Patent engineers are spending significant time analyzing patent claim structures to grasp the range of technology covered or to compare similar patents in the same patent family. Though patent claims are the most important section in a patent, it is hard for a human to examine them. In this paper, we propose an information-extraction-based technique to grasp the patent claim structure. We confirmed that our approach is promising through empirical evaluation of entity mention extraction and the relation extraction method. We also built a preliminary interface to visualize patent structures, compare patents, and search similar patents.

Gallium nitride high-electron mobility transistors (GaN HEMTs) have attractive properties, low on-resistances and fast switching speeds. This paper presents the characteristics of a normally-on GaN HEMT that we fabricated. Further, the circuit operation of a Class-E amplifier is analyzed. Experimental results demonstrate the excellent performance of the gate drive circuit for the normally-on GaN HEMT and the 13.56MHz radio frequency (RF) power amplifier.

This paper describes the installation of a mathematical formula recognition module into an open source OCR system: OCRopus. In particular we consider the identification of inline formulas utilizing existing modules. Text lines including math formulas are first processed using a N-gram language model to reduce the number of formula candidates by thresholding the conditional probability of words. Then the formula candidates are classified into formulas and texts by SVM using geometric features associated with the bounding boxes of symbols.

This paper describes a method to extract the magnetic ink characters (MICR E-13B font) printed on bank-checks for the purpose of using OCR as supporting MICR. In the case of OCR, the colorful background patterns and the overlapped signatures on MICR characters make it difficult to extract characters respectively by using simple binarization and labeling. Our method estimates the color and pitch of MICR characters in order to separate the characters in contact with sign strokes, then the remaining sign strokes are removed by tracing them. In the experiment, we use circulated bank-checks and samples provided by SEIKO EPSON and show the performance of our method.

This paper represents the practical evaluations of time-sharing high-frequency resonant soft-switching inverter composed of two multi-resonant sub inverter cells in parallel for common series resonant load for all metal induction heating (IH) cooking appliances. Together with a fixed switching frequency resonant current, IH power regulation is carried out by phasor angle control scheme with phase-shifted PWM scheme. In particular, this high-frequency multi-resonant inverter can efficiently operate under the conditions of load resonant frequency changing mode; frequency doubler mode (100 kHz) for low resistivity metal IH loads and fundamental switching frequency mode (50 kHz) for high resistivity metal IH loads, respectively. The feasible performance of the time-sharing high-frequencys inverter are demonstrated and discussed in detail.

In tandem with the penetration of PHV as well as EV, contact-less power supply system in parking towers have become important subjects. In parking towers, EV and PHV are charged on the traveling pallets, supply power is to be transmitted from the building to travelling pallets. In this paper, high-frequency link AC-AC converter for contact-less power supply system in parking tower is newly proposed and its basic operation and performance have been demonstrated from computer aided simulation and experiment.

We present a system that supports review of a knowledge work lifelog as an activity history. Since knowledge workers often review their own activity histories, gathering each user's activities on his/her terminal as a lifelog is a promising approach. However, readability of the stored lifelog is a large problem of lifelog-based application. We propose a term extraction method to add annotation labels to the stored lifelog for supporting knowledge workers, exploiting text data acquired from desktop activities. Our prototype system monitors a user's desktop activities after combining raw events, and then extracts possible annotation labels with LDA and C-value techniques from documents and text data in sensor events. In this paper, we introduce a lifelogging module and a lifelog annotation method based on term extraction techniques. According to an empirical evaluation for three weeks, we found that the current method is useful for one-week review.

This paper describes embedding a mathematical formula recognition module into the OCR system OCRopus aiming at developing a OCR system for scientific and technical documents which include mathematical formulas. OCRopus is a open source OCR system emphasizing modularity, easy extensibility, and reuse. This system has several basic components such as preprocessing, layout analysis, and text line recognition, so it is a challenging project to embed the mathematical formula recognition module into the OCRopus system. We have developed the math OCR module, then report how to embed our module into the OCRopus system in order to realize a math OCR which can deal with wide variety of documents including mathematical formulas.

