Processors configured to detect objects and methods of detecting objects
An electronic device, to detect an object, may continuously obtain a driving image and convert the driving image into a top-view image, generate a change value including a horizontal change value based on a steering angle and a vertical change value based on a vehicle speed, compensate the top-view image based on a plurality of lanes to generate first data and compensate the first data based on the change value to generate second data, generate third data based on performing a subtraction operation on a first frame and a second frame of the second data, and a group the third data and extract a grouped object.
1. An electronic device, comprising:
processing circuitry configured to
continuously obtain a driving image and convert the driving image into a top-view image,
generate a change value including a horizontal change value based on a steering angle and a vertical change value based on a vehicle speed,
compensate the top-view image based on a plurality of lanes to generate first data and compensate the first data based on the change value to generate second data,
generate third data based on performing a subtraction operation on a first frame of the second data and a second frame of the second data, and
group the third data and extract a grouped object.
2. The electronic device of claim 1 , wherein the processing circuitry is configured to generate the first data based on compensating the plurality of lanes to be parallel to each other in the top-view image.
3. The electronic device of claim 2 , wherein the processing circuitry is configured to generate the second data based on removing the horizontal change value and the vertical change value from the first data.
4. The electronic device of claim 1 , further comprising:
a memory configured to store frames of the top-view image,
wherein the processing circuitry is configured to remove the change value from an n-th frame of the top-view image loaded in the memory and compensates the plurality of lanes to be parallel to each other, and
wherein the plurality of lanes are detected from an (n+1)-th frame of the top-view image and n is a natural number.
5. The electronic device of claim 1 , wherein the processing circuitry is configured to verify an object of interest (OOI) based on comparing the OOI provided from a neural processing circuitry with the grouped object.
6. The electronic device of claim 5 , wherein the OOI is extracted from a plurality of objects included in the driving image, based on a deep-learning scheme.
7. The electronic device of claim 5 , wherein the processing circuitry is configured to generate a verification signal indicating whether the grouped object is included in the OOI.
8. The electronic device of claim 7 , wherein, in response to the verification signal, the neural processing circuitry is configured to add the grouped object as a new OOI in response to a determination that the grouped object is not included in the OOI.
9. The electronic device of claim 1 , wherein the processing circuitry is configured to group a plurality of data values included in the third data into at least one set.
10. The electronic device of claim 1 , wherein
the steering angle is measured based on a movement of a steering wheel, and
the vehicle speed is measured based on at least one of a quantity of revolutions of an engine, pedaling of a driver, of a wheel speed over a particular period of elapsed time.
11. A method of detecting an object, the method comprising:
continuously obtaining a driving image and converting the driving image into a top-view image;
generating a change value including
a horizontal change value based on a steering angle, and
a vertical change value based on a vehicle speed,
generating first data based on compensating the top-view image based on a plurality of lanes detected in the top-view image;
generating second data based on compensating the first data based on the change value;
generating third data based on performing a subtraction operation on a first frame of the second data and a second frame of the second data; and
grouping the third data and extracting a grouped object.
12. The method of claim 11 , wherein the generating of the first data includes
detecting the plurality of lanes in the top-view image; and
compensating the plurality of lanes to be parallel to each other.
13. The method of claim 11 , wherein the generating of the second data includes
removing the horizontal change value and the vertical change value from the first data.
14. The method of claim 11 , further comprising:
obtaining an object of interest (OOI) from a neural processing circuitry,
wherein the OOI is extracted from a plurality of objects included in the driving image, based on a deep-learning scheme.
15. The method of claim 14 , further comprising:
generating the grouped object based on grouping a plurality of data values included in the third data into at least one set.
16. The method of claim 15 , further comprising:
verifying the OOI based on the grouped object.
17. The method of claim 16 , wherein the verifying of the OOI comprises:
ascertaining whether the grouped object is included in the OOI; and
generating a verification signal indicating whether the OOI is included in the third data.
18. The method of claim 17 , wherein the verification signal adds the grouped object as a new OOI when the grouped object is not included in the OOI.
19. The method of claim 18 , further comprising:
adjusting at least one of the vehicle speed or the steering angle in response to the verification signal.