专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

71. 发明专利

JP5885956B2 フォントマッチング有权
标题翻译：字体匹配
公开(公告)号：JP5885956B2
公开(公告)日：2016-03-16
申请号：JP2011168993
申请日：2011-08-02
申请人：セラーテムインク.
发明人：ハモンドブルースクインシー
IPC分类号： G06K9/62
CPC分类号： G06K9/6828

72. 发明专利

JP4886098B1 The portable information terminal device 有权
公开(公告)号：JP4886098B1
公开(公告)日：2012-02-29
申请号：JP2011532461
申请日：2011-07-29
申请人：義一中村
发明人：義一中村
IPC分类号： G09G5/22 , G06F17/21 , G09G5/00 , G09G5/24
CPC分类号： G06F17/214 , G06F3/04883 , G06K9/6828 , G09G5/24 , G09G2370/022
摘要： A text writing device is provided with a code specification means for specifying a character code representing a character inputted by a user key operation, a storage means in which a handwriting font representing a character shape according to a user's handwriting is stored in advance for each character code, a font extraction means for extracting from the storage means a handwriting font corresponding to a character code specified by the code specification means, and an output means for outputting the handwriting font extracted by the font extraction means, wherein the storage means stores a plurality of mutually different types of handwriting fonts for a single character code, the font extraction means extracts one handwriting font from the plurality of types of handwriting fonts stored by the storage means in accordance with a prescribed extraction procedure so that the same type of handwriting font corresponding to each character code is at least not consecutive for a text comprising a plurality of character codes specified by the code specifying means.

73. 发明专利

JP2006155380A Image processing apparatus, its method, and its control method 有权
标题翻译：图像处理设备，其方法及其控制方法
公开(公告)号：JP2006155380A
公开(公告)日：2006-06-15
申请号：JP2004347235
申请日：2004-11-30
申请人： Canon Inc , キヤノン株式会社
发明人： SATOU MINEKO
IPC分类号： G06F17/30 , G06T7/00 , H04N1/00
CPC分类号： G06K9/2063 , G06F17/30271 , G06K9/00456 , G06K9/6828 , G06K2209/01 , H04N1/2179 , H04N1/32133 , H04N1/4413 , H04N1/4433 , H04N2201/3249 , H04N2201/3295
摘要： PROBLEM TO BE SOLVED: To address the fact that information digitization has made it easier to share and make use of information over a wide range, making it very important to manage electronic documents and prints of electronic documents. SOLUTION: After a conference is ended, a participant causes an MFP to read a document image in a distributed material (S2101). The MFP references information in a storage part to search for a data file that matches the read document image (S2105), and deletes the detected data file (S2108). If the number of pages for the read image is insufficient (S2102, S2103) or if the data file cannot be detected or if part of the read document image is lacking (S2106, S2107), an alarm is issued. When notification that collection has ended is received (S2109), the MFP references information in the storage part to determine whether or not all the data files related to the conference have been deleted (S2110); if there are data files yet to be deleted and if these data files have been printed (S2111), an alarm is issued (S2113). COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：为了解决信息数字化使得更广泛地分享和使用信息的事实，使得管理电子文档和电子文档的打印非常重要。
解决方案：会议结束后，参与者使MFP以分布式资料读取文档图像（S2101）。 MFP参照存储部中的信息来搜索与读取的文档图像相匹配的数据文件（S2105），并删除检测到的数据文件（S2108）。如果读取图像的页数不足（S2102，S2103），或者如果无法检测到数据文件或者部分读取的文档图像缺少（S2106，S2107），则发出报警。当接收到收集结束的通知（S2109）时，MFP参考存储部分中的信息，以确定与会议相关的所有数据文件是否已被删除（S2110）; 如果有数据文件尚未被删除，并且如果这些数据文件已被打印（S2111），则发出报警（S2113）。版权所有（C）2006，JPO＆NCIPI

