System and method for reducing surveillance detection errors
A method is disclosed. The method includes providing an imaging apparatus, recording image data of an imaging location using the imaging apparatus, displaying the image data to a user via a user device, selecting an image object from the image data based on a selection criteria, and determining whether or not a selection criteria error of the image object is to be checked. The method also includes displaying a bounding shape, which bounds the image object, to the user via the user device when the selection criteria error is to be checked, prompting the user to enter user input indicating whether or not the selection criteria error is present, and storing data of the image object in a cache when the user input indicates that the selection criteria error is present.
1. A method, comprising:
providing an imaging apparatus;
recording image data of an imaging location using the imaging apparatus;
displaying the image data to a user via a user device;
selecting, using a machine learning model, a first image object from the image data based on a selection criteria;
determining, using the machine learning model, whether or not a selection criteria error of the first image object is to be checked;
displaying a bounding shape, which bounds the first image object, to the user via the user device when the selection criteria error is to be checked;
prompting the user to enter user input indicating whether or not the selection criteria error is present;
storing data of the first image object in a cache as a negative training set in response to the user input indicating that the selection criteria error is present;
updating the machine learning model in real-time to remove the selection criteria error using the negative training set to generate an updated machine learning model;
selecting, using the updated machine learning model, a second image object from the image data based on the selection criteria; and
determining, using the updated machine learning model, whether or not a selection criteria error of the first image object is to be checked.
2. The method of claim 1 , further comprising:
selecting a third image object from the first image data based on the selection criteria; and
subsequently deselecting the third image object based on comparing the third image object to the data of the first image object stored in the cache.
3. The method of claim 2 , wherein the third image object is similar to the first image object.
4. The method of claim 1 , further comprising:
storing data of a plurality of image objects in the cache;
selecting a plurality of second image objects from the image data based on the selection criteria; and
subsequently deselecting the plurality of second image objects based on comparing the plurality of second image objects to data of the plurality of image objects stored in the cache.
5. The method of claim 1 , wherein displaying the bounding shape includes processing the first image data using non-maximal suppression and identifying the bounding shape that is a single bounding box.
6. The method of claim 1 , wherein prompting the user to enter user input includes displaying a graphical element via the user device to the user to select whether or not the first image object is a false positive error selected based on the selection criteria.
7. The method of claim 1 , wherein the selection criteria includes selecting at least one first image object selected from the group of an image of a human, an image of a vehicle, and combinations thereof.
8. The method of claim 1 , wherein the imaging apparatus is a still video camera and the image data includes a still video stream.
9. The method of claim 1 , wherein:
the data of the first image object stored in the cache is data of a validated false positive error stored in a short-term memory; and
the validated false positive error is validated based on the user entering the user input indicating that the selection criteria error is present.
10. The method of claim 1 , wherein selecting the first image object from the image data based on the selection criteria includes performing object detection of the first image object using convolutional neural network object detection.
11. A surveillance detection error reduction system, comprising:
an imaging apparatus;
a user device;
a surveillance detection error reduction module, comprising computer-executable code stored in non-volatile memory; and
a processor;
wherein the imaging apparatus, the user device, the surveillance detection error reduction module, and the processor are configured to:
record image data of an imaging location using the imaging apparatus;
display the image data to a user via the user device;
select, using a machine learning model, a first image object from the image data based on a selection criteria;
determine, using the machine learning model, whether or not a selection criteria error of the first image object is to be checked;
display a bounding shape, which bounds the first image object, to the user via the user device when the selection criteria error is to be checked;
prompt the user to enter user input indicating whether or not the selection criteria error is present;
store data of the first image object in a cache as a negative training set in response to the user input indicating that the selection criteria error is present;
update the machine learning model in real-time to remove the selection criteria error using the negative training set to generate an updated machine learning model;
select, using the updated machine learning model, a second image object from the image data based on the selection criteria; and
determine, using the updated machine learning model, whether or not a selection criteria error of the first image object is to be checked
select a third image object from the image data based on the selection criteria and then subsequently deselect the third image object based on comparing the third image object to the data of the first image object stored in the cache.
12. The surveillance detection error reduction system of claim 11 , wherein:
the selection criteria includes selecting at least one image object selected from the group of an image of a human, an image of a vehicle, and combinations thereof; and
both the first image object and the third image object do not meet the selection criteria.
13. The surveillance detection error reduction system of claim 11 , wherein:
selecting the first image object, the second image object, and the third image object from the image data based on the selection criteria includes performing object detection of the first image object, the second image object, and the third image object using convolutional neural network object detection; and
displaying the bounding shape includes processing the image data using non-maximal suppression.
14. A method, comprising:
providing a still video camera;
recording image data of an imaging location using the still video camera;
displaying the image data to a user via a user device;
selecting, using a machine learning model, a first image object from the image data based on a selection criteria using convolutional neural network object detection;
determining, using the machine learning model, whether or not a selection criteria error of the first image object is to be checked;
determining a bounding box using non-maximal suppression and displaying the bounding box, which bounds the first image object, to the user via the user device when the selection criteria error is to be checked;
prompting the user to enter user input indicating whether or not the selection criteria error is present;
storing data of the first image object in a cache as a negative training set in response to the user input indicating that the selection criteria error is present;
updating the machine learning model in real-time to remove the selection criteria error using the negative training set to generate an updated machine learning model;
selecting, using the updated machine learning model, a second image object from the image data based on the selection criteria; and
determining, using the updated machine learning model, whether or not a selection criteria error of the first image object is to be checked.
15. The method of claim 14 , wherein prompting the user to enter user input includes displaying a dialog box via the user device to the user to select whether or not the first image object is a false positive error selected based on the selection criteria.
16. The method of claim 15 , wherein the first image object is the false positive error when the image object is not an image of a human or an image of a vehicle.
17. The method of claim 14 , further comprising preventing identification of a third image object to the user based on comparing the third image object to the data of the first image object stored in the cache.
18. The method of claim 14 , wherein comparing the third image object to the data of the first image object stored in the cache includes comparing the third image object to a similarity model constructed using the data of the first image object stored in the cache.
19. The method of claim 14 , further comprising transferring alert data based on the data of the first image object when the user input indicates that the selection criteria error is not present.