TAHIR0110 / ThereForYou

ThereForYou: Your mental health ally. Kai, our AI assistant, offers compassionate support. Track your mood trends, find solace in a secure community, and access crisis resources swiftly. We're here to empower your journey towards improved well-being, leveraging technology for a brighter tomorrow.
Other
85 stars 95 forks source link

Title :- Indian Medicinal Leaves and Plants Classification #156

Closed aaradhyasinghgaur closed 2 months ago

aaradhyasinghgaur commented 3 months ago

Is your feature request related to a problem? Please describe. The problem of identifying, categorizing, and preserving the vast and diverse knowledge of Indian medicinal plants can be effectively addressed through a deep learning-based classification system. By training neural networks on extensive datasets comprising images, morphological descriptions, regional names, and medicinal uses, deep learning models can accurately identify and classify plants even with subtle visual differences. This system can streamline the documentation and preservation of traditional knowledge, validate medicinal properties through cross-referencing scientific studies, and make the information widely accessible. This integration not only supports scientific research and conservation efforts but also facilitates the safe and effective use of traditional medicinal practices in modern healthcare.

dataset I'll use ;- https://www.kaggle.com/datasets/crypticfate5/medicinal-plants

Describe the solution you'd like

  1. Data Preparation: Collect and preprocess a balanced dataset of real and fake face images, including normalization, resizing, and augmentation.
  2. Base Model Selection: EfficientNetB0,VGG16 , Xception , InceptionV3 like 5 different models excluding its top layers, to leverage its learned features.
  3. Model Construction: Add custom layers on top of the base model for binary classification, compiling with appropriate loss and metrics.
  4. Initial Training: Train the model with the base layers frozen to only update the new layers.
  5. Fine-Tuning: Unfreeze some or all of the base model layers and continue training with a lower learning rate to fine-tune the entire network. 6.) EDA analysis. 7.) Comaprioson using performance matrices such as accuracy scores , confusion matrix etc.

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context Add any other context or screenshots about the feature request here.

aaradhyasinghgaur commented 3 months ago

@TAHIR0110 kindly assign this issue to me and merge my pull requests in case of no problems please . Thank you.

TAHIR0110 commented 3 months ago

@aaradhyasinghgaur assigned! do one thing that the model would first predict the leaf/plant and then would briefly describe the use case of the plant/ leaf as well.