Skip to content

Latest commit

 

History

History
35 lines (24 loc) · 1.03 KB

README.md

File metadata and controls

35 lines (24 loc) · 1.03 KB

CarDetection

Descriptioin

This project will detect cars in one image using the YOLO model. Many of the ideas in this notebook are described in the two YOLO papers: Redmon et al., 2016 (https://arxiv.org/abs/1506.02640) and Redmon and Farhadi, 2016 (https://arxiv.org/abs/1612.08242).

Because the YOLO model is very computationally expensive to train, I will load pre-trained weights to use.

In this project I used the dataset from drive.ai

Two main work have done:

  1. Use object detection on a car detection dataset
  2. Deal with bounding boxes

Model

the architecture as follows: image

Requirements

  1. TensorFlow
  2. python 3 or later
  3. python packages : Keras, numpy, scipy, pandas, PIL
  4. download the weights yolo.h5 on the website

Usage

python cardetection.py

Resualts

the resuals as follows: image