Repository hosted by TU Delft Library

Home · Contact · About · Disclaimer ·

Perceptual evaluation of different image fusion schemes

Publication files not online:

Author: Toet, A. · IJspeert, J.K.
Publisher: The International Society for Optical Engineering
Place: Bellingham, WA
Institution: TNO Technische Menskunde
Source:Kadar, I., Signal Processing, Sensor Fusion, and Target Recognition X, 4380, 436-441
Proceedings of SPIE
Identifier: 10972
doi: doi:10.1117/12.436969
Keywords: Vision · Detection · Image fusion · Infrared · Intensified visual · Recognition · Situational awareness · Visual performance · Cameras · Charge coupled devices · Color image processing · Image analysis · Image sensors · Object recognition · Photomapping · Sensitivity analysis · Image fusion schemes · Sensor data fusion


Human perceptual performance was tested with images of nighttime outdoor scenes. The scenes were registered both with a dual band (visual and near infrared) image intensified low-light CCD camera (DII) and with a thermal middle wavelength band (3-5 μm) infrared (IR) camera. Fused imagery was produced through a pyramid image merging scheme, in combination with different colour mappings. For all (individual and fused) image modalities, small patches of the scenes, displaying a range of different objects and materials, were briefly presented to human observers. The sensitivity of human observers was tested for different recognition tasks. The results show that greyscale image fusion yields improved performance levels for most perceptual tasks investigated here. When an appropriate colour mapping scheme is applied, the addition of colour to greyscale fused imagery significantly increases observer sensitivity for a given condition and a certain task. However, inappropriate use of colour significantly decreases observer performance compared to straightforward greyscale image fusion. This suggests that colour mapping should adapt to the visual task and the conditions (scene content) at hand