公开(公告)号：US20170249519A1

公开(公告)日：2017-08-31

申请号：US15594224

申请日：2017-05-12

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Pil-seung YANG ,  Da-hye PARK ,  Seol-hye WON ,  In-kuk YUN ,  Yong-gook PARK

IPC: G06K9/00 ,  G06F3/0482 ,  G06F3/16 ,  G06F3/0488

CPC classification number: G06K9/00872 ,  G06F3/0482 ,  G06F3/04842 ,  G06F3/0488 ,  G06F3/04883 ,  G06F3/04886 ,  G06F3/167 ,  G06F2203/0381 ,  G06F2203/04803 ,  G06F2203/04807 ,  G06F2203/04808

Abstract: Provided is a device including: a display unit configured to display handwritten content based on an analog handwritten input of a user; a user input unit that receives a user input of selecting a portion of the handwritten content displayed on the display unit; and a control unit reproduces a segment of multimedia content, which corresponds to the portion of the handwritten content, from the multimedia content synchronized with the handwritten content.

公开(公告)号：US20160378195A1

公开(公告)日：2016-12-29

申请号：US15190484

申请日：2016-06-23

Applicant: ORANGE

Inventor： Gregoire Lefebvre

IPC: G06F3/01 ,  G06K9/00

CPC classification number: G06F3/017 ,  G06K9/00402 ,  G06K9/00416 ,  G06K9/00852 ,  G06K9/00872 ,  G06K9/00879 ,  G06K9/224 ,  G06K9/48 ,  G06K9/6289 ,  G06K2209/01

Abstract: The invention relates to a method for recognizing handwriting on a physical surface on the basis of three-dimensional signals originating from sensors of a terminal, the method being characterized in that the signals are obtained on the basis of at least 3 different types of sensors, and in that it comprises steps of sampling, according to 3 axes and over a sliding time window, of inertial signals originating from the sensors, fusing the sampled signals into a 9-dimensional vector for each sampling period, converting the fused signals into a sequence of characteristic 9-dimensional vectors, and, when a signal characteristic of an input start has been detected, storing the sequence of characteristic vectors in a list of sequences of characteristic vectors, the preceding steps being repeated until the detection of a signal characteristic of an input end, the method furthermore comprising, on detection of said signal characteristic of an input end, a step of recognizing a word on the basis of the list of sequences of characteristic vectors created over the time window.

Abstract translation: 本发明涉及一种用于基于源自终端的传感器的三维信号在物理表面上识别笔迹的方法，其特征在于，基于至少3种不同类型的传感器获得信号， 并且其包括以下步骤：根据3轴和滑动时间窗口对源自传感器的惯性信号进行采样，在每个采样周期将采样信号融合成9维向量，将融合信号转换为序列 特征9维向量的特征向量的特征向量的特征，当已经检测到输入开始的信号特性时，将特征向量的序列存储在特征向量序列的列表中，重复前述步骤，直到检测到 输入端，所述方法还包括在检测到输入端的所述信号特性时，识别所述基底上的单词的步骤 s在时间窗口上创建的特征向量序列列表。

公开(公告)号：US20150339525A1

公开(公告)日：2015-11-26

申请号：US14760236

申请日：2014-01-11

Applicant: NATURAL INTELLIGENT TECHNOLGIES s.r.l.

Inventor： Angelo Marcelli ,  Adolfo Santoro ,  Claudio De Stefano ,  Antonio Parziale ,  Rosa Senatore

IPC: G06K9/00 ,  G06K9/62 ,  G06K9/72 ,  G06K9/34

CPC classification number: G06K9/00416 ,  G06K9/00429 ,  G06K9/00852 ,  G06K9/00859 ,  G06K9/00872 ,  G06K9/00926 ,  G06K9/34 ,  G06K9/6255 ,  G06K9/723 ,  G06K2209/01

Abstract: Process, and related apparatus, that exploits psycho-physiological aspects involved in generation and perception of handwriting for directly inferring from the trace on the paper (or any other means on which the author writes by hand) the interpretation of writing, i.e. the sequence of characters that the trace is intended to represent.

Abstract translation: 过程和相关的设备，利用了生成和感知笔迹的心理生理方面，从纸上的痕迹（或作者用手写的任何其他手段）直接推断出书写的解释，即序列 该曲线旨在表示的字符。

公开(公告)号：US20140355835A1

公开(公告)日：2014-12-04

申请号：US13903218

申请日：2013-05-28

Applicant: Xerox Corporation

Inventor： Jose Antonio Rodriguez-Serrano ,  Florent C. Perronnin

IPC: G06K9/72

CPC classification number: G06K9/72 ,  G06K9/00872 ,  G06K9/3258 ,  G06K9/6232 ,  G06K2209/15

Abstract: A system and method for computing confidence in an output of a text recognition system includes performing character recognition on an input text image with a text recognition system to generate a candidate string of characters. A first representation is generated, based on the candidate string of characters, and a second representation is generated based on the input text image. A confidence in the candidate string of characters is computed based on a computed similarity between the first and second representations in a common embedding space.

