1. A computer-implemented method of automatically detecting a face in an image, said method comprising:
generating an integral image based upon said image;
sub-sampling said integral image to generate a plurality of sub-sampled integral images, wherein said plurality of sub-sampled integral images comprises a plurality of regions;
applying a plurality of classifiers to a region of said plurality of regions of said plurality of sub-sampled integral images to generate classification data; and
determining whether said region is associated with a face based upon said classification data,
wherein said plurality of sub-sampled integral images comprise different scales of said integral image.
2. The method of claim 1 further comprising:
applying a plurality of classifiers to a second region of said plurality of regions of said plurality of sub-sampled integral images to generate second classification data;
determining whether said second region is associated with a face based upon said second classification data; and
wherein said region and said second region are associated with localized data of said plurality of sub-sampled integral images, and wherein said applying a plurality of classifiers to said region and said applying a plurality of classifiers to said second region involve localized data processing.
3. The method of claim 1 further comprising:
generating a second image from said image based upon results of determining that said region is associated with said face.
4. The method of claim 3, wherein said second image comprises graphical data for rendering a graphical object around said face, and wherein said graphical object is selected from a group consisting of a box, a circle, a polygon, and a pair of brackets.
5. The method of claim 1, wherein said applying a plurality of classifiers further comprises:
applying a first plurality of classifiers in a first stage to said region of said plurality of sub-sampled integral images to generate first classification data; and
if said region is determined to be associated with said face based upon said first classification data, applying a second plurality of classifiers in a second stage to said region of said plurality of sub-sampled integral images to generate second classification data.
6. The method of claim 5, wherein said determining whether said region is associated with a face further comprises determining said region is associated with said face if said region is determined to be associated with said face based upon said second classification data.
7. The method of claim 1, wherein said plurality of classifiers comprises at least two classifiers with a different characteristic, wherein said different characteristic is selected from a group consisting of a size and a shape.
8. The method of claim 1, wherein said determining whether said region is associated with a face further comprises:
comparing a first portion of said classification data with a predetermined threshold to determine that a region of said image is associated with a portion of said face, wherein said first portion of said classification data is associated with said region of said image.
9. A non-transitory computer-usable medium having computer-readable program code embodied therein for causing a computer system to perform a method of automatically detecting a face in an image, said method comprising:
generating an integral image based upon said image;
sub-sampling said integral image to generate a plurality of sub-sampled integral images, wherein said plurality of sub-sampled integral images comprises a plurality of regions;
applying a plurality of classifiers to a region of said plurality of regions of said plurality of sub-sampled integral images to generate classification data; and
determining whether said region is associated with a face based upon said classification data.
10. The non-transitory computer-usable medium of claim 9, wherein said method further comprises:
generating a second image from said image based upon results of determining that said region is associated with said face.
11. The non-transitory computer-usable medium of claim 10, wherein said second image comprises graphical data for rendering a graphical object around said face, and wherein said graphical object is selected from a group consisting of a box, a circle, a polygon, and a pair of brackets.
12. The non-transitory computer-usable medium of claim 9, wherein said applying a plurality of classifiers further comprises:
applying a first plurality of classifiers in a first stage to said region of said plurality of sub-sampled integral images to generate first classification data; and
if said region is determined to be associated with said face based upon said first classification data, applying a second plurality of classifiers in a second stage to said region of said plurality of sub-sampled integral images to generate second classification data.
13. The non-transitory computer-usable medium of claim 12, wherein said determining whether said region is associated with a face further comprises determining said region is associated with said face if said region is determined to be associated with said face based upon said second classification data.
14. The non-transitory computer-usable medium of claim 9, wherein said plurality of classifiers comprises at least two classifiers with a different characteristic, wherein said different characteristic is selected from a group consisting of a size and a shape.
15. The non-transitory computer-usable medium of claim 9, wherein said determining whether said region is associated with a face further comprises:
comparing a first portion of said classification data with a predetermined threshold to determine that a region of said image is associated with a portion of said face, wherein said first portion of said classification data is associated with said region of said image.
