WE CLAIM:
1. A computer-implemented optical polling method comprising:
receiving an image of augmented reality (AR) tag markers associated with a poll,
wherein each AR tag marker is encoded with data, the data comprising a user identifier
and a polling option;
processing the image to identify the AR tag markers, wherein the processing
comprises:
detecting a first set of edges in the image by traversing vertically along a plurality of columns in the image, wherein the columns are separated by a predetermined length;
identifying final edges from the detected first set of edges based on contrast in pixels of the detected first set of edges and associated neighbouring pixels; identifying potential AR tag markers in the image based on a length of the identified final edges;
identifying AR tag markers from the identified potential AR tag markers by sampling bit areas in each of the obtained potential AR tag markers; decoding the AR tag markers to determine user identifier and polling option
selected by the user identifier.
2. The method of claim 1, wherein the predetermined length is less than or equivalent to a minimum AR tag marker length in the image.
3. The method of claim 1, wherein the detecting the first set of edges comprises:

i. identifying an edge by traversing every xth column in the image, wherein “x” is
minimum AR tag marker length;
ii. identifying a line by recursively traversing the identified edge to corners;
iii. identifying side edges of the marker by traversing downwards from corners of the
identified line; and
iv. traversing horizontally from an identified bottom corner to another bottom corner.
4. The method of claim 1, wherein identifying the final edges comprises:
i. accepting edge pixels having higher contrast than neighboring pixels; and ii. accepting edge pixels with contrast higher than a threshold value.
5. The method of claim 1, wherein identifying potential AR tag markers comprises:
i. comparing the length of the final edges with a predetermined minimum AR tag
marker length;
ii. rejecting the final edges if the length is less than the predetermined minimum marker length; and
iii. accepting the final edges if the length is greater than the predetermined minimum marker length to obtain the potential AR tag marker.
6. The method of claim 1, wherein identifying AR tag markers from the identified
potential AR tag markers comprises:
i. checking whether sampled bit areas are in the same range of either close to white
or black color; and
ii. checking whether the sampling value match with an actual sampling value.

7. The method of claim 1, wherein color of sampled bit area is one of black or white.
8. The method of claim 1, wherein decoding the AR tag markers comprises:
i. determining a binary code for each of sampled bit areas based on the color of the sampled bit area;
ii. decoding the binary bit areas in each of the potential AR tag markers based on the position of the most significant bit and the least significant bit in the AR tag markers.
9. The method of claim 1, further comprising identifying the AR tag markers in poor lighting conditions using adaptive thresholding.
10. The method of claim 1, further comprising calculating position of center of the bit representation to check whether the center is within or outside the edges.
11. The method of claim 1, wherein the processing further comprises splitting the
received image into multiple parts prior to finding a first set of edges.
12. An optical polling system, comprising:
a camera unit (112) configured to capture an image of augmented reality (AR) tag markers (118), wherein each AR tag marker is encoded with data, the data comprising a user identifier and a polling option;
a computing device (102) for evaluating and assessing responses, the computing device comprising:
a processing unit (404);

a memory unit (402) coupled to the processing unit, the memory unit comprises: an image reception module (416) configured to receive the image of AR tag markers;
an edge detection module (418) configured to detect a first set of edges in the image by traversing vertically along a plurality of columns in the image, wherein the columns are separated by a predetermined length; an edge filtering module (420) configured to identify final edges from the detected first set of edges based on contrast in pixels of the detected first set of edges and associated neighboring pixels;
a marker identification module (422) configured to identify potential AR tag markers based on a length of the identified final edges, and identify AR tag markers from the identified potential AR tag markers by sampling bit areas in each of the obtained potential AR tag markers; and
a decoding module (424) configured to decode the AR tag markers to determine user identifier and polling option selected by the user identifier.
13. The system of claim 12, wherein the memory unit further comprises a summary generation module (426) configured to generate a summary of the response of each user.
14. The system of claim 12, wherein the memory unit further comprises a learning module (404) configured to provide a plurality of subject courses.
15. The system of claim 12, wherein the decoding module is configured to:
determine a binary code for each of sampled bit areas based on the color of the
sampled bit area; and

decode the binary bit areas in each of the potential AR tag markers based on the position of the most significant bit and the least significant bit in the AR tag markers.
16. The system of claim 12, further comprising a second computing device (104) communicatively coupled to the computing device over a network.
17. The system of claim 12, wherein the system is configured to decode AR tag markers at a near real-time.
18. The system of claim 12, further comprising a plurality of user devices (116) for responding to poll, wherein the user devices display the AR tag marker indicating a user identity and user response.
19. The system of claim 12, wherein the camera unit (112) is configured to adjust the exposure time based on an average grayscale value of the image.
20. The system of claim 12, wherein the memory unit further comprises an image splitting module configured to split the received image prior to finding a first set of edges.