US 9,866,768 B1Granted

Computer vision qualified infrared temperature sensor

AssigneeThe United States of America as represented by the Secretary of Agriculture

InventorsSusan A. O'Shaughnessy, Joaquin J. Casanova, Steven R. Evett, Paul D. Colaizzi

Life of this patent

Estimating the term…

Loading patent…

Abstract

Plant canopy images and temperature are acquired with a wireless computer vision instrument, and the temperature data are qualified and pixels within the image are classified. The instrument includes sensors for measuring plant radiation due to canopy temperature and an imaging sensor which collects data over four bands, a microprocessor to receive and store and process measured data from the sensors, and a wireless transmitter for transmitting data from the microprocessor to a remote receiver. The data are used to detect plant stress and canopy cover percentage and to qualify temperature data accordingly. The data from the sensors provide information for decision support algorithms related to the initiation of automatic irrigation scheduling as a function of crop canopy cover and water stress, the qualification of temperature data used in automatic irrigation scheduling algorithms, and the detection of diseased crops for the purpose of withholding irrigations when yield potential is compromised.

A method for managing the irrigation of plants comprising the steps of:

a) providing a computer vision instrument in the vicinity of vegetation within a target control area subject to irrigation;
the computer vision instrument comprising an infrared thermometer (IRT), and a digital camera for image acquisition, the digital camera detecting red, green, blue or near infrared light, wherein said IRT and said digital camera have substantially coincident fields of view;
b) utilizing said digital camera for collecting a digital image comprising the near infrared, red, green and blue bands in said target control area, and using the IRT to collect a temperature measurement;
c) said digital camera communicating with a microprocessor, said microprocessor calculating a hue of each pixel of the digital image and classifying each pixel as soil or vegetation;
d) repeating step c) for all pixels in the digital image;
and
e) comparing the percent vegetation in the digital image to a predetermined percent vegetation threshold value to classify the digital image data as vegetation or soil, said predetermined percent vegetation threshold value comprising a subjective value/decision by an irrigation operator based on at least water availability and irrigation cost;
and
1) if the data are from vegetation, calculating a mean hue of the vegetation, and evaluating the mean hue of the vegetation to determine if the vegetation is diseased or healthy;
2) if the vegetation is diseased, optionally withholding irrigation;
3) if the vegetation is healthy, using qualified plant (for example vegetation) temperature data to determine if the vegetation is water-stressed wherein said qualified vegetation temperature is calculated by obtaining an estimate of the soil temperature, Ts;
and shadow temperature, Tg;
further evaluating a digital image data to determine a percent shadow in the image, fg;
a percent soil in the image, fs;
and a percent vegetation, fp in the image;
and calculating the qualified vegetation temperature, Tp, as
T
p
=
[
(
T
IRT
4
-
f
s
⁢
T
s
4
-
f
g
⁢
T
g
4
)
f
p
]
1
/
4
wherein Tirt is the temperature measured by the IRT;
4) based on the Tp value calculated in step 3), determining if the vegetation is water-stressed;
and,
5) generating an irrigation signal if the vegetation is water-stressed.

46 more elements

Facts

Application

14/263,451Filed 28 Apr 2014 · us provisional application

Priority

29 Apr 2013

Granted

9 Jan 2018

Term ends

Not estimatedThe source records don’t support an estimate.

Classification6 codes

G06V 10/143
Optical characteristics of the device performing the acquisition or on the illumination arrangements: sensing or illuminating at different wavelengths
G06V 20/60
Scenes; Scene-specific elements: type of objects
H04N 23/11
For generating image signals from different wavelengths: for generating image signals from visible and infrared light wavelengths
H04N 23/62
Control of cameras or camera modules: control of parameters via user interfaces

Abstract

Claims8