Image analysis – Applications – Animal – plant – or food inspection
Reexamination Certificate
2000-12-01
2003-03-04
Wu, Jingge (Department: 2623)
Image analysis
Applications
Animal, plant, or food inspection
C382S191000, C047S048500, C047S05810R, C702S005000
Reexamination Certificate
active
06529615
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method for determining the health of crops in a field, and more particularly relates to a method for determining the same by aerial photographic analysis allowing for real time or near real time correction of deficiencies in the health of the crop at any time during the growing season.
BACKGROUND OF THE INVENTION
It is well known that monitoring of agricultural crops is desirable to determine existing growing conditions so as to allow for improvement and maximization of yields. While there is considerable interest in remote sensing of crops in fields, most of the current attention is directed towards identification of atypical areas of fields that may indicate crop stress due to water, insect, disease or weed pressures. Although some attention has been directed toward identification of nutrient stress using remote sensing, the success of such identifications is generally ineffective unless nutrient stresses are sufficiently severe to reduce overall crop biomass. Moreover, as the current focus in remote sensing is directed towards anomaly detection, the most typical image product is a false color image, or a three-spectral (wavelength) image combining near infrared, red and green plane images with no better than approximately one meter spatial resolution per pixel. While interest is growing for better spatial resolution, the products necessary for such are not currently available because of the increase in cost and because file size increases exponentially with resolution, the cost being disproportionate to any increase in benefit obtained by an increase in information.
Other prior art regarding remote sensing for crop nitrogen characterization has demonstrated that the intensity of red, green and blue values of scanned color photographs is closely related to corn yields on plots that had varying nitrogen fertilizer rates. This relationship was determined during the late growing season, well after the normal times of nitrogen fertilization. While the results suggested that the different intensity values were due to nitrogen rates this was not clearly demonstrated. Moreover, the results did not demonstrate that the different intensity values were related to crop nitrogen levels.
Researchers have demonstrated that nitrogen stress can be observed using RGB aerial photos, but such observations have been qualitative in nature, and have not yet been used as a basis for immediate further nitrogen applications.
Still other prior art has demonstrated that hand-held chlorophyll meters such as the “SPAD” sold by Minolta Corporation gives measurements which are closely related to the nitrogen concentration in individual plants, and that these readings from a hand-held chlorophyll meter could be employed as a basis for supplementing required nutrients such as nitrogen. Chlorophyll, which is the related green pigments found in photosynthetic organisms, has been found to be closely related to the amount of nitrogen present in plants or crops. Thus, low chlorophyll concentration levels have been found to be indicative of slower growth in plants and ultimately lessening of yield in the crops in the field.
One of the real discoveries that would allow for remote sensing and analysis of the nitrogen contained in a plant in the field was the ability to segment between soil and crop pixels so as to permit discarding of the soil pixels and thus permit analysis only of the green vegetation or crops. The advantage of this may readily be seen because an analysis can start soon after the plants have started their growth and are emerging from the soil. In other words, a large mass of plant growth, which is usually exhibited only very late in the growing season, is unnecessary and analysis of the health of the crops may be obtained throughout the growing season. This permits correction almost immediately upon discovery of low nitrogen levels in the plants.
One way to analyze the health of the crop is to compare current images with prior images to determine whether the nitrogen or nutrient level is either the same, lower, or higher than in the previous image. A major problem with that kind of analysis is that the light present at different times of the day is difficult to monitor effectively. Shadows that are cast upon the field or the plants vary with light conditions, and in general lighting conditions in which the photographs were being taken all affect the green intensity. Therefore, comparison between previously taken photographic images and/or data in recently taken photographs is not conclusive as to whether or not the plants or crops are healthy, or need additional nitrogen. Moreover prior aerial photographic work was restricted to later in the growing season because of the inability of distinguishing between soil and crop pixels as well as mixed soil/crop pixels. Moreover, even when restricting the aerial photography to late in the growing season, shadows and the like, due to differences in lighting give inconclusive results as to the health of the biomass or canopy photographed. Additional discussions of prior art may be found in the parent application of which this is a continuation in part, the parent application being herein incorporated by reference.
Moreover, in many instances the time consumed in making a determination of the health of the crop, where stress in the crops due to lack of nitrogen poses a very real threat to yield. It is essential that corrective action be taken as quickly as possible, (i.e. within hours or days). Thus it is preferable in these instances to immediately process, in real time, the aerial photographs so that large amounts of data retention are unnecessary and all that is required is to convey the resultant information on the location of observed nitrogen deficiencies to the ground so that nitrogen may be applied to the areas requiring treatment in the field.
SUMMARY OF THE INVENTION
The present invention provides a process for determining not only the health of crops in the field but the ability to take corrective action as soon as realistically practicable. The process comprises the steps of establishing, in a field in which the health of the crops is to be determined, at least one predetermined reference area having a determined crop nutrient or nitrogen content. Aerial photographs of at least a portion of the field are obtained and georeferenced to known ground positions. A relative greenness map of the field based upon the nutrient or nitrogen referenced area is then derived, the relative greenness map indicating places in the field where the crops require immediate attention. The relative greenness map may include a gray scale value applied in the georeferenced file. The gray value map may be employed as the basis for the nitrogen prescriptions, higher gray values indicating lower crop nutrient or nitrogen concentrations which would require higher levels of additional nutrients to reach maximum yields. Concomitantly higher nutrient or nitrogen fertilizer rates would be applied as the gray values increase.
The georeferenced “relative” greenness map provides crop status information with spatial resolution equivalent to the spatial resolution of the original aerial images. This “relative” greenness map may then be converted directly to a nitrogen recommendation map through the use of a look up table based upon the relative greenness values and stage of crop development. For example, areas of the field with relative greenness values similar to those of the reference rows would receive little or no additional fertilizer, while areas showing significant nutrient stress would receive much higher rates. The actual rate recommended would also be based upon stage of crop growth, with higher nitrogen levels required for stress observed early in the season as compared to stress observed closer to, for example, harvest. In crops such as small grains, georeferenced maps of canopy extent can also be used to adjust nitrogen recommendations. Other factors such as soil type, available moisture, previous crop and manure
Dickson Monte A.
Han Shufeng
Hendrickson Larry L.
Case Corporation
Wu Jingge
LandOfFree
Method of determining and treating the health of a crop does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of determining and treating the health of a crop, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of determining and treating the health of a crop will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3025013