Claims:WE CLAIM
1. A method for identifying a vacant parking space in a parking lot, the method comprising:
receiving, by a parking space identifier, at least one image frame from an imaging device, wherein the at least one image frame is associated with a portion of the parking lot;
determining, by the parking space identifier, one or more bounding boxes indicative of one or more vehicles in the at least one image frame;
executing, by the parking space identifier, vehicle classification on the detected one or more bounding boxes in the at least one image frame to determine at least one first dimension corresponding to each of the one or more vehicles in the portion of the parking lot, wherein the at least one first dimension represents a longest dimension of each of the one or more vehicles;
comparing, by the parking space identifier, at least one second dimension of space in between pairs of non-overlapping bounding boxes and a predefined threshold corresponding to the vehicle classification, wherein the at least one second dimension is different from the at least one first dimension; and
determining, by the parking space identifier, the vacant parking space associated with the vehicle classification based on the at least one first dimension and the comparison.

2. The method of claim 1, further comprises computing vehicle parking orientation based on a comparison between an optical axis associated with the imaging device and a plane perpendicular to the one or more vehicles.

3. The method of claim 2, wherein determining the vehicle parking orientation comprises:
comparing the at least one first dimension with a corresponding at least one predefined orientation threshold associated with each of the one or more bounding boxes; and
determining the vehicle parking orientation of the one or more vehicles relative to the optic axis based on the comparison.

4. The method of claim 1, further comprising:
pre-training the parking space identifier to detect a parking area, wherein the pre-training comprises:
segmenting the at least one image frame into one or more blocks;
determining trajectory density map (TDM) for the one or more vehicles detected in each of the one or more blocks; and
detecting the parking area based on the TDM and a vehicle speed map (VSM) for the one or more vehicles detected in each of the one or more blocks.

5. A system for identifying a vacant parking space in a parking lot, comprising:
at least one processor;
an imaging device; and
a memory, communicatively coupled to the at least one processor, wherein the memory stores processor-executable instructions, which on execution, causes the at least one processor to:
receive at least one image frame from the imaging device, wherein the at least one image frame is associated with a portion of the parking lot;
determine one or more bounding boxes indicative of one or more vehicles in the at least one image frame;
execute vehicle classification on the detected one or more bounding boxes in the at least one image frame to determine at least one first dimension corresponding to each of the one or more vehicles in the portion of the parking lot, wherein the at least one first dimension represents a longest dimension of each of the one or more vehicles;
compare at least one second dimension of space in between pairs of non-overlapping bounding boxes and a predefined threshold corresponding to the vehicle classification, wherein the at least one second dimension is different from the at least one first dimension; and
determine the vacant parking space associated with the vehicle classification based on the at least one first dimension and the comparison.

6. The system as claimed in claim 5, further comprises computing vehicle parking orientation based on a comparison between an optical axis associated with the imaging device and a plane perpendicular to the one or more vehicles.

7. The system as claimed in claim 6, wherein determining the vehicle parking orientation comprises:
comparing the at least one first dimension with a corresponding at least one predefined orientation threshold associated with each of the one or more bounding boxes; and
determining the vehicle parking orientation of the one or more vehicles relative to the optic axis based on the comparison.

8. The system as claimed in claim 5, further comprising:
pre-training the system to detect a parking area, wherein the pre-training comprises:
segmenting the at least one image frame into one or more blocks;
determining trajectory density map (TDM) for the one or more vehicles detected in each of the one or more blocks; and
detecting the parking area based on the TDM and a vehicle speed map (VSM) for the one or more vehicles detected in each of the one or more blocks.

Dated this 16th day of February 2017

R Ramya Rao
Of K&S Partners
Agent for the Applicant
, Description:TECHNICAL FIELD
The present disclosure generally relates to image processing, and more particularly, to a method and a system for identifying a vacant parking space in a parking lot.