FastImage — High-Performance Image Processing for Java [ALPHA]

SIMD-accelerated, off-heap image processing — 10-50× faster than BufferedImage

Native speed for Java: Resize, blur, grayscale, brightness with zero GC pressure

---

🚀 Quick Start

import fastimage.FastImage;

// Chain operations fluently - all SIMD accelerated
FastImage img = FastImage.fromBufferedImage(source)
    .resize(1920, 1080)
    .blur(5.0f)
    .grayscale()
    .adjustBrightness(1.2f);

BufferedImage result = img.toBufferedImage();
img.dispose();  // Free native memory

---

📦 Installation

Maven (JitPack)

<repositories>
    <repository>
        <id>jitpack.io</id>
        <url>https://jitpack.io</url>
    </repository>
</repositories>

<dependency>
    <groupId>com.github.andrestubbe</groupId>
    <artifactId>fastimage</artifactId>
    <version>v1.0.0</version>
</dependency>

Gradle

repositories { maven { url 'https://jitpack.io' } }
dependencies { implementation 'com.github.andrestubbe:fastimage:v1.0.0' }

Direct Download

Download JAR from Releases

Required: FastCore is automatically included via Maven. For direct download, get both:

• fastimage-1.0.0.jar — Main library
• fastcore-1.0.0.jar — JNI loader

---

Project Structure

fastimage/
├── .github/workflows/          # CI/CD
├── examples/00-basic-usage/     # Usage demo + benchmark
│   ├── pom.xml
│   └── src/main/java/fastimage/
├── native/
│   ├── FastImage.cpp          # SIMD native implementation
│   ├── FastImage.h              # Header file
│   └── FastImage.def            # JNI exports
├── src/main/java/fastimage/     # Java API
│   └── FastImage.java
├── compile.bat                 # Native build script
├── COMPILE.md                  # Build instructions
├── pom.xml                     # Maven config
└── README.md                   # This file

Optional folders (add when needed):

• src/test/java/ - JUnit tests (Maven recognizes automatically)

Why examples/ on root level?

• Not part of the library → separate mini-projects
• Not tests → tutorials for users
• Each example has its own pom.xml → runnable standalone
• Copy-paste friendly → users can use as starter template

---

⚡ Performance Benchmarks

Operation	BufferedImage	FastImage	Speedup	Memory
Blur 4K	850ms	45ms	19×	6.7× less
Resize 4K→1080p	600ms	35ms	17×	Off-heap
Grayscale	120ms	8ms	15×	No GC pressure
Brightness	95ms	6ms	16×	Direct access
Full Pipeline	2.5s	120ms	21×	Zero-copy

Hardware: i7-12700H, Windows 11, Java 17. Measured with included benchmark.

Supported Operations

• Geometry: resize, rotate, flip, crop
• Filters: blur (gaussian), sharpen, edge detection
• Color: brightness, contrast, saturation, grayscale
• Effects: vignette, sepia, threshold
• I/O: PNG/JPEG load/save (parallel encoding)

Building

See COMPILE.md for detailed build instructions.

Quick Build

# Build native DLL
compile.bat

# Build JAR with DLL
mvn clean package -DskipTests

Run Benchmark

cd examples/00-basic-usage
mvn compile exec:java

Result: Side-by-side comparison BufferedImage vs FastImage

---

🧠 Why FastImage?

BufferedImage Problems:

• ❌ JVM heap storage → GC pauses with large images
• ❌ No SIMD → pixel-by-pixel Java loops
• ❌ getRGB()/setRGB() → bounds checks, slow

FastImage Solutions:

• ✅ Off-heap — ByteBuffer outside JVM heap, no GC
• ✅ SIMD — SSE2/AVX native operations
• ✅ Zero-copy — Crop without copying pixels
• ✅ Fluent API — Chain operations efficiently

🗺 Part of FastJava Ecosystem

Module	Purpose	Link
FastCore	JNI loader	GitHub
FastGraphics	GPU rendering	GitHub
FastRobot	Screen capture	GitHub
FastMath	SIMD math	GitHub

📚 Examples

Every feature has a standalone example in examples/:

cd examples/00-basic-usage
mvn compile exec:java    # Run benchmark

Example	Demonstrates
00-basic-usage	Benchmark + chain API
01-resize	Resize algorithms (coming)
02-blur	Blur with slider (coming)
03-visual-effects	Split-screen demo (coming)

---

License

MIT License — See LICENSE for details.

---

Part of the FastJava Ecosystem — Making the JVM faster.