Plant.health's new feature Symptoms provides clearer and more actionable insights into plant health. Using thousands of phytosanitary expert-annotated photos, we've created 11 key symptoms, see table below.
Symptom
Description
Typical Cause(s)
Blight
Chlorosis, browning, and often rotting of plant tissues (leaves, twigs, branches, or flowers)
Pathogenic infection
Browning Leaf Tips
Dead, dried-out leaf tissue, most commonly located at leaf tips or leaf margins
Overwatering or uneven watering
Curled Leaves
Leaves tend to roll
Lack of water, pathogenic infection
Discoloration
Leaves turn pale green and yellow
Nutrient deficiencies (nitrogen, iron, magnesium, or potassium), under or over watering, low light, pests, viruses, or natural aging
Holes in the Leaves
Gaps or cuts
Insects or snails
Leaf Cuts
Typically straight cuts or light streaks on leaves
Human activity
Leaf Dieback
Whole leaves die, turn dark, and often appear wilted
Disease or age
Leaf Spots
Small dark spots on leaves
Bacterial infection
Pest Traces or Occurrence
Presence of insects or other pests or their products, such as mines (meandering tunnels), webs, eggs, or excrement
Insects or other pests
White Patches
Whitish or grayish coatings or spots on the surface of plants
Fungi
Wilted Leaves
Drooping leaves
Lack of water
In the API, you can set up this feature by using the parameter "symptoms":true.
The API returns heatmaps of all identified symptoms, highlighting the most affected plant areas. Additionally, there is a score for each of the symptoms indicating its severity and an overall defect score, giving a quick overview of the plant's health status.
Try this feature on the Plant.id demo. Here is an example (health assessment - Symptoms - Visualize symptoms).
We believe this feature will enable users to quickly and accurately diagnose plant issues, improving their understanding and ability to address underlying problems, and enhancing their overall experience.
We tested LLMs to improve multilingual descriptions of plants, insects, and fungi where Wikipedia entries were inadequate. GPT-4o outperformed smaller models such as GPT-4o mini and Mistral Pixtral 12B, especially in grammar and factual accuracy. It performed better in widely spoken languages and common species. In the end, GPT-4o was used to generate descriptions in 28 languages, providing better coverage for the Kindwise API.
The Plant.id model upgrade resulted in significant improvements in accuracy. It reached over 85% accuracy in the first response (TOP 1). The increase is greatest for European and North American plants, with 6.0% and 3.6%, respectively.
Plant.id now allows plant species to be filtered by various criteria such as geographical distribution, morphology, ecology or practical use. This increases identification accuracy and confidence in the model.
Multimodal Large Language models have recently made it possible to identify various objects from images, including plants. This became a competition for narrowly focused neural network models, such as that of Plant.id. Here we compare the accuracy of these models by focusing on the proportion of wrong answers.
As the world of houseplants and ornamental varieties continues to flourish, many people now seek more than species identification. Our new 'Varieties' model responds to this demand and offers you a unique opportunity to gain a competitive advantage by enabling the identification of almost 1,000 plant varieties.
We have upgraded the Plant.id model, increasing the number of classes from 33,325 to 35,756. The new version, released in October 2023, shows significant performance improvements over the previous models, especially in Europe and North America.
By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.