Objective: To build object Detection Model to detect Mushrooms from image using yolov3
Dataset: Images Scrap from Google
1. Web Scraping images and annotations using LabelImg
2. Cloning Github Repo of Darknet
!git clone 'https://github.com/AlexeyAB/darknet.git'**
3. Setting Google colab GPU using Make file
4. Making Changes in yolov3.config file
5. Generating label.data file using python script file creating-files-data-and-name.py
6. Genrating train.txt and test.txt using python script file creating-train-and-test-txt-files.py
7. Download pre-trained Yolov3 weights for the convolutional layers.
8. Training Cutom Object Detector !darknet/darknet detector train Custom_data/labelled_data.data darknet/cfg/yolov3.cfg custom_weights/darknet53.conv.74 -dont_show
9. Testing Custom Object Detector with Test image (Applying Non-max separation)