Palm Tree Detection

Detect palm tree using YOLO V8 model and having .gpkj as the final output.

Step by step

1) Installation requirement

• Make sure you have install PYTHON and GIT on your computer.

• Follow the step to install Anaconda from the documentation.

• Press WINDOWS button and find Anaconda Prompt.

• 

• Check if you have install CUDA by running this line in the prompt.

nvidia-smi

[OUTPUT]

+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.98                 Driver Version: 535.98       CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                     TCC/WDDM  | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA GeForce RTX 3080 Ti   WDDM  | 00000000:01:00.0  On |                  N/A |
| 31%   59C    P2             208W / 400W |   2416MiB / 12288MiB |     54%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+

• If you did not get the output, you can go to this link. Once you have install, check the output again.
• Then you may change to the desired directory.

cd folder/directory

• Download this repository

git clone https://github.com/pkcktkksh98/palmtree_detection.git
cd palmtree_detection

• Download Yolo V8 repository here using git clone.
• Now install ultralytics for Yolo V8 usage.

conda install -c conda-forge ultralytics

• Now you should be able to run YOLO on your desktop. Next we are going to install Slicing Aided Hyper Inference(SAHI) by following the Installation details.

2) Folder Management

• In the palmtree_detection folder should contain

  • adds
  • image
  • sdpr.pt
  • README.md
  • Tree_Counting.ipynb
  • ultralytics

• Move Tree_Counting.ipynb, sdpr.pt and adds into ultralytics folder.

3) Execution

• To execute Jupyter Notebook, you can run jupyter notebook in Anaconda Prompt terminal and it will open in one of the browser.

• You may open the Tree Counting.ipynb file in Jupyter Notebook.

• Make sure to change the name of the estate.

• 

• The "source" variable is the directory to the Folder that contains IMG folder and in it is the grided image of the estate.

• Then you may run the program by clicking Restart & Run All.

• 

• There should be two (2) new folder created in the source directory.

  • predict
  • GPKJ