苹果成熟度轻量化实时检测模型GCA-YOLOv8n 的设计与实现

打开文本图片集

中图分类号：S661.1;TP391.41 文献标志码：A 文章编号：1001-411X（2026）01-0128-11

Abstract： 【Objective】 To address issues with traditional apple ripeness detection models， including excessive size， slow inference speed and low detection accuracy. 【Method】 We constructed a lightweight real-time detection model， GCA-YOLOv8n， based onan improved YOLOv8n. First， the C3Ghost module replaced the C2f module of original model to achieve lightweight design and enhance inference speed. Second， the GhostConv module substituted the original Conv layer to improve information extraction eficiency and reduce redundancy in convolutional layers.Finally， the ACmix atention mechanism was integrated into the original model architecture to boost feature extraction capability and detection accuracy. The improved GCA-YOLOv8n model was applied to apple ripeness detection experiments. 【Result】Experimental results showed that the GCA-YOLOv8n model achieved 2.0×106 parameters， 5.7×109 floating point operations， and a weight file size of 4.4MB ，representing reductions of 33.1% ， 29.6% ，and 30.2% respectively compared to YOLOv8n. The inference speed reached 130.8 frames per second， a 21.5% improvement over YOLOv8n. The mean average precision and F1 scorewere 89.2% and 82.5% respectively， demonstrating high detection accuracy and inference speed. 【Conclusion】 The constructed GCA-YOLOv8n model significantly reduces model complexity and computational load while maintaining detection accuracy，achieving lightweight and eficient performance.The model demonstrates high and real-time detection capability and can operate stably on edge computing devices （including mobile devices）， providing technical support for automated harvesting.

Key words： Object detection; Apple ripeness; YOLOv8n; Lightweight model; Inference speed

我国的苹果产量常年蝉联世界首位[1]，传统的人工采摘存在效率低、成本高等问题，随着智慧农业技术的快速发展，自动化采摘成为现代农业领域的研究热点。（剩余15227字）