hdr

HDR Image Capture and Processing Example

This example demonstrates how to acquire multiple images with different exposure times using an IDS camera and combine them into a single High Dynamic Range (HDR) image. It supports both live acquisition from a connected camera and offline HDR generation from previously saved exposure sequences.

Requirements

This example depends on the following non-Python components:

• A GenTL producer (e.g. installed via the IDS peak Setup).
• A compatible IDS camera.

Required third-party Python packages

• opencv-python
• matplotlib
• numpy>2

Required IDS peak Python packages:

• ids-peak-common>=1.3.0
• ids-peak-icv>=1.3.0
• ids-peak>=1.14.0

Install Python dependencies:

pip install -r requirements.txt

Running the Example

The application can either:

• Capture a sequence of images from a connected IDS camera and generate an HDR image
• Generate an HDR image from previously saved exposure images

Capture HDR Images from Camera

Basic example:

python main.py --min 0.5 --max 50 -n 4

This will:

• Capture 4 image with different exposures
• Exposure range from 0.5 ms to 50 ms
• Space exposure values logarithmically across the range

If --min and --max are not specified, a default range is used.

Save Results to Disk

python main.py --min 0.5 --max 50 -o output_folder

Saved files include:

• seqX_YmsZZZ.png → individual exposure images
  • X = image index
  • Y = full milliseconds
  • ZZZ = decimals of the milliseconds
• hdr_preview_8bit.png → 8-bit preview image
• hdr_image.tiff → full HDR image
• hdr_image.hdr → Radiance HDR format

Select HDR Acquisition Mode

python main.py --min 1 --max 30 --acquisition-mode MODE_NAME

By default, the most suitable mode is selected automatically. With --acquisition-mode, a specific mode can be enforced.

Available modes:

• AUTO
  Automatically selects the best available acquisition mode.
• PROGRESSIVE
  In progressive mode the exposure value is changed programmatically before each image is captured. This mode works for every camera that changes exposure instantly.
• SEQUENCER
  Sequencer mode uses the camera's sequencer feature which is preconfigured for the given exposure times.
• QUAD_EXPOSURE
  Sensor feature for IMX900-based cameras that capture a single image using four distinct exposure times arranged in a 2×2 pixel pattern.
  Note: When using QUAD_EXPOSURE, set -n 4.
• CLEAR_HDR
  Sensor feature for IMX675-based cameras that capture a single image using two gain factors in alternating lines starting with lower gain.
  Note: If you choose CLEAR_HDR, you also need to set -n 2.

Available modes are printed at runtime based on camera support.

Run Without GUI

For console-only or headless environments:

python main.py --min 1 --max 30 -o output_folder --no-gui

• No display window is shown
• Errors and progress are printed to the console

Note: If no output directory is specified in --no-gui mode, the results are not saved.

Process Existing Image Sequence

Instead of capturing from a camera, process images from a directory:

python main.py -i input_folder -o output_folder

This will:

• Load the exposure sequence from input_folder (typically a previously captured set)
• Generate the HDR image
• save results to output_folder (optional)
• Display the HDR result (unless --no-gui is set)

Command Line Arguments

Argument	Description
--min	Minimum exposure in milliseconds
--max	Maximum exposure in milliseconds
-n, --num-exposures	Number of exposures (default: 4)
-s, --serial	Camera serial number
--no-gui	Disable GUI elements
-o, --out-dir	Output directory for images
-i, --in-dir	Use images from directory instead of camera
-a, --acquisition-mode	HDR acquisition mode

Notes

• --min and --max must always be provided together.
• Exposure steps are logarithmically spaced across the specified range.
• Exposure values are provided in milliseconds.
• The camera is reset to default settings before and after acquisition.
• If no acquisition mode is specified, the preferred HDR provider is selected automatically.