公开(公告)号：US4624011A

公开(公告)日：1986-11-18

申请号：US462042

申请日：1983-01-28

申请人： Sadakazu Watanabe ,  Hidenori Shinoda ,  Tsuneo Nitta ,  Yoichi Takebayashi ,  Shouichi Hirai ,  Tomio Sakata ,  Kensuke Uehara ,  Yasuo Takahashi ,  Haruo Asada

发明人： Sadakazu Watanabe ,  Hidenori Shinoda ,  Tsuneo Nitta ,  Yoichi Takebayashi ,  Shouichi Hirai ,  Tomio Sakata ,  Kensuke Uehara ,  Yasuo Takahashi ,  Haruo Asada

IPC分类号： G10L15/10 ,  G10L11/00 ,  G10L15/00 ,  G10L15/28 ,  G10L5/00

CPC分类号： G10L15/00

摘要： An acoustic signal processing circuit extracts input speech pattern data and subsidiary feature data from an input speech signal. The input speech pattern data comprise frequency spectra, whereas the subsidiary feature data comprise phoneme and acoustic features. These data are then stored in a data buffer memory. The similarity measures between the input speech pattern data stored in the data buffer memory and reference speech pattern data stored in a dictionary memory are computed by a similarity computation circuit. When the largest similarity measure exceeds a first threshold value and when the difference between the largest similarity measure and the second largest measure exceeds a second threshold value, category data of the reference pattern which gives the largest similarity measure is produced by a control circuit to correspond to an input speech. When recognition cannot be performed, the categories of the reference speech patterns which respectively give the largest to mth similarity measures are respectively compared with the subsidiary feature data. In this manner, subsidiary feature recognition of the input voice is performed by a subsidiary feature recognition section.

摘要翻译： 声信号处理电路从输入语音信号中提取输入语音模式数据和辅助特征数据。 输入语音模式数据包括频谱，而辅助特征数据包括音素和声学特征。 然后将这些数据存储在数据缓冲存储器中。 存储在数据缓冲存储器中的输入语音模式数据与字典存储器中存储的参考语音模式数据之间的相似性度量由相似度计算电路计算。 当最大相似性度量超过第一阈值时，当最大相似性度量与第二最大量度之间的差异超过第二阈值时，通过控制电路产生给出最大相似性度量的参考模式的类别数据，以对应于 输入语音。 当不能执行识别时，将分别给出最大到第m个相似性度量的参考语音模式的类别分别与辅助特征数据进行比较。 以这种方式，由辅助特征识别部执行输入语音的辅助特征识别。

公开(公告)号：US4888823A

公开(公告)日：1989-12-19

申请号：US101789

申请日：1987-09-28

申请人： Tsuneo Nitta ,  Kensuke Uehara ,  Sadakazu Watanabe

发明人： Tsuneo Nitta ,  Kensuke Uehara ,  Sadakazu Watanabe

IPC分类号： G10L11/00 ,  G06F3/16 ,  G10L15/02 ,  G10L15/06 ,  G10L15/10 ,  G10L15/14 ,  G10L15/18

CPC分类号： G10L15/187 ,  G10L15/142

摘要： Phoneme feature parameters are extracted from input digital speech signals by means of LPC analysis. Phonetic segments having phonetical meanings are obtained together with similarities to prescribed basic phonetic segments from the feature parameters to be passed through nodes of transition networks provided for each word. In passing the nodes, scores for similarity Sj of predetermined segments of the corresponding phonetic segments are made in selective scoring and the accumulation of the scores is used for recognition of continuous word speech.

公开(公告)号：US4881266A

公开(公告)日：1989-11-14

申请号：US19781

申请日：1987-02-27

申请人： Tsuneo Nitta ,  Kensuke Uehara ,  Sadakazu Watanabe

发明人： Tsuneo Nitta ,  Kensuke Uehara ,  Sadakazu Watanabe

IPC分类号： G10L11/00 ,  G10L11/02 ,  G10L15/02 ,  G10L15/04 ,  G10L15/10 ,  G10L15/20 ,  G10L21/02

CPC分类号： G10L25/87 ,  G10L15/04 ,  G10L15/20

摘要： In a speech recognition system for recognizing speeches uttered by non-specific speakers, start and end points of a word or speech interval are determined by a novel preprocessor for searching a sound power level to obtain speech boundary candidates and for determining likelihoods of speech or word intervals on the basis of the boundary candidates. Since likelihoods (probabilities) are determined for speech interval candidates, the similarity rate between feature parameters and reference pattern set of a speech signal are calculated for only the higher likelihood candidates, thus improving the accuracy and the speed of speech recognition. A percentage of erroneous boundary decision is about 0.5% when two speech interval candidates of the first and second likelihoods are adopted.

