专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US5255349A Electronic neural network for solving "traveling salesman" and similar global optimization problems 失效
标题翻译：电子神经网络解决“旅行推销员”和类似的全球优化问题
公开(公告)号：US5255349A
公开(公告)日：1993-10-19
申请号：US789567
申请日：1991-11-08
申请人： Anilkumar P. Thakoor , Alexander W. Moopenn , Tuan A. Duong , Silvio P. Eberhardt
发明人： Anilkumar P. Thakoor , Alexander W. Moopenn , Tuan A. Duong , Silvio P. Eberhardt
IPC分类号： G06N3/04 , G06N3/063 , G06Q10/04 , G06F15/46 , G06G7/12
CPC分类号： G06Q10/047 , G06N3/0445 , G06N3/063 , G06N3/0635
摘要： This invention is a novel high-speed neural network based processor for solving the "traveling salesman" and other global optimization problems. It comprises a novel hybrid architecture employing a binary synaptic array whose embodiment incorporates the fixed rules of the problem, such as the number of cities to be visited. The array is prompted by analog voltages representing variables such as distances. The processor incorporates two interconnected feedback networks, each of which solves part of the problem independently and simultaneously, yet which exchange information dynamically.
摘要翻译：本发明是一种用于解决“旅行推销员”等全球优化问题的新型高速神经网络处理器。它包括采用二进制突触阵列的新型混合架构，其实施例结合了问题的固定规则，例如要访问的城市的数量。该阵列由表示诸如距离等变量的模拟电压提示。该处理器集成了两个互连的反馈网络，每个都能独立和同时地解决问题的一部分，而动态地交换信息。

2. 发明授权

US5479579A Cascaded VLSI neural network architecture for on-line learning 失效
标题翻译：用于在线学习的级联VLSI神经网络架构
公开(公告)号：US5479579A
公开(公告)日：1995-12-26
申请号：US316711
申请日：1994-09-22
申请人： Tuan A. Duong , Taher Daud , Anilkumar P. Thakoor
发明人： Tuan A. Duong , Taher Daud , Anilkumar P. Thakoor
IPC分类号： G06N3/063 , G06F15/18
CPC分类号： G06N3/0635 , G06N3/063
摘要： High-speed, analog, fully-parallel and asynchronous building blocks are cascaded for larger sizes and enhanced resolution. A hardware-compatible algorithm permits hardware-in-the-loop learning despite limited weight resolution. A computation-intensive feature classification application has been demonstrated with this flexible hardware and new algorithm at high speed. This result indicates that these building block chips can be embedded as application-specific-coprocessors for solving real-world problems at extremely high data rates.
摘要翻译：高速，模拟，全并行和异步构建块级联以实现更大的尺寸和更高的分辨率。硬件兼容算法尽管有限的重量分辨率允许硬件在环学习。已经通过这种灵活的硬件和新算法在高速度下演示了计算密集型特征分类应用。该结果表明，这些构建块芯片可以嵌入作为特定于应用程序的协处理器，以极高的数据速率解决现实问题。

3. 发明授权

US08204927B1 System and method for cognitive processing for data fusion 有权
标题翻译：数据融合认知处理系统与方法
公开(公告)号：US08204927B1
公开(公告)日：2012-06-19
申请号：US13042196
申请日：2011-03-07
申请人： Tuan A. Duong , Vu A. Duong
发明人： Tuan A. Duong , Vu A. Duong
IPC分类号： G06G7/02
CPC分类号： G06N3/063 , G06N3/0454 , G11C11/54
摘要： A system and method for cognitive processing of sensor data. A processor array receiving analog sensor data and having programmable interconnects, multiplication weights, and filters provides for adaptive learning in real-time. A static random access memory contains the programmable data for the processor array and the stored data is modified to provide for adaptive learning.
摘要翻译：一种用于认知处理传感器数据的系统和方法。接收模拟传感器数据并具有可编程互连，乘法重量和滤波器的处理器阵列实时提供自适应学习。静态随机存取存储器包含处理器阵列的可编程数据，修改存储的数据以提供自适应学习。