In the dark place photographing, the increase of noise is one of the major problems to be addressed. Although using a flash is effective to reduce the noise, natural colors are faded away due to increase of specular lights. In this paper, we present a method that generates a no-flash like flash image by approximating colors of a flash image by those of a no-flash image. Our method is based on the ”color-line” feature of images, a linear distribution of colors in a local region is transformed by a set of transform matrices automatically. This method is able to deal correctly with the occluded regions where the colors are saturated by reflections or shadows and the color distribution is squashed.

Inputting query terms on a mobile terminal is frustrating because of device limitations though mobile web search is becoming popular. Query prediction is a promising approach for mobile search. In the literature, however, little attention has been paid to the first query term, though there are many reports on query expansion or second query recommendation. In this paper, we propose a first query prediction method that enables users to search related information for the current browsed webpage with an easier interface such as touch operation without any background search processes. The proposed method consists of body-text extraction, candidate query term extraction, and a scoring process. We also implemented a one-touch search application that works for Japanese webpages at a practical speed on a smartphone. According to our closed evaluation with 299 webpages and open evaluation with 14 users, our method achieved practical quality.

This paper describes a method for extracting character strings from scene images. Most characters on scene images appear with the same color and font size at every word or text line. In our algorithm, a scene image is divided into several blocks based on edges in the color space at first. Then the blobs, which consist of similar color pixels, are extracted by a clustering in a color space for each block. Although these blobs are correspond to characters or background patterns, after connecting them using these aspect ratios and pitches, SVM (Support Vector Machine) on several textural features of these blobs will classify each connected blob into character or background patterns. Testing with 251 images from ICDAR 2003 Text Locating Competition shows effectiveness of our algorithm.

This poster presents an overview of the characteristics of a one-button information retrieval interface with closed captions from TV watching activities, which is intended to lighten the burden of remembering and entering query terms while watching TV. We investigated this interface with an experimental system named Video Bookmarking Search, which estimates query terms from closed captions with named-entity recognition and sentence labeling techniques. According to an empirical evaluation for 1,138 search queries from 206 bookmarks using seven actual TV shows on city life, travel, health, and cuisine, we found wider queries and search results are acceptable through the query-input-free interface, despite the fact that the number of queries and search results that are directly relevant to the users' original intentions is not high. The main reason is a watching user's interest is wider than what is expressed with query terms.

We proposed two methods for mathematical expression recognition. One is based on projection profile cutting and the other uses top-down and bottom-up strategies to analyze the two-dimensional structure of expressions. The paper describes the improvement of the latter method in terms of structural analysis robustness and application to matrix recognition. To evaluate the performance of our method, intensive experiments were carried out on a large variety of mathematical expression images which were collected from many mathematical journals.

A wide variety of mathematical expressions printed in documents can be recognized by analyzing the two-dimensional layout structures of the symbols within the expressions. Our method consists of two basic strategies. One is a specific structure analysis where local structures of subexpressions are checked by a bottom-up method. The other is a basic structure analysis in which the global structure of a whole expression is checked by using a top-down method. The structure of a recognized expression is represented by a tree structure and the original expression can be reproduced by using a suitable formatter like T/sub E/X. Experiments have been carried out for many types of expressions found in printed journals and our recognition method has shown favorable results.

A method of page segmentation using field separators and white streams is described and applied to the layout analysis of various types of printed pages which may have horizontal and vertical textlines. In complex page layouts, text columns which are printed closely together are often separated by thin black lines (field separators) or long white spaces (white streams). These separators are first extracted by horizontal and vertical scanning of a page, and then a global partitioning of the page into blocks is performed. Next in each block, black connected components are merged into textlines along the directions of separators horizontally or vertically. In experimental trials on various types of page layouts, such techniques produced robust and fast results.>