74. 发明授权

US10878298B2 Tag-based font recognition by utilizing an implicit font classification attention neural network 有权
公开(公告)号：US10878298B2
公开(公告)日：2020-12-29
申请号：US16294417
申请日：2019-03-06
申请人： Adobe Inc.
发明人： Zhaowen Wang , Tianlang Chen , Ning Xu , Hailin Jin
IPC分类号： G06N20/00 , G06K9/68 , G06K9/62 , G06K9/46 , G06F16/906 , G06N3/08 , G06F16/903 , G06F40/109
CPC分类号： G06K9/6828 , G06F16/906 , G06F16/90335 , G06F40/109 , G06K9/46 , G06K9/6255 , G06K9/6257 , G06K9/6262 , G06K9/6277 , G06N3/084
摘要： The present disclosure relates to a tag-based font recognition system that utilizes a multi-learning framework to develop and improve tag-based font recognition using deep learning neural networks. In particular, the tag-based font recognition system jointly trains a font tag recognition neural network with an implicit font classification attention model to generate font tag probability vectors that are enhanced by implicit font classification information. Indeed, the font recognition system weights the hidden layers of the font tag recognition neural network with implicit font information to improve the accuracy and predictability of the font tag recognition neural network, which results in improved retrieval of fonts in response to a font tag query. Accordingly, using the enhanced tag probability vectors, the tag-based font recognition system can accurately identify and recommend one or more fonts in response to a font tag query.

75. 发明申请

US20190213408A1 DOCUMENT AUTHENTICITY DETERMINATION 审中-公开
公开(公告)号：US20190213408A1
公开(公告)日：2019-07-11
申请号：US16245453
申请日：2019-01-11
申请人： ONFIDO LTD
发明人： Jacques Cali , Joao Silva Gomes
IPC分类号： G06K9/00 , G06F17/21 , G06N3/04
CPC分类号： G06K9/00456 , G06F17/214 , G06K9/00463 , G06K9/3233 , G06K9/6828 , G06K2209/01 , G06N3/0454
摘要： A computer-implemented method for assessing if characters in a sample image are formed from a predefined font. The method comprises forming a first embedded space representation for the predefined font, extracting sample characters from the sample image, forming a second embedded space presentation of the sample characters, and comparing the first and second embedded space representation to assess if the sample characters are of the predefined font.

76. 发明申请

US20180247386A1 SYSTEMS AND METHODS FOR STEGANOGRAPHY BASED ON TEXT FONTS 审中-公开
公开(公告)号：US20180247386A1
公开(公告)日：2018-08-30
申请号：US15903888
申请日：2018-02-23
申请人： The Trustees of Columbia University in the City of New York
发明人： Changxi ZHENG , Chang XIAO , Cheng ZHANG
IPC分类号： G06T1/00 , H03M13/15 , G06K9/68
CPC分类号： G06T1/0021 , G06K9/6828 , G06T1/0028 , G06T2201/0061 , G06T2201/0062 , H03M13/09 , H03M13/159
摘要： Disclosed are methods, systems, devices, apparatus, media, and other implementations, including a method that includes obtaining input visual data comprising a sequence of symbols, selected from a symbol set, with each of the symbols associated with a glyph representation. The method also includes obtaining a code message comprising code message symbols, and modifying at least one of the symbols of the input visual data to a different glyph representation associated with a respective at least one of the code message symbols to generate, at a first time instance, a resultant coded visual data.

77. 发明授权

US09875429B2 Font attributes for font recognition and similarity 有权
公开(公告)号：US09875429B2
公开(公告)日：2018-01-23
申请号：US14876667
申请日：2015-10-06
申请人： Adobe Systems Incorporated
发明人： Zhaowen Wang , Luoqi Liu , Hailin Jin
IPC分类号： G06K9/62 , G06K9/66
CPC分类号： G06K9/6257 , G06K9/4628 , G06K9/6272 , G06K9/6277 , G06K9/66 , G06K9/6828 , G06K2209/01 , G06N3/0454
摘要： Font recognition and similarity determination techniques and systems are described. In a first example, localization techniques are described to train a model using machine learning (e.g., a convolutional neural network) using training images. The model is then used to localize text in a subsequently received image, and may do so automatically and without user intervention, e.g., without specifying any of the edges of a bounding box. In a second example, a deep neural network is directly learned as an embedding function of a model that is usable to determine font similarity. In a third example, techniques are described that leverage attributes described in metadata associated with fonts as part of font recognition and similarity determinations.