4. 发明授权

US07711157B2 Artificial intelligence systems for identifying objects 有权
标题翻译：用于识别物体的人工智能系统
公开(公告)号：US07711157B2
公开(公告)日：2010-05-04
申请号：US11498531
申请日：2006-08-01
申请人： Tuan A. Duong , Vu A. Duong , Allen R. Stubberud
发明人： Tuan A. Duong , Vu A. Duong , Allen R. Stubberud
IPC分类号： G06K9/00 , G06K9/46
CPC分类号： G06K9/6247 , G06K9/32
摘要： A process for object identification comprising extracting object shape features and object color features from digital images of an initial object and storing the extracted object shape features and object color features in a database where said extracted object shape features and object color features are associated with a unique identifier associated with said object and repeating the first step for a plurality of different objects. Then extracting object shape features and object color features from a digital image of an object whose identity is being sought and correlating the extracted object shape features and object color features of the object whose identity is being sought with the extracted object shape features and object color features previously stored in the database. If a first correlation of the extracted object shape features is better than a first threshold value for a given object associated with an identifier in the database and if a second correlation of the extracted object color features is better than a second threshold value for the given object, then making a determination that the object whose identity is being sought is said given object.
摘要翻译：一种用于对象识别的过程，包括从初始对象的数字图像中提取对象形状特征和对象颜色特征，并将提取的对象形状特征和对象颜色特征存储在数据库中，其中所提取的对象形状特征和对象颜色特征与唯一标识符，并且针对多个不同的对象重复第一步骤。然后从正在寻找其身份的对象的数字图像中提取对象形状特征和对象颜色特征，并将所提取的对象形状特征和正在寻找其身份的对象的对象颜色特征与提取的对象形状特征和对象颜色特征相关联以前存储在数据库中。如果提取的对象形状特征的第一相关性优于与数据库中的标识符相关联的给定对象的第一阈值，并且如果提取的对象颜色特征的第二相关性优于给定对象的第二阈值，然后确定正在寻找其身份的对象是所述给定对象。

5. 发明授权

US08340435B2 Method and system for object recognition search 有权
标题翻译：对象识别搜索的方法和系统
公开(公告)号：US08340435B2
公开(公告)日：2012-12-25
申请号：US12796631
申请日：2010-06-08
申请人： Tuan A. Duong , Vu A. Duong , Allen R. Stubberud
发明人： Tuan A. Duong , Vu A. Duong , Allen R. Stubberud
IPC分类号： G06K9/46
CPC分类号： G06K9/6247 , G06K9/342 , G06K9/4652
摘要： A method for object recognition using shape and color features of the object to be recognized. An adaptive architecture is used to recognize and adapt the shape and color features for moving objects to enable object recognition.
摘要翻译：一种用于识别物体的形状和颜色特征的方法。自适应架构用于识别和调整用于移动对象的形状和颜色特征以实现对象识别。

6. 发明授权

US06307587B1 Input circuit for an image processor 失效
标题翻译：图像处理器的输入电路
公开(公告)号：US06307587B1
公开(公告)日：2001-10-23
申请号：US08954600
申请日：1997-10-20
申请人： Tuan A. Duong
发明人： Tuan A. Duong
IPC分类号： H04N5225
CPC分类号： H04N5/335 , G06T1/0007
摘要： A circuit is used with an imager that is capable of capturing an image. The circuit includes a memory that is configured to during a first time interval, store a first representation of a first subimage of the image from the imager and during a second time interval, receive an update from the imager and use the update and the first representation to store a second representation of a second subimage of the image. The first subimage partially overlaps the second subimage, and the update represents a portion of the second subimage that is not present in the first subimage. The circuit also has an output circuit that is configured to during the first time interval, use the first representation to generate output signals representative of the first subimage and during the second time interval, use the second representation to generate output signals representative of the second subimage.
摘要翻译：电路与能够捕获图像的成像器一起使用。电路包括被配置为在第一时间间隔期间存储来自成像器的图像的第一子图像的第一表示，并且在第二时间间隔期间，从成像器接收更新并使用更新和第一表示以存储图像的第二子图像的第二表示。第一子图像部分地与第二子图像重叠，并且更新表示第二子图像中不存在于第一子图像中的部分。电路还具有被配置为在第一时间间隔期间的输出电路，使用第一表示来生成表示第一子图像的输出信号，并且在第二时间间隔期间，使用第二表示来生成表示第二子图像的输出信号。

7. 发明授权

US08426806B2 Differential mobility spectrometer with spatial ion detector and methods related thereto 有权
标题翻译：具有空间离子检测器的差示迁移率光谱仪及其相关方法
公开(公告)号：US08426806B2
公开(公告)日：2013-04-23
申请号：US12973843
申请日：2010-12-20
申请人： Tuan A. Duong , Isik Kanik , Vu A. Duong
发明人： Tuan A. Duong , Isik Kanik , Vu A. Duong
IPC分类号： B01D59/44
CPC分类号： G01N27/624
摘要： Differential mobility spectrometer with spatial ion detector and methods related thereto are disclosed. The use of one or more spatial detector within differential mobility spectrometry can provide for the identification and separation of ions with similar mobility and mass.
摘要翻译：公开了具有空间离子检测器的差分迁移率光谱仪及其相关方法。在差分迁移率光谱法中使用一个或多个空间检测器可以提供具有相似移动性和质量的离子的鉴定和分离。

