Masahide Yoneyama

Millimeter-wave active imaging using neural networks for signal processing

A neural network has been successfully implemented in an active-mode millimeter-wave (60 GHz) imaging system with a Yagi-Uda antenna array in order to recognize objects and reconstruct images that appear distorted under coherent millimeter-wave illumination. With 10 /spl times/ 10 sampling points and five teaching trials, a recognition rate of 98% has been obtained for ten dissimilar alphabetical letters used as objects. The success rate of reconstruction of distorted millimeter-wave images was 80% when five dissimilar letters were used for the reconstruction. The recognition rate after changing the spatial resolution of the optical system and sampling interval of the image is also discussed.

A method to classify emotional expressions of text and synthesize speech

Representing emotional expressions on text-to-speech synthesis has been a interesting subject. We aim to add emotional expressions to synthetic speech. This system recognizes emotion based on the words in a sentence and synthesizes emotional speech.

Robustness against instability of sensory judgment in a human interface to draw a facial image using a psychometrical space model

We proposed a human interface to search and draw an image that a user imagines, as a cartoon. An IGA (Interactive Genetic Algorithm) was applied to search for a target image in a users psychometrical space. Fuzzy reasoning was adopted to reduce the users burden in assigning the fitness in the IGA. Furthermore, fuzzy reasoning was executed in the psychometrical space to reflect human senses on the fitness assignment process. This resulted not only in a lighter burden and the effective reflection of human senses but also robustness against instability of sensory judgment in the human interface that we proposed.

An ultrasonic robot eye for object identification using neural network

To construct an ultrasonic robot eye, a new system which combines acoustical imaging with a neural network was developed. In this system, by measuring sound pressures of scattered waves by a receiver array, 3-D object categories are identified, their locations estimated, and their images improved. The structure of the system is described, and some experimental results are reported.<<ETX>>

Image Restoration For Millimeter Wave Images by Hopfield Neural Network

Publisher Summary A vision system is a key technology to develop intelligent robots working in the extremely severe circumstances in place of a human worker. This robot vision is required to reconstruct an objects image and telecommunicate this image to an operator for existence. To create such a vision system, many technologies have been proposed and researched. This chapter proposes a discrete type hopfield neural network for use as a post processor for an imaging radar using 60GHz millimeter wave. The millimeter wave image obtained by the imaging radar is generally much degraded because of design limitation of the imaging system. The role of the post processor is to restore the degraded millimeter wave images. The hopfield neural network is capable of recalling the correct original images from the degraded input image by means of its associative memory effect. Some experimental results using two dimensional objects are reported in the chapter.

A Nonlinear Ultrasonic Imaging Method Based on the Modified Information Criterion

It has been shown that acoustical imaging is useful for three-dimensional object recognition in robotic sensing and factory automation, and that it is refined by neural information processing1,2. The 3-D shape information which is extracted from the backscattered acoustic waves is precisely restored and identified by artificial neural networks. However, there remains a problem of the neuro-ultrasonic recognition system. It is still difficult to optimize several parameters automatically, resulting that a lot of trials and errors are needed to construct a reliable system for practical uses.

3D face synthesis based on the information of words expressing facial features

Our aim is to synthesize faces based on freely-elicited expressions by expanding the range of words describing the shape of facial elements to include abstract or metaphorical expressions. We realize this by defining the synthesizing process of a human face as a mapping from a word space to a physical model space. The use of whole words existing in the word space has made it possible to synthesize a human face based on freely-elicited expressions.

Reconstruction of the Ultrasonic Image by the Combination of Genetic Programming and Constructive Solid Geometry

We propose a method for improving the highly-degraded images obtained by ultrasonic imaging system. To reconstruct 3-dimensional images from highly-degraded acoustic images, strongly typed genetic programming including constructive solid geometry which is an expression method of 3-dimensional object is used as post processor. In the tree structure of genetic programming, the temporally node is established in order to enable the individual change of type and position information of the primitive. By the experimental results, it is possible to specify type and position of the primitive from the 2-dimensional ultrasonic wave image

Hybrid fitness assignment strategy in IGA

We have been developing a hybrid fitness assignment strategy to realize a natural interaction in IGA. The strategy allows a user to select some individuals and evaluate a grade that shows how the selected individual resembles a target image. In this paper, we will show a method to compose fitness when a user selects two individuals in the hybrid fitness assignment strategy. It is known that better performance is obtained when two individuals are selected in the generations limited with a condition. The condition is equivalent to the actual situation in which it is difficult for a user to select only one individual. The hybrid strategy is useful to realize a more natural interaction in the actual situation.

Reconstruction for Artificial Degraded Image Using Constructive Solid Geometry and Strongly Typed Genetic Programming

Acoustic imaging is effective in extreme environments to take images without being influenced by optical properties. However, such images tend to deteriorate rapidly because acoustic impedance in air is low. It is thus necessary to restore the image of the object from a deteriorated image so that the object can be recognized in a search. We used a neural network in the previous work as a postprocessor and tried to reconstruct the original object image. However, this method needs to learn the original object image. In this work, we propose combining Constructive Solid Geometry (CSG) with Genetic Programming (GP) as a new technique that does not require learning. To confirm the effectiveness of this technique, we reconstruct the image of an object from a deteriorated image created by applying a 2-dimensional sinc filter to the original image.