GBiDC-PEST: A novel lightweight model for real-time multiclass tiny pest detection and mobile platform deployment

Deep learning-based intelligent recognition algorithms are increasingly recognized for their potential to address the labor-intensive challenge of manual pest detection. However, their deployment on mobile devices has been constrained by high computational demands.

To that end, a team of researchers from China and the United States developed GBiDC-PEST, a mobile application that incorporates an improved, lightweight detection algorithm based on the You Only Look Once (YOLO) series single-stage architecture for real-time detection of four tiny pests (wheat mites, sugarcane aphids, wheat aphids, and rice planthoppers).

“Insect pests significantly threaten global food security, resulting in substantial economic losses, particularly in major food-producing countries such as China and the United States” explains the study’s senior author, Qiong Su, a professor at Clemson University, United States.

“Deep learning techniques offer unique advantages in pest detection by automatically learning, however, most deep learning-based pest detection studies were conducted in controlled environments with stable lighting and uniform backgrounds,” says first author Weiyue Xu, a researcher at Changzhou University, China. “Optimizing these algorithms for accurate detection in complex, natural environments remain a significant challenge, particularly for tiny objects.”

The GBiDC-PEST model specifically addresses the practical needs of modern agricultural production of three major crops—sorghum, wheat, and rice—which are critical for global food security. The model targets four significant pests (sugarcane aphid, wheat mite, wheat aphid, and rice planthopper) that cause substantial damage to these crops.

“GBiDC-PEST incorporates several innovative modules, including GhostNet for lightweight feature extraction and architecture optimization by reconstructing the backbone, the Bi-directional Feature Pyramid Network (BiFPN) for enhanced multiscale feature fusion, Depthwise convolution (DWConv) layers to reduce computational load, and the Convolutional Block Attention Module (CBAM) to enable precise feature focus,” shares Xu.

The GBiDC-PEST algorithm achieves a balanced, lightweight design while maintaining high detection accuracy (mAP=80.1%) and a fast-processing speed (FPS=161.3).

GBiDC-PEST was successfully deployed as an Android application for real-time pest detection in the field.

“The optimization and App deployment approach of the GBiDC-PEST algorithm proposed in this study enhances the applicability of mobile devices for the automatic detection of multiple pests in complex field environments,” adds Su.

The team's findings, published in KeAi’s Journal of Integrative Agriculture, offer a robust technical framework for the rapid, onsite identification and localization of tiny pests. This advancement provides valuable insights for effective pest monitoring, counting, and control in various agricultural settings.

###

Contact the author: Weiyue Xu, E-mail: wyxu@cczu.edu.cn; #Correspondence Qiong Su, E-mail: qsu@clemson.edu

The publisher KeAi was established by Elsevier and China Science Publishing & Media Ltd to unfold quality research globally. In 2013, our focus shifted to open access publishing. We now proudly publish more than 200 world-class, open access, English language journals, spanning all scientific disciplines. Many of these are titles we publish in partnership with prestigious societies and academic institutions, such as the National Natural Science Foundation of China (NSFC).