/PEDApp

PEDApp - Power Electronics Detection Applications using image processing and deep learning techniques to identify the electronic components and perform auto-simulation of the identified hand-sketched power converter circuits.

Primary LanguagePython

Instructions to run the project

Step 1: Clone the project -> git clone --recursive https://github.com/Varat7v2/PEDApp.git
Step 2: Intsall required packages -> pip install -r requirements.txt
Step 3: Run the main program ->python PEDApp_main.py

Description of the project

This project has been trained on both TensorFlow (SSD MobileNet V2) and PyTorch (YOLOV5 and YOLOR) frameworks. For training the model, dataset have been prepared by ourselves and then annotated. Trained on YOLOR weights till 300 epcochs, model accuracy of 91% is achieved.

Results

alt text

buck_converter_simulation-2023-05-10_01.04.13.mp4