Abstract translation: 用于计算文本识别系统的输出的置信度的系统和方法包括使用文本识别系统对输入文本图像执行字符识别以生成候选字符串。 基于候选字符串生成第一表示，并且基于输入的文本图像生成第二表示。 基于在公共嵌入空间中计算出的第一和第二表示之间的相似度来计算候选字符串的置信度。

公开(公告)号：US20100278430A1

公开(公告)日：2010-11-04

申请号：US12831244

申请日：2010-07-06

Applicant: Jonathon Leigh Napper

Inventor： Jonathon Leigh Napper

IPC: G06K9/46

CPC classification number: G06K9/00872 ,  G06K9/6292 ,  G06K9/723 ,  G06K2209/01

Abstract: A method of identifying a string formed from a number of hand-written characters is disclosed. The method starts by determining character probabilities for each hand-written character in the string. Each character probability represents the likelihood of the respective hand-written character being a respective one of a number of predetermined characters. Next, template probabilities for the string are determined. Each template probability represents the likelihood of the string corresponding to a respective one of a number of templates. Each template represents a respective combination of character types. The step of determining the template probabilities for the string includes the sub-steps of determining the number of characters in the string, selecting templates having an identical number of characters, and obtaining a template probability for each selected template.

Abstract translation: 公开了一种识别由多个手写字符形成的字符串的方法。 该方法通过确定字符串中每个手写字符的字符概率开始。 每个字符概率表示各个手写字符是多个预定字符中的相应一个的可能性。 接下来，确定字符串的模板概率。 每个模板概率表示对应于多个模板中的相应一个的字符串的可能性。 每个模板表示字符类型的相应组合。 确定字符串的模板概率的步骤包括确定字符串中的字符数的子步骤，选择具有相同字符数的模板，以及为每个所选模板获得模板概率。

公开(公告)号：US07489819B2

公开(公告)日：2009-02-10

申请号：US11747418

申请日：2007-05-11

Applicant: Rodney L. Sheets

Inventor： Rodney L. Sheets

IPC: G06K9/00

CPC classification number: G06K9/2063 ,  G06K9/00436 ,  G06K9/00872 ,  G06K9/6253 ,  G06K2009/226 ,  G06K2209/01

Abstract: Systems and methods are provided for recognizing handwritten characters drawn on a paper form using a digital pen that records stroke coordinates corresponding to respective pen strokes. In one embodiment, a field on the paper form is assigned a lexical inference level. For example, the field may be assigned a word level, a word prefix level, and/or a word stem level. The assigned lexical inference level is used to recognize one or more stroke coordinates corresponding to pen strokes written in the field. Recognized characters are then used to create or modify an inference lexicon used to perform handwriting recognition for the entire field. In one embodiment, the inference lexicon is used for handwriting recognition in the same field on subsequently processed forms.

Abstract translation: 提供了系统和方法，用于识别使用记录与各笔画笔相对应的笔画坐标的数字笔在纸上形成的手写字符。 在一个实施例中，纸张表格上的字段被分配词汇推理级别。 例如，该字段可以被分配一个词级别，一个词前缀级别和/或一个词干级别。 分配的词汇推理级别用于识别对应于在字段中写入的笔划笔划的一个或多个笔划坐标。 然后，识别的字符用于创建或修改用于对整个字段执行手写识别的推理词典。 在一个实施例中，推论词典用于在随后处理的表单上的相同字段中的手写识别。

公开(公告)号：US20070286486A1

公开(公告)日：2007-12-13

申请号：US11443302

申请日：2006-05-30

Applicant: Ira P. Goldstein

Inventor： Ira P. Goldstein

IPC: G06K9/00 ,  G06K9/72

CPC classification number: G06K9/00872 ,  G06K9/00422 ,  G06K9/2054 ,  G06K9/6292 ,  G06K2209/01

Abstract: A system for automatically recognizing a handwriting image and converting such image to text data including a sequence of validated words, has an image input device, a number of handwriting recognition engines, and control unit. A first handwriting recognition engine is responsive to the image input device, for analyzing the data file and providing one or more possible text words for each successive word in the data file. The first handwriting recognition engine further provides a resemblance indication for each possible text word indicating a level of resemblance between its appearance and the appearance of the handwritten word in the data file. In the event that there is not a high level of confidence in the selection of the first handwriting recognition engine, a selection of a validated word is based on the selections of one or more of the other handwriting recognition engines.

