US 11,948,289 B2
Systems and methods for precisely and accurately identifying a meter reading from an analog meter using image processing techniques
Hitesh K. Patel, San Diego, CA (US); Benjamin D. Allen, New York, NY (US); and Gaurav Goel, Los Gatos, CA (US)
Assigned to Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed by Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed on Jul. 19, 2021, as Appl. No. 17/379,413.
Prior Publication US 2023/0019755 A1, Jan. 19, 2023
Int. Cl. G06T 7/00 (2017.01); G06Q 10/0631 (2023.01); G06T 5/00 (2006.01); G06T 7/62 (2017.01); G06T 7/64 (2017.01); G06T 7/70 (2017.01); G08B 5/22 (2006.01); G08B 21/18 (2006.01); H04Q 9/00 (2006.01)
CPC G06T 7/0004 (2013.01) [G06Q 10/0631 (2013.01); G06T 5/002 (2013.01); G06T 5/009 (2013.01); G06T 7/62 (2017.01); G06T 7/64 (2017.01); G06T 7/70 (2017.01); G08B 5/22 (2013.01); G08B 21/18 (2013.01); H04Q 9/00 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/30164 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, by a device, video data identifying frames of a video capturing an analog meter with a dial and a needle;
processing, by the device, each of the frames, with a Hough transform model, to identify a center, a radius, and a perimeter of the dial;
determining, by the device, calibrated values for the dial based on the center, the radius, and the perimeter;
applying, by the device, a Gaussian blur model to one of the frames to create a base mask;
applying, by the device, thresholding for a dynamic hue saturation value (HSV) bounding value, to the base mask and the frames, to create masked frames;
identifying, by the device, contours for each of the masked frames;
processing, by the device, the contours, with a point polygon model, to identify and remove noisy polygons from the contours;
identifying, by the device, a quantity of points for each of the contours of each of the frames;
estimating, by the device, angles of the needle of the analog meter based on the quantity of points for each of the contours;
averaging, by the device, the estimated angles to determine an averaged needle angle and to identify anomalies;
determining, by the device, a needle direction based on the averaged needle angle and the anomalies;
calculating, by the device, a meter reading for the analog meter based on the averaged needle angle, the needle direction, and the calibrated values; and
providing, by the device, the meter reading for display.