Researcher

Hiroshi Sako

Also known as 迫宏

Hiroshi Sako is a machine-vision researcher whose work spans factory automation, handwriting recognition and medical-image screening. At Hitachi's Central Laboratory, he helped develop a vision-based transistor assembly system used in 1973. A 1997 patent named him as a co-inventor of a method and apparatus for processing facial-expression images. IAPR named him a Fellow in 2004, and he coauthored a study on handwritten numeral recognition. He cochaired the 2008 IAPR Document Analysis Systems workshop, wrote an IEICE machine-vision foreword in July 2008, coauthored a 2009 paper on retinal screening, and served on MVA's advisory board in 2017.

Works
69
Citations
1,588
h-index
16
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 1973

    Transistor assembly by vision

    At Hitachi's Central Laboratory, Sako is among the engineers behind a vision-based transistor assembly system that enters use in 1973. It uses pattern recognition to locate electrodes on tiny chips.

  2. 1997

    Facial-expression image processing

    A 1997 patent names him as a co-inventor of a method and apparatus for processing facial-expression images.

  3. 2004

    An IAPR Fellowship

    IAPR names him a Fellow in 2004 for contributions to developing and applying machine-vision technology for factory and office automation.

  4. 2004

    Handwritten numeral recognition

    In 2004, he coauthors “Effects of Classifier Structures and Training Regimes on Integrated Segmentation and Recognition of Handwritten Numeral Strings” with Cheng-Lin Liu and Hiromichi Fujisawa.

  5. 2008

    DAS workshop chair

    He co-chairs IAPR's Document Analysis Systems workshop in 2008.

  6. July 2008

    Machine-vision foreword

    In July 2008, he writes the foreword to IEICE's Special Section on Machine Vision and its Applications.

  7. 2009

    Fundus-lesion screening

    In 2009, he coauthors a paper on a system that screens fundus images for diabetic-retinopathy lesions. The system selects classifiers by candidate-lesion position and reports 98% sensitivity and 81% specificity.

  8. 2017

    MVA advisory board

    For MVA 2017, he serves on the conference advisory board.

Most-cited works

  1. Handwritten digit recognition: benchmarking of state-of-the-art techniques2003 · Pattern Recognition · 543 citationsWith Cheng-Lin Liu and Hiromichi Fujisawa
  2. Handwritten digit recognition: investigation of normalization and feature extraction techniques2003 · Pattern Recognition · 297 citationsWith Cheng-Lin Liu and Hiromichi Fujisawa
  3. Discriminative Learning Quadratic Discriminant Function for Handwriting Recognition2004 · IEEE Transactions on Neural Networks · 112 citationsWith Hiromichi Fujisawa
  4. Performance evaluation of pattern classifiers for handwritten character recognition2002 · International Journal on Document Analysis and Recognition (IJDAR) · 96 citationsWith Cheng-Lin Liu and Hiromichi Fujisawa
  5. Handwritten Chinese character recognition: alternatives to nonlinear normalization2004 · Proceedings of the International Conference on Document Analysis and Recognition · 52 citationsWith Cheng-Lin Liu and Hiromichi Fujisawa

Sources and credits

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

  1. Transistor Assembly System robogaku.jp. Checked 21 October 2026.
  2. US5689575A - Method and apparatus for processing images of facial expressions patents.google.com. Checked 21 October 2026.
  3. Alphabetical List of IAPR Fellows iapr.org. Checked 21 October 2026.
  4. Chenglin Liu csauthors.net. Checked 21 October 2026.
  5. Workshop Report: DAS 2008 iapr.org. Checked 21 October 2026.
  6. FOREWORD jstage.jst.go.jp. Checked 21 October 2026.
  7. Detection of Fundus Lesions Using Classifier Selection jstage.jst.go.jp. Checked 21 October 2026.
  8. Organizing Committee | IAPR MVA2017 mva-org.jp. Checked 21 October 2026.