Computers and Electronics in Agriculture·8月27日 · 周四论文农业人工智能与决策模型阅读 2

Lavender electrical signal classification

Montserrat Alvarado-González · Edwin Bryan Salas-López

Computers and Electronics in AgricultureDOI 10.1016/j.compag.2026.112328开放获取 PDF ↗
论文解读
66接近精选线
速览

研究薰衣草对不同触觉刺激的电生理反应,实现高精度分类,探索植物触觉传感。

方法

记录薰衣草电信号,分析不同刺激和部位,使用分类算法评估。

发现

植物产生可重复的电信号模式,单指刺激分类精度超0.95,脱水下仍可检测。

机会点

可探索植物电信号用于环境监测或生物混合传感器,结合机器学习提升复杂刺激下的分类鲁棒性。

影响力12
信息增量20
专业深度18
信源权威12
时效性4

研究植物电信号用于触觉传感,方法新颖,数据详实,但应用尚在实验室阶段,影响有限。

原文摘要

Bioinspired approaches offer promising strategies for the development of advanced sensing systems. In this work, we investigate the electrophysiological responses of Lavandula dentata to controlled tactile stimuli applied to different anatomical regions. Electrical signals were recorded under multiple stimulation conditions, including single-finger contact, two-finger contact, and inflorescence compression. The results show that the plant generates distinct and reproducible electrical potential patterns depending on both the stimulation type and the anatomical location. Classification performance reached precision and recall values above 0.95 under single-finger stimulation. Although accuracy decreased under more complex interactions, such as two-finger contact and inflorescence squeezing, responses remained robust for stimulations applied to the stem, peduncle, and leaves. Observed voltage ranges further indicate inter-individual variability across plant samples. In addition, electrical conductivity analysis suggests that detectable responses are preserved under dehydration conditions. These observations suggest that residual ionic conduction and remaining water content may contribute to the persistence of electrical responses under dehydration. The possible influence of essential oils stored in glandular trichomes remains speculative and requires further investigation. These findings support the feasibility of using plant electrophysiology for tactile sensing and highlight the potential of L. dentata as a biohybrid sensing element. The results contribute to the development of plant-based interfaces and open new avenues for integrating living substrates into next-generation tactile sensing technologies.

生物混合传感 植物电信号 智慧农业 触觉传感相关资讯

Zenodo (CERN European Organization for Nuclear Research)·10月21日 · 周三69 分 论文社会化服务与规模经营农业人工智能与决策模型阅读 0

Evaluation of Concept-Rank Collapse Audits for Computer Vision Agriculture Disease Classifiers Under Low-Label Regimes: An Interpretability Capacity Bound for Smallholder Fungicide Selectivity

提出可解释性容量上界,评估低标签下概念秩崩溃审计的可靠性。

Zenodo (CERN European Organization for Nuclear Research)·10月21日 · 周三70 分 论文社会化服务与规模经营农业人工智能与决策模型阅读 0

Evaluation of Concept-Probe Temporal-Stability Audits for Computer Vision Agriculture Disease Classifiers Under Seasonal Phenology Drift: An Interpretability Decay Bound for Smallholder Fungicide Timing

提出概念探针时间稳定性审计与可解释性衰减界,判断何时需重新审计。

Zenodo (CERN European Organization for Nuclear Research)·10月21日 · 周三65 分 论文农业AI与大模型农业人工智能与决策模型阅读 0

Calibration-Shift Inference for Computer Vision Agriculture Yield Models Under Retrofit Sensor Replacement: An Artificial Intelligence Post-Deployment Recalibration Guarantee

提出改造传感器替换后视觉产量模型校准漂移的检测与置信区间保证方法。