ZeaMays-Seed-Detection-LowLight

A deep learning-based object detection model for identifying Zea mays seeds in low-light conditions using SSD MobileNet. The model is optimized for accuracy in challenging lighting environments.

Pre-trained model : SSDMobileNetV2-with FPN
API : Tensorflow Object Detection API
Dataset: GermPredDataset.zip (data.mendeley.com)

Traied checkpoints are given in

zeamays-model/

Requirements

Install following dependencies:

pip install tensorflow opencv-python numpy matplotlib jupyter

Additionally, install the TensorFlow Object Detection API

git clone https://github.com/tensorflow/models.git
cd models/research
protoc object_detection/protos/*.proto --python_out=.
cp object_detection/packages/tf2/setup.py .
pip install .

Running the Code

Clone this repository:

git clone https://github.com/PubDe/ZeaMays-Seed-Detection-LowLight.git
cd ZeaMays-Seed-Detection-LowLight

To run the seed detection script:

python seed-detection.py

Using Jupyter Notebook

To use the provided Jupyter Notebook for object detection:

Start Jupyter Notebook:
```
jupyter notebook
```
Open the provided seed-detection.ipynb file and run the cells step by step

Sample Predictions

Model Configuration

To train a custom model, configure the pipeline.config file to match your dataset and environment settings. Ensure the paths to your dataset and pre-trained model are correct.

Downloads last month: -; Downloads are not tracked for this model. How to track