78. 发明授权

US09824269B2 Enhanced interpretation of character arrangements 有权
公开(公告)号：US09824269B2
公开(公告)日：2017-11-21
申请号：US15352485
申请日：2016-11-15
申请人： Microsoft Technology Licensing, LLC
发明人： Yi Zhang , Daniel Parish , Aaron Lamar Wilson , Shangwei Fang
IPC分类号： G06K9/18 , G06K9/00 , G06K9/68
CPC分类号： G06K9/00456 , G06K9/00442 , G06K9/00449 , G06K9/18 , G06K9/6828 , G06K2209/01
摘要： Technologies are described herein for interpreting character arrangements. An image including an arrangement of characters may be received or captured by a computing device. Techniques described herein generate data representative of the characters. Characteristics and other information interpreted from the image may be processed to determine a data type. The data representative of the characters may be arranged into a data structure based on the data type, an arrangement type and/or other information interpreted from the image. The data type may indicate one or more attributes of the arranged data such as a format, font, date, language, or currency. The data type may also indicate how data is used in a process, equation or calculation. In addition, the data type may identify an anchor that may be used to merge data generated from the image with other data generated from another image.

79. 发明申请

US20170098141A1 Determination of Font Similarity 有权
公开(公告)号：US20170098141A1
公开(公告)日：2017-04-06
申请号：US14876660
申请日：2015-10-06
申请人： Adobe Systems Incorporated
发明人： Zhaowen Wang , Hailin Jin
IPC分类号： G06K9/68 , G06K9/62 , G06T7/60 , G06K9/66
CPC分类号： G06K9/6828 , G06K9/6267 , G06K9/627 , G06K9/66 , G06K2209/01 , G06K2209/27 , G06N3/0454 , G06T7/60 , G06T2207/20081 , G06T2207/20084
摘要： Font recognition and similarity determination techniques and systems are described. In a first example, localization techniques are described to train a model using machine learning (e.g., a convolutional neural network) using training images. The model is then used to localize text in a subsequently received image, and may do so automatically and without user intervention, e.g., without specifying any of the edges of a bounding box. In a second example, a deep neural network is directly learned as an embedding function of a model that is usable to determine font similarity. In a third example, techniques are described that leverage attributes described in metadata associated with fonts as part of font recognition and similarity determinations.

80. 发明授权

US09501724B1 Font recognition and font similarity learning using a deep neural network 有权
标题翻译：使用深层神经网络的字体识别和字体相似性学习
公开(公告)号：US09501724B1
公开(公告)日：2016-11-22
申请号：US14734466
申请日：2015-06-09
申请人： ADOBE SYSTEMS INCORPORATED
发明人： Jianchao Yang , Zhangyang Wang , Jonathan Brandt , Hailin Jin , Elya Shechtman , Aseem Omprakash Agarwala
IPC分类号： G06K9/00 , G06K9/36 , G06K9/66 , G06K9/62 , G06K9/68 , G06T3/40
CPC分类号： G06T3/40 , G06K9/6255 , G06K9/6828
摘要： A convolutional neural network (CNN) is trained for font recognition and font similarity learning. In a training phase, text images with font labels are synthesized by introducing variances to minimize the gap between the training images and real-world text images. Training images are generated and input into the CNN. The output is fed into an N-way softmax function dependent on the number of fonts the CNN is being trained on, producing a distribution of classified text images over N class labels. In a testing phase, each test image is normalized in height and squeezed in aspect ratio resulting in a plurality of test patches. The CNN averages the probabilities of each test patch belonging to a set of fonts to obtain a classification. Feature representations may be extracted and utilized to define font similarity between fonts, which may be utilized in font suggestion, font browsing, or font recognition applications.
摘要翻译：对卷积神经网络（CNN）进行字体识别和字体相似学习。在训练阶段，通过引入差异来合成具有字体标签的文本图像，以最小化训练图像与真实世界文本图像之间的差距。生成训练图像并将其输入到CNN中。根据CNN正在训练的字体数量，输出被输入到N-way softmax函数中，产生N类标签上分类文本图像的分布。在测试阶段，每个测试图像的高度被标准化，并以纵横比挤压，从而产生多个测试贴片。 CNN对属于一组字体的每个测试补丁的概率进行平均，以获得分类。可以提取和利用特征表示来定义可以在字体建议，字体浏览或字体识别应用中使用的字体之间的字体相似性。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式