TokyoTech 5-band Multispectral Image Dataset  [Download]

     This dataset contains 5-band multispectral images of 12 scenes, which were used for the evaluation of multispectral demosaicking algorithms in our published papers. The dataset originally consists of 16 scenes, but we removed 4 scenes that include woman's faces due to a license limitation. The dataset is available only for research purpose and you may not re-distribute the dataset.


Multispectral Demosaicking Codes  [Package Download]

     This package contains MATLAB codes of several multispectral demosaicking algorithms that we have proposed for our proposed five-band multispectral filter array. Please see included readme files for the usage of the codes and see the following references for the descriptions of our project and algorithms.


  • Adaptive Residual Interpolation for Color and Multispectral Image Demosaicking   [PDF]
    Yusuke Monno, Daisuke Kiku, Masayuki Tanaka and Masatoshi Okutomi
    Sensors, vol.17, no.12, pp.2787-1-21, December, 2017.

  • A Practical One-Shot Multispectral Imaging System Using a Single Image Sensor   [Project]
    Yusuke Monno, Sunao Kikuchi, Masayuki Tanaka and Masatoshi Okutomi
    IEEE Transactions on Image Processing (TIP), vol.24, no.10, pp.3048-3059, October, 2015.

  • Multispectral Demosaicking with Novel Guide Image Generation and Residual Interpolation   [Project]
    Yusuke Monno, Daisuke Kiku, Sunao Kikuchi, Masayuki Tanaka and Masatoshi Okutomi
    IEEE International Conference on Image Processing (ICIP2014), pp.645-649, October, 2014.

  • Multispectral Demosaicking Using Guided Filter   [Project]
    Yusuke Monno, Masayuki Tanaka and Masatoshi Okutomi
    IS&T/SPIE Electronic Imaging (EI2012), Digital Photography VIII (Proc. SPIE 8299), pp.82990O-1-7, January, 2012.

  • Multispectral Demosaicking Using Adaptive Kernel Upsampling   [Project]
    Yusuke Monno, Masayuki Tanaka, and Masatoshi Okutomi
    IEEE International Conference on Image Processing (ICIP2011), pp.3218-3221, September, 2011.

See Also

    TokyoTech 5-band Multispectral RAW Dataset
    TokyoTech 31-band Hyperspectral Image Dataset
    TokyoTech 59-band Visible-NIR Hyperspectral Image Dataset

Contact Information

     If you have any questions, please feel free to ask Yusuke Monno (ymonno(at)