Optimal multi-class classifier threshold-offset estimation with particle swarm optimization for visual object recognition
    31.
    发明授权
    Optimal multi-class classifier threshold-offset estimation with particle swarm optimization for visual object recognition 有权
    用于视觉对象识别的粒子群优化的最优多类分类器阈值偏移估计

    公开(公告)号:US08768868B1

    公开(公告)日:2014-07-01

    申请号:US13440881

    申请日:2012-04-05

    IPC分类号: G06N5/00

    CPC分类号: G06N5/00

    摘要: Described is a system for multi-class classifier threshold-offset estimation for visual object recognition. The system receives an input image with input features for classifying. A pair-wise classifier is trained for each pair of a plurality of object classes. A set of classification responses is generated, and a multi-class receiver-operating-characteristics (ROC) curve is computed for a set of threshold-offsets. An objective function of classification performance is computed from the ROC curve and optimized using particle swarm optimization (PSO) to generate a set of optimized threshold-offsets. The optimized threshold-offsets are then applied to the classification responses. The resulting classification responses are compared to a predetermined value to classify each input feature as belonging to one object class or another. The tuning of the threshold-offsets with (PSO) improves classification performance in a visual object recognition system.

    摘要翻译: 描述了用于视觉对象识别的多类分类器阈值偏移估计的系统。 系统接收具有输入特征进行分类的输入图像。 针对多对象类的每一对训练一对成对的分类器。 生成一组分类响应,并计算一组阈值偏移量的多类接收器操作特性(ROC)曲线。 从ROC曲线计算分类性能的目标函数,并使用粒子群优化(PSO)进行优化,以生成一组优化的阈值偏移。 然后将优化的阈值偏移应用于分类响应。 将所得分类响应与预定值进行比较,以将每个输入特征分类为属于一个对象类或另一对象类。 使用(PSO)调整阈值偏移可提高视觉对象识别系统中的分类性能。

    Hierarchical spatial representation for multimodal sensory data
    32.
    发明授权
    Hierarchical spatial representation for multimodal sensory data 有权
    多模态感觉数据的分层空间表示

    公开(公告)号:US08693729B1

    公开(公告)日:2014-04-08

    申请号:US13585356

    申请日:2012-08-14

    IPC分类号: G06T7/00 B25J9/16

    摘要: The present invention creates and stores target representations in several coordinate representations based on biologically inspired models of the human vision system. By using biologically inspired target representations a computer can be programmed for robot control without using kinematics to relate a target position in camera eyes to a target position in body or head coordinates. The robot sensors and appendages are open loop controlled to focus on the target. In addition, the invention herein teaches a scenario and method to learn the mappings between coordinate representations using existing machine learning techniques such as Locally Weighted Projection Regression.

    摘要翻译: 本发明基于人类视觉系统的生物学启发模型创建并存储几个坐标表示中的目标表示。 通过使用生物启发的目标表示,计算机可以被编程用于机器人控制,而不使用运动学来将相机眼睛中的目标位置与身体或头部坐标中的目标位置相关联。 机器人传感器和附件是开环控制的,专注于目标。 另外,本发明教导了使用现有机器学习技术(例如局部加权投影回归)来学习坐标表示之间的映射的情景和方法。

    Bio-inspired actionable intelligence method and system
    33.
    发明授权
    Bio-inspired actionable intelligence method and system 有权
    生物启发的可操作智能方法和系统

    公开(公告)号:US08515160B1

    公开(公告)日:2013-08-20

    申请号:US12316941

    申请日:2008-12-17

    IPC分类号: G06K9/62 G06K9/00

    CPC分类号: G06K9/4623 G06K9/3241

    摘要: A bio-inspired actionable intelligence method and system is disclosed. The actionable intelligence method comprises recognizing entities in an imagery signal, detecting and classifying anomalous entities, and learning new hierarchal relationships between different classes of entities. A knowledge database is updated after each new learning experience to aid in future searches and classification. The method can accommodate incremental learning via Adaptive Resonance Theory (ART).

