YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
-# Aries 3.5 🌱
Aries 3.5 is an AI-powered plant health classification model designed to identify whether a plant is healthy or unhealthy from an input image.
The model achieves 90.7% classification accuracy and is optimized for efficient deployment using a C++ inference library and Edge Impulse EON Compiler, making it suitable for edge and embedded AI applications where performance and resource efficiency are important.
Key Features
- 🌿 Plant Health Classification: Detects whether a plant appears healthy or unhealthy.
- 🎯 90.7% Accuracy: Achieves 90.7% accuracy on the evaluated dataset.
- ⚡ Edge-Optimized: Designed with efficient inference in mind for embedded and edge devices.
- 💻 C++ Support: Uses a C++ library for integration and deployment.
- 🔧 EON Compiler: Built/optimized using the Edge Impulse EON Compiler for efficient machine-learning deployment.
- 📷 Image-Based Detection: Uses plant images as input for classification.
Intended Use
Aries 3.5 can be used in applications such as:
- Smart agriculture and precision farming
- Plant monitoring systems
- Automated greenhouse monitoring
- Educational AI and computer vision projects
- IoT-based plant health detection
- Edge AI and embedded systems
Performance
Accuracy: 90.7%
Accuracy was measured on the evaluation data used during model development. Actual performance may vary depending on image quality, plant species, lighting conditions, camera hardware, and environmental factors.
Technology
Aries 3.5 combines computer vision with efficient edge-AI deployment techniques. The model is intended to work with a C++ inference environment and leverages the EON Compiler to reduce resource usage and improve deployment efficiency on supported hardware.
Limitations
Aries 3.5 should be considered a plant-health classification tool rather than a replacement for professional agricultural or botanical diagnosis. The model may perform differently on plant species, diseases, environmental conditions, or image types that were not represented in its training data.