✦ AstroView

Lightweight astronomical image viewer for Android

Open and visualize FITS, XISF, and TIFF files from your phone.

---

About

AstroView is a free, open-source Android app designed for astrophotographers who need a quick way to preview astronomical image files on their phone or tablet. No bloat, no ads, no internet — just open your files and inspect them.

Built entirely with Jetpack Compose and zero external image processing libraries. All FITS/XISF/TIFF decoding and the auto-stretch algorithm are implemented natively in Kotlin.

---

Features

Feature	Details
📂 File Support	FITS (.fits, .fit, .fts), XISF (.xisf), TIFF (.tiff, .tif)
🎨 Bit Depths	8-bit, 16-bit, 32-bit integer, 32-bit float, 64-bit float
🌈 Color Modes	Monochrome and RGB
✨ Auto Stretch (STF)	Midtone Transfer Function targeting median 0.25 with MAD-based shadow clipping
🔄 Transforms	Rotate ±90°, flip horizontal, flip vertical
🔍 Zoom & Pan	Pinch-to-zoom (0.5x–20x) with smooth panning
📊 Info Bar	Filename, dimensions, bit depth, color mode, format, STF indicator
📱 Modern UI	Material 3, dark space theme, adaptive navigation bar support
🪶 Lightweight	~4 MB APK, no internet permissions, no external dependencies

---

Auto Stretch Algorithm

AstroView implements a Midtone Transfer Function (MTF) auto-stretch inspired by PixInsight's Screen Transfer Function (STF):

1. Median is computed per channel
2. MAD (Median Absolute Deviation) is calculated and normalized (×1.4826)
3. Shadow clipping is set at median - 2.8 × MAD
4. The midtone balance parameter is computed to map the normalized median to a target of 0.25
5. The MTF curve is applied to every pixel:

MTF(x, m) = (m - 1) × x / ((2m - 1) × x - m)

This ensures that any linear astronomical image — regardless of its original histogram distribution — becomes clearly visible with a single toggle.

---

Installation

From APK (Recommended)

1. Go to the Releases page
2. Download the latest AstroView-vX.X.X.apk
3. Transfer to your Android device and install
4. You may need to enable "Install from unknown sources" in your device settings

Build from Source

Prerequisites:

• Android Studio Hedgehog (2023.1.1) or newer
• Android SDK 34
• JDK 17

Steps:

# Clone the repository
git clone https://github.com/Ft2801/AstroView.git
cd AstroView

# Open in Android Studio
# File → Open → select the AstroView folder

# Or build from command line
./gradlew assembleDebug

The debug APK will be at:

app/build/outputs/apk/debug/app-debug.apk

For a release build:

./gradlew assembleRelease

Note: Release builds require a signing key. See Android documentation for details.

---

Usage

1. Open the app and tap the folder icon or the "Open File" button
2. Select a FITS, XISF, or TIFF file from your device
3. The image loads and displays in its original linear state
4. Tap STF to toggle the auto-stretch — the image becomes visible instantly
5. Use Rot L / Rot R to rotate and Flip H / Flip V to mirror
6. Pinch to zoom in/out, drag to pan when zoomed
7. Tap Reset to restore original orientation
8. The info bar at the top shows file details and current STF state

---

Supported Formats

FITS (Flexible Image Transport System)

BITPIX	Type	Supported
8	Unsigned 8-bit integer	✅
16	Signed 16-bit integer	✅
32	Signed 32-bit integer	✅
-32	32-bit IEEE float	✅
-64	64-bit IEEE double	✅

• NAXIS2 (2D mono) and NAXIS3 (3D color) images
• BZERO/BSCALE applied automatically
• Big-endian per FITS standard

XISF (Extensible Image Serialization Format)

Sample Format	Supported
UInt8	✅
UInt16	✅
UInt32	✅
Float32	✅
Float64	✅

• Uncompressed attachments
• XML header parsing
• Planar and pixel-interleaved layouts

TIFF (Tagged Image File Format)

Configuration	Supported
8-bit unsigned	✅
16-bit unsigned	✅
32-bit unsigned	✅
32-bit float	✅
Uncompressed	✅
Compressed (LZW, ZIP, etc.)	❌

---

Architecture

com.astroview/
├── model/           Data classes (AstroImage, ViewerState)
├── decoder/         Format-specific decoders (FITS, XISF, TIFF)
├── engine/          Image processing (AutoStretch, Transforms, Bitmap conversion)
├── viewmodel/       MVVM state management
└── ui/
    ├── theme/       Material 3 dark space theme
    ├── components/  Reusable UI (ImageViewer, BottomToolbar)
    └── screens/     Main viewer screen

Design decisions:

• No external libraries for image decoding — everything is native Kotlin
• MVVM with StateFlow for reactive UI updates
• Pre-computed bitmaps — both original and stretched versions are cached in memory for instant toggle
• Background processing — decoding and stretch computation run on Dispatchers.IO / Dispatchers.Default
• Adaptive system bars — systemBarsPadding() handles gesture nav, button nav, and notches

---

Tech Stack

Component	Technology
Language	Kotlin 1.9
UI Framework	Jetpack Compose with Material 3
Architecture	MVVM + StateFlow
Min SDK	API 26 (Android 8.0 Oreo)
Target SDK	API 34 (Android 14)
Build System	Gradle 8.2 with Kotlin DSL
Image Decoding	Custom native Kotlin decoders
External Dependencies	None beyond AndroidX

---

Requirements

• Android 8.0 (API 26) or higher
• ~4 MB storage
• RAM depends on image size (a 4096×4096 RGB float image uses ~200 MB)

---

Roadmap

• Compressed TIFF support (LZW, ZIP)
• XISF compressed blocks
• Histogram display
• Per-channel stretch controls
• Debayer for raw CFA images
• Dark/Flat reference overlay

---

License

This project is licensed under the MIT License — see the LICENSE file for details.