ランキング (合計) - Ceek.jp Altmetrics

1 0 0 0 視覚障害者のためのカメラ映像内の文字情報読み上げシステム

著者: 清田公保島川学本田幸代江崎修央砂崎由樹
出版者: 一般社団法人情報処理学会
雑誌: 情報処理学会研究報告ユビキタスコンピューティングシステム(UBI) (ISSN:09196072)
巻号頁・発行日: vol.2006, no.14, pp.139-143, 2006-02-16
被引用文献数: 2

This paper proposes a text reading system by using active micro camera. This system is designed as information navigation for visually impaired person. The system is composed with a laptop PC Network Camera and a command input device. At first the system extracts character areas automatically and characters are recognized by offline-character recognition application. Next The text database of the layered structure is composed from captured menu image by using the system. A technical factor in this system is as follows. Because the user of the system is the blind the captured menu image is not certain whether to contain an appropriate character area. Some characters are transformed such as inclination too big or small. Therefore the proposed system extracts the character area including a small character first. The camera zooms to each candidate character area and then character area of high resolution is extracted. The proposed system can be used to check items of menu and prices. We apply this system to restaurant menu reading system for blind person. The user opens the first page of the menu and pushes [capture] button on the portable PC. In this time the camera turned in the direction of the menu. Then system tries to make database of the menu automatically. Then a menu item database is composed of the menu items that are extracted by offline character recognition. The database items are name of menu and its price. The system announces the menu items by voice synthesizer. The user is able to select the candidate item by using [yes] or [No] button on the control interface board in the system. This operation is very easy for the blind. From the viewpoint that the visually impaired person uses it we discuss that the system design of menu reading system and the presentation method of an efficient voice reading out.This paper proposes a text reading system by using active micro camera. This system is designed as information navigation for visually impaired person. The system is composed with a laptop PC, Network Camera and a command input device. At first, the system extracts character areas automatically, and characters are recognized by offline-character recognition application. Next, The text database of the layered structure is composed from captured menu image by using the system. A technical factor in this system is as follows. Because the user of the system is the blind, the captured menu image is not certain whether to contain an appropriate character area. Some characters are transformed such as inclination, too big or small. Therefore the proposed system extracts the character area including a small character first. The camera zooms to each candidate character area, and then character area of high resolution is extracted. The proposed system can be used to check items of menu and prices. We apply this system to restaurant menu reading system for blind person. The user opens the first page of the menu, and pushes [capture] button on the portable PC. In this time, the camera turned in the direction of the menu. Then system tries to make database of the menu automatically. Then, a menu item database is composed of the menu items that are extracted by offline character recognition. The database items are name of menu and its price. The system announces the menu items by voice synthesizer. The user is able to select the candidate item by using [yes] or [No] button on the control interface board in the system. This operation is very easy for the blind. From the viewpoint that the visually impaired person uses it, we discuss that the system design of menu reading system, and the presentation method of an efficient voice reading out.

2008-09-26 03:15:23
1 はてなブックマーク

https://ci.nii.ac.jp/naid/110004710191

1 0 0 0 無線LANによる船内LAN構築と船内監視システム

著者: 三輪誠若林伸和矢野吉治
出版者: 公益社団法人日本航海学会
雑誌: 日本航海学会論文集 (ISSN:03887405)
巻号頁・発行日: vol.114, pp.243-250, 2006
被引用文献数: 2 2

The regulation to enforce improvement of security level was executed in 2004. Ships which are obliged should take the cost for introduction of new security system. We tried to construct on-board LAN using wireless technology aboard existent ship in operation, in order to utilize ship's monitoring system. In this paper, we describe outline of implementation of on-board LAN system aboard T. S. Fukae-maru and conclusion of the radio intensity measurement of wireless LAN, also describe ship monitoring system utilizing network camera as a practical application of the LAN.

