Back to results
US 2016/0239718 A1Published application
Image analysis systems and methods
AssigneeNetspark Ltd.
InventorsElyasaf KORENWAITZ, Haim EAOSHYHES
Life of this patent
Estimating the term…
Abstract
A system including: (a) a network hub or port adapted to detect image files in transit according to their file designations; (b) an object detector configured to identify one or more regions of interest (ROI) in each image file as potentially containing an object of interest (OOI); (c) a feature analyzer adapted to express one or more General Classification Features (GCF) of each ROI as a vector; and (d) a decision module adapted accept or reject each ROI as containing said OOI based upon the one or more GCF vectors.
Claims20
A system comprising:
- (a) a network hub or port adapted to detect image files in transit according to their file designations;
- (b) an object detector configured to identify one or more regions of interest (ROI) in each image file as potentially containing an object of interest (OOI);
- (c) a feature analyzer adapted to express one or more General Classification Features (GCF) of each ROI as a vector;
- and
- (d) a decision module adapted accept or reject each ROI as containing said OOI based upon said one or more GCF vectors.
2 more elements
A method comprising:
- (a) scanning a plurality of images with an object detector to identify one or more regions of interest (ROI) in image file as potentially containing an object of interest (OOI);
- (b) manually reviewing each ROI to determine true positives (TP) which actually contain the OOI and false positives (FP) which do not contain the OOI;
- (c) analyzing each ROI with respect to one or more General Classification Features (GCF) and expressing each GCF of each ROI as a vector;
- and
- (d) determining a range of vector values for each GCF indicative of TP ROI which retains 95.8% of ROIs identified by an object detector and retains only 5.8% of ROIs which should have been rejected.
2 more elements
A method comprising:
- (a) identifying a region of interest (ROI) in a blue/green/red (BGR) image;
- (b) transforming BGR data in the ROI to hue-saturation-value (HSV) or another separative color space (e.g. HSL, XYZ, LUV, LAB);
- and
- (c) generating edge histogram vectors from at least one of the H, S and V channels.
1 more element
Dependencies are inferred from the claim text.