YOLOv11-Litchi: Efficient Litchi Fruit Detection based on UAV-Captured Agricultural Imagery in Complex Orchard Environments

TL;DR

The paper addresses the real-time detection of litchi fruits in UAV imagery, where small targets and occlusions hamper accuracy. It introduces YOLOv11-Litchi, a lightweight extension of YOLOv11 that integrates a multi-scale residual C3-MSR, a compact F3 fusion neck, and a Litchi-Head with SEAM occlusion attention to robustly detect litchi under occlusion. Experimental results show a compact model of $6.35$ MB achieving $mAP$ of $90.1\%$ and $F1$ of $85.5\%$ at $57.2$ FPS, outperforming several baselines while meeting real-time requirements, and generalizes well to Laboro Tomato and Citrus datasets. Overall, the method promises practical impact for precision agriculture by enabling accurate, real-time UAV based fruit detection with strong occlusion handling and scale adaptation.

Abstract

Litchi is a high-value fruit, yet traditional manual selection methods are increasingly inadequate for modern production demands. Integrating UAV-based aerial imagery with deep learning offers a promising solution to enhance efficiency and reduce costs. This paper introduces YOLOv11-Litchi, a lightweight and robust detection model specifically designed for UAV-based litchi detection. Built upon the YOLOv11 framework, the proposed model addresses key challenges such as small target size, large model parameters hindering deployment, and frequent target occlusion. To tackle these issues, three major innovations are incorporated: a multi-scale residual module to improve contextual feature extraction across scales, a lightweight feature fusion method to reduce model size and computational costs while maintaining high accuracy, and a litchi occlusion detection head to mitigate occlusion effects by emphasizing target regions and suppressing background interference. Experimental results validate the model's effectiveness. YOLOv11-Litchi achieves a parameter size of 6.35 MB - 32.5% smaller than the YOLOv11 baseline - while improving mAP by 2.5% to 90.1% and F1-Score by 1.4% to 85.5%. Additionally, the model achieves a frame rate of 57.2 FPS, meeting real-time detection requirements. These findings demonstrate the suitability of YOLOv11-Litchi for UAV-based litchi detection in complex orchard environments, showcasing its potential for broader applications in precision agriculture.

Figures (12)

• Figure 1: Tilt shooting method of the UAV.
• Figure 2: Example of litchi images captured by the UAV.
• Figure 3: Different occlusion types observed in litchi images.
• Figure 4: Image enhancement methods applied to litchi images.
• Figure 5: Sample image from the Laboro Tomato public dataset.