# Rice disease detection method based on multi-scale dynamic feature fusion

## TL;DR

This paper introduces an optimized deep learning model for detecting rice leaf diseases in complex environments, suitable for mobile deployment with reduced computational needs.

## Contribution

A novel YOLOv11-based model with multi-scale dynamic feature fusion and pruning techniques for efficient rice disease detection on mobile devices.

## Key findings

- The model reduces computational complexity, parameters, and memory usage by 50.7%, 49.6%, and 36.9% compared to YOLOv11n.
- The model achieves a 1.7% improvement in mAP@0.5:0.9 while maintaining real-time performance on mobile devices.
- The optimized model is suitable for practical agricultural use with hardware limitations.

## Abstract

In order to enhance the accuracy of rice leaf disease detection in complex farmland environments, and facilitate the deployment of the deep learning model onto mobile terminals for rapid real-time inference, this paper introduces a disease detection network titled YOLOv11 Multi-scale Dynamic Feature Fusion for Rice Disease Detection (YOLOv11-MSDFF-RiceD). The model adopts the concept of ParameterNet to design the FlexiC3k2Net module, which replaces the neck feature extraction network, thereby bolstering the model's feature learning capabilities without significantly increasing computational complexity. Additionally, an efficient multi-scale feature fusion module (EMFFM) is devised, improving both the computational efficiency and feature extraction capabilities of the model, while simultaneously reducing the number of parameters and memory footprint. The bounding box regression loss function, inner-WIoU, utilizes auxiliary bounding boxes and scale factors. Finally, the Dependency Graph (DepGraph) pruning model is employed to minimize the model's size, computational load, and parameter count, with only a moderate sacrifice in accuracy. Compared to the original YOLOv11n model, the optimized model achieves reductions in computational complexity, parameter scale, and memory usage by 50.7%, 49.6%, and 36.9%, respectively, with only a 1.7% improvement in mAP@0.5:0.9. These optimizations enable efficient deployment on resource-constrained mobile devices, making the model highly suitable for real-time disease detection in practical agricultural scenarios where hardware limitations are critical. Consequently, the improved model proposed in this paper effectively detects rice disease targets in complex environments, providing theoretical and technical support for the deployment and application of mobile terminal detection devices, such as rice disease detectors, in practical scenarios.