Hitachi, Ltd.

Norio Tanaka

Masao Takatoo

Tadaaki Kitamura

Hiroshi Takenaga

Kazuyoshi Asada

Yoshiyuki Okuyama

Kuniyuki Kikuchi

A binary processing method for a variable density image the method includes the steps of binary processing an image having variable density, according to which brightness of an image having variable density to be binary processed is given as 1/n (where n is an integer of 2 or above), and expansion processing and smoothening processing the image having the brightness of 1/n so that a threshold value image for binary processing is generated. A difference is obtained between the threshold value image and the variable density image to be binary processed, such that it is possible to carry out binary processing even for an image having a complex background or variance in brightness.

Method and unit for binary processing in image processing unit and

An information processing apparatus using a neural network learning function has, in one embodiment, a computer system and a pattern recognition apparatus associated with each other via a communication cable. The computer system includes a learning section having a first neural network and serves to adjust the weights of connection therein as a result of learning with a learning data signal supplied thereto from the pattern recognition apparatus via the communication cable. The pattern recognition apparatus includes an associative output section having a second neural network and receives data on the adjusted weights from the learning section via the communication cable to reconstruct the second neural network with the data on the adjusted weights. The pattern recognition apparatus with the associative output section having the reconstructed second neural network performs pattern recognition independently of the computer system with the communication cable being brought into an electrical isolation mode.

Information processing system using neural network learning function

A method and apparatus for binarizing a variable density image wherein brightness of the variable density image is changed by 1/n (where n is an integer of 2 or above), and an expansion processing is carried out on the brightness changed image to generate an expanded image. A smoothing processing is performed on the expanded image to generate a smoothed image and a differential image is generated by taking a difference between the smoothed image and the variable density image. The differential image is binarized by comparing the differential image to a predetermined threshold.

Growing community of inventors

Hitachi, Japan

Norio Tanaka