    摘要翻译: 公开了一种生物灵感的可操作智能方法和系统。 可操作的智能方法包括识别图像信号中的实体,检测和分类异常实体,以及学习不同类别的实体之间的新的层次关系。 每个新的学习经验之后更新知识数据库,以帮助未来的搜索和分类。 该方法可以通过自适应共振理论(ART)来适应增量学习。

    System for allocating resources to optimize transition from a current state to a desired state
    34.
    发明授权
    System for allocating resources to optimize transition from a current state to a desired state 有权
    用于分配资源以优化从当前状态到期望状态的转换的系统

    公开(公告)号:US08458715B1

    公开(公告)日:2013-06-04

    申请号:US12070865

    申请日:2008-02-21

    IPC分类号: G06F9/46 G06F19/00 G06F15/00

    CPC分类号: G06F9/50 G06Q10/06

    摘要: Described is a Distributed Resource Allocation System (DRAS) for sensor control and planning. The DRAS comprises an information framework module that is configured to specify performance goals, assess current performance state, and includes sensor models to achieve the performance goals. The DRAS is configured to further allocate the sensors to achieve the performance goals. Once allocated, the DRAS then reassesses the current performance state and continues reallocating the sensors until the current performance state is most similar to the performance goals.

    摘要翻译: 描述了用于传感器控制和规划的分布式资源分配系统(DRAS)。 DRAS包括一个信息框架模块,配置为指定性能目标,评估当前性能状态,并包括传感器模型以实现性能目标。 DRAS被配置为进一步分配传感器以实现性能目标。 一旦分配,DRAS然后重新评估当前的性能状态,并继续重新分配传感器,直到当前的性能状态与性能目标最相似。

    Hierarchical spatial representation for multimodal sensory data
    36.
    发明授权
    Hierarchical spatial representation for multimodal sensory data 有权
    多模态感觉数据的分层空间表示

    公开(公告)号:US08311317B1

    公开(公告)日:2012-11-13

    申请号:US12192918

    申请日:2008-08-15

    IPC分类号: G06K9/00

    摘要: The present invention creates and stores target representations in several coordinate representations based on biologically inspired models of the human vision system. By using biologically inspired target representations a computer can be programmed for robot control without using kinematics to relate a target position in camera eyes to a target position in body or head coordinates. The robot sensors and appendages are open loop controlled to focus on the target. In addition, the invention herein teaches a scenario and method to learn the mappings between coordinate representations using existing machine learning techniques such as Locally Weighted Projection Regression.

    摘要翻译: 本发明基于人类视觉系统的生物学启发模型创建并存储几个坐标表示中的目标表示。 通过使用生物启发的目标表示,计算机可以被编程用于机器人控制,而不使用运动学来将相机眼睛中的目标位置与身体或头部坐标中的目标位置相关联。 机器人传感器和附件是开环控制的,专注于目标。 另外,本发明教导了使用现有机器学习技术(例如局部加权投影回归)来学习坐标表示之间的映射的情景和方法。

    Method and system for directed area search using cognitive swarm vision and cognitive Bayesian reasoning
    37.
    发明授权
    Method and system for directed area search using cognitive swarm vision and cognitive Bayesian reasoning 有权
    使用认知群体视觉和认知贝叶斯推理的定向区域搜索的方法和系统

    公开(公告)号:US08213709B1

    公开(公告)日:2012-07-03

    申请号:US12590110

    申请日:2009-11-03

    IPC分类号: G06K9/62

    摘要: A method and system for a directed area search using cognitive swarm vision and cognitive Bayesian reasoning is disclosed. The system comprises a domain knowledge database, a top-down reasoning module, and a bottom-up module. The domain knowledge database is configured to store Bayesian network models comprising visual features and observables associated with various sets of entities. The top-down module is configured to receive a search goal, generate a plan of action using Bayesian network models, and partition the plan into a set of tasks/observables to be located in the imagery. The bottom-up module is configured to select relevant feature/attention models for the observables, and search the visual imagery using a cognitive swarm for the at least one observable. The system further provides for operator feedback and updating of the domain knowledge database to perform better future searches.

