Edge Detection with Sobel Method

A high-performance Java application that implements edge detection using the Sobel operator. This project showcases the power of modern Java features by implementing both standard CPU processing and hardware-accelerated processing using TornadoVM.

The project is implemented as a PoC for Medium article Accelerating Java with TornadoVM.

Overview

This application provides edge detection functionality using the Sobel operator with two different implementation approaches:

• Standard CPU-based processing (SobelStandard)
• Hardware-accelerated processing using TornadoVM (SobelTornado)

The application features a JavaFX-based GUI for easy interaction and real-time image processing visualization.

Prerequisites

• Java Development Kit (JDK) 21 or later
• Maven 3.x
• TornadoVM 1.1.0 runtime
• Graphics card supporting OpenCL (for TornadoVM acceleration)

Installing TornadoVM

Instructions from the official site can be used.

As an example for existing Eclipse Temurin JDK 21 on a macOS:

1. Clone TornadoVM repo and run installation script:

   cd /usr/local/share/
   git clone https://github.com/beehive-lab/TornadoVM.git
   cd TornadoVM
   ./bin/tornadovm-installer --jdk temurin-jdk-21 --javaHome $JAVA_HOME --backend opencl

2. The following prompt should be displayed on console:

   ###########################################################################
   Tornado build success
   Updating PATH and TORNADO_SDK to tornado-sdk-....
   Commit  : ....
   ###########################################################################
   ------------------------------------------
          TornadoVM installation done        
   ------------------------------------------
   Creating source file ......................
   ........................................[ok]
   

3. Append the following line to ~/.zshrc for environment settings:

   # Set TornadoVM variables
   source /usr/local/share/TornadoVM/setvars.sh
   

4. On a new terminal shell list accelerator devices available to TornadoVM:

   tornado --devices

   An output like this should be displayed:

   Number of Tornado drivers: 1
   Driver: OpenCL
   Total number of OpenCL devices  : 3
   Tornado device=0:0  (DEFAULT)
   OPENCL --  [Apple] -- Intel(R) UHD Graphics 630
   Global Memory Size: 1,5 GB
   Local Memory Size: 64,0 KB
   Workgroup Dimensions: 3
   Total Number of Block Threads: [256]
   Max WorkGroup Configuration: [256, 256, 256]
   Device OpenCL C version: OpenCL C 1.2
   
   Tornado device=0:1
   OPENCL --  [Apple] -- AMD Radeon Pro 5500M Compute Engine
   Global Memory Size: 4,0 GB
   Local Memory Size: 64,0 KB
   Workgroup Dimensions: 3
   Total Number of Block Threads: [256]
   Max WorkGroup Configuration: [256, 256, 256]
   Device OpenCL C version: OpenCL C 1.2
   
   Tornado device=0:2
   OPENCL --  [Apple] -- Intel(R) Core(TM) i9-9880H CPU @ 2.30GHz
   Global Memory Size: 32,0 GB
   Local Memory Size: 32,0 KB
   Workgroup Dimensions: 3
   Total Number of Block Threads: [1024]
   Max WorkGroup Configuration: [1024, 1, 1]
   Device OpenCL C version: OpenCL C 1.2
   

5. Run TornadoVM unit tests with the following JVM parameters as a workaround if you are using Turkish locale (issue filed to project):

   tornado-test -V --jvm="-Duser.country=US -Duser.language=en"

   Run a specific example to test installation:

   tornado --jvm="-Duser.country=US -Duser.language=en" -m tornado.examples/uk.ac.manchester.tornado.examples.compute.MonteCarlo

Features

• Dual Implementation: Choose between standard CPU processing or TornadoVM acceleration
• Modern Java Features:
  • Module system integration
  • Latest JavaFX controls and features
• Real-time Processing: Live preview of edge detection results
• Hardware Acceleration: TornadoVM support for GPU/FPGA acceleration
• Configurable Parameters: Adjust Sobel operator parameters in real-time
• Comprehensive Logging: Detailed logging using SLF4J and Logback

Running the Application

Using IDE

To run a TornadoVM app, one should use output of tornado --printJavaFlags CLI command. The following lines should be modified as this project is a JavaFx app.

