Optical character recognition using specialized confidence functions
Systems and methods for optical character recognition using specialized confidence functions. An example method comprises: receiving a grapheme image; computing a feature vector representing the grapheme image in a space of image features; and computing a confidence vector associated with the grapheme image, wherein each element of the confidence vector reflects a distance, in the space of image features, between the feature vector and a center of a class of a set of classes.
1. A method, comprising:
receiving, by a computer system, a grapheme image;
computing a feature vector representing the grapheme image in a space of image features;
computing a confidence vector associated with the grapheme image, wherein each element of the confidence vector is produced by a monotonically decreasing function of a distance, in the space of image features, between the feature vector and a center of a class of a set of classes, wherein the monotonically decreasing function is based on a series of predefined confidence values, such that intervals between the predefined confidence values increase towards a lowest confidence value;
identifying an element having a maximum value among elements of the confidence vector; and
associating the grapheme image with a grapheme class corresponding to the identified element of the confidence vector.
2. The method of claim 1 , further comprising:
responsive to determining that the maximum value falls below a threshold, returning an error code indicating that the grapheme image is not recognizable.
3. The method of claim 1 , further comprising:
performing, in view of the confidence vector, segmentation of an original image to produce a second grapheme image; and
repeating, for the second grapheme image, operations of computing the feature vector and computing the confidence vector.
4. The method of claim 1 , wherein each class of the set of classes corresponds to a character of an alphabet.
5. The method of claim 1 , further comprising:
computing a set of class weights, such that each class weight characterizes a degree of association of the grapheme image with a certain class of a set of classes; and
computing, using a normalized exponential transformation, a set of probabilities corresponding to the set of class weights, such that each probability characterizes a hypothesis of the grapheme image representing an instance of a certain class of the set of classes.
6. The method of claim 5 , wherein the confidence vector is determined for a subset of classes associated with highest probability values.
7. The method of claim 1 , wherein computing the feature vector is performed by a neural network, the method further comprising:
training the neural network using a loss function represented by a linear combination of a cross entropy loss function and a center loss function.
8. A system, comprising:
a memory;
a processor, coupled to the memory, the processor configured to:
receive a grapheme image;
compute a feature vector representing the grapheme image in a space of image features;
compute a confidence vector associated with the grapheme image, wherein each element of the confidence vector is produced by a monotonically decreasing function of a distance, in the space of image features, between the feature vector and a center of a class of a set of classes, wherein the monotonically decreasing function is based on a series of predefined confidence values, such that intervals between the predefined confidence values increase towards a lowest confidence value;
identify an element having a maximum value among elements of the confidence vector; and
associate the grapheme image with a grapheme class corresponding to the identified element of the confidence vector.
9. The system of claim 8 , wherein each class of the set of classes corresponds to a character of an alphabet.
10. The system of claim 8 , wherein the processor is further configured to:
compute a set of class weights, such that each class weight characterizes a degree of association of the grapheme image with a certain class of a set of classes; and
compute a set of probabilities corresponding to the set of class weights, such that each probability characterizes a hypothesis of the grapheme image representing an instance of a certain class of the set of classes.
11. The system of claim 10 , wherein the confidence vector is determined for a subset of classes associated with highest probability values.
12. The system of claim 8 , wherein the processor is further configured to:
perform, in view of the confidence vector, segmentation of an original image to produce a second grapheme image; and
repeat, for the second grapheme image, operations of computing the feature vector and computing the confidence vector.
13. The system of claim 8 , wherein the processor is further configured to:
responsive to determining that the maximum value falls below a threshold, return an error code indicating that the grapheme image is not recognizable.
14. A computer-readable non-transitory storage medium comprising executable instructions that, when executed by a computer system, cause the computer system to:
receive a grapheme image;
compute a feature vector representing the grapheme image in a space of image features;
compute a confidence vector associated with the grapheme image, wherein each element of the confidence vector is produced by a monotonically decreasing function of a distance, in the space of image features, between the feature vector and a center of a class of a set of classes, wherein the monotonically decreasing function is based on a series of predefined confidence values, such that intervals between the predefined confidence values increase towards a lowest confidence value;
identify an element having a maximum value among elements of the confidence vector; and
associate the grapheme image with a grapheme class corresponding to the identified element of the confidence vector.
15. The computer-readable non-transitory storage medium of claim 14 , further comprising executable instructions causing the computer system to:
perform, in view of the confidence vector, segmentation of an original image to produce a second grapheme image; and
repeat, for the second grapheme image, operations of computing the feature vector and computing the confidence vector.
16. The computer-readable non-transitory storage medium of claim 14 , wherein each class of the set of classes corresponds to a character of an alphabet.
17. The computer-readable non-transitory storage medium of claim 14 , further comprising executable instructions causing the computer system to:
responsive to determining that the maximum value falls below a threshold, return an error code indicating that the grapheme image is not recognizable.
18. The computer-readable non-transitory storage medium of claim 14 , further comprising executable instructions causing the computer system to:
computing a set of class weights, such that each class weight characterizes a degree of association of the grapheme image with a certain class of a set of classes; and
computing, using a normalized exponential transformation, a set of probabilities corresponding to the set of class weights, such that each probability characterizes a hypothesis of the grapheme image representing an instance of a certain class of the set of classes.
19. The computer-readable non-transitory storage medium of claim 14 , wherein the confidence vector is determined for a subset of classes associated with highest probability values.
20. The computer-readable non-transitory storage medium of claim 14 , wherein computing the feature vector is performed by a neural network that is trained using a loss function represented by a linear combination of a cross entropy loss function and a center loss function.