发明名称 Orientation-aware average intensity histogram to indicate object boundary depth in ultrasound images
摘要 Linear candidate lines at different angles are used to determine an average intensity for each pixel level of an ultrasound image. The resultant average intensities are collected into a histogram, and the histogram is used to determine the depth positions of tissue boundary lines within an ultrasound image.
申请公布号 US9014452(B2) 申请公布日期 2015.04.21
申请号 US201313971912 申请日期 2013.08.21
申请人 Seiko Epson Corporation 发明人 Xiao Jing;Wu Chenyu
分类号 G06K9/00;G06T7/00;G06K9/46 主分类号 G06K9/00
代理机构 代理人
主权项 1. A method of identifying candidate inter-tissue boundary lines in an ultrasound image, comprising: using a data processing device to implement the following steps: accessing an input ultrasound image, said input ultrasound image including rows and columns of pixels; identifying candidate boundary lines by: (a) dividing said input ultrasound image into discrete, sequential depth positions, each discrete depth position being defined by a fixed number of rows of pixels; (b) defining a plurality of horizontal band regions, each horizontal band region corresponding to a separate one of said discrete depth positions and fully encompassing its corresponding depth position, each horizontal band region spanning a plurality of sequential depth positions; (c) progressing, one discrete depth position at a time, along said input ultrasound image, and at each, current, depth position executing the following steps: (i) defining a plurality of potential boundary lines within the current horizontal band region that corresponds to the current depth position, each of the defined potential boundary lines traversing a common point within the current horizontal band at a different angle relative to the horizontal;(ii) defining a candidate band region corresponding to each potential boundary line following the trajectory of its corresponding boundary line and encompassing its corresponding boundary line;(iii) determining an intensity value for each potential boundary line based at least in part on the intensity of pixels within its corresponding candidate boundary region;(iv) recording the potential boundary line of highest intensity value as the candidate line for the current depth position; (d) collecting the intensity values of candidate lines recorded at the different depth positions into an intensity-depth histogram; (e) identifying as said candidate inter-tissue boundary lines the candidate lines that substantially correspond to peaks in said intensity-depth histogram.
地址 Tokyo JP