AN IMAGE PROCESSING SYSTEM AND A METHOD FOR EXTRACTING SPATIAL RELATIONSHIP BETWEEN OBJECTS IN AN IMAGE

    公开(公告)号:MY181673A

    公开(公告)日:2020-12-31

    申请号:MYPI2012701161

    申请日:2012-12-13

    Applicant: MIMOS BERHAD

    Abstract: The present invention relates to an image processing system (100). The image processing system (100) is able to compute and analyse spatial relationship between objects detected in an image. The image processing system (100) comprises of an image segmentation and labelling component (110), a blob detection component (120), a spatial relationship extractor component (130), and a domain knowledge base (140). The image processing system (100) extracts spatial relationship between objects in an image by performing a surface subdivision computation, two-dimensional spatial relation computation, three-dimensional spatial relation computation and spatial relation extender. (Figure 1)

    USER ASSISTED DEDUCTIVE REASONING
    2.
    发明申请
    USER ASSISTED DEDUCTIVE REASONING 审中-公开
    用户辅助减免理由

    公开(公告)号:WO2010114356A3

    公开(公告)日:2011-04-07

    申请号:PCT/MY2010000037

    申请日:2010-03-29

    CPC classification number: G06F17/3064

    Abstract: A method (100) and an apparatus for user assisted deductive reasoning using a plurality of semantic network refinement iterations (102) comprises performing a user assisted query (104) to ascertain at least one valid node from a plurality of nodes of a semantic network, reducing a solution space of the semantic network (106) by disregarding at least one invalid node from the plurality of nodes of the semantic network and pruning the semantic network (108) based on the at least one valid node from the plurality of nodes of the semantic network. Pruning the semantic network (108) further comprises querying a knowledge base (200) to ascertain a plurality of hypothesis and a corresponding plurality of antecedents for the at least one valid node and constructing a refined semantic network (214) based on the plurality of hypothesis and the corresponding plurality of antecedents for the at least one valid node.

    Abstract translation: 使用多个语义网络细化迭代(102)的方法(100)和用于用户辅助演绎推理的装置包括执行用户辅助查询(104)以从语义网络的多个节点中确定至少一个有效节点, 通过从所述语义网络的所述多个节点中忽略至少一个无效节点来减少所述语义网络(106)的解空间,并且基于来自所述语义网络的所述多个节点的所述至少一个有效节点来修剪所述语义网络(108) 语义网络 修剪语义网络(108)还包括:查询知识库(200)以确定用于所述至少一个有效节点的多个假设和对应的多个前提,并且基于所述多个假设构建精简语义网络(214) 以及用于所述至少一个有效节点的对应的多个前提。

Patent Agency Ranking