English/Japanese

What's New

2023/09/28 We accepted an exchange Master course student from Autonomous University of Baja California (Mexico). He will work on a project on the following topic: "Developing AI using spiking neural networks."
2023/09/08 An exchange Master course student from Tampere University (Finland) gave his final presentation of his internship study. He developed a method for increasing the learning efficiency of a spiking neural network. (Photos taken at the presentation are here)
2023/08/01 A Doctor course student Mr. Morita joined Prof. Hsin Chen's group at National Tsing Hua University in Taiwan as an internship student. He will develop a system for analizing the relationship between the brain electrical activity and the behavior of the whole body of a rat, by combining the brain implant electrode developed by Prof. Hsin Chen's group and neuromorphic image recognition technique developed by our lab.
2023/04/03 We accepted an exchange Master course student from Tampere University (Finland). He will work on a project on the following topic: "Information coding and decoding in spiking neural networks."
2023/02/09 -12 Master course students, Mr. Kobayashi, Mr. Sato, and Mr. Ikeda gave online presentations at International conference on Artificial Life and Robotics (ICAROB2023). The titles are "A Figure-Ground Discrimination Algorithm Inspired by Border-Ownership Selective Cells", "Binocular Disparity Estimation Algorithm Using Multiple Spatial Frequency Information and a Neural Network", "Increasing Selectivity to a Feature Combination Using Inhibitory Synaptic Plasticity in a Spiking Neural Network".
2022/12/03 A Doctor course student, Mr. Morita and a Master course student, Mr. Yamaji gave oral presentations at a workshop of the Technical Committee on Neurocomputing held at Osaka, Japan. The titles are "Image Classification Using Gabor Filters as Preprocessing for CNNs", "Spatio-temporal filtering circuit simulating the characteristics of the visual nervous system". (in Japanese)
2022/11/24 Dr. Okuno gave an invited talk at a lecture serieas "Neural circuits providing functions" produced by the Total Biomimetics Research Group of The Japan Society of the Applied Physics. The title was "Development of artificial intelligence inspired by the visual nervous system." (in Japanese)
2022/07/24 A student project "AI smart car development project", to which our laboratory provided an image recognition system that is insusceptible to the intensity and the color of illumination, participated in Nakanoshima robot challenge, held at Osaka, Japan. The travel distance of our robot was 330 m (the course was 477.25 m), and the robot cleared "finding a human test", and "object classification test". The performance of our robot is shown in the following movie.
2022/01/20 -23 Master course students, Mr. Morita, Mr. Hisamitsu, and Mr. Yamaji gave online presentations at International conference on Artificial Life and Robotics (ICAROB2022). The titles are "The Effect of Preprocessing with Gabor Filters on Image Classification Using CNNs", "An Image Coding Algorithm with Color Constancy Using the Retinex Theory and the Naka-Rushton Equation", "A Fast Image Sensor System with an Efficient Multi-Scale Gaussian Filtering Circuit".
2021/12/08 -12 A master course student, Mr. Horiguchi gave an online presentation at International conference on Neural Information Processing (ICONIP2021). The title is "A Visual Feature Detection Algorithm Inspired by Spatio-temporal Properties of Visual Neurons".
2021/06/06 -08 Master course students, Mr. Hayashi, Mr. Hisamitsu, and Mr. Yamaji gave online presentation at the Robotics and Mechatronics Conference 2021. The titles are "A real-time binocular disparity detection system inspired by the spatio-temporal properties of the visual nervous system", "An image sensor system with color constancy using the Retinex theory and the Michaelis-Menten equation", "FPGA implementation of a neuro-inspired algorithm for spatio-temporal visual feature extraction", respectively.
2021/01/21 -23 Master course students, Mr. Morita, Mr. Horiguchi, and Mr. Hayashi, gave online presentations at 2021 International Symposium on Artificial Life and Robotics (AROB2021). The titles are "Motion and shape classification using spiking neural network" "A spatio-temporal feature extraction algorithm that simulates a physiological model of a neuron in the early visual cortex" "A binocular disparity detection algorithm inspired by the spatio-temporal properties of binocular cells".
2020/12/19 -20 The robocup team "O.I.T. Challenger", to which our laboratory provided a binocular image sensor system that is insusceptible to the intensity and the color of illumination, won the second prize at the open platform league of Robocup at home, Robocup Japan Open 2020, held online.
2020/08/05 A master course student, Mr. Horiguchi, gave an online presentation at Meeting on Image Recognition and Understanding 2020. The title is "A visual feature extraction algorithm inspired by the spatio-temporal property of neurons in the visual cortex.".
2020/05/29 Master course students, Mr. Kurose, Mr. Suenaga, Mr. Hayashi, and Mr. Morita gave online presentation at the Robotics and Mechatronics Conference 2020. The titles are "Estimation of drone motion state using opitical flow and echo state network", "Gesture classification utilizing reservoir computing and spatio-temporal features inspired by the visual nervous systems", "A neuro-inspired binocular stereo vision system that uses texture boundaries for corresponding point search", "Human action classification using the motion detection model of the insect visual system and the echo state network", respectively.
Before FY2019 Information is here

You can scroll the list above.

Research goals

    This laboratory has the following two goals.
   The first goal is to develop efficient systems for visual perception and cognition inspired by the visual nervous systems. The visual nervous systems perceive and recognize visual signals efficiently in terms of both energy and time, while understanding of visual information is one of the most difficult task for conventional computers even though its processing speed is much faster than that of neurons. Learning from the advantages of nervous systems is a promissing way to develop noble visual systems.
    The second goal is to reveal how the efficient visual signal processing is accomplished in the visual nervous systems, including the retina and the visual cortex. We have to reveal it before learning from the visual nervous systems because they remain poorly understood.
    Understanding the visual nervous systems and developing noble systems for visual perception and cognition inspired by the nervous systems: these two points are the keys of this laboratory.

Student projects that our laboratory supports

1. Escape room utilizing IT development project
2. AI smart car development project
   (Introduction movies: preparation, competition)

Keywords

neuromorphic, retina, brain, visual cortex, AI, robot, image sensor, image processing, neural network, FPGA, circuit design, algorithm development

Contact

Affiliation:Osaka Institute of Technology, Faculty of Information Science and Technology
Location:1-79-1 Kitayama, Hirakata City, Osaka, 573-0196 Japan
 Buliding number 1, Room number 515
E-mail:hirotsugu.okuno [ATMARK] oit.ac.jp

Copyright © 2016 Neuromorphic systems laboratory All Rights Reserved.