Hirotada Ueda

IMPACT: an interactive natural-motion-picture dedicated multimedia authoring system

A new approach to achieving a natural-motion-picture dedicated multi-media authoring system is proposed. The main point of this approach, discussed in this paper, is that the user’s environment or interface is improved to encourage user’s creativity, with image processing and recognition technology. According to the discussion, a prototype motion picture authoring system dtat has several image-processing functions is developed. The newly developed functions include object extraction of the picture, semi-automatic visualization of motion pictures structure, and certain descriptions of the scene. Result of using the prototype shows the appropriateness of the proposed approach.

Automatic structure visualization for video editing

We developed intelligent functions for the automatic description of video structure, and visualization methods for temporal-spatial video structures obtained by these functions as well as for the functions. The functions offer descriptions of cut separations, motion of the camera and filmed objects, tracts and contour lines of objects, existence of objects, and periods of existence. Furthermore, identical objects are automatically linked. Thus the visualization methods supported by object-links allow users to freely browse and directly manipulate the structure including descriptions and raw video data.

A multiprocessor system utilizing enhanced DSPs for image processing

A general-purpose image processor (GPIP) consisting of 64 digital signal processors (DSPs) in a 0.31-m/sup 3/ box is proposed to perform a wide range of image processing tasks. A high-speed DSP called DSP-i has been especially developed for this purpose. It has a highly parallel architecture with a two-level instruction hierarchy, multibank cache, and multiprocessor interface. The DSP-i machine cycle is 50 ns. A novel ring shift register bus architecture offers a flexible structure and an efficient data-exchange method for the system. Along with four proposed operation modes, it cuts the multiprocessing overhead to as little as 20%. The performance of the GPIP is 1000 MOPS (million operations per second).<<ETX>>

Automatic scene separation and tree structure GUI for video editing

INTRODUCTION Through the discussion with the users of cument nonlinear editing system, we found that they have a common complaint. That is a manual scene cutting process as a preparation for the nonlinear icon based editing. Users also request a method to see the total video structure in several levels of detail for more effective video editing. We propose an approach to integrate image recognition technology into nonlinear video editing system. It consists of three concepts; automatic scene separation [2] during disk recording, quick review with micon [1], and tree structure story editing. This approach is the exact sohttion for the complaint of the current system users. The automatic scene separation removes time consuming manual separation process and user can immediately start editing after the hard disk recording. The quick review with micon is a quick and convenient shot comparison method. The tree structure story editing provides a method to edit video structure in several different detail levels and relieves the constrain of limited size of the window.

A 50ns DSP with parallel processing architecture

This report will cover a programmable DSP that avoids I/O bottlenecks through a two-level hierarchy of instructions. The IC contains 430K transistors on a 149mm2die.

FRIEND21 project: a construction of 21st century human interface

The FRIEND21 project is a research project for the human interface of the 21st century computing machinery. The target of the project is to come up with principles that should be embodied into those machinery forthosepeople who use them in the 21st century information society. We propose a new design principles of constructing such human interface using multiple metaphors forpeople’s direct engagement into the information environment called Contextual Metaphors and anew software architedurecalled Agency Model for implementing it.

The application of image processing technology to industrial automation

Abstract Some typical applications of image processing technology in Japanese industry are introduced in this paper. A steady growth of application has been seen in a number of industrial fields over the past decade. The motivation for such applications has always been the demand for more efficient production systems and product quality improvement by eliminating human error. Results so far are promising, though the barriers presented by existing limitations to binary image processing remain to be overcome.

An Industrial Eye that Recognizes hole Positions in a Water Pump Testing Process

In the production of small water pumps, an automatic machine was requested for water pressure testing of the final products, in which hoses are automatically connected to inlet and outlet flange holes of the pumps. For this purpose, a visual device has been developed to detect the hole positions by means of CCD type TV cameras. This “industrial eye” also includes special electronic circuitry for pattern matching, which utilizes quarter patterns of a circular hole image as templates.

FRIEND21 Project: two-tiered architecture for 21st-century human interfaces

The FRIEND21 Project, which ended on March 31,1994, is briefly described and its final results are presented. The FRIEND21 Project conducted research into human interfaces having two-tiered architecture consisting of “metaware” and “agency model.” This approach attempts to deal with human-interface issues from the very top layer of a symbolic and cognitive environment constructed between the user and the machine to the bottom layer of machine functions organized into the agency model.

Direction-invariant methodology for bill pattern recognition

This paper presents a new pattern recognition method for the automatic discrimination of various kinds of bills. The pattern of the object input from the sensor fluctuates due to the a posteriori factors such as the orientation, position and rotation errors, as well as the a priori factors such as wrinkles and smears. The proposed method is to cope mostly with the fluctuations due to the a posteriori factors. It is noted first that the patterns obtained by the two sensing elements at the point-symmetrical positions in regard to the center of the arrayed sensors are equivalent in terms of the orientation. Then, as the feature parameter for the discrimination, some Hadamard features are adopted which are suited to the processing by the microprocessors. Two feature parameters are derived from the bill pattern, which have a higher discrimination ability, and the “mutually paired domain judgment method” is proposed, where the matching discrimination is applied by exchanging those two features as feature points in the two-dimensional space. The decision region to be used in the discrimination has the orientation-invariant property due to the above equivalence, which helps to reduce greatly the memory capacity. Finally, the result of the evaluation experiment using the actual samples is presented. It is shown that the required misrecognition rate can be cleared if the number of elements with the point-symmetry is increased. It is anticipated that the proposed method can also be used in the automatic recognition of other sheet-shaped printings.