Methods and systems for estimating distance of an object from a moving vehicle

1. A method, comprising:

detecting, by a camera module in one or more image frames, an object on a road on which a vehicle is moving;

electronically determining a pair of lane markings associated with the road;

electronically determining a lane width between the pair of lane markings in an image coordinate of the one or more image frames, the lane width determined at a location of the object on the road; and

electronically determining a real-world distance of the object from the vehicle based at least on:

a number of pixels corresponding to the lane width in the image coordinate,

a pre-defined lane width associated with the road, and

at least one camera parameter of the camera module.

2. The method as claimed in claim 1 , wherein electronically determining the pair of lane markings comprises:

detecting, by the camera module, a predefined object above the road or on a side of the road;

determining, at a first time instant of detection of the predefined object, at least a first pair of points in the image coordinate, spaced at a pre-determined distance to each other, on a first orthogonal line from a base of the predefined object to across the road;

determining, at a second time instant of detection of the predefined object, at least a second pair of points in the image coordinate, spaced at the pre-determined distance to each other, on a second orthogonal line from the base of the predefined object to across the road; and

defining the pair of lane markings based on joining corresponding points of the first pair of points and the second pair of points.

3. The method as claimed in claim 2 , wherein when the camera module is orthogonal to the road, determining at least the first pair of points and determining at least the second pair of points comprise determining the pre-determined distance in the image coordinate based on a number of pixels associated with the predefined object in the image coordinate.

4. The method as claimed in claim 2 , wherein when the camera module is not orthogonal to the road, determining at least the first pair of points and determining at least the second pair of points comprise:

comparing an image of the predefined object with a template image of the predefined object to determine a transformation matrix; and

determining the pre-determined distance in the image coordinate based at least on the transformation matrix and a number of pixels associated with the predefined object in the image coordinate.

5. The method as claimed in claim 3 , wherein when the camera module is not orthogonal to the road, determining at least the first pair of points and determining at least the second pair of points comprise:

comparing an image of the predefined object with a template image of the predefined object to determine a transformation matrix; and

facilitating correction in orientation of the camera module to make the camera module orthogonal to the road based on the transformation matrix.

6. The method as claimed in claim 2 , wherein the camera module is a front camera mounted on a front side of the vehicle and the object is a front vehicle, and wherein the method further comprises:

calculating a velocity of the vehicle; and

calculating a safe braking distance of the vehicle for the front vehicle based on the real-world distance of the front vehicle from the vehicle and the velocity of the vehicle.

7. The method as claimed in claim 2 , wherein the camera module is a rear camera mounted on a rear side of the vehicle and the object is a rear vehicle, and wherein the method further comprises:

calculating a velocity of the vehicle; and

calculating a safe braking distance of the vehicle for the rear vehicle based on the real-world distance of the rear vehicle from the vehicle and the velocity of the vehicle.

8. The method as claimed in claim 2 , wherein the camera module comprises a front camera mounted on a front side of the vehicle and a rear camera mounted on a rear side of the vehicle, and wherein the method further comprises:

calculating a distance of the predefined object from the front camera at a time instant t 1 ;

calculating a distance of the predefined object from the rear camera at a time instant t 2 , the time instant t 2 chronologically later than the time instant t 1 ; and

calculating a length of the vehicle based at least on:

the distance of the predefined object from the front camera;

the distance of the predefined object from the rear camera; and

a current speed of the vehicle.

9. The method as claimed in claim 2 , further comprising correcting a location information of the vehicle obtained from a satellite signal receiver, based on the distance of the vehicle determined from the object.

10. The method as claimed in claim 1 , wherein the object is a front vehicle, and the method further comprises:

determining a vanishing point associated with the pair of lane markings; and

determining a time to collision with the front vehicle based on positioning of each of the vehicle and the front vehicle with respect to the vanishing point in the image coordinate.

11. The method as claimed in claim 1 , wherein the object is an anomalous object on the road, and the method further comprises:

detecting the anomalous object based on road discontinuity in the one or more image frames captured by the camera module;

determining a vanishing point associated with the pair of lane markings; and

determining a time to collision with the anomalous object based on positioning of each of the vehicle and the anomalous object with respect to the vanishing point in the image coordinate.

12. The method as claimed in claim 1 , wherein the object is an approaching vehicle on a main road when the vehicle joins from a side road to the main road, and wherein the method further comprises:

checking for a yield sign; and

detecting for a yield compliance followed by the vehicle based on determining a distance of the approaching vehicle from a rear portion of the vehicle.

13. The method as claimed in claim 1 , further comprising identifying surrounding vehicles of the vehicle in the one or more image frames using a Vehicle to Vehicle (V2V) communication.

14. A system, comprising:

a camera module configured to:

capture one or more image frames of a road on which a vehicle is moving; and

a memory configured to:

store the one or more image frames;

store image processing instructions; and

a processor configured to execute the image processing instructions on the one or more image frames and cause the system to at least:

detect, in the one or more image frames, an object on the road;

determine a pair of lane markings associated with the road;

determine a lane width between the pair of lane markings in an image coordinate of the one or more image frames, the lane width determined at a location of the object on the road; and

determine a real-world distance of the object from the vehicle based at least on a number of pixels corresponding to the lane width in the image coordinate, a pre-defined lane width associated with the road and at least one camera parameter of the camera module.

15. The system as claimed in claim 14 , wherein to determine the pair of lane markings associated with the road, the system is further caused to at least:

detect a predefined object above the road or on a side of the road;

determine at a first time instant of detection of the predefined object, at least a first pair of points in the image coordinate, spaced at a pre-determined distance to each other, on a first orthogonal line from a base of the predefined object to across the road;

determine at a second time instant of detection of the predefined object, at least a second pair of points in the image coordinate, spaced at the pre-determined distance to each other, on a second orthogonal line from the base of the predefined object to across the road; and

define the pair of lane markings based on joining corresponding points of the first pair of points and the second pair of points.

16. The system as claimed in claim 15 , wherein the camera module comprises a front camera mounted on a front side of the vehicle and a rear camera mounted on a rear side of the vehicle, and wherein the system is further caused to at least:

calculate a distance of the predefined object from the front camera at a time instant t 1 ;

calculate a distance of the predefined object from the rear camera at a time instant t 2 , the time instant t 2 chronologically later than the time instant t 1 ; and

calculate a length of the vehicle based at least on:

the distance of the predefined object from the front camera;

the distance of the predefined object from the rear camera; and

a current speed of the vehicle.

17. The system as claimed in claim 15 , wherein if the camera module is not orthogonal to the road, the system is further caused to at least:

compare an image of the predefined object, captured by the camera module not orthogonal to the road, with a template image of the predefined object stored in the memory to determine a transformation matrix; and

facilitate correction in orientation of the camera module to make the camera module orthogonal to the road based on the transformation matrix.

18. The system as claimed in claim 15 , further comprising one or more inertial sensors to determine if the camera module is orthogonal to the road.

19. The system as claimed in claim 15 , wherein the system is further caused to at least:

calculate a velocity of the vehicle; and

calculate a safe braking distance of the vehicle for the object based on the real-world distance of the object from the vehicle and the velocity of the vehicle.

20. The system as claimed in claim 15 , wherein the system is further caused to at least:

determine a vanishing point associated with the pair of lane markings; and

determine a time to collision with the object on the road based on positioning of each of the vehicle and the object with respect to the vanishing point in the image coordinate.