Cott-ADNet: Lightweight Real-Time Cotton Boll and Flower Detection Under Field Conditions

2025-09-15

AI Summary

Cotton harvesting relies heavily on manual labor, resulting in low efficiency and frequent missed optimal harvest windows—necessitating high-precision, real-time detection of cotton bolls and flowers. To address this, we propose a lightweight YOLOv11n variant incorporating a NeLU-enhanced global attention mechanism to strengthen weak feature responses, coupled with an expanded-receptive-field SPPF module for robust multi-scale contextual modeling. Evaluated on a custom field dataset of 4,966 images, the model achieves 91.5% precision, 93.3% mAP₅₀, and 71.3% mAP while requiring only 7.5 GFLOPs—effectively balancing accuracy and computational efficiency. This approach directly supports automated harvesting decision-making and high-throughput cotton phenotyping, delivering a deployable, vision-based solution for intelligent crop sensing.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Hardware-aware ML

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Cotton is one of the most important natural fiber crops worldwide, yet harvesting remains limited by labor-intensive manual picking, low efficiency, and yield losses from missing the optimal harvest window. Accurate recognition of cotton bolls and their maturity is therefore essential for automation, yield estimation, and breeding research. We propose Cott-ADNet, a lightweight real-time detector tailored to cotton boll and flower recognition under complex field conditions. Building on YOLOv11n, Cott-ADNet enhances spatial representation and robustness through improved convolutional designs, while introducing two new modules: a NeLU-enhanced Global Attention Mechanism to better capture weak and low-contrast features, and a Dilated Receptive Field SPPF to expand receptive fields for more effective multi-scale context modeling at low computational cost. We curate a labeled dataset of 4,966 images, and release an external validation set of 1,216 field images to support future research. Experiments show that Cott-ADNet achieves 91.5% Precision, 89.8% Recall, 93.3% mAP50, 71.3% mAP, and 90.6% F1-Score with only 7.5 GFLOPs, maintaining stable performance under multi-scale and rotational variations. These results demonstrate Cott-ADNet as an accurate and efficient solution for in-field deployment, and thus provide a reliable basis for automated cotton harvesting and high-throughput phenotypic analysis. Code and dataset is available at https://github.com/SweefongWong/Cott-ADNet.

Problem

• Automating cotton harvesting through accurate boll detection
• Enabling real-time recognition under complex field conditions
• Providing lightweight solution for yield estimation and breeding

Innovation

• Lightweight detector based on YOLOv11n
• NeLU-enhanced Global Attention Mechanism
• Dilated Receptive Field SPPF module