Abstract translation: 一种用于自动识别手写图像并将该图像转换成包括一系列有效字词的文本数据的系统，具有图像输入装置，多个手写识别引擎和控制单元。 第一手写识别引擎响应于图像输入设备，用于分析数据文件并为数据文件中的每个连续字提供一个或多个可能的文本字。 第一手写识别引擎还为每个可能的文本字提供一个相似性指示，指示其外观与数据文件中的手写字的外观相似程度。 在对第一手写识别引擎的选择没有高度的置信度的情况下，所选择的验证词是基于一个或多个其他手写识别引擎的选择。

公开(公告)号：US06738519B1

公开(公告)日：2004-05-18

申请号：US09590261

申请日：2000-06-09

Applicant: Daisuke Nishiwaki

Inventor： Daisuke Nishiwaki

IPC: G06K962

CPC classification number: G06K9/723 ,  G06K9/00872 ,  G06K2209/01

Abstract: There are provided: an n-fold-character recognizing part for collectively recognizing an unmatched portion without segmenting character candidate patterns character by character for an image of a read-wise skipped portion, i.e., the unmatched portion upon word verification; and an n-fold-character recognizing dictionary referred to by the n-fold-character recognizing part upon recognition; to thereby conduct re-recognition independent of instability of character segmentation even when the portion read-wise skipped by the word verification includes two or more characters.

Abstract translation: 提供：用于共同识别不匹配部分的n折字符识别部分，而不分割字符候选图案，用于逐字跳过部分的图像，即，字验证时的不匹配部分; 以及在识别时由所述n倍字符识别部分提及的n倍字符识别字典; 从而即使当通过字验证跳过的部分包括两个或更多个字符时，也可以独立于字符分段的不稳定性进行重新识别。

公开(公告)号：US06393395B1

公开(公告)日：2002-05-21

申请号：US09227170

申请日：1999-01-07

Applicant: Angshuman Guha ,  Patrick M. Haluptzok ,  James A. Pittman

Inventor： Angshuman Guha ,  Patrick M. Haluptzok ,  James A. Pittman

IPC: G10L1504

CPC classification number: G06K9/222 ,  G06K9/00872 ,  G06K9/723 ,  G06K2209/01

Abstract: A method and system for recognizing user input information including cursive handwriting and spoken words. A time-delayed neural network having an improved architecture is trained at the word level with an improved method, which, along with preprocessing improvements, results in a recognizer with greater recognition accuracy. Preprocessing is performed on the input data and, for example, may include resampling the data with sample points based on the second derivative to focus the recognizer on areas of the input data where the slope change per time is greatest. The input data is segmented, featurized and fed to the time-delayed neural network which outputs a matrix of character scores per segment. The neural network architecture outputs a separate score for the start and the continuation of a character. A dynamic time warp (DTW) is run against dictionary words to find the most probable path through the output matrix for that word, and each word is assigned a score based on the least costly path that can be traversed through the output matrix. The word (or words) with the overall lowest score (or scores) are returned. A DTW is similarly used in training, whereby the sample ink only need be labeled at the word level.

Abstract translation: 用于识别用户输入信息的方法和系统，包括草书手写和口语。 具有改进的架构的时间延迟的神经网络在词级上被用一种改进的方法进行训练，该方法与预处理改进一起产生具有更高识别精度的识别器。 对输入数据执行预处理，例如，可以包括基于二阶导数的采样点重新采样数据，以将识别器聚焦在输入数据的每时间斜率变化最大的区域上。 输入数据被分段，特征化并馈送到时间延迟的神经网络，其输出每个段的字符得分矩阵。 神经网络架构为字符的开始和继续输出单独的分数。 针对词典运行动态时间扭曲（DTW），以通过该词的输出矩阵找到最可能的路径，并且基于可以通过输出矩阵遍历的最便宜的路径为每个单词分配一个分数。 返回总分最低分（或分数）的单词（或单词）。 DTW类似地用于训练，因此样品油墨只需要在单词级别进行标记。

公开(公告)号：US20010016074A1

公开(公告)日：2001-08-23

申请号：US09769378

申请日：2001-01-26

Inventor： Tomoyuki Hamamura

IPC: G06K009/72

CPC classification number: G06K9/00872

Abstract: In word recognition using the character recognition result, recognition processing is performed for an input character string that corresponds to a word to be recognized, a probability at which characteristics obtained as the result of character recognition are generated by conditioning characters of words contained in a word dictionary that stores in advance candidates of words to be recognized. The thus obtained probability is divided by a probability at which characteristics obtained as the result of character recognition are generated, and each of the division results obtained relevant to the characters of the words contained in the word dictionary is multiplied relevant to all the characters. The recognition results of the above words are obtained based on the multiplication results.

Abstract translation: 在使用字符识别结果的字识别中，对与要识别的字对应的输入字符串执行识别处理，通过对包含在字中的单词的字符进行调整而生成作为字符识别结果而获得的特性的概率 预先存储要识别的单词的候选字典的字典。 将由此获得的概率除以生成作为字符识别结果获得的特性的概率，并且与包含在单词辞典中的单词的字符相关的每个分割结果与所有字符相乘相乘。 基于乘法结果得到上述单词的识别结果。