Field Image Correlation Differential Change Detection & Alerting System
AssigneeFarmers Edge Inc.
InventorGordon Logie
Life of this patent
Estimating the term…
Abstract
A system and method detect a change within an agricultural field during a growing season. The system and method involve the computation of correlation coefficients between pairs of images in a time series of images over an area, for example an agricultural field. Through comparisons of correlation coefficients over time, sudden field changes can be detected. When these field changes meet specified criteria (such as a minimum value threshold), a notification (alert) can be sent to an agricultural grower, informing them of the change in crop condition, and allowing them to view the changes in the field.
Claims15
A method of detecting a change within an agricultural field during a growing season, the method comprising:
- (a) receiving a plurality of remotely sensed images of the agricultural field acquired during the growing season;
- (b) for each one of a plurality of the remotely sensed images, using a processor,
- (i) designating a currently selected one of the remotely sensed images as a master image;
- (ii) identifying a set of past images among the remotely sensed images that were acquired before the master image;
- (iii) computing a past correlation coefficient between each past image and the master image;
- (iv) identifying a maximum past correlation value among the past correlation coefficients;
- (v) identifying a set of future images among the remotely sensed images that were acquired after the master image;
- (vi) computing a future correlation coefficient between each future image and the master image;
- (vii) identifying a maximum future correlation value among the future correlation coefficients;
- and
- (viii) computing a correlation differential value corresponding to a difference between the maximum past correlation value and the maximum future correlation value;
- and
- (c) comparing each correlation differential value to a notification criterium and determining a notification condition if one of the correlation differential values meets the notification criterium.
A system for detecting a change within an agricultural field during a growing season, the system comprising:
- a memory storing programming instructions thereon;
- at least one processor operably coupled to the memory so as to execute the programming instructions whereby said at least one processor is configured to:
- (a) receive a plurality of remotely sensed images of the agricultural field acquired during the growing season;
- (b) for each one of a plurality of the remotely sensed images:
- (i) identify a currently selected one of the remotely sensed images as a master image;
- (ii) identify a set of past images among the remotely sensed images that were acquired before the master image;
- (iii) compute a past correlation coefficient between each past image and the master image;
- (iv) identify a maximum past correlation value among the past correlation coefficients;
- (v) identify a set of future images among the remotely sensed images that were acquired after the master image;
- (vi) compute a future correlation coefficient between each future image and the master image;
- (vii) identify a maximum future correlation value among the future correlation coefficients;
- and
- (viii) compute a correlation differential value corresponding to a difference between the maximum past correlation value and the maximum future correlation value;
- and
- (c) compare each correlation differential value to a notification criterium and determine a notification condition if one of the correlation differential values meets the notification criterium.
Dependencies are inferred from the claim text.