https://github.com/jingew/morning-glory-detection

A detection tool with Matlab UI
https://github.com/jingew/morning-glory-detection

agriculture detection image-processing matlab ui

Last synced: 1 day ago
JSON representation

A detection tool with Matlab UI

• Host: GitHub
• URL: https://github.com/jingew/morning-glory-detection
• Owner: JingeW
• License: mit
• Created: 2020-06-15T16:25:47.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2024-06-01T00:59:44.000Z (8 months ago)
• Last Synced: 2024-12-12T10:08:06.733Z (about 2 months ago)
• Topics: agriculture, detection, image-processing, matlab, ui
• Language: MATLAB
• Homepage:
• Size: 20.1 MB
• Stars: 4
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Morning Glory Detection

[![License: MIT](https://img.shields.io/badge/License-MIT-brightgreen.svg)](LICENSE)

A detection tool with Matlab UI.

📃 [Read the Full Paper](https://www.mdpi.com/2624-7402/6/1/34).

## Instruction:

Need to install Matlab first.\

Run "main.m" to activate the app interface.  

Play with the following steps:

1. Click "Open New Image" to load the image you want to detect. Browse and choose the image in the pop-up window.\

   

2. Choose the segmentation method for shadow removal, and click "Segment" to apply it.

	- 2.1 K-means

		The parameter of K-means is the number of the cluster. The default k equals 3. Small k's can avoid most of the noise in the shadow. Larger k provides more details.

	- 2.2 Mean shift

		The parameter of Mean shift is the bandwidth of the kernel. The default bw equals to 0.2. It's a faster method for segmentation. The detection result is very similar to the K-means when k = 3.

    

3. Click "Mask" to generate the binary mask for shadow removal.\

   

4. Click "Output" to get the detection result.\

   

5. (optional) Save the image.

6. Click "Count" to count the number of detected clusters.\

   

* After each step you might need to wait for a few seconds to see the image change, which indicates the current step finished. 

  The waiting time depends on the input file size and computing speed.

* This app referenced the mean shift method from *K. Fukunaga and L.D. Hosteler, 

  "The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition".* [PDF](https://ieeexplore.ieee.org/document/1055330)

## Citation

@article{valicharla2024morning,

  title={Morning Glory Flower Detection in Aerial Images Using Semi-Supervised Segmentation with Gaussian Mixture Models},

  author={Valicharla, Sruthi Keerthi and Wang, Jinge and Li, Xin and Gururajan, Srikanth and Karimzadeh, Roghaiyeh and Park, Yong-Lak},

  journal={AgriEngineering},

  volume={6},

  number={1},

  pages={555--573},

  year={2024},

  publisher={MDPI}

}