8. 发明申请

US20090116747A1 Artificial intelligence systems for identifying objects 有权
标题翻译：用于识别物体的人工智能系统
公开(公告)号：US20090116747A1
公开(公告)日：2009-05-07
申请号：US11498531
申请日：2006-08-01
申请人： Tuan A. Duong , Vu A. Duong , Allen R. Stubberud
发明人： Tuan A. Duong , Vu A. Duong , Allen R. Stubberud
IPC分类号： G06K9/46
CPC分类号： G06K9/6247 , G06K9/32
摘要： A process for object identification comprising extracting object shape features and object color features from digital images of an initial object and storing the extracted object shape features and object color features in a database where said extracted object shape features and object color features are associated with a unique identifier associated with said object and repeating the first step for a plurality of different objects. Then extracting object shape features and object color features from a digital image of an object whose identity is being sought and correlating the extracted object shape features and object color features of the object whose identity is being sought with the extracted object shape features and object color features previously stored in the database. If a first correlation of the extracted object shape features is better than a first threshold value for a given object associated with an identifier in the database and if a second correlation of the extracted object color features is better than a second threshold value for the given object, then making a determination that the object whose identity is being sough is said given object.
摘要翻译：一种用于对象识别的过程，包括从初始对象的数字图像中提取对象形状特征和对象颜色特征，并将提取的对象形状特征和对象颜色特征存储在数据库中，其中所提取的对象形状特征和对象颜色特征与唯一标识符，并且针对多个不同的对象重复第一步骤。然后从正在寻找其身份的对象的数字图像中提取对象形状特征和对象颜色特征，并将所提取的对象形状特征和正在寻找其身份的对象的对象颜色特征与提取的对象形状特征和对象颜色特征相关联以前存储在数据库中。如果提取的对象形状特征的第一相关性优于与数据库中的标识符相关联的给定对象的第一阈值，并且如果提取的对象颜色特征的第二相关性优于给定对象的第二阈值，然后确定其身份正确的对象是所述给定对象。

9. 发明授权

US06513023B1 Artificial neural network with hardware training and hardware refresh 失效
标题翻译：具有硬件训练和硬件刷新的人造神经网络
公开(公告)号：US06513023B1
公开(公告)日：2003-01-28
申请号：US09412199
申请日：1999-10-01
申请人： Tuan A. Duong
发明人： Tuan A. Duong
IPC分类号： G06N302
CPC分类号： G06N3/08 , G06N3/063
摘要： A neural network circuit is provided having a plurality of circuits capable of charge storage. Also provided is a plurality of circuits each coupled to at least one of the plurality of charge storage circuits and constructed to generate an output in accordance with a neuron transfer function. Each of a plurality of circuits is coupled to one of the plurality of neuron transfer function circuits and constructed to generate a derivative of the output. A weight update circuit updates the charge storage circuits based upon output from the plurality of transfer function circuits and output from the plurality of derivative circuits. In preferred embodiments, separate training and validation networks share the same set of charge storage circuits and may operate concurrently. The validation network has a separate transfer function circuits each being coupled to the charge storage circuits so as to replicate the training network's coupling of the plurality of charge storage to the plurality of transfer function circuits. The plurality of transfer function circuits may be constructed each having a transconductance amplifier providing differential currents combined to provide an output in accordance with a transfer function. The derivative circuits may have a circuit constructed to generate a biased differential currents combined so as to provide the derivative of the transfer function.
摘要翻译：提供具有能够进行电荷存储的多个电路的神经网络电路。还提供了多个电路，每个电路都耦合到多个电荷存储电路中的至少一个，并被构造为根据神经元传递函数产生输出。多个电路中的每一个耦合到所述多个神经元传递函数电路中的一个，并被构造为生成所述输出的导数。权重更新电路基于来自多个传递函数电路的输出并从多个微分电路输出来更新电荷存储电路。在优选实施例中，单独的训练和验证网络共享同一组电荷存储电路并且可以同时操作。验证网络具有各自耦合到电荷存储电路的单独的传递函数电路，以便将多个电荷存储器的训练网络的耦合复制到多个传递函数电路。可以构造多个传递函数电路，每个具有跨导放大器，其提供组合的差分电流，以根据传递函数提供输出。导数电路可以具有构造为产生组合的偏置差分电流以便提供传递函数的导数的电路。

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