2008-09-26 03:15:18
1 はてなブックマーク

1 0 0 0 3513 駅設備遠隔監視制御システム(SS5,SS5 設備・サービス向上,J-Rail 2006)

著者: 上島貞浩道下文雄
出版者: 一般社団法人日本機械学会
雑誌: 鉄道技術連合シンポジウム(J-Rail)講演論文集
巻号頁・発行日: vol.2006, no.13, pp.359-362, 2006-12-13

In the Keihanna line of Kintetsu Corporation opened for service this spring, we have developed and introduced a remote monitoring and control system for automatic fare collection equipment and other station equipment such as elevators, escalators etc., in order to reduce railway station duties. This system uses the network camera product and is applying the VoIP technology, and data streams of video, voice and control are integrated in digital form and transmitted on the IP network, so the system can be simply constructed. And the consoles are placed at two centers and six stations, so remote operation is available from every station, as well as from center.

2008-09-26 03:15:04
1 はてなブックマーク

https://ci.nii.ac.jp/naid/110006639443

We propose a new method for determining the correspondence between audio and visual events based on the selective attention which is observed in a living organism. For auditory attention, our system recognizes a target sound in noisy environments using an auditory filter that focuses on a specific period corresponding to visual events. It was confirmed that the success rate of correspondence between audio and visual events increased from 78.6% without the auditory filter, to 95.2% with the auditory filter. This certifies the importance of visual information for auditory attention. We also realized a visual attention of localizing a sound source and adjusting the line of sight and the iris of a camera when the system perceived a sound, but could not obtain visual information. The success rate of experiments based on visual attention is 93.3%. These results show the effectiveness of the proposed method.

2008-09-26 03:12:03
1 はてなブックマーク

1 0 0 0 890万画素撮像素子を用いた放送用単板カラーハイビジョンカメラの試作

著者: 島本洋三橋政次峯忠己船津良平橘川俊哉三谷公二浜田浩行羽鳥好律
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.7, pp.1104-1109, 2008-07-01
被引用文献数: 5 1

We have developed a prototype single-chip color HDTV camera for broadcasting. This camera uses a 1.25-inch 8.9-M-pixel CMOS image sensor for obtaining enough resolution characteristics and signal-to-noise ratio (SNR) in a singlechip color system. The camera has an HD-SDI output and a body weight of only 1.4 kg (3.1 lb). The measured resolution was more than 35% on a modulation transfer function (MTF) (45% on a square wave response) at 800 TV-lines. The sensitivity was an f-stop of 11 at 2000 lux with a dynamic range of 400%, and the SNR was 60 dB. These characteristics were as high as those with conventional 2.2-M-pixel 3-CCD HDTV cameras. The measured color difference (delta-E) between this camera and a 3CCD camera on a CIELAB color space was about 9.3. Our simulation showed that the delta-E was about 1.8 with a linear matrix color correction.

2008-09-26 03:11:34
1 はてなブックマーク

1 0 0 0 角膜反射を利用した瞳孔位置検出の高精度化

著者: 海老澤嘉伸中島彩
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.7, pp.1122-1126, 2008-07-01
被引用文献数: 4 11

In general, the relative position between the centers of the pupil and corneal reflection are used for precise eye-gaze detection. Recently, for stable pupil detection under various ambient light conditions, the pupil has been detected by the difference between the bright and dark pupil images. However, due to the difference in acquisition timing between the bright and dark images, the precision of the pupil position easily decreases when the head moves. We propose a compensation method which differentiates those pupil images after shifting one of the images so that the positions of corneal reflections in both images should match, and show its effectiveness.

2008-09-26 03:11:26
1 はてなブックマーク

1 0 0 0 ホログラフィ入力に適したインテグラルフォトグラフィ撮影のための複数要素画像の重複記録低減に関する検討

著者: 三科智之山本健詞大井隆太朗奥井誠人
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.7, pp.1132-1137, 2008-07-01
被引用文献数: 5

We describe how to reduce interference between elemental images suitable for production of holograms during integral photography pickup that uses a lens array composed of typical convex lenses used as elemental lenses. We propose the use of an optical system that consists of both a field lens and an aperture that is located in the focal plane of the field lens. The system is arranged between the lens array and the camera used for taking elemental images. Diffraction patterns in the focal plane of the field lens of beams going through elemental lenses were calculated with the Fresnel-Kirchhoff diffraction formula. Consequently, we clarified that the system was able to limit the incident angle of the beam passing through the elemental lenses with the size of the aperture located in the focal plane. In addition, we carried out an optical experiment and proved that using the optical system reduced the interference between the elemental images.

