https://github.com/minsithu/podoco
Deep Learning App for Plant Disease Classification
https://github.com/minsithu/podoco
flutter tensorflow tensorflowlite
Last synced: 3 months ago
JSON representation
Deep Learning App for Plant Disease Classification
- Host: GitHub
- URL: https://github.com/minsithu/podoco
- Owner: MinSiThu
- Created: 2023-05-25T07:45:13.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2023-05-28T10:40:04.000Z (almost 2 years ago)
- Last Synced: 2025-01-09T05:45:24.453Z (4 months ago)
- Topics: flutter, tensorflow, tensorflowlite
- Language: C++
- Homepage:
- Size: 42.7 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Podoco - Plant Disease Classifier App
Flutter App that can classify cassava plant diseases offline, using tflite.
## About Model
This model is trained to classify an input image into one of 6 cassava disease classes: Bacterial Blight, Brown Streak Disease, Green Mite, Mosaic Disease, Healthy, and Unknown.
The training dataset is curated by the Mak-AI team at the Makerere University.
## Suitable usecases
This model is trained to recognize four Cassava diseases.
## Unsuitable usecases
This model does not detect diseases in other plants.
## Known limitations
- This model assumes that the input contains a well-cropped image of a Cassava plant. If the image contains some other plant, the output results may be meaningless.
- In some cases, the model may misdiagnose or falsely diagnose a cassava disease. A misdiagnosed result might lead to the wrong treatment course being applied to cassava plants in the field.
- This model should generally work on cassava plants from around the world, but was trained only on photos of diseased cassava plants taken in Uganda