Credentials

The following list presents a non-exhaustive list of papers and applications pertaining to CogniMem chip or its ancestor the ZISC chip. For your convenience, the comparison between these two generations of chips is made at the end of this document.

Applications: Face recognition, Artificial nose, Target tracking , Machine vision, Classification, Lidar imaging, Bioinformatics, Adaptive control

Face recognition #1

Face recognition using the RBF classifier of the ZISC chip mounted a PCMCIA card.

• Paper:              Eide_faceRec.pdf, presented in 1994

• Main author:     Pr. Thomas Lindblad at KTH Royal Institut of technology

Face recognition #2

Implementation of a RBF neural network on embedded systems: Real time face tracking and identity verification.

• Short paper:     face346.pdf  (complete version published in IEEE Transactions on Neural Networks, Vol.14 (N°5), Sep 2003)

• Presentation:    Real-time multi-face recognition

• Company:         University of Bourgogne, Laboratoire Electronique, Informatique et Image

• Authors:           Fan Yang, Michel Paindavoine

Face recognition #3

A combination of adaptive neural network systems and statistical method for integrating speech and face image information for person identification. The face recognition subnetwork described in this paper is based on the ZISC chip.

• Paper:              davidicarcv.pdf, presented at 8th International Conference on Control, Automation, Robotics and Vision in 2004

• Company:         Khe Knowledge Engineering and Discovery Research Institute (KEDRI), New Zealand

• Author:             David Zhang, Akbar Ghobakhlou and Nikola Kasabov

Artificial Nose

Pr. Lindblad and his team presented a case at the Excitera Innovation Challenge  2009 for an artificial nose trained to detect cancer. The classifier used for the detection is an RBF neural network. Pr. Lindblad, one of the first advocates of the ZISC chip, intends to deploy the above solution on hardware with the new CogniMem chip. (http://innovationchallenge.se/cases/enose)

Machine Vision

Fish Inspection System Using a Parallel Neural Network Chip and the Image Knowledge Builder Application (published in 2007 for the Association for the Advancement of Artificial Intelligence)

• Paper:              http://www.aaai.org/ojs/index.php/aimagazine/article/viewArticle/2084

• Company:         General Vision

Target Tracking

Infrared Target-Flare Discrimination using a ZISC Hardware Neural Network ()

• Paper:             Labonte_Deck.pdf (published in Journal of Real-Time Processing in April 2009)

• Company:        Royal Military College

Classification

Data Classification using ZISC-Digital Neural network and applied to a variety of data sources

• Paper:              neucom_Mahesh.pdf (Academic presentation)

• Company:         Institute of Integrated Sensor Systems

• Authors:            Mahesh Poolakaparambil and Prof.Dr.Ing Andreas König

Lidar Imaging

Onboard Feature Indexing from Satellite Lidar Images using the RBF classifier of the ZISC chip

• Paper:              iwadc-nasa.pdf (published in Proceedings of IEEE IWADC in 2003)

• Company:         Carnegie Mellon University and NASA Langley Research Center

BioInformatics

Design of a fast, accurate, and inexpensive way to detect pathogens using a biosensor. The envisioned biosensor is based on the patented “Intelliglass” concept which will allow to embed a CogniMem chip into a microscope slide.

• Presentation:    Microarrays for Pathogens Detection with IntelliGlass

• Company:         University of Hawaii, Bioinformatics Laboratory

• Author:             Guylaine Poisson

Adaptive Control

Design of a new electronic equipment to help handicapped patients walk again without having to check the electric stimuli levels. The proposed equipment is based on the CogniMem chip to provide a real-time closed loop regulation necessary to adapt the electrical levels to the angular position of each leg.

• Sponsors:         European Laboratory for Sensory Intelligence (created in January 2009)

Comparison between the ZISC and CogniMem chips

ZISC stands for Zero Instruction Set Computer and is a neural network chip which was invented jointly by a team of engineers at IBM France and Guy Paillet in 1993. The chip was manufactured by IBM between 1993 and 1999 and marketed consecutively as the ZISC36 and ZISC78 chips.

Following the termination of the manufacturing of the ZISC chip by IBM, Guy Paillet has built a new team to work on the development of the next generation of the neural network chip. As a result, CogniMem LTD ( Hong Kong ) released the CogniMem chip (CM1K) in August 2007. CogniMem stands for Cognitive memory and implements the same patents as the ZISC chip and several significant improvements.

Following is a specification chart comparing the ZISC and the CogniMem CM1K chip.