Creative Results with openExposureFusion: Tips, Presets, and Troubleshooting

openExposureFusion Workflow: Fast Techniques for Natural-Looking MergesopenExposureFusion is a lightweight, scriptable tool for merging bracketed exposures into a single image that preserves highlight and shadow detail without the often-overdone look of conventional HDR. This guide walks you step-by-step through a practical workflow, focusing on speed and natural results. It covers preparation, alignment and registration, parameter choices and ordering, local adjustments, batch processing, and final polishing.

---

Why choose openExposureFusion?

• Fast, non-destructive blending: openExposureFusion combines exposures using multi-scale fusion techniques that are computationally efficient and tend to avoid extreme tone-mapping artifacts.
• Scriptable and automatable: perfect for batch jobs and integration into a larger processing pipeline.
• Natural output: designed to blend exposures for realistic, film-like results rather than exaggerated HDR contrast.

---

1. Preparation: capture and raw conversion

Good results start with good source images.

• Shoot a bracketed sequence with consistent framing (tripod preferred). Typical brackets: -2, 0, +2 EV or -1, 0, +1 for scenes with modest dynamic range.
• Use mirror lock-up or electronic shutter for minimal motion blur.
• Keep ISO low to reduce noise in shadow areas.
• Convert RAW to linear (or near-linear) TIFFs when possible. openExposureFusion works best with high-bit-depth files (16-bit) because it relies on subtle luminance differences for fusion.

Practical tip: If you must shoot handheld, enable in-camera auto exposure bracketing and use a 3-5 frame sequence with small EV steps; alignment tools will handle residual motion.

---

2. Alignment and registration

Even with a tripod, slight shifts may occur. Align before fusing.

• Use an image alignment tool (e.g., built-in openExposureFusion alignment if available, or external tools like Hugin align_image_stack, or align via a raw processor).
• Check for ghosting from moving subjects (people, leaves, water). If present, consider:
  • Masking problematic regions manually, or
  • Using weighted fusion parameters that favor the median or reference frame for motion areas.

Example workflow:

1. Choose the middle exposure (0 EV) as the reference.
2. Align other frames to the reference.
3. Inspect at 100% for edges and small misalignments; re-run alignment with finer control if necessary.

---

3. Choosing the right parameters

openExposureFusion exposes several parameters that control how pixels from different exposures are weighted and combined. Typical parameter groups include:

• Weight maps: exposure, contrast, saturation, and well-exposedness.
• Pyramid scales: number of scales for the multi-scale fusion.
• Sigma values: for local contrast boosting or smoothing.

Guidelines for a natural look:

• Increase exposure weight slightly for midtones to retain natural brightness.
• Keep saturation weight moderate; pushing it high can produce oversaturated colors.
• Use contrast weight conservatively — too much makes the image look “HDR-ish.”
• Use more pyramid scales for fine detail preservation in landscapes; fewer for faster processing.

Quick starting presets:

• Landscape natural: exposure weight 1.0, contrast 0.6, saturation 0.4, scales 6.
• Interiors/architecture: exposure 1.2, contrast 0.5, saturation 0.3, scales 5.
• Fast preview: exposure 1.0, contrast 0.4, saturation 0.3, scales 3 (lower quality but faster).

---

4. Handling moving subjects and ghosting

Motion is the main challenge in exposure fusion.

• Ghost detection: openExposureFusion may include ghost detection—enable it if available. It tries to detect inconsistencies and downweight frames causing artifacts.
• Manual masks: for stubborn cases, paint masks in your editor to force certain exposures to dominate an area (e.g., use the darker frame for highlights or the mid frame for faces).
• Reference-frame locking: lock the mid or best-exposed frame for problematic regions to preserve natural texture and avoid doubled edges.

Example: In a street scene with passing people, use the mid exposure as the reference; for areas where people moved, replace fused result with the reference frame using a soft mask.

---

5. Local adjustments after fusion

After the fusion, treat the image like any RAW edit but with more recovered dynamic range.

• Global tone: small gamma/exposure tweaks if the fusion leans too bright/dark. Avoid heavy global curves that counteract the fusion balance.
• Local contrast: use gentle localized dodge & burn or a low-opacity clarity layer—avoid extreme clarity/structure filters.
• Color grading: correct white balance and perform modest color grading. Since fusion preserves color well, subtle filmic color shifts often look best.
• Noise reduction: apply shadow noise reduction carefully; fusion can amplify shadow noise from the underexposed frames. Use luminance-only denoising and preserve details with edge-preserving methods.

Quick sequence:

1. White balance and exposure fine-tune.
2. Noise reduction on shadows.
3. Local contrast and dodge/burn.
4. Color grading and sharpening (final step—sharpen after resizing).

---

6. Batch processing and automation

One strength of openExposureFusion is scriptability.

• Create presets for typical scenarios (landscape, interior, handheld) and run in batch.
• Preprocess RAW to TIFF in bulk with tools like RawTherapee or dcraw, then feed into openExposureFusion.
• Use a simple shell script or Python wrapper to:
  • Detect bracket sets by filename/exif,
  • Align stacks,
  • Apply preset parameters,
  • Export fused TIFFs.

Example shell pseudo-command:

for stack in $(find . -name "*_bracket_*"); do   align_stack $stack   openExposureFusion --preset landscape --input ${stack}_aligned --output ${stack}_fused.tif done 

---

7. Performance tips

• Use 16-bit processing to avoid banding; but preview in 8-bit for speed.
• Limit pyramid scales for quick previews, then increase for final outputs.
• If your machine has multiple cores, run parallel jobs for separate stacks.

---

8. Final output and export

• Export to 16-bit TIFF for further editing; only convert to 8-bit JPEG for final delivery.
• Resize and sharpen for intended display size (different sharpening levels for web vs print).
• For prints, check dynamic range on a calibrated monitor and soft-proof if necessary.

---

9. Example pipeline (concise)

1. Shoot bracketed RAWs (-2/0/+2 EV).
2. Convert RAW → 16-bit TIFFs.
3. Align images to middle exposure.
4. Run openExposureFusion with landscape preset.
5. Fix ghosting with masks if needed.
6. Apply noise reduction, local contrast, color grading.
7. Export 16-bit TIFF → sharpen & export JPEG.

---

10. Common mistakes to avoid

• Overusing contrast and saturation weights—this creates artificial HDR appearance.
• Skipping alignment—small shifts cause halos.
• Neglecting shadow noise—fusion can exaggerate it without denoising.
• Applying strong global tone curves before fusion—better after.

---

Closing note

openExposureFusion is best used as part of a thoughtful, repeatable workflow: good capture technique, careful alignment, conservative weighting, and subtle local edits yield fast, natural-looking merges.