Final Project BoofCV

A collection of computer vision applications built with BoofCV library, focusing on corn plant image analysis and processing.

Overview

This repository contains multiple Java-based computer vision applications that utilize the BoofCV (BoofCV Computer Vision) library for image processing tasks. The primary focus is on agricultural applications, specifically corn plant detection, segmentation, and analysis.

Project Structure

The repository is organized into several independent modules:

Main Applications

• CornPlantColorSegment - Color segmentation for corn plant images
• CornPlantLevel1 - Basic corn plant image processing
• CornPlantTunnelGuideColorSegmentationV1/V2/V3 - Progressive versions of tunnel guide color segmentation for corn plants
• WebCamColorSegmentation - Real-time color segmentation using webcam
• WebCamColorSegmentation-CornPlant - Specialized webcam segmentation for corn plants with linear regression analysis

Computer Vision Examples

• DetectLine - Line detection algorithms
• ExamplePlanarImages - Planar image processing examples
• FiducialSquareBinary - Fiducial marker detection
• ImageBasic - Basic image processing operations
• ImageSegmentation - Image segmentation techniques
• InterestPoint - Interest point detection and feature extraction
• SuperPixcel - Superpixel segmentation
• WatershedWithSeeds - Watershed segmentation with seed points

Utilities

• Webcam - Webcam capture utilities
• WebcamViewerExample - Example webcam viewer implementation
• TakeAPicture - Simple image capture application

Prerequisites

• Java Development Kit (JDK) 8 or higher
• Apache Ant (for building projects)
• Webcam (for webcam-based applications)

Dependencies

The project uses the following libraries (included in lib_018 and lib_023 directories):

• BoofCV - Computer vision library (versions 0.18 and 0.23)
  • Core modules: ip, feature, geo, calibration, io, recognition, sfm, visualize
• Webcam Capture - Version 0.3.10
• DDogleg - Dense matrix operations
• GeoRegression - Geometric operations
• XStream - XML serialization

Building the Projects

Each module is a separate NetBeans project with its own build configuration. To build a specific module:

cd <module-name>
ant clean
ant build

For example, to build the CornPlantColorSegment module:

cd CornPlantColorSegment
ant clean
ant build

Running the Applications

After building, you can run the applications using:

cd <module-name>
ant run

Or execute the JAR files directly from the dist directory of each module.

Key Features

• Color Segmentation: HSV-based color segmentation for plant detection
• Real-time Processing: Webcam integration for live image analysis
• Image Processing: Various algorithms including watershed, superpixels, and interest point detection
• Agricultural Focus: Specialized tools for corn plant analysis and tunnel guide detection
• Multiple Versions: Iterative improvements in corn plant detection algorithms (V1, V2, V3)

Sample Images

Several modules include sample images (PNG and JPG format) for testing and demonstration purposes.

License

This is a final project repository. Please refer to the original project documentation for license information.

Author

pacozaa

Acknowledgments

This project is built using:

• BoofCV - Open source computer vision library
• Webcam Capture - Library for webcam access