Detecting species diversity by image texture analysis

• US 9,858,661 B2
• Filed: 06/13/2014
• Issued: 01/02/2018
• Est. Priority Date: 06/13/2013
• Status: Expired due to Fees

• Current Assignee: Charles Stark Draper Laboratory Incorporated, University of Florida Research Foundation, Incorporated (State University System of Florida)
• Original Assignee: Charles Stark Draper Laboratory Incorporated
• Inventors: Mangoubi, Rami, Convertino, Matteo, Lowry, Nathan, Desai, Mukund
• Primary Examiner(s): Liew, Alex
• Application Number:
  US14/304,531
• Publication Number:
  US 20140369568A1
• Time in Patent Office: 1,299 Days
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/10032 Satellite or aerial image; Remote sensing
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 2207/30188 Vegetation; Agriculture
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/41 based on statistical description of texture

Abstract

A method of measuring species diversity is provided. The method includes receiving a first image of a landscape and receiving a second image of a second landscape. The method also includes representing a portion of the first image as a first region of interest comprising a multiplicity of pixels and representing a portion of the second image as a second region of interest comprising a multiplicity of pixels. The method further includes comparing at least one textural feature of the first region of interest and the second region of interest and calculating the species diversity between the first landscape and the second landscape based on the comparison of the at least one textural features of the regions of interest.

Claims

1. A system for measuring species diversity, comprising:

a memory storing processor executable instructions; and

one or more processors coupled to the memory, wherein execution of the processor executable instructions by the one or more processors causes the one or more processors to;

receive a first image of a first landscape;

receive a second image of a second landscape;

represent a portion of the first image as a first region of interest comprising a multiplicity of pixels, and

represent a portion of the second image as a second region of interest comprising a multiplicity of pixels;

calculate at least one textural characteristic of two or more spectral bands of each region of interest; and

calculate the species diversity between the first landscape and the second landscape based on the textural features.

2. The system of claim 1, wherein calculating at least one textural characteristic of each region of interest comprises performing a multi-scale wavelet decomposition on the regions of interest.

3. The system of claim 2, wherein calculating at least one textural characteristic of each region of interest comprises generating a probability density function of pixel intensities in the regions of interest based on the multi-scale wavelet decomposition.

4. The system of claim 3, wherein calculating the species diversity comprises calculating the Kullback-Leibler divergence between probability density functions of pixel intensities in the regions of interest.

5. The system of claim 4, wherein calculating the species diversity comprises correlating the Kullback-Leibler divergence with species beta-diversity by regression analysis using species diversity data.

6. The system of claim 1, wherein calculating at least one textural characteristic of each region of interest comprises calculating the Shannon entropy of the two or more spectral bands of each region of interest.

7. The system of claim 6, wherein calculating the species diversity comprises calculating the difference between calculated Shannon entropy of each region of interest.

8. The system of claim 7, wherein calculating the species diversity comprises correlating the difference between calculated Shannon entropy of each region of interest with species diversity data by regression analysis.

9. The system of claim 1, wherein receiving a first image of a first landscape comprises receiving a first image of a microscopic landscape containing multiple species of eukaryotic or prokaryotic cells, and receiving a second image of a second landscape comprises receiving a second image of a microscopic landscape containing multiple species of eukaryotic or prokaryotic cells.

10. The system of claim 1, wherein receiving a first image of a first landscape comprises receiving a first satellite image of an area of land, and receiving a second image of a second landscape comprises receiving a second satellite image of an area of land.

11. The system of claim 1, wherein receiving a first image of a first landscape comprises receiving a first medical image of a brain, and receiving a second image of a second landscape comprises receiving a second medical image of a brain.

12. The system of claim 1, wherein receiving a first image of a first landscape comprises receiving a first image of a landscape at a first point in time, and receiving a second image of a second landscape comprises receiving a second image of the landscape at a second point in time.

13. A method of measuring species diversity, comprising:

receiving a first image of a first landscape;

receiving a second image of a second landscape;

representing a portion of the first image as a first region of interest comprising a multiplicity of pixels;

representing a portion of the second image as a second region of interest comprising a multiplicity of pixels;

comparing at least one textural characteristic of two or more spectral bands of the first region of interest and the second region of interest; and

calculating the species diversity between the first landscape and the second landscape based on the comparison of the at least one textural features of the regions of interest.

14. The method of claim 13, wherein calculating at least one textural characteristic of each region of interest comprises performing a multi-scale wavelet decomposition on the regions of interest.

15. The method of claim 14, wherein calculating at least one textural characteristic of each region of interest comprises generating a probability density function of pixel intensities in the regions of interest based on the multi-scale wavelet decomposition.

16. The method of claim 15, wherein comparing textural features comprises calculating the Kullback-Leibler divergence between probability density functions of pixel intensities in the regions of interest.

17. The method of claim 16, wherein calculating the species diversity comprises correlating the Kullback-Leibler divergence with species beta-diversity by regression analysis using species diversity data.

18. The method of claim 13, wherein comparing at least one textural characteristic of two or more spectral bands of each region of interest comprises calculating the Shannon entropy of the two or more spectral bands of each region of interest.

19. The method of claim 18, wherein comparing textural features comprises calculating the difference between calculated Shannon entropy of each region of interest.

20. The method of claim 19, wherein calculating the species diversity comprises correlating the difference between calculated Shannon entropy of each region of interest with species diversity data by regression analysis.

21. The method of claim 13, wherein receiving a first image of a first landscape comprises receiving a first image of a microscopic landscape containing multiple species of eukaryotic or prokaryotic cells, and receiving a second image of a second landscape comprises receiving a second image of a microscopic landscape containing multiple species of eukaryotic or prokaryotic cells.

22. The method of claim 13, wherein receiving a first image of a first landscape comprises receiving a first satellite image of an area of land, and receiving a second image of a second landscape comprises receiving a second satellite image of an area of land.

23. The method of claim 13, wherein receiving a first image of a first landscape comprises receiving a first medical image of a brain, and receiving a second image of a second landscape comprises receiving a second medical image of a brain.

24. The method of claim 13, wherein receiving a first image of a first landscape comprises receiving a first image of a landscape at a first point in time, and receiving a second image of a second landscape comprises receiving a second image of the landscape at a second point in time.