li,Xingqun Qi,Muyi Sun,Huiling Zhou

2024-09-03

Abstract:Pest infestations pose a significant risk to both the quality and quantity of stored grain, resulting in substantial economic losses. Accurate and timely pest monitoring is essential, but traditional methods are time-consuming and labor-intensive, and often ineffective at detecting pests beneath the grain surface. In this paper, we introduce a systematic framework: Pest Manager for precise pest counting and identification within the invisible grain pile environment. The framework consists of three components: an improved grain probe trap PestMoni, a pest drop dataset PestSet collected by PestMoni, and a multi-task Transformer-based architecture PestFormer for pest counting and identification. Specifically, PestMoni uses an asymmetric layout of infrared diodes and photodiodes, enabling precise recording of pest drops. Based on PestMoni, we develop PestSet, an infrared-perception dataset for pest drops, including five common grain storage pests. Furthermore, we propose PestFormer, a multi-task model with a conditional modification module to reduce deviations across different PestMonis and thereby normalize the data processing workflow. Extensive experiments validate the design, dataset, and model, with PestFormer achieving state-of-the-art accuracies of 99.2% in counting and 86.9% in identification, highlighting its potential for effective pest management in invisible storage environment.

Signal Processing

What problem does this paper attempt to address?

The paper attempts to address the problem of accurately monitoring and identifying pests in invisible grain storage environments. Specifically:

1. **Background Issues**:
- Pest infestations pose a significant threat to the quality and quantity of stored grain, leading to substantial economic losses.
- Traditional pest monitoring methods rely on manual sampling strategies, which are time-consuming, labor-intensive, and lack accuracy.
- Existing intelligent monitoring methods mainly focus on the visible surface of the grain, lacking the ability to penetrate into the interior of the grain heap.

2. **Research Objectives**:
- Propose a system framework "Pest Manager" for accurately counting and identifying pests in invisible grain heap environments.
- The framework includes three components:
- **PestMoni**: An improved electronic grain probe trap that uses asymmetrically orthogonal layouts of infrared diodes and photodiodes to accurately record the process of pest falling.
- **PestSet**: A multi-class pest falling dataset collected by PestMoni, containing data on 5 common stored grain pests.
- **PestFormer**: A multi-task Transformer-based architecture for pest counting and identification, which reduces bias between different devices through a conditional modification module, thereby standardizing the data processing flow.

3. **Innovations**:
- The design innovation of **PestMoni** lies in its asymmetrically orthogonal layout of infrared components and custom circuits, which can accurately record the complete waveform of pest falling.
- **PestSet** is a comprehensive pest falling dataset containing 12,632 waveforms, covering 5 common stored grain pests.
- **PestFormer** is a conditionally modified Transformer model that can handle device-sensitive data, achieving high-precision time series classification.

4. **Experimental Validation**:
- Extensive experiments validated the rationality of the PestMoni design, the effectiveness of the PestSet dataset, and the effectiveness of the PestFormer model.
- Experimental results show that PestFormer achieved 99.2% accuracy in pest counting and 86.9% accuracy in species identification, demonstrating great potential for pest management in invisible storage environments.

In summary, this paper aims to address the challenge of accurately monitoring and identifying pests in invisible grain heap environments by proposing a systematic framework, Pest Manager, thereby improving the safety and economic efficiency of stored grain.