Device, system, and method for indexing digital image frames
Methods and apparatus are disclosed to index digital frames. An example method includes identifying channel types associated with a plurality of image frames, splitting each one of the plurality of image frames into a respective color channel based on the identified channel types, applying a local binary pattern to each of the respective color channels to generate a respective pattern number, generating a spatial representation of each respective pattern number to determine transition probabilities for each channel type, and identifying a degree of similarity between the plurality of image frames based on the transition probabilities.
1. A method to identify a degree of similarity between image frames, comprising:
identifying channel types associated with image frames;
splitting respective ones of the image frames into respective color channels based on the identified channel types;
applying a local binary pattern to pixel neighborhoods of the respective color channels to generate respective pattern numbers;
generating spatial representations of respective ones of the pattern numbers to determine transition probabilities for the channel types when transitioning between the pixel neighborhoods; and
recombining the color channels to identify a degree of similarity between the image frames based on the transition probabilities.
2. A method as defined in claim 1 , further comprising selecting an index frame from among the plurality of images frames.
3. A method as defined in claim 1 , wherein the pattern numbers for respective ones of the pixel neighborhoods correspond to relative values of pixels in the corresponding pixel neighborhood.
4. A method as defined in claim 1 , further comprising applying a uniform local binary pattern operator to the pattern numbers.
5. A method as defined in claim 1 , further comprising combining adjacent pixel neighborhoods to form a single pixel neighborhood.
6. A method as defined in claim 1 , wherein the spatial representation represents a co-occurrence matrix between respective pattern numbers.
7. A method as defined in claim 1 , wherein the transition probabilities represent a Markov transition matrix.
8. A tangible machine readable storage device or storage disk comprising instructions that, when executed, cause a machine to at least:
identify channel types associated with image frames;
split ones of the image frames into respective color channels based on the identified channel types;
apply a local binary pattern to pixel neighborhoods of the respective color channels to generate respective pattern numbers;
generate spatial representations of respective ones of the pattern numbers to determine transition probabilities for the channel types when transitioning between the pixel neighborhoods; and
recombine the color channels to identify a degree of similarity between the plurality of image frames based on the transition probabilities.
9. A machine readable storage device or storage disk as defined in claim 8 , wherein the instructions, when executed, cause the machine to select an index frame from among the plurality of images frames.
10. A machine readable storage device or storage disk as defined in claim 8 , wherein the pattern numbers for respective ones of the pixel neighborhoods correspond to relative values of pixels in the corresponding pixel neighborhood.
11. A machine readable storage device or storage disk as defined in claim 8 , wherein the instructions cause the machine to apply a uniform local binary pattern operator to the pattern numbers.
12. A machine readable storage device or storage disk as defined in claim 8 , wherein the instructions cause the machine to combine adjacent pixel neighborhoods to form a single pixel neighborhood.
13. A machine readable storage device or storage disk as defined in claim 8 , wherein the spatial representation represents a co-occurrence matrix between respective pattern numbers.
14. A machine readable storage device or storage disk as defined in claim 8 , wherein the transition probabilities represent a Markov transition matrix.
15. An apparatus comprising:
a memory comprising machine readable instructions; and
a processor to execute the instructions to:
identify channel types associated with image frames;
split ones of the image frames into respective color channels based on the identified channel types;
apply a local binary pattern to pixel neighborhoods of the respective color channels to generate respective pattern numbers;
generate spatial representations of respective ones of the pattern numbers to determine transition probabilities for the channel types when transitioning between the pixel neighborhoods; and
recombine the color channels to identify a degree of similarity between the image frames based on the transition probabilities.
16. An apparatus as defined in claim 15 , wherein the processor is to select an index frame from among the plurality of images frames.
17. An apparatus as defined in claim 15 , wherein the pattern numbers for respective one of the pixel neighborhoods correspond to relative values of pixels in the corresponding pixel neighborhood.
18. The apparatus as defined in claim 15 , wherein the processor is to apply a uniform local binary pattern operator to the pattern numbers.
19. An apparatus as defined in claim 15 , wherein the processor is to combine adjacent pixel neighborhoods to form a single pixel neighborhood.
20. An apparatus as defined in claim 15 , wherein the transition probabilities represent a Markov transition matrix.