16. A computer system comprising a processor and a memory, wherein said memory comprises instructions that when executed by said processor perform a method of detecting a face in an image, said method comprising:
generating an integral image based upon said image;
sub-sampling said integral image to generate a plurality of sub-sampled integral images, wherein said plurality of sub-sampled integral images comprises a plurality of regions;
applying a plurality of classifiers to a region of said plurality of regions of said plurality of sub-sampled integral images to generate classification data; and
determining whether said region is associated with a face based upon said classification data.
17. The computer system of claim 16, wherein said method further comprises:
generating a second image from said image based upon results of determining that said region is associated with said face.
18. The computer system of claim 17, wherein said second image comprises graphical data for rendering a graphical object around said face, and wherein said graphical object is selected from a group consisting of a box, a circle, a polygon, and a pair of brackets.
19. The computer system of claim 16, wherein said applying a plurality of classifiers further comprises:
applying a first plurality of classifiers in a first stage to said region of said plurality of sub-sampled integral images to generate first classification data; and
if said region is determined to be associated with said face based upon said first classification data, applying a second plurality of classifiers in a second stage to said region of said plurality of sub-sampled integral images to generate second classification data.
20. The computer system of claim 19, wherein said determining whether said region is associated with a face further comprises determining said region is associated with said face if said region is determined to be associated with said face based upon said second classification data.
21. The computer-usable medium of claim 16, wherein said plurality of classifiers comprises at least two classifiers with a different characteristic, wherein said different characteristic is selected from a group consisting of a size and a shape.
22. The computer system of claim 16, wherein said determining whether said region is associated with a face further comprises:
comparing a first portion of said classification data with a predetermined threshold to determine that a region of said image is associated with a portion of said face, wherein said first portion of said classification data is associated with said region of said image.
23. The computer system of claim 16, wherein said processor comprises a graphics processing unit.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A method of biosensor-based detection of toxins, comprising the steps of:
concentrating a plurality photosynthetic organisms in a fluid to be analyzed into a concentrated region using biased AC electro-osmosis;
obtaining a measured photosynthetic activity of said photosynthetic organisms in said concentrated region, wherein chemical, biological or radiological agents reduce a nominal photosynthetic activity of said photosynthetic organisms, and
determining a presence of at least one of said chemical, biological or radiological agents, or precursors thereof, in said fluid based on said measured photosynthetic activity.
2. The method of claim 1, wherein said plurality of photosynthetic organisms are naturally-occurring, free-living, indigenous organisms in said fluid.
3. The method of claim 1, wherein said photosynthetic activity comprises chlorophyll fluorescence induction.
4. The method of claim 1, wherein a lab-on-a-chip system is used for said concentrating step.
5. The method of claim 1, wherein said fluid is drawn from a source of primary-source drinking water.
6. The method of claim 5, further comprising the step of refreshing a supply of said photosynthetic organisms by drawing a fresh supply of said drinking water and repeating said method.
7. The method of claim 1, wherein said determining step further comprises the steps of:
providing at least one time-dependent control signal generated by said photosynthetic organism is said fluid;
obtaining a time-dependent biosensor signal from said biosensor in said fluid for the presence of one or more of said agents;
processing said time-dependent biosensor signal to obtain a plurality of feature vectors using at least one of amplitude statistics and a time-frequency analysis, and
determining said presence of at least one of said chemical, biological or radiological agents, or precursors thereof, from said feature vectors based on reference to said control signal.
8. The method of claim 7, wherein said time-frequency analysis comprises wavelet coefficient analysis.
9. The method of claim 7, wherein both said amplitude statistics and time-frequency analysis are used in said processing step.
10. The method of claim 7, further comprising the step of identifying which of said agents are present in said fluid.
11. The method of claim 10, wherein a linear discriminant method is used for said identifying step.
12. The method of claim 11, wherein said linear discriminant method comprises support vector machine (SVM) classification.
13. The method of claim 1, wherein a DC bias for said biased ACEO is from 1 to 10 volts.