This project is a result of my participation in KaggleX BIPOC Mentorship Program. During this program, I had the opportunity to take courses, attend talks, and work on a project that interests me. I chose to work on image classification. The dataset, from Kaggle, contains over 4500 images of healthy and diseased plants from 12 different types. Using both MobilenetV2 and VGG16 pre-trained models as a base, a classifier with 22 classes was attached as a head. In using the provided test images, an accuracies of 88% and 89% were achieved, respectively.
MobilenetV2 code --> PlantClassification_TF_MobilenetV2.ipynb. This notebook also includes code for a Streamlit App.
VGG16 code --> PlantClassification_TF_vgg16.ipynb
Courses: Kaggle: Computer Vision DataCamp:
Image Processing with Keras in Python Introduction to Deep Learning in Keras
OpenCV: Deep Learning with Tensorflow and Keras
Data: https://www.kaggle.com/datasets/csafrit2/plant-leaves-for-image-classification
Streamlit App:
