KhetGo ← KhetGo ऐप

KhetGo · AI tool · आलू, टमाटर, मक्का

फसल डॉक्टर

पत्ती की फोटो लें, और AI बताएगा कि उसमें कौन-सा रोग हो सकता है, साथ में क्या करें।

Photograph a potato, tomato or maize leaf. A neural network recognises 17 diseases (or healthy) right on your phone. The photo is never uploaded.

1. फसल चुनें Choose the crop (helps the model)

2. पत्ती की फोटो Photo of one leaf

एक पत्ती की साफ़ फोटो यहाँ आएगी
Drag a photo here, or use the buttons

सबसे अच्छा नतीजा: एक ही पत्ती, दिन की रोशनी, पत्ती फ्रेम में भरी हुई, सादा पीछे का हिस्सा।

फोटो नहीं है? ये आज़माएँ (test leaves the model never saw)

AI मॉडल लोड हो रहा है… 0%

How it works

Transfer learning on a laptop CPU, tested honestly on leaves the model never saw, and run entirely in the browser.

1. Training

MobileNetV3-Large, pretrained on ImageNet, fine-tuned in PyTorch on about 24,000 PlantVillage photos of potato, tomato and maize leaves, with random crops, flips, rotations and colour changes so it copes with phone photos.

2. Honest testing

PlantVillage has about four photos of each physical leaf. A random split would put the same leaf in training and test and inflate the score, so the split is by leaf: every test photo shows a leaf the model never saw.

3. On your phone

The model is exported to ONNX and runs with ONNX Runtime Web (WebAssembly). Your photo stays on your device. Choosing the crop restricts the answer to that crop’s diseases.

What it can and can’t do

Recognises 17 classes

Lab photos vs. the field

PlantVillage photos are single leaves on a plain background. Research shows models trained on them are much less accurate on cluttered field photos, so treat the answer as a first hint, and photograph one leaf against a plain background.

Not a replacement for an expert

  • Only potato, tomato and maize; any other photo still gets one of these answers.
  • Advice is general and non-chemical. Medicines and doses: KVK or agriculture officer.