Bio-security system based on multi-spectral sensing

1. A bio-security based face recognition system for recognizing a face, wherein the recognition system comprising:

an imaging module, wherein the imaging module comprising:

an RGB camera unit to detect the face, wherein the RGB camera unit comprising:

an offline calibrator, wherein the offline calibrator calibrates the face with a calibration board to generate a first calibrated face; and

an online calibrator, wherein the online calibrator rectifies errors in the first calibrated face to generate a second calibrated face; and

an IR thermal sensor unit, wherein the IR thermal sensor unit comprising;

a thermal extractor, wherein the thermal extractor extracts a first thermal image of the face;

a plurality of filters, wherein the plurality of filters comprising:

a temporal filter for eliminating noise from the first thermal image to generate a second thermal image; and

a spatial filter for dividing the second thermal image to a plurality of grid cells, wherein the spatial filter identifies temperature distribution among the plurality of grid cells;

a calibrator, wherein the calibrator identifies a grid cell with highest temperature from the plurality of grid cells, further wherein the calibrator eliminates a temperature drift from the grid cell to generate a calibrated grid cell;

a Convolutional Neural Network (CNN) based module, wherein the Convolutional Neural Network (CNN) based module comprising:

a temperature measurement unit for comparing the temperature distribution of the plurality of grid cells with a predefined template to authenticate the face; and

an anti-spoofing unit, wherein the anti-spoofing unit analyzes the calibrated grid cell and the second calibrated face to authenticate the face.

2. The face recognition system in accordance with claim 1 , wherein the face recognition system is a multi-spectral fusion security system.

3. The face recognition system in accordance with claim 2 , wherein the multi-spectral fusion security systems is based on anti-spoofing and temperature measurement of the face.

4. The face recognition system in accordance with claim 1 , wherein the face recognition system further authenticates the face based on one or more bio-attributes.

5. The face recognition system in accordance with claim 4 , wherein the bio-attributes include either of a forehead, a tear gland and a neck.

6. The face recognition system in accordance with claim 1 , wherein the calibration board is a multi-modal calibration board.

7. The face recognition system in accordance with claim 6 , wherein the multi-modal calibration board is a black-white calibration board with one or more beating devices.

8. The face recognition system in accordance with claim 7 , wherein the one or more heating devices adjusts temperature of the plurality of grid cells.

9. The face recognition system in accordance with claim 7 , wherein the multi-modal calibration board forms one or more calibration patterns.

10. The face recognition system in accordance with claim 9 , wherein the one or more calibration patterns includes either of a rectangle pattern, a circle pattern and a chess pattern.

11. The face recognition system in accordance with claim 1 , wherein the RGB camera unit detects one or more features from the face including gender, glasses, pose , hair mask or face mask.

12. The face recognition system in accordance with claim 1 , wherein the face recognition system generates an alarm based on temperature of the calibrated grid cell above the pre-defined threshold.

13. The face recognition system in accordance with claim 1 , wherein the face recognition system authenticates a human face.

14. A bio-security based multispectral face recognition system for recognizing a human face, wherein the multispectral face recognition system comprising:

an imaging module, wherein the imaging module comprising:

an RGB camera unit to detect the human face, wherein the RGB camera unit includes an offline calibrator and an online calibrator, further wherein the offline calibrator calibrates the human face based on one or more features of the human face to generate a first calibrated face, further wherein the online calibrator rectifies errors in the first calibrated face to generate a second calibrated face; and

an IR thermal sensor unit for extracting first thermal image of the human face, wherein IR thermal sensor unit;

a pair of filters including a temporal filter and a spatial filter, wherein the temporal filter eliminates noise from the first thermal image to generate a second thermal image, further wherein the spatial filter divides the second thermal image to a plurality of grid cells further wherein the spatial filter identifies temperature distribution among the plurality of grid cells; and

a calibrator, wherein the calibrator identifies a grid cell with highest temperature from the plurality of grid cells and eliminates a temperature drift from the grid cell to generate a calibrated grid cell;

a Convolutional Neural Network (CNN) based module, wherein the Convolutional Neural Network (CNN) based module comprising;

a temperature measurement unit for comparing the temperature distribution of the plurality of grid cells with a predefined template to authenticate the human face; and

an anti-spoofing unit, wherein the anti-spoofing unit analyzes the calibrated grid cell and the second calibrated face to authenticate the human face.

15. A method, for recognizing a face, wherein the method comprising:

detecting the face by an RGB camera unit;

calibrating the face based on one or more features of the face to generate a first calibrated face;

rectifying errors in the first calibrated face to generate a second calibrated face;

extracting a first thermal image of the face by a thermal extractor;

eliminating noise from tide first thermal image to generate a second thermal image;

dividing the second thermal image into a number of grid cells,

identifying temperature distribution among the number of grid cells;

identifying a grid cell with highest temperature from the number of grid cells,

eliminating a temperature drift from the number of grid cells to generate a calibrated grid cell;

comparing the temperature distribution of the number of grid cells with a predefined template to authenticate the face; and

analyzing the calibrated grid cell and the second calibrated face to authenticate the face.

16. A method for face recognition based on multispectral sensing, wherein the method comprising:

detecting the face by an RGB camera unit;

calibrating the face based on one or more features of the face and rectifying the error in the calibration of the face to generate a calibrated face;

extracting a first thermal image of the face by a thermal extractor;

eliminating noise from the first thermal image to generate a second thermal image;

dividing the second thermal image into a number of grid cells and identifying temperature distribution among the number of grid cells;

identifying a grid cell with highest temperature from the number of grid cells,

eliminating a temperature drift from the number of grid cells to generate a calibrated grid cell;

analyzing the calibrated grid cell and the calibrated face to authenticate the face; and

comparing the temperature distribution of the number of grid cells with a predefined template to authenticate the face.