CoinDetector

KO | EN

gitlite — search

#javascript #hacktoberfest #typescript #go #android #deep-learning #windows #cli #kubernetes #vue #pytorch #mcp #openai #open-source

(★ 16)

An image based coin counter using Python and Tensorflow

CoinDetector 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# A VGG16-based coins detector

Authors: [Khanh Vu](https://github.com/khanhvu207), [Thimo Blom](https://github.com/thim0o), Hannah Stone.  
Institute: Vrije Universiteit Amsterdam.

## Description
### 1. Images acquisition (tools: OpenCV, Python)
1. Capture raw images from the webcam.
2. Convert RGB to HSV color space.  
![HSV](https://github.com/thim0o/CoinDetector/blob/master/img/hsv.JPG)  
3. Apply [median blur](https://docs.opencv.org/master/d4/d13/tutorial_py_filtering.html) in order to cancel noises.
4. Apply [image thresholding](https://docs.opencv.org/master/d7/d4d/tutorial_py_thresholding.html) & [background subtraction](https://docs.opencv.org/3.4/d1/dc5/tutorial_background_subtraction.html) to capture existing coins.  
![params](https://github.com/thim0o/CoinDetector/blob/master/img/params.JPG)![thresholding](https://github.com/thim0o/CoinDetector/blob/master/img/img_thresholding.JPG)   
5. Make bounding boxes and crop out coins.  
![extracted_coins](https://github.com/thim0o/CoinDetector/blob/master/img/extracted_coins.JPG)

### 2. Convolutional neural network - The VGG16 (Pre-trained model)
***From https://neurohive.io/en/popular-networks/vgg16/***
> VGG16 is a convolutional neural network model proposed by K. Simonyan and A. Zisserman from the University of Oxford in the paper “Very Deep Convolutional Networks for Large-Scale Image Recognition”. The model achieves 92.7% top-5 test accuracy in ImageNet, which is a dataset of over 14 million images belonging to 1000 classes. It was one of the famous model submitted to ILSVRC-2014. It makes the improvement over AlexNet by replacing large kernel-sized filters (11 and 5 in the first and second convolutional layer, respectively) with multiple 3×3 kernel-sized filters one after another.  
![VGG16](https://neurohive.io/wp-content/uploads/2018/11/vgg16-1-e1542731207177.png)

We made few adjustions to the existing VGG16 model from Keras:  
1. We kept only the convolutional layers so we removed the last 4 layers (flatten + 2 x 4096 fully-connected + 1000 classes prediction layers).   
2. Added some fully-connected layers together with a batch_normalization layer and a dropout layer.
![model](https://github.com/thim0o/CoinDetector/blob/master/model_plot.png)

### 3. Model training (tools: Google Colab)
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/15Yoca54IK-rvgcD8L71iymNVYCUOtcKG?usp=sharing)
* Dataset:
  * Total images: **2923 (~500 images per class)**.
  * 6 classes: '10cent', '1euro', '20cent', '2euro', '50cent', '5cent'.
  * Train/Validation ratio: 7:3.
  * Data augmentation: Rotation and Flip.

* Model:
  * Optimizer: Adam - Learning rate = 0.0001.
  * Loss function: 'categorical_crossentropy'.
  * **Transfer learning:** We froze all the layers except for the last 7 hidden layers to accelerate the training process.
  * Metrics tracker: Tensorboard.

### 4. Deploy the model and classify coins (locally, tools: Jupyter notebook)
We downloaded the model params (model.h5) and ran it in our machine - [source code](https://github.com/thim0o/CoinDetector/blob/master/demo.ipynb).  
![demo](https://github.com/thim0o/CoinDetector/blob/master/img/demo.gif)

## Evaluation
See our [project report](https://github.com/thim0o/CoinDetector/blob/master/Real-time-CNN-coin-classification-with-OpenCV-data-acquisition.pdf) for more details.

CoinDetector

CoinDetector 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

CoinDetector

CoinDetector 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README