Crack evaluation of roofing membrane by artificial neural networks

    公开(公告)号:US11861871B2

    公开(公告)日:2024-01-02

    申请号:US16980647

    申请日:2019-03-28

    CPC classification number: G06V10/764 G06F18/214 G06F18/2415 G06N3/08 G06V10/82

    Abstract: A method and system for evaluating crack intensity on polymeric sheet based on predetermined scale of crack intensity grades, which includes the steps of a) recording digital image of at least portion of surface of polymeric sheet using apparatus for recording digital images; and b) automatic classification of crack intensity by computer-implemented program for pattern recognition by means of trained artificial neural network, including 1) inputting digital image or one or more subareas of digital image to trained artificial neural network as input data, 2) classification by artificial neural network by assigning grade from predetermined scale of crack intensity grades to digital image or one or more subareas and 3) outputting assigned grade or grades for digital image and/or one or more subareas as output data, wherein artificial neural network is trained in advance in learning phase with plurality of digital images or subareas thereof of polymeric sheet surface portions.

    Tire filling based on acrylic hydrogels

    公开(公告)号:US11001019B2

    公开(公告)日:2021-05-11

    申请号:US15737074

    申请日:2016-08-09

    Abstract: A method for producing a tire filled with a (meth)acrylic hydrogel, the method comprises a) providing a mixture comprising at least one water-soluble (meth)acrylic compound, water and an initiator, and b) filling the mixture in a tire in which the mixture polymerizes to form the (meth)acrylic hydrogel. The tire filling material is suitable for producing flat proof tires, enables fast and controlled cure and is insensitive towards dosage errors. Moreover, the tire filling material is environmental friendly and cost effective.

Patent Agency Ranking