Architectural Lighting Design: Image Processing and Analysis Software

Architectural lighting design is the field of work or study that is concerned with lighting buildings properly. To light a building properly, there are different requirements: the lighting should be visually pleasing, but it also needs to be well quantified to enable the building users to see properly while avoiding any type of discomfort. Glare is one such type of visual discomfort.

Glare can be quantified through a lighting measurement equipment called a luminance camera. Unfortunately, a luminance camera is a very expensive piece of equipment that few lighting designers or researchers can afford. To overcome this problem, researchers developed a technique called High Dynamic Range Imaging (HDRI) that can quantify glare in a similar way that luminance cameras do, but using commercially available cameras (such as Canon/Nikon DSLR cameras, or even Raspberry Pi camera modules). The HDRI technique requires several steps of image processing and analysis before a glare evaluation can be derived from the images.

This EECS project consists in further improving an open-source and user-friendly desktop application that was created by EECS Capstone teams over the past two years, and which automates those image processing and analysis steps. The desktop app currently reads a series of inputs (such as .jpg images and other text files), applies the image processing and analysis steps (which have been summarized in a publicly available written tutorial), and writes the output file (which is going to be a .hdr image file). The app is built upon the image processing strengths of an existing open-source program (Radiance) and provides a Graphical User Interface for creating and analyzing HDR images. The app design was developed to be user-friendly and intuitive for the users.

Objectives

The main objective is to improve (new features expected) an open-source and user-friendly desktop application to create and analyze High Dynamic Range (HDR) images based on a series of inputs. HDR images are the lighting measurements used for glare quantification in architectural lighting design. There is an existing backlog of potential improvements, but it needs to be reviewed.

The deliverables include an improved desktop application from our existing GitHub repository, together with installation instructions and binaries working on Windows, macOS, and Linux, and documentation.

Motivations

The development of the High Dynamic Range Imaging (HDRI) technique means that hardware for glare quantification and glare research is now economically accessible for the majority of potential users. This first improvement was a huge step forward in enabling lighting designers and researchers to do their jobs better.

However, the HDRI technique still requires the application of several image processing and analysis steps. While a tutorial was recently published explaining all these steps, it is a tedious process and people who want to apply it need to have programming skills and some knowledge of Radiance (a command-line-based light calculation software). This tediousness can be directly observed by the number of posts on the Radiance online forum asking for help to use the HDRI technique.

Providing an open-source and user-friendly desktop application that automates all these steps would make glare quantification and glare research accessible to the majority of potential users. Such an improvement would be the crowning stroke for glare quantification and research, and very much appreciated by the community. By enabling lighting designers and researchers to do their jobs better, it might also have an impact on how well buildings are lit, hence reducing discomfort and accidents (e.g., tripping or falling) due to glare.

Qualifications

Minimum Qualifications:
  • Web development (HTML, CSS, JS)
  • Rust or similar (Tauri backend is Rust)
  • Git
  • Accessible UI/UX design

Preferred Qualifications:
  • Tauri
  • Next.js (React)
  • TailwindCSS
  • GitHub Actions
  • Image manipulation and processing


Details

Project Partner:

Clotilde Pierson

NDA/IPA:

No Agreement Required

Number Groups:

1

Project Status:

Accepting Applicants

Website:
https://github.com/radiantlab/HDRICalibrationTool
Keywords:
WebHCI / UX / UIReactCross-PlatformRustFOSS
Card Image Capstone