OpenCV Notes

My awesome notes about computer vision

Description

This is the notes I took in my computer vision journey. Contents are from image basics and processing to object detection and tracking, even deep learning.

Contents

These notes consist of six topics. You can easily access all the content below.

1. Image Basics

• 1.1 Numpy Basics np.array, np.zeros, np.ones, np.arange, np.linspace, np.random.randint, np.reshape
• 1.2 OpenCV Basics cv2.imread, cv2.resize, cv2.flip, cv2.rotate, cv2.imwrite
• 1.3 Opening Images with OpenCV
• 1.4 Drawing on Images cv2.rectangle, cv2.circle, cv2.line, cv2.putText, cv2.polylines
• 1.5 Drawing on Images with a Mouse

2. Image Processing

• 2.1 Colorspaces cv2.cvtColor
• 2.2 Blending and Pasting Images cv2.addWeighted
• 2.3 Image Thresholding cv2.threshold, cv2.adaptiveThreshold
• 2.4 Image Blurring cv2.blur, cv2.GaussianBlur, cv2.medianBlur, cv2.bilateralFilter, cv2.filter2D
• 2.5 Morphological Operators cv2.getStructuringElement, cv2.morphologyEx
• 2.6 Gradients cv2.Sobel, cv2.Scharr, cv2.Laplacian
• 2.7 Histograms cv2.calcHist, cv2.equalizeHist, cv2.convertScaleAbs
• 2.8 Transformations cv2.getAffineTransform, cv2.warpAffine, cv2.getPerspectiveTransform, cv2.warpPerspective

3. Video Basics

• 3.1 Connecting to Camera
• 3.2 Saving Video
• 3.3 Opening Video Files
• 3.4 Drawing on Video

4. Object Detection

• 4.1 Template Matching cv2.matchTemplate, cv2.minMaxLoc
• 4.2 Corner Detection cv2.cornerHarris, cv2.goodFeaturesToTrack
• 4.3 Edge Detection cv2.Canny
• 4.4 Grid Detection cv2.findChessboardCorners, cv2.drawChessboardCorners, cv2.findCirclesGrid
• 4.5 Contour Detection cv2.findContours, cv2.drawContours
• 4.6 Feature Matching
• 4.7 Watershed Algorithm
• 4.8 Custom Seeds with Watershed Algorithm
• 4.9 Face Detection

5. Object Tracking

• 5.1 Lucas-Kanade Optical Flow
• 5.2 Dense Optical Flow
• 5.3 MeanShift Tracking
• 5.4 CAMShift Tracking

6. Deep Learning

• 6.1 Keras Basics
• 6.2 Keras CNN (MNIST)
• 6.3 Working with Custom Images

Installation

In order to play around with the notes, you can set up this repository locally following these simple steps.

Note that in order to avoid potential conflicts with other packages, it is strongly recommended to use a virtual environment or a conda environment.

Install using pip

1. Clone the repo

   git clone https://github.com/hasan-alper/opencv-notes.git

2. Move inside your copy

   cd opencv-notes

3. Create a virtual environment

   virtualenv [my_virtual_env_name]

4. Active the virtual environment

   source [my_virtual_env_name]/bin/activate

5. Install necessary packages

   pip install -r requirements.txt

Install using conda

1. Clone the repo

   git clone https://github.com/hasan-alper/opencv-notes.git

2. Move inside your copy

   cd opencv-notes

3. Create a conda environment

   conda create --name [my_conda_env_name]

4. Active the conda environment

   conda activate [my_conda_env_name]

5. Install necessary packages

   pip install -r requirements.txt

   ⚠️ For Apple silicon Macs, you must follow these steps instead:

   conda install -c apple tensorflow-deps

   pip install -r requirements-apple-silicon.txt

Contributing

Any contributions are greatly appreciated. Feel free to create a pull request to improve one or two things. You can also simply open an issue.

License

Distributed under the MIT License. See LICENSE for more information.