Imaging system with object recognition feedback
Examples provide an imaging system including an image sensor; and an electronic processor in communication with the image sensor. The electronic processor is configured to determine a set of imaging parameters, and, for each iteration included in an iteration set, update the set of imaging parameters by: receiving an image captured by the image sensor using the set of imaging parameters, performing object recognition on the image to detect an object of interest, determining an object recognition confidence score, determining a set of external factors based on the image and/or information from a second sensor, linking the set of imaging parameters and the set of external factors to the object recognition confidence score, and training a predictive model for updating the set of imaging parameters using the object recognition confidence score, the set of imaging parameters, and the set of external factors.
1. An imaging system comprising:
an image sensor; and
an electronic processor in communication with the image sensor and configured to:
determine a set of imaging parameters; and
for each iteration included in an iteration set, update the set of imaging parameters by:
receiving an image captured by the image sensor using the set of imaging parameters;
executing an object recognition program to detect an object of interest in the image;
determining an object recognition confidence score;
determining a set of external factors based on the image and/or information from a second sensor;
linking the set of imaging parameters and the set of external factors to the object recognition confidence score; and
training a predictive model distinct from the object recognition program, for updating the set of imaging parameters using the object recognition confidence score, the set of imaging parameters, and the set of external factors.
2. The imaging system of claim 1 , wherein
the iteration set includes a plurality of iteration subsets, and
the electronic processor is configured to, for each iteration subset,
determine a selected imaging parameter to update, and
update the selected imaging parameter based on the object recognition confidence score.
3. The imaging system of claim 2 , wherein the electronic processor is configured to determine the selected imaging parameter using the predictive model.
4. The imaging system of claim 2 , wherein the electronic processor is configured to determine a number of iterations included in each iteration subset using the predictive model.
5. The imaging system of claim 1 , wherein the set of imaging parameters includes at least one selected from the group consisting of: aperture settings, shutter speed settings, focus settings, illumination settings, image capture timing settings, and lens cleaning settings.
6. The imaging system of claim 1 , wherein the set of external factors include at least one selected from the group consisting of: ambient light information, a weather condition, a reflectivity of the object of interest, a location of the object of interest in the image, and a speed of the object of interest.
7. The imaging system of claim 1 , wherein the object of interest is a license plate, and performing object recognition includes at least one selected from the group consisting of: detecting a location of the license plate in the image, performing optical character recognition (“OCR”) on a set of characters included in the license plate, and determining a state or region associated with the license plate.
8. The imaging system of claim 1 , wherein the electronic processor is configured to determine the set of imaging parameters using the predictive model.
9. The imaging system of claim 1 , wherein the electronic processor is configured to determine the set of imaging parameters using automatic camera tuning.
10. A method performed in an imaging system, the method comprising:
determining a set of imaging parameters; and
for each iteration included in an iteration set, updating the set of imaging parameters by:
receiving an image captured by an image sensor using the set of imaging parameters;
executing an object recognition program to detect an object of interest in the image;
determining an object recognition confidence score;
determining a set of external factors based on the image and/or information from a second sensor;
linking the set of imaging parameters and the set of external factors to the object recognition confidence score; and
training a predictive model distinct from the object recognition program, for updating the set of imaging parameters using the object recognition confidence score, the set of imaging parameters, and the set of external factors.
11. The method of claim 10 , wherein
the iteration set includes a plurality of iteration subsets; and
the method further comprises:
for each iteration subset, determining a selected imaging parameter to update, and
updating the selected imaging parameter based on the object recognition confidence score.
12. The method of claim 11 , wherein the selected imaging parameter is determined using the predictive model.
13. The method of claim 11 , further comprising:
determining a number of iterations included in each iteration subset using the predictive model.
14. The method of claim 10 , wherein the set of imaging parameters includes at least one selected from the group consisting of: aperture settings, shutter speed settings, focus settings, illumination settings, image capture timing settings, and lens cleaning settings.
15. The method of claim 10 , wherein the set of external factors include at least one selected from the group consisting of: ambient light information, a weather condition, a reflectivity of the object of interest, a location of the object of interest in the image, and a speed of the object of interest.
16. The method of claim 10 , wherein the object of interest is a license plate, and performing object recognition includes at least one selected from the group consisting of: detecting a location of the license plate in the image, performing optical character recognition (“OCR”) on a set of characters included in the license plate, and determining a state or region associated with the license plate.
17. The method of claim 10 , wherein the set of imaging parameters is determined using the predictive model.
18. The imaging system of claim 10 , wherein the set of imaging parameters is determined using automatic camera tuning.
19. A non-transitory computer readable medium storing a program that, when executed by an electronic processor, causes the electronic processor to perform a set of operations comprising:
determining a set of imaging parameters; and
for each iteration included in an iteration set, updating the set of imaging parameters by:
receiving an image captured by an image sensor using the set of imaging parameters;
executing an object recognition program to detect an object of interest in the image;
determining an object recognition confidence score;
determining a set of external factors based on the image and/or information from a second sensor;
inking the set of imaging parameters and the set of external factors to the object recognition confidence score; and
training a predictive model distinct from the object recognition program, for updating the set of imaging parameters using the object recognition confidence score, the set of imaging parameters, and the set of external factors.
20. The non-transitory computer readable medium of claim 19 , wherein the set of imaging parameters includes at least one selected from the group consisting of: aperture settings, shutter speed settings, focus settings, illumination settings, image capture timing settings, and lens cleaning settings.