2008-09-26 03:11:21
1 はてなブックマーク

1 0 0 0 3-3. 映像表現およびコンピュータグラフィックス(3.メディア情報,<特集>映像情報メディア年報)

著者: 吉川浩津田貴生檜山茂雄新谷幹夫小黒久史
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.8, pp.1255-1258, 2008-08-01
被引用文献数: 1

2008-09-26 03:10:20
1 はてなブックマーク

1 0 0 0 第8回画像の本質を引き出す : バイラテラルフィルタによるノイズ除去とイラスト風画像の生成(講座次世代ディジタルカメラ/ディジタルムービーを予測する)

著者: 浦浜喜一
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.8, pp.1268-1273, 2008-08-01
被引用文献数: 8 7

2008-09-26 03:10:13
1 はてなブックマーク

1 0 0 0 動画像特徴量と位相限定相関を用いた番組識別実験

著者: 大竹剛真島恵吾合志清一藤井亮介伊藤浩鈴木光義高井重典寛座武智
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.8, pp.1292-1300, 2008-08-01
被引用文献数: 1

As digital broadcasting has become widespread and the available bandwidth for Internet access has become wider, more high-quality digital content has been distributed to users, making illegal copying a critical issue in the content business. New technology for identifying illegally distributed content is necessary to deter it. We developed two methods for identifying video content: the extraction of characteristic information from video and the use of phase-only correlation. A system was developed to apply these methods to digital TV programs. Experiments on four weeks of broadcast content gave good results, which suggests that a combination of the two methods would enable us to identify video content distributed illegally over the Internet precisely and quickly.

2008-09-26 03:10:07
1 はてなブックマーク

1 0 0 0 局所境界要素法によるイメージセンサの3次元波動光学シミュレーション

著者: 武藤秀樹須川成利
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.8, pp.1319-1325, 2008-08-01
被引用文献数: 1 1

A novel wave optical simulation method (a localized boundary element method) has been developed. This method enables us to execute 3-D wave optical simulation with much smaller memory space and much shorter calculation time than conventional boundary element methods or FDTD(Finite Difference Time Domain) methods. The color shading characteristics of inner lens structures in CCD image sensors and the light-gathering power and crosstalk of light waveguide structures in CMOS image sensors can be analyzed by this method. This method was found to be powerful and useful for wave optical analysis of image sensors.

2008-09-26 03:09:59
1 はてなブックマーク

1 0 0 0 重心ボロノイ図とそのドロネー三角形分割によるノンフォトリアリスティックハーフトーニング

著者: 井上光平浦浜喜一
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.8, pp.1326-1329, 2008-08-01
被引用文献数: 1 1

Non-photorealistic rendering(NPR) methods are presented for generating halftone images using a centroidal Voronoi diagram(CVD) and its Delaunay triangulation(DT). On the basis of the approximation of a CVD by hexagonal lattice, relations of brightness between an input image and its CVD and DT are derived and applied to brightness correction of halftone images. Experimental results showed that the halftone images generated by the proposed methods reproduced the continuous tone of the input image well.

2008-09-26 03:09:38
1 はてなブックマーク

1 0 0 0 動作特徴を考慮したモーションキャプチャデータベースの構築と動作生成への応用

著者: 佐藤浩司青木工太長橋宏
出版者: 一般社団法人映像情報メディア学会
雑誌: 映像情報メディア学会誌 : 映像情報メディア (ISSN:13426907)
巻号頁・発行日: vol.62, no.8, pp.1330-1332, 2008-08-01
被引用文献数: 1

Motion capture data is widely used to easily reproduce natural-looking human motions. However, it is very difficult to search a huge motion database for any particular human motion. A variety of motion retrieval techniques that enable users to find accurately corresponding motions have been proposed. We developed a construction and retrieval technique for the human motion database that considers the overall features of motion by using a dimension reduction technique and a machine learning algorithm. This technique enables efficient retrieval of a set of motions similar to the query. It is also possible to apply our technique to motion synthesis by creating a hierarchical motion graph based on the human motion database.