Vision language models enable fast crop health diagnosis on devices
PlantVLM-Edge: Quantized Vision-Language Models for On-Device Crop Health Diagnosis
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Diagnosing crop health often needs expert knowledge that many farmers don't have easy access to. The authors created PlantVLM-Edge, a version of vision-language AI models that runs on small devices without needing internet. This lets farmers ask questions about plant diseases or pests and get quick, helpful answers directly on their device. The system uses smart data tricks to keep answers almost as good as bigger models while using much less memory and computing power.
What this means in practice
- •For agriculture technology teams: Provide farmers with offline AI tools to identify crop diseases and pests quickly on handheld edge devices.$Commercial implications: Enables sale of portable crop health diagnostic devices that operate without cloud dependency, improving access for growers.
- •For mobile device developers: Implement efficient vision-language models on resource-limited devices for real-time interactive applications without internet.
Authors
Gaurab Wagle, Bishal Adhikari, Jiajia Li, Xiaolei Huang, Dong Chen
Abstract
Timely and accurate crop health diagnosis is essential for sustainable agricultural production, yet accurate diagnosis often requires expert knowledge that is not readily available. Recent vision-language models (VLMs) have demonstrated remarkable capabilities in visual understanding, reasoning, and natural-language interaction, creating new opportunities for intelligent crop health assessment. However, deploying VLMs on resource-constrained devices remains challenging due to their substantial computational and memory requirements. This paper presents PlantVLM-Edge, a quantized vision-language framework for conversational crop health diagnosis that can run entirely on resource-constrained edge devices. PlantVLM-Edge combines agricultural domain adaptation, post-training quantization, and on-device serving with an offline chat interface, allowing growers to ask open-ended questions and receive diagnoses, symptom explanations, and management guidance without cloud connectivity. The framework is backbone-agnostic: we validate it with two different VLM types with varying scale and architecture (Qwen3-VL-2B and LLaVA-1.5-7B), four quantization schemes, and two diagnostic domains (plant diseases and insect pests), and we deploy all variants on an NVIDIA Jetson AGX Orin. Eight-bit quantization affects answer quality minimally relative to the full-precision model, and 4-bit activation-aware weight quantization (AWQ) retains 98.6% of full-precision answer quality on both domains while reducing the weight memory of the 2B backbone by 49%. In its most compact form, PlantVLM-Edge requires 2.01 GiB of weight memory, returns its first token in about 0.12 s, and completes a 64-token diagnostic response in 0.78 s. These results demonstrate that timely and accurate crop health diagnosis is practical on edge devices, providing a foundation for next-generation crop monitoring and decision support systems.