According to one example for segmenting image data, image data comprising color pixel data, IR data, and depth data is received from a sensor. The image data is segmented into a first list of objects based on at least one computed feature of the image data. At least one object type is determined for
According to one example for segmenting image data, image data comprising color pixel data, IR data, and depth data is received from a sensor. The image data is segmented into a first list of objects based on at least one computed feature of the image data. At least one object type is determined for at least one object in the first list of objects. The segmentation of the first list of objects is refined into a second list of objects based on the at least one object type. In an example, the second list of objects is output.
대표청구항▼
1. A method of segmenting image data, comprising: receiving image data from a sensor comprising color pixel data, infrared data, and depth data, wherein the image data is associated with an image of a plurality of different objects placed on a mat of a device that are within a field of view of the s
1. A method of segmenting image data, comprising: receiving image data from a sensor comprising color pixel data, infrared data, and depth data, wherein the image data is associated with an image of a plurality of different objects placed on a mat of a device that are within a field of view of the sensor;segmenting the image data into a first list of objects based on at least one computed feature of the image data, wherein the first list of objects includes each one of the plurality of different objects;determining at least one object type for at least one object in the first list of objects;refining the first list of objects that is segmented into a second list of objects based on the at least one object type; andoutputting the second list of objects. 2. The method according to claim 1, wherein the determining at least one object type for at least one object in the first list of objects comprises accessing the depth data received from the sensor and comparing the depth data to the first list of objects. 3. The method according to claim 1, wherein the segmenting the image data into a first list of objects based on the at least one computed feature of the image data comprises computing gradient data in the image data. 4. The method according to claim 1, wherein the segmenting the image data into a first list of objects based on the at least one computed feature of the image data comprises applying an edge-detection algorithm to the image data. 5. The method according to claim 1, wherein the segmenting the image data into a first list of objects based on the at least one computed feature of the image data comprises analyzing an intensity of at least one pixel value in the image data. 6. The method according to claim 1, wherein the segmenting the image data into a first list of objects based on the at least one computed feature of the image data comprises converting a colorspace of at least one pixel value in the image data. 7. The method according to claim 1, wherein the segmenting the image data into a first list of objects based on the at least one computed feature of the image data comprises detecting a texture of at least one pixel and at least one adjacent pixel in the image data to create a texture map. 8. The method according to claim 1, wherein the segmenting the image data into a first list of objects based on the at least one computed feature of the image data comprises detecting a region of similar pixels in the image data. 9. The method according to claim 1, wherein the refining the segmentation of the first list of objects into a second list of objects based on the at least one object type comprises applying one or more of a text sharpening filter, a contrast adjustment, a compensation, and a graph cut segmentation. 10. A system for capturing and segmenting image data, comprising: a support structure including a base;a sensor attachable to the support structure, wherein the sensor an image of a plurality of different objects placed on a mat of the system that are within a field of view of the sensor; anda computer communicatively coupled to the sensor;wherein the computer is to receive image data associated with the image comprising at least RGB data and depth data from the sensor, segment the image data into a first list of objects based on at least one computed feature the image data, wherein the first list of objects includes each one of the plurality of different objects, and refine the first list of objects that is segmented into a second list of objects based on at least one object type; andwherein the computer is to output the second list of objects. 11. The system according to claim 10, wherein the at least one computed feature of the image data comprises one or more of a computed gradient, edge, intensity, color, texture, and region in the image data. 12. The system according to claim 10, wherein the computer is to compute gradient data for two or more RGB channels in the image data and combine the channels into a single gradient map. 13. The system according to claim 10, wherein the image data further comprises data relating to a known background. 14. A non-transitory computer readable storage medium on which is embedded a computer program, said computer program to segment three-dimensional image data, said computer program comprising a set of instructions to: receive three-dimensional image data from a sensor, wherein the three-dimensional image data is associated with an image of a plurality of different objects placed on a mat of a device that are within a field of view of the sensor;segment the image data into a first list of objects based on one or more of a gradient, an edge, an intensity, a color, a texture, and a region of the image data, wherein the first list of objects includes each one of the plurality of different objects;determine at least one object type for at least one object in the first list of objects;refine the first list of objects that is segmented into a second list of objects based on the at least one object type; andoutput the second list of objects. 15. The non-transitory computer readable storage medium according to claim 14, further comprising instructions to apply one or more of a text sharpening filter, a contrast adjustment, a compensation, and a graph cut segmentation to refine the segmentation of the first list of objects into a second list of objects.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (10)
Izadi, Shahram; Sellen, Abigail J; Banks, Richard M; Taylor, Stuart; Hodges, Stephen E; Butler, Alex, Archive for physical and digital objects.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.