Automated Mapping of Land Cover Using Sequences of Aerial Imagery
AssigneeRegents of the University of Minnesota
InventorsShyam Boriah, Ankush Khandelwal, Vipin Kumar, Varun Mithal, Karsten Steinhaeuser
Life of this patent
Estimating the term…
Abstract
A system has an aerial image database containing sensor data representing a plurality of aerial images of an area having multiple sub-areas. A processor applies a classifier to the sensor values to identify a label for each sub-area in each aerial image and to thereby generate an initial label sequence for each sub-area. The processor identifies a most likely land cover state for each sub-area based on the initial label sequence, a confusion matrix and a transition matrix. For each sub-area, the processor stores the most likely land cover state sequence for the sub-area.
Claims20
A system comprising:
- an aerial image database containing sensor data representing a plurality of aerial images of an area having multiple sub-areas;
- a processor applying a classifier to the sensor values to identify a label for each sub-area in each aerial image and to thereby generate an initial label sequence for each sub-area;
- the processor identifying a most likely land cover state for each sub-area based on the initial label sequence, a confusion matrix and a transition matrix;
- and
- for each sub-area, the processor storing the most likely land cover state sequence for the sub-area.
A method comprising:
- retrieving from memory features for a set of temporally separated images of an area, the area divided into a plurality of sub-areas and each sub-area having separate features for each of the temporally separated images, wherein the set of temporally separated images comprises more than two temporally separated images;
- classifying the features of each sub-area of each image such that a sequence of classes is identified for each sub-area, wherein the sequence of classes is ordered based on the temporal order of the temporally separated images;
- for each sub-area, computing a change score that indicates a confidence in a change in state for the sub-area between two images using the sub-area's entire sequence of classes, a confusion matrix and a transition matrix;
- in response to the change score, generating a user interface for a display to indicate which sub-areas changed state between two images.
A computer-readable storage medium having stored thereon computer-executable instructions that when executed by a processor cause the processor to perform steps comprising:
- classifying each pixel of a set of images into one of a set of classes, each class indicating a type of ground cover present in a geographic sub-area shown in a pixel;
- and
- for a geographic sub-area, identifying ground cover states of the geographical sub-area based on the classes of the geographic sub-area's pixels in the set of images, wherein at least one of the ground cover states is not in the set of classes.
Dependencies are inferred from the claim text.