Method and apparatus for the automated detection and classification of defects in sewer pipes
First Claim
1. A method for detecting a plurality of defects in an item under inspection comprising:
- acquiring at least one image of said item;
providing a plurality of neural networks, at least one of said plurality of neural networks corresponding to each one of said plurality of defects to be detected, wherein each one of said plurality of defects is selected from at least one of cross-sectional reductions, misalignments, infiltration, and cracks;
processing said at least one image to produce a processed image having objects isolated from an image background of said image; and
inputting said processed image into said plurality of neural networks to obtain information concerning corresponding defects.
0 Assignments
0 Petitions
Accused Products
Abstract
In accordance with a first aspect of the present invention, there is provided a method for detecting a defect on a portion of an element comprising the steps of: acquiring an image of said portion; analyzing said image to highlight problematic regions of said portion; calculating a probability that said problematic region is a defect; if said probability is higher than a threshold value, determining a position of said defect on said element. Another method for classifying a defect on an element is provided. The method comprises: acquiring an image of said defect; calculating a probability that said defect corresponds to one of a series of types of defects; if said probability is higher than a threshold value, determining that said defect is a defect of that particular type. Another method for recommending a most suitable rehabilitation technique for a defect is provided. The method comprises: identifying a series of parameters corresponding to said defect; calculating a relative utility for each of a series of potential rehabilitation techniques using rehabilitation profiles; determining a most suitable rehabilitation technique for said defect corresponding to a highest value of said relative utility.
-
Citations
35 Claims
-
1. A method for detecting a plurality of defects in an item under inspection comprising:
-
acquiring at least one image of said item; providing a plurality of neural networks, at least one of said plurality of neural networks corresponding to each one of said plurality of defects to be detected, wherein each one of said plurality of defects is selected from at least one of cross-sectional reductions, misalignments, infiltration, and cracks; processing said at least one image to produce a processed image having objects isolated from an image background of said image; and inputting said processed image into said plurality of neural networks to obtain information concerning corresponding defects. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
-
20. A method for detecting a selected defect in an item under inspection comprising:
-
acquiring an image of said item; providing a neural network for detecting said selected defect, wherein said selected defect is selected from at least one of deposits, cross-sectional reductions, misalignments, infiltration, and cracks; selecting a set of image analysis techniques as a function of said selected defect; processing said image according to said selected set of image analysis techniques for said selected defect to produce a processed image having objects isolated from an image background of said image; inputting said processed image to said neural network to obtain information corresponding to said selected defect. - View Dependent Claims (21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35)
-
Specification