    摘要翻译: 公开了一种使用认知群体视觉和认知贝叶斯推理的定向区域搜索的方法和系统。 该系统包括域知识数据库,自上而下推理模块和自下而上模块。 域知识数据库被配置为存储包括与各组实体相关联的视觉特征和可观察性的贝叶斯网络模型。 自顶向下模块被配置为接收搜索目标,使用贝叶斯网络模型生成行动计划,并将计划分成一组要在图像中的任务/可观察值。 自下而上模块被配置为选择可观察的相关特征/关注模型,并且使用用于至少一个可观察的认知群搜索视觉图像。 该系统进一步提供操作者反馈和更新领域知识数据库以执行更好的未来搜索。

    Visual attention and object recognition system
    38.
    发明授权
    Visual attention and object recognition system 有权
    视觉注意和对象识别系统

    公开(公告)号:US08165407B1

    公开(公告)日:2012-04-24

    申请号:US11973161

    申请日:2007-10-04

    IPC分类号: G06K9/62 G06K9/00

    CPC分类号: G06K9/4623 G06K9/3241

    摘要: Described is a bio-inspired vision system for object recognition. The system comprises an attention module, an object recognition module, and an online labeling module. The attention module is configured to receive an image representing a scene and find and extract an object from the image. The attention module is also configured to generate feature vectors corresponding to color, intensity, and orientation information within the extracted object. The object recognition module is configured to receive the extracted object and the feature vectors and associate a label with the extracted object. Finally, the online labeling module is configured to alert a user if the extracted object is an unknown object so that it can be labeled.

    摘要翻译: 描述了一种用于物体识别的生物启发视觉系统。 该系统包括注意模块,对象识别模块和在线标签模块。 注意模块被配置为接收表示场景的图像,并从图像中发现和提取对象。 注意模块还被配置为生成与所提取的对象内的颜色,强度和方向信息相对应的特征向量。 对象识别模块被配置为接收提取的对象和特征向量并将标签与所提取的对象相关联。 最后,在线标签模块被配置为提醒用户,如果提取的对象是未知对象,以便可以对其进行标记。

    System and method for multi-mission prioritization using cost-based mission scheduling
    39.
    发明授权
    System and method for multi-mission prioritization using cost-based mission scheduling 有权
    使用基于成本的任务调度的多任务优先级的系统和方法

    公开(公告)号:US07895071B2

    公开(公告)日:2011-02-22

    申请号:US11504227

    申请日:2006-08-14

    IPC分类号: G06F9/46

    摘要: Described is a system for multi-mission scheduling. The system is configured to compile a list of missions, where each mission includes at least one task. Additionally, each mission has a mission value associated with it such that the mission value reflects an ordering priority of the mission. The system also compiles a list of available resources that can be utilized to complete the tasks. The resources have, varying capabilities of completing tasks. Based on the lists, the system allocates and schedules the resources to complete tasks within the missions to maximize a total mission value of completed missions. Thus, the system schedules multiple missions to maximize the value of completed missions given available resources, whereby a mission is scheduled when the totality of its tasks have been allocated a sufficient amount of resources.

    摘要翻译: 描述了一种用于多任务调度的系统。 该系统被配置为编制任务列表,其中每个任务包括至少一个任务。 此外,每个任务都有与之相关的任务值,使任务值反映出任务的顺序优先级。 该系统还编译可用于完成任务的可用资源列表。 资源有不同的完成任务的能力。 根据这些清单,系统分配和安排资源,以完成特派团内的任务,以最大限度地发挥已完成任务的总任务价值。 因此,系统安排多个任务,以最大限度地发挥现有资源的已完成特派团的价值,从而在其任务总数已分配足够资源的情况下安排任务。

    Information Processing System for Classifying and/or Tracking an Object
    40.
    发明申请
    Information Processing System for Classifying and/or Tracking an Object 有权
    用于分类和/或跟踪对象的信息处理系统

    公开(公告)号:US20080235318A1

    公开(公告)日:2008-09-25

    申请号:US12027588

    申请日:2008-02-07

    IPC分类号: G06F15/16

    摘要: According to one embodiment, a computing system includes a computing node coupled to a number of sensors. The sensors are operable to generate records from received information and transmit these records to the computing node. The computing node is operable to bind the plurality of records in a plurality of classifications using a multiple level classifier such that each classification has a differing level of specificity.

    摘要翻译: 根据一个实施例,计算系统包括耦合到多个传感器的计算节点。 传感器可操作以从接收到的信息生成记录,并将这些记录传送到计算节点。 计算节点可操作以使用多级分类器来以多个分类结合多个记录,使得每个分类具有不同的特异性水平。