摘要翻译： 在用于识别由非特定扬声器发出的讲话的语音识别系统中，用于搜索声功率电平以获得语音边界候选的新型预处理器确定单词或语音间隔的开始和结束点，并且用于确定语音或单词的可能性 基于边界候选人的间隔。 由于为语音间隔候选确定了可能性（概率），所以仅针对较高似然候选来计算特征参数和语音信号的参考模式集合之间的相似度，从而提高语音识别的准确性和速度。 当采用第一和第二可能性的两个语音间隔候选时，错误边界决定的百分比约为0.5％。

公开(公告)号：US4653097A

公开(公告)日：1987-03-24

申请号：US870309

申请日：1986-05-23

申请人： Sadakazu Watanabe ,  Hidenori Shinoda

发明人： Sadakazu Watanabe ,  Hidenori Shinoda

IPC分类号： G07F7/12 ,  G06Q20/40 ,  G06Q40/00 ,  G06Q40/02 ,  G07C9/00 ,  G07D9/00 ,  G07F7/10 ,  G10L15/00 ,  G10L17/00 ,  G10L17/24 ,  G10L5/00

CPC分类号： G07F7/10 ,  G07C9/00142 ,  G07C9/00158

摘要： Speaker verification is tested in a sequence of steps: speech recognition of the spoken identification code (key code) is followed by speaker verification using the sounds of the spoken identification code. If verification fails, the speaker is urged by a speech synthesizer to utter his or her name for speaker verification.

摘要翻译： 扬声器验证以一系列步骤进行测试：语音识别码（键码）的语音识别之后是使用语音识别码的声音进行说话人验证。 如果验证失败，讲话者将被语音合成器强制发出说话人验证的姓名。

公开(公告)号：US3963350A

公开(公告)日：1976-06-15

申请号：US523056

申请日：1974-11-12

申请人： Sadakazu Watanabe ,  Hidenori Shinoda

发明人： Sadakazu Watanabe ,  Hidenori Shinoda

IPC分类号： A61B5/145 ,  A61B5/1455 ,  G01N15/14 ,  G01N21/27 ,  G01N33/48 ,  G01N33/49 ,  G06M11/00 ,  G06T1/00 ,  G01N33/16 ,  G01N21/06

CPC分类号： G01N15/1468 ,  G01N33/49 ,  G01N33/491

摘要： The segmentation apparatus comprises means for illuminating a blood smear mounted on a slide and subjected to WRIGHT staining successively with lights having wavelengths of 410 nm and 530 nm respectively for obtaining two blood corpuscle images produced by monochromatic lights having respective wavelengths, means for producing two electric signals having levels corresponding to the tones of respective blood corpuscle images, means for converting the two electric signals into binary signals in accordance with two different threshold valves, means for obtaining the exclusive OR of the two binary signals for detecting the white blood corpuscles in the blood, and means for obtaining the logical product of said two binary signals the segmenting the red blood corpuscles in the blood.

摘要翻译： 分割装置包括用于照射安装在载玻片上的血液涂片的装置，并分别用波长为410nm和530nm的光进行WRIGHT染色，以获得由具有各自波长的单色光产生的两个血球图像;用于产生两个电 具有与各个血球图像的音调相对应的电平的信号，根据两个不同阈值阀将两个电信号转换为二进制信号的装置，用于获得用于检测所述二维信号中的白细胞的两个二进制信号的异或 血液和用于获得分割血液中的红色血球的所述两个二进制信号的逻辑积的装置。

公开(公告)号：US4677673A

公开(公告)日：1987-06-30

申请号：US563755

申请日：1983-12-21

申请人： Teruhiko Ukita ,  Tsuneo Nitta ,  Sadakazu Watanabe

发明人： Teruhiko Ukita ,  Tsuneo Nitta ,  Sadakazu Watanabe

IPC分类号： G10L11/00 ,  G10L15/00 ,  G10L15/10 ,  G10L5/00

CPC分类号： G10L15/00

摘要： Continuous speech signal is recognized using "rough" and "detail" parameters derived from prestored reference speech and current unknown speech. The detail parameters are 16 spectral coefficients, the rough parameters 2 or 4 spectral coefficients representing the signal. A word interval detector decides segmentation based on rough parameter similarity.

