US 7,551,783 B2
Methods, functional data, and systems for image feature translation
Michael Johnson, Phoenix, Ariz. (US); Rida M. Hamza, Inver Grove Heights, Minn. (US); Au W. Kwong, Bloomington, Minn. (US); and Thomas A. Henderson, Sammamish, Wash. (US)
Assigned to Honeywell International Inc., Morristown, N.J. (US)
Filed on May 30, 2007, as Appl. No. 11/807,742.
Application 11/807742 is a continuation of application No. 10/055876, filed on Jan. 23, 2002, granted, now 7,242,806.
Prior Publication US 2007/0237397 A1, Oct. 11, 2007
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/62 (2006.01)
U.S. Cl. 382—209  [382/217; 382/218; 348/160] 20 Claims
OG exemplary drawing
 
1. A method of processing an image, comprising:
using a computer to carry out the method of identifying one or more sub images which are associated with an image of one or more instrument panels of one or more devices;
identifying values represented by one or more located features within each sub image by using one or more templates, wherein each template includes one or more rules to assist in identifying the values and each template associated with a feature vector and the feature vector representing a shape of a particular sub image and coverage area of the particular sub image and angular position of the particular sub image within the image, and wherein each feature vector is represented as a collection of three pixel coordinates; and
using the values to generate one or more readings for one or more of the instrument panels of one or more of the devices, wherein the values are used to determine locations of the features within each of the sub images, each location for each feature is then calculated using an initial calibration coordinate points and the angular position to resolve the one or more readings.