• Install JavaFx runtime to your PC and prepend the lib path to module path setting:

  --module-path [JAVA_FX_PATH/lib]:[TORNADO CLI COMMAND MODULE PATH]
  

• Add modules for JavaFX:

  --add-modules [TORNADO CLI COMMAND MODULES],javafx.controls,javafx.fxml
  

• Add the following as a workaround for host with Turkish locale

  -Duser.country=EN
  -Duser.language=us
  

An example VM option set:

-server
-XX:-UseCompressedOops
-XX:+UnlockExperimentalVMOptions
-XX:+EnableJVMCI
-XX:-UseCompressedClassPointers
--enable-preview
-Djava.library.path=/usr/local/share/TornadoVM/bin/sdk/lib
--module-path
/usr/local/share/JavaFX/javafx-sdk-21.0.7/lib:/usr/local/share/TornadoVM/bin/sdk/share/java/tornado
-Dtornado.load.api.implementation=uk.ac.manchester.tornado.runtime.tasks.TornadoTaskGraph
-Dtornado.load.runtime.implementation=uk.ac.manchester.tornado.runtime.TornadoCoreRuntime
-Dtornado.load.tornado.implementation=uk.ac.manchester.tornado.runtime.common.Tornado
-Dtornado.load.annotation.implementation=uk.ac.manchester.tornado.annotation.ASMClassVisitor
-Dtornado.load.annotation.parallel=uk.ac.manchester.tornado.api.annotations.Parallel
--upgrade-module-path
/usr/local/share/TornadoVM/bin/sdk/share/java/graalJars
-XX:+UseParallelGC
@/usr/local/share/TornadoVM/bin/sdk/etc/exportLists/common-exports
@/usr/local/share/TornadoVM/bin/sdk/etc/exportLists/opencl-exports
--add-modules
ALL-SYSTEM,tornado.runtime,tornado.annotation,tornado.drivers.common,tornado.drivers.opencl,javafx.controls,javafx.fxml
-Dtornado.debug=False
-Dtornado.printKernel=False
-Dtornado.threadInfo=False
-Duser.country=EN
-Duser.language=us

Implementation Details

Standard Implementation (CPU)

• Implements classical Sobel operator algorithm

TornadoVM Implementation

• Leverages GPU/FPGA acceleration
• Parallel processing of image data
• Dynamic hardware selection based on availability

Dependencies

• JavaFX (21.0.5): UI framework
• TornadoVM (1.1.0): Hardware acceleration
• ControlsFX (11.2.1): Enhanced JavaFX controls
• SLF4J/Logback (2.0.17/1.5.17): Logging framework
• JUnit (5.10.2): Testing framework

Configuration

Logging

• Log files are stored in the logs directory
• Main log file: data.log
• Configure logging levels in logback.xml

TornadoVM Settings

• Default device selection: First available GPU
• Fallback to CPU if no GPU is available
• Configure TornadoVM settings through environment variables

Performance Considerations

• GPU acceleration provides significant speedup for large images
• Memory usage is optimized for both implementations

Troubleshooting

Common Issues

1. TornadoVM not detected

   • Ensure TornadoVM is properly installed
   • Check OpenCL installation
   • Verify GPU drivers are up to date

2. Performance Issues

   • Try switching between implementations
   • Check system resource usage
   • Verify image size constraints

Error Logging

Check the logs/data.log file for detailed error information and stack traces.

Future Enhancements

• Add support for additional edge detection algorithms
• Implement batch processing capability
• Add more hardware acceleration options
• Enhance UI with additional controls and features

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Acknowledgments

• TornadoVM team for the acceleration framework
• JavaFX community for UI components