FAHRIS Algorithm for Fast High-Resolution Image Selection
Author Information
Author(s): Åsebø Bjørnar, Cavazzani Stefano, Bertolin Chiara, Bettanini Carlo, Fusato Giampaolo, Bertolo Andrea, Fiorentin Pietro
Primary Institution: Norwegian University of Science and Technology
Hypothesis
The FAHRIS algorithm aims to improve the classification, grouping, and stitching of high-resolution images acquired from aerial instruments.
Conclusion
The FAHRIS algorithm effectively classifies and stitches images from aerial data, although it faces challenges with image distortion and processing limitations.
Supporting Evidence
- FAHRIS was validated using datasets from three flights over Italy.
- The algorithm allows for quick classification and grouping of images.
- Challenges included computational constraints and image distortion during stitching.
- FAHRIS can operate with low-cost instrumentation and limited computing power.
Takeaway
This study created a smart computer program that helps pick the best pictures taken from drones and balloons, making it easier to see what's happening in the sky.
Methodology
The FAHRIS algorithm classifies images based on motion blur, groups them by various metadata, and stitches them together using feature-based alignment techniques.
Limitations
The algorithm struggles with classifying images due to the absence of EXIF metadata and has a processing limit of six images at a time to avoid memory issues.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website