CLIAMS:1. A method for monitoring corrosion level of a corroded object; said method comprising:

a. acquiring a plurality of optical and thermal image of the corroded object using a multiagent system (MAS) sensor module (202);
b. performing three dimensional reconstruction and a calibration metrics from a robot operating system (ROS) using optical flow on acquired plurality of optical image for obtaining at least one point cloud using an image reconstruction module (204);
c. processing acquired plurality of thermal image using a corrosion attention model (CAM) based on experiential sampling for generating an attention graph using a corrosion attention model (CAM) module (206);
d. superimposing at least one processed thermal image out of the plurality of acquired thermal image with the at least one three dimensionally reconstructed point cloud out of the plurality of acquired optical image using an image superimposition module (208);
e. back projecting a plurality of spike point from a three dimensional to a two dimensional image for ranking corroded part of the corroded object using a back projection module (210); and
f. representing corroded part of the corroded object based on the corrosion attention model ranking; and monitoring corrosion level of the corroded object by highlighting and visualizing the corroded part of the corroded object on an opto-thermal mapper using a corrosion monitoring module (212).

2. The method as claimed in claim 1, wherein the multiagent system (MAS) sensor module (202) comprises of at least one optical sensor for acquiring the plurality of optical image of the corroded object with simultaneous robot pose estimation using the robot operating system.

3. The method as claimed in claim 1, wherein the multiagent system (MAS) sensor module (202) comprises of at least one thermal sensor for acquiring the plurality of thermal image of the corroded object.

4. The method as claimed in claim 1, further comprises of estimating thermal gradient of the corroded object using acquired thermal image of the corroded object.

5. The method as claimed in claim 1, further comprises of defining the corrosion attention model (CAM) module (206) based on spike curvatures, statistical analysis using a Gaussian mask.

6. The method as claimed in claim 1, wherein the spike point is a highest value point in the attention graph of the corrosion attention model.

7. The method as claimed in claim 1, wherein the corrosion attention model ranking is based on top n values in the attention graph of the corrosion attention model at particular instant of time.

8. The method as claimed in claim 1, further comprises of identifying three dimensional coordinate position of the corroded object using focal length, resolution and field of view (FOV) of the at least one optical sensor of the multiagent system (MAS) sensor module (202).

9. A system (200) for monitoring corrosion level of a corroded object; said system (200) comprising:

a. a multiagent system (MAS) sensor module (202) adapted for acquiring a plurality of optical and thermal image of the corroded object;
b. an image reconstruction module (204) adapted for performing three dimensional reconstruction and a calibration metrics from a robot operating system (ROS) using optical flow on acquired plurality of optical image for obtaining at least one point cloud;
c. a corrosion attention model (CAM) module (206) adapted for processing acquired plurality of thermal image based on experiential sampling for generating an attention graph;
d. an image superimposition module (208) adapted for superimposing at least one processed thermal image out of the plurality of acquired thermal image with the at least one three dimensionally reconstructed point cloud out of the plurality of acquired optical image;
e. a back projection module (210) adapted for back projecting a plurality of spike point from three dimension to a two dimensional image for ranking corroded part of the corroded object; and
f. a corrosion monitoring module (212) adapted for representing corroded part of the corroded object based on the corrosion attention model ranking; and monitoring corrosion level of the corroded object by highlighting and visualizing the corroded part of the corroded object on an opto-thermal mapper.

10. The system (200) as claimed in claim 9, wherein the multiagent system (MAS) sensor module (202) is integrated on an autonomous grounded vehicle (AGV) robot, navigating around the corroded object for acquiring the plurality of optical and thermal image of the corroded object.

11. The system (200) as claimed in claim 9, wherein the multiagent system (MAS) sensor module (202) further comprises of at least one optical sensor for acquiring the plurality of optical image of the corroded object with simultaneous robot pose estimation using the robot operating system.

12. The system (200) as claimed in claim 9, wherein the multiagent system (MAS) sensor module (202) comprises of at least one thermal sensor for acquiring the plurality of thermal image of the corroded object.

13. The system (200) as claimed in claim 9, wherein the multiagent system (MAS) sensor module (202) is further adapted to identify three dimensional coordinate position of the corroded object using focal length, resolution and field of view (FOV) of the at least one optical sensor.

14. The system (200) as claimed in claim 9, wherein the corrosion attention model (CAM) module (206) is defined based on spike curvatures, statistical analysis using a Gaussian mask.
,TagSPECI:As Attached