Cervical Cancer

Introduction:

The aim of the project is to detect and classify whether the person has cervical cancer or not and to detect the area where cancer is detected.We Will be using tiny yolov3 architecture inorder to achieve the goal.

Steps:

1. Prepare the Dataset
2. Preprocess and Augment the Dataset
3. Train the model
4. Test the model
5. Fine Tune the model to achieve better accuracy
6. Generate a weights file
7. Convert weights to intermediate representation
8. Deploy the generated files on camera using OpenVino toolkit

Model Architecture:

Output:

Check Out the Project here:

1. Google Colab
2. GitHub

References:

1. https://www.researchgate.net/figure/The-network-structure-of-Tiny-YOLO-V3_fig1_338162578
2. https://github.com/qqwweee/keras-yolo3
3. https://www.pyimagesearch.com/2020/01/27/yolo-and-tiny-yolo-object-detection-on-the-raspberry-pi-and-movidius-ncs/

Contributors:

1. Vansh Shah
2. Viral Prajapati