Dynamic image annotation using infrared-identifiable wearable articles
A method for tracking a target within a field of view, includes: receiving, from an input device, a target identifier uniquely identifying the target; receiving, from an infrared camera, a sequence of images resulting from infrared radiation received from the target and comprising a first region of radiation received from the target at a first intensity and an adjacent region of radiation received from the field of view at a second intensity; detecting, in a reference image from the sequence, a mask defined by a border between the first region and the adjacent region; storing an association between the target identifier and the mask; detecting the mask in a further image from the sequence; and annotating the further image with the target identifier according to the stored association.
1. A method for tracking a target within a field of view, the method comprising:
receiving, from an input device, a target identifier uniquely identifying the target;
receiving, from an infrared camera, a sequence of images resulting from infrared radiation received from the target and comprising a first region of radiation received from the target at a first intensity and an adjacent region of radiation received from the field of view at a second intensity;
detecting, in a reference image from the sequence of images, a mask defined by a border between the first region and the adjacent region;
storing an association between the target identifier and the mask;
detecting the mask in another image from the sequence of images; and
annotating the another image with the target identifier according to the stored association.
2. The method of claim 1 , wherein the infrared radiation received from the field of view at the second intensity less than the first intensity.
3. The method of claim 1 , further comprising:
storing a plurality of mask attribute sets, and a respective article identifier for each mask attribute set;
wherein detecting the mask includes detecting one of the mask attribute sets.
4. The method of claim 3 , wherein the plurality of mask attribute sets includes a classification model; and
wherein detecting the mask includes applying the classification model to the another image.
5. The method of claim 3 , wherein detecting the mask in the another image from the sequence of images includes:
for each image in the sequence of images, determining whether the image contains any of the mask attribute sets; and
when the determination is affirmative, determining that the mask attribute set corresponds to the article identifier associated with the target identifier.
6. The method of claim 1 , further comprising:
detecting the target in a region of the another image; and
annotating the another image by storing the target identifier in association with the region of the another image.
7. The method of claim 1 , further comprising:
receiving a task completion indicator; and
discarding the association between the target identifier and the mask.
8. The method of claim 7 , wherein the task completion indicator includes receiving the target identifier from the input device.
9. The method of claim 1 , wherein the input device includes a barcode scanner, the barcode scanner having a read range overlapping with the field of view.
10. A computing device for tracking a target within a field of view, the computing device comprising:
a memory; and
a processor configured to:
receive, from an input device, a target identifier uniquely identifying the target;
receive, from an infrared camera, a sequence of images resulting from infrared radiation received from the target and comprising a first region of radiation received from the target at a first intensity and an adjacent region of radiation received from the field of view at a second intensity;
detect, in a reference image from the sequence of images, a mask defined by a border between the first region and the adjacent region;
store an association between the target identifier and the mask;
detect the mask in another image from the sequence of images; and
annotate the another image with the target identifier according to the stored association.
11. The computing device of claim 10 , wherein the infrared radiation received from the field of view at the second intensity is less than the first intensity.
12. The computing device of claim 10 , wherein the processor is further configured to:
store a plurality of mask attribute sets, and a respective article identifier for each mask attribute set; and
detect the mask by detecting one of the mask attribute sets.
13. The computing device of claim 12 , wherein the plurality of mask attribute sets includes a classification model; and
wherein the processor is configured to detect the mask by applying the classification model to the another image.
14. The computing device of claim 12 , wherein the processor is configured to detect the mask in the another image from the sequence of images by:
for each image in the sequence, determining whether the image contains any of the mask attribute sets; and
when the determination is affirmative, determining that the mask attribute set corresponds to the article identifier associated with the target identifier.
15. The computing device of claim 10 , wherein the processor is further configured to:
detect the target in a region of the another image; and
annotate the another image by storing the target identifier in association with the region of the another image.
16. The computing device of claim 10 , wherein the processor is further configured to:
receive a task completion indicator; and
discard the association between the target identifier and the mask.
17. The computing device of claim 16 , wherein the task completion indicator includes receiving the target identifier from the input device.
18. The computing device of claim 10 , wherein the input device includes an RFID reader, the RFID reader having a read range overlapping with the field of view.