User-in-the-loop object detection and classification systems and methods
A detection device is adapted to traverse a search area and generate sensor data associated with an object that may be present in the search area, the detection device comprising a first logic device configured to detect and classify the object in the sensor data, communicate object detection information to a control system when the detection device is within a range of communications of the control system, and generate and store object analysis information for a user of the control system when the detection device is not in communication with the control system. A control system facilitates user monitoring and/or control of the detection device during operation and to access the stored object analysis information. The object analysis information is provided in an interactive display to facilitate user detection and classification of the detected object by the user to update the detection information, trained object classifier, and training dataset.
1. A system comprising:
a detection logic device configured to:
detect and classify an object in sensor data at an unmanned device, wherein the unmanned device comprises an unmanned ground vehicle (UGV), and unmanned aerial vehicle (UAV), and/or an unmanned marine vehicle (UMV);
generate and store object analysis information for a user of a control system, wherein the object analysis information facilitates user detection and classification of the detected object; and
a second logic device configured to:
receive real-time communications from the unmanned device relating to the detected object;
access the stored object analysis information during a period when the unmanned device is in a communication range of the control system;
display at least a portion of the object analysis information for the user to facilitate detection and classification of the detected object by the user;
update object detection information in accordance with user input;
generate a training data sample from the updated object detection information for use in training an object classifier;
retrain the object classifier using a dataset that includes the training data sample; and
determine whether to replace a trained object classifier with the retrained object classifier, determination based at least in part on a comparative accuracy of the trained object classifier and the retrained object classifier in classifying a test dataset.
2. The system of claim 1 , wherein the unmanned device further comprises a sensor configured to generate the sensor data, the sensor comprising a visible light image sensor, an infrared image sensor, a radar sensor, and/or a Lidar sensor.
3. The system of claim 1 , wherein the detection logic device is further configured to execute a trained neural network configured to receive a portion of the sensor data and output a location of the object in the sensor data and the classification for the object.
4. The system of claim 3 , wherein the trained neural network is configured to generate a confidence factor associated with the classification.
5. The system of claim 1 , further comprising the control system, wherein the control system is configured to facilitate user monitoring and/or control of the unmanned device during operation and comprises:
a display screen configured to display at least the portion of the object analysis information; and
a user interface configured to receive the detection and the classification of the detected object.
6. The system of claim 1 , wherein the second logic device is further configured to:
download the retrained object classifier to the unmanned device to replace the trained object classifier; and
add the training data sample to a training dataset.
7. The system of claim 1 , wherein the detection logic device is further configured to construct a map based on generated sensor data.
8. The system of claim 1 , wherein the unmanned device is adapted to traverse a search area and generate the sensor data associated with one or more objects that may be present in the search area.
9. A method comprising:
operating an unmanned device, wherein the unmanned device comprises an unmanned ground vehicle (UGV), and unmanned aerial vehicle (UAV), and/or an unmanned marine vehicle (UMV);
detecting and classifying an object in sensor data sensed at the unmanned device;
generating and storing object analysis information for a user of a control system, wherein the object analysis information is generated to facilitate user detection and classification of the detected object;
receiving real-time communications from the unmanned device relating to the detected object;
accessing the stored object analysis information during a period when the unmanned device is in communication range of the control system;
displaying at least a portion of the object analysis information for the user to facilitate detection and classification of the detected object by the user;
updating object detection information in accordance with user input;
generating a training data sample from the updated object detection information for use in training an object classifier;
retraining the object classifier using a dataset that includes the training data sample; and
determining whether to replace a trained object classifier with the retrained object classifier, determination based at least in part on a comparative accuracy of the trained object classifier and the retrained object classifier in classifying a test dataset.
10. The method of claim 9 , further comprising:
generating the sensor data comprising a visible light image, an infrared image, a radar signal, and/or a Lidar signal.
11. The method of claim 9 , further comprising:
receiving, at a neural network, a portion of the sensor data, an output a location of the detected object in the sensor data, and the classification for the detected object; and
generating, at the neural network, a confidence factor associated with the classification.
12. The method of claim 9 , further comprising:
downloading the retrained object classifier to the unmanned device to replace the trained object classifier; and
adding the training data sample to a training dataset.
13. The method of claim 9 , further comprising constructing a map based on generated sensor data.
14. The method of claim 9 , wherein the operating the unmanned device further comprises operating the unmanned device to traverse a search area and generate the sensor data associated with one or more objects that may be present in the search area.
15. A non-transitory computer-readable medium having instructions stored thereon, that when executed by a processor cause the processor to perform operations, the operations comprising:
operating an unmanned device, wherein the unmanned device comprises an unmanned ground vehicle (UGV), and unmanned aerial vehicle (UAV), and/or an unmanned marine vehicle (UMV);
detecting and classifying an object in sensor data sensed at the unmanned device;
generating and storing object analysis information for a user of a control system, wherein the object analysis information is generated to facilitate user detection and classification of the detected object;
receiving real-time communications from the unmanned device relating to the detected object;
accessing the stored object analysis information during a period when the unmanned device is in communication range of the control system;
updating object detection information based on the object analysis information;
generating a training data sample from the updated object detection information for use in training an object classifier;
retraining the object classifier using a dataset that includes the training data sample; and
determining whether to replace a trained object classifier with the retrained object classifier, determination based at least in part on a comparative accuracy of the trained object classifier and the retrained object classifier in classifying a test dataset.
16. The non-transitory computer-readable medium of claim 15 , further comprising:
generating the sensor data comprising a visible light image, an infrared image, a radar signal, and/or a Lidar signal.
17. The non-transitory computer-readable medium of claim 15 , further comprising:
receiving, at a neural network, a portion of the sensor data, an output a location of the detected object in the sensor data, and the classification for the detected object; and
generating, at the neural network, a confidence factor associated with the classification.
18. The non-transitory computer-readable medium of claim 15 , further comprising:
downloading the retrained object classifier to the unmanned device to replace the trained object classifier; and
adding the training data sample to a training dataset.
19. The non-transitory computer-readable medium of claim 15 , further comprising:
constructing a map based on generated sensor data.
20. The non-transitory computer-readable medium of claim 15 , further comprising:
operating the unmanned device to traverse a search area and generate the sensor data associated with one or more objects that may be present in the search area.