Discrete Kalman Filter for 1D and 2D object tracking

---

Description

This project aims to implement the Kalman Filter (Linear Quadratic Estimation) for Object-tracking in 1D and 2D in C++ using this reference

Through this port, I got a brush up on using CMake and gained exposure to an amazing graph plotting library for C++, Matplot++ along with the Eigen C++ library used for linear algebra and the OpenCV C++ library used for image processing and computer vision techniques.

System Information

Type	Info
Operating System	Ubuntu 22.04
Processor	AMD® Ryzen 9 5900hx with radeon graphics × 16
Graphics card	NVIDIA Corporation GA106M [GeForce RTX 3060 Mobile / Max-Q] / NVIDIA GeForce RTX 3060 Laptop GPU/PCIe/SSE2
Installed RAM	16.0 GB (15.7 GB usable)
System type	64-bit operating system, x64-based processor
Pen and touch	No pen or touch input is available for this display

---

Installation

Prerequisites

• Ubuntu 22.04
• CMake ≥ 3.20
• C++17-compatible compiler
• OpenCV (with video codec support)
• Matplot++ (installed via vcpkg)

Project structure:

project_root/ ├── build/ ├── data/ ├── include/ └── src/

Installation Steps

1. After cloning the repository, create a build directory with mkdir build and then go inside the build directory with cd build

2. Run cmake .. from inside the /build directory (which you already should be in) if you are building for the first time.
   This step just tells CMake to look for CMakeLists.txt in the /project_root directory.

3. Run cmake --build . from inside the build directory.
   This will create the KalmanFilter executable in the ./build folder Then run the executable through the commandline with ./KalmanFilter

matplotplusplus library was installed on this system using vcpkg.

Refer to https://github.com/alandefreitas/matplotplusplus#embed-with-cpmcmake for CMake configuration details

If opencv package is not getting included, then move the "opencv2" folder out of the "opencv4" folder:

sudo mv /usr/local/include/opencv4/opencv2/ /usr/local/include/

Also, make sure that opencv4.pc is present in /usr/local/lib/pkgconfig/
If it is not, then:

1. If opencv4.pc exists in /usr/lib/x86_64-linux-gnu/pkgconfig/, then execute:

sudo cp /usr/lib/x86_64-linux-gnu/pkgconfig/opencv4.pc /usr/local/lib/pkgconfig/

2. Otherwise, follow: https://forums.developer.nvidia.com/t/compiling-opencv-c-program/160402

Make sure that you have all the available codecs installed for running video files and that you have built and installed opencv ONLY AFTER installing said codecs:

1. sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-dev

2. Then go to the directory where you have opencv and opencv-contrib unzipped and create a build folder

3. Go inside this build directory and open terminal

4. In terminal, execute: cmake -DFFMPEG=1 -DWITH_OPENCL=1 -DOPENCV_EXTRA_MODULES_PATH=../opencv_contrib-4.x/modules ../opencv-4.x

5. Build the opencv install file with: cmake --build .

6. Install opencv with: sudo make install

• All data files should be put in the /data folder and provide the relative path to them in functions as "../data/filename.type"

• ".h" files are commonly associated with C files.

• Either declare classes and functions in ".h" and define them in corresponding ".cpp" files OR define as well as declare them in ".hpp" files which are associated with C++