Researcher

Atsushi Imiya

Also known as 井宮 淳

Atsushi Imiya is a researcher in computer vision and pattern recognition whose work spans signal smoothing, randomized Hough methods, optical flow and discrete geometry. He published a generalized Lanczos method for signal smoothing in 1986 and applied the randomized Hough transform to piecewise-linear signal detection in 1996. His work on featureless robot navigation used optical flow in 2005, and he became an IAPR Fellow in 2008 for contributions to randomized model fitting and discrete geometry. At Chiba University, he led projects on computing mean organ images for clinical diagnosis in 2009–2011 and model-based longitudinal analysis of dynamic multi-modal higher-dimensional data in 2020–2024; he held a Doctor of Engineering from Tokyo Institute of Technology by 2024. As of October 2026, he is a Chiba University emeritus professor.

Works
297
Citations
1,510
h-index
17
Badges

Publication numbers from OpenAlex, as of 18 October 2026.

Timeline

  1. 1986

    Generalized Lanczos method

    He publishes a paper on a generalized Lanczos method for signal smoothing in 1986.

  2. 1996

    Randomized Hough transform

    His 1996 paper applies the randomized Hough transform to piecewise-linear signal detection.

  3. 2005

    Featureless robot navigation

    He coauthors a paper on featureless robot navigation using optical flow in 2005.

  4. 2008

    An IAPR Fellow

    He becomes an IAPR Fellow in 2008 for contributions to randomized model fitting in pattern recognition and computer vision, and to discrete geometry.

  5. 2009–2011

    Computing mean organ images

    As principal investigator, he leads a Chiba University project on computing mean organ images for clinical diagnosis from 2009 to 2011.

  6. 2013

    Curvature flow in digital space

    His 2013 paper introduces curvature flow for isothetic polytopes defined in digital space.

  7. 2020–2024

    Model-based data analysis

    He leads a Chiba University project on model-based longitudinal analysis of dynamic multi-modal higher-dimensional data from 2020 to 2024. By 2024, he holds a Doctor of Engineering from Tokyo Institute of Technology.

  8. As of October 2026

    Emeritus professor at Chiba

    He is a Chiba University emeritus professor.

Career

  1. Since 2009Chiba UniversityPrincipal Investigator

Education

  1. 2024Institute of Science TokyoDoctor of Engineering

Most-cited works

  1. Linear Scale-Space has First been Proposed in Japan1999 · Journal of Mathematical Imaging and Vision · 180 citationsWith Joachim Weickert
  2. Digital and Image Geometry2001 · Lecture notes in computer science · 93 citations
  3. On the History of Gaussian Scale-Space Axiomatics1997 · Computational imaging and vision · 93 citationsWith Joachim Weickert
  4. Fast Spectral Clustering with Random Projection and Sampling2009 · Lecture notes in computer science · 61 citations
  5. Dominant plane detection from optical flow for robot navigation2006 · Pattern Recognition Letters · 51 citations

Sources and credits

Photo: We have no published photo of Atsushi Imiya yet. Send us a free-licensed photo.

  1. Atsushi Imiya csauthors.net. Checked 18 October 2026.
  2. Atsushi Imiya - Misc. - researchmap researchmap.jp. Checked 18 October 2026.
  3. Alphabetical List of IAPR Fellows – International Association for Pattern Recognition iapr.org. Checked 18 October 2026.
  4. KAKEN — Research Projects | Computation of Mean Organs and Its Application to Clinical Diagnosis (KAKENHI-PROJECT-21500158) kaken.nii.ac.jp. Checked 18 October 2026.
  5. Curvature and Flow in Digital Space numdam.org. Checked 18 October 2026.
  6. KAKEN — Research Projects | Model-based longitudinal analysis of dynamic multi-modal higher-dimensional data (KAKENHI-PROJECT-20K11881) kaken.nii.ac.jp. Checked 18 October 2026.
  7. Atsushi Imiya - My portal - researchmap researchmap.jp. Checked 18 October 2026.