摘要翻译： 使用从预存储的参考语音和当前未知语音导出的“粗略”和“详细”参数识别连续语音信号。 详细参数是16个频谱系数，粗略参数2或4个表示信号的频谱系数。 字间隔检测器根据粗略的参数相似度来决定分割。

公开(公告)号：US3973725A

公开(公告)日：1976-08-10

申请号：US523077

申请日：1974-11-12

申请人： Sadakazu Watanabe ,  Hidenori Shinoda ,  Yuichi Imasato

发明人： Sadakazu Watanabe ,  Hidenori Shinoda ,  Yuichi Imasato

IPC分类号： G01N33/48 ,  G01N15/14 ,  G01N33/49 ,  G06M11/00 ,  G06M11/02 ,  G06T1/00

CPC分类号： G06M11/02 ,  G01N15/1468 ,  G06F19/366

摘要： The apparatus comprises an illuminating system including a rotary filter for spectrum-analyzing a sample containing white blood corpuscles subjected to WRIGHT staining process, for example, and mounted on a glass slide, into monochromatic lights having different wavelengths in the visible light range, a video camera for scanning a predetermined area of the spectrum-analyzed sample for producing a picture image, a photoelectric transducer for producing an electric signal corresponding to the tone of the picture image at the sampling point in the predetermined area, a digital-analogue converter for converting the electric signal into a digital signal, a memory device for storing the output from the digital-analogue converter, a circuit for normalizing the information stored in the memory device, similarity calculating circuit for determining the similarity between the output from the normalyzing circuit and reference data, means for detecting and storing the types of the white blood corpuscles respectively corresponding to the maximum value and the value next to the maximum of the similarity thus determined, means for classifying the sampling points of each type of the white blood corpuscles and counting the number of the sampling points, and means for estimating the number of the white blood corpuscles of each type in the predetermined area based on the nunber of the sampling points.

摘要翻译： 该装置包括一个照明系统，该照明系统包括旋转滤光器，用于对包含经过WRIGHT染色处理的白血球的样品进行光谱分析，并将其安装在载玻片上，并将其安装在可见光范围上具有不同波长的单色光，视频 用于扫描用于产生图像的频谱分析样本的预定区域的相机，用于产生与预定区域中的采样点处的图像的色调相对应的电信号的光电变换器，用于转换的数模转换器 将电信号转换为数字信号，存储装置，用于存储来自数模转换器的输出;存储装置中存储的信息的归一化电路;相似度计算电路，用于确定来自正常分析电路的输出与参考 数据，用于检测和存储白血军的类型的装置 分别对应于最大值和接近于如此确定的相似度的最大值的值，用于对每种类型的白血球进行分类的采样点和对采样点的数量进行计数的装置，以及用于估计 基于采样点的正数的预定区域中的每种类型的白血球。

公开(公告)号：US4908704A

公开(公告)日：1990-03-13

申请号：US282257

申请日：1988-12-09

申请人： Arisa Fujioka ,  Sadakazu Watanabe

发明人： Arisa Fujioka ,  Sadakazu Watanabe

IPC分类号： G01B11/00 ,  G01B11/03 ,  G01C15/00 ,  G06T7/20 ,  G08B13/196 ,  G08B21/00 ,  H04N7/18

CPC分类号： G06K9/00771 ,  G08B13/19604 ,  B60G2400/823 ,  B60G2401/142 ,  B60G2800/222 ,  B60G2800/242

摘要： An apparatus for monitoring a moving object comprises an image memory for storing a reference monitor image of a designated monitor region, a distance map memory for storing a distance map, the monitor image of the designated monitor region and the distance map comprising a plurality of blocks having distance data from a predetermined reference point in the monitor region to points corresponding to the blocks, an object image detector for detecting an object image based on an input monitor image of the designated monitor region and the reference monitor image read out from the image memory, and a distance detector for detecting a distance from the reference point to the moving object in accordance with the detected object image and the distance map read out from the distance map memory.

摘要翻译： 一种用于监测移动物体的装置，包括用于存储指定监视区域的参考监视图像的图像存储器，用于存储距离图的距离图存储器，指定监视区域的监视图像，以及包括多个块的距离图 具有从监视区域中的预定参考点到对应于块的点的距离数据;基于指定监视区域的输入监视图像和从图像存储器读出的参考监视图像来检测对象图像的对象图像检测器 以及距离检测器，用于根据检测到的对象图像和从距离图存储器读出的距离图来检测从参考点到移动物体的距离。

公开(公告)号：US4723290A

公开(公告)日：1988-02-02

申请号：US608208

申请日：1984-05-08

申请人： Sadakazu Watanabe ,  Teruhiko Ukita

发明人： Sadakazu Watanabe ,  Teruhiko Ukita

IPC分类号： G10L15/00 ,  G10L15/02 ,  G10L15/10 ,  G10L5/00

CPC分类号： G10L15/10

摘要： Speech recognition is improved by first recognizing a sequence of phonemes, then generating a compatibility coefficient between two continuous phonemes to modify the similarity measurements. The compatibility coefficient is varied for a plurality of modified similarity measurements to further improve recognition accuracy.

摘要翻译： 通过首先识别音素序列，然后产生两个连续音素之间的相容性系数来修改相似度测量来改善语音识别。 为了进一步提高识别精度，对于多个修改的相似度测量来改变兼容性系数。