Information processing apparatus and non-transitory computer readable medium
An information processing apparatus includes a processor configured to obtain, for each character of plural characters recognized from an image, (a) position of the character in the image, (b) size of the character, and (c) confidence level of a character recognition result of the character; and determine whether to regard the character as a noise based on a distance between the character and its nearest character, the size of the character, and the confidence level of the character recognition result of the character.
1. An information processing apparatus comprising:
a processor configured to
for each character of a plurality of characters recognized from an image,
obtain,
(a) position of the character in the image,
(b) size of the character, and
(c) confidence level of a character recognition result of the character, and
determine whether to regard the character as a noise based on a distance between the character and its nearest character, the size of the character, and the confidence level of the character recognition result of the character.
2. The information processing apparatus according to claim 1 ,
wherein the determination of whether to regard the character as a noise is based on whether a point representing the distance, the size, and the confidence level of the character in a three-dimensional space is in a noise region defined in the three-dimensional space.
3. The information processing apparatus according to claim 2 , wherein the processor is further configured to
receive sample images,
obtain, for each sample image character of a plurality of sample image characters recognized from the sample images, a distance between the sample image character and its nearest sample image character, a size of the sample image character, and a confidence level of a character recognition result of the sample image character, and
receive correct answer data that indicates whether to regard an element corresponding to the sample image character in the sample images as a noise or an actual character, and
wherein the noise region is defined based on the correct answer data and the distance, the size, and the confidence level of the sample image character.
4. The information processing apparatus according to claim 1 ,
wherein the determination is made by using a neural network that receives the distance, the size, and the confidence level of the character, and outputs information indicating whether to regard the character as a noise or an actual character.
5. The information processing apparatus according to claim 1 ,
wherein the character is regarded as a noise if a value of at least one of the distance, the size, or the confidence level of the character is outside a corresponding predetermined range.
6. The information processing apparatus according to claim 1 ,
wherein the character is regarded as a noise if a score calculated based on values of the distance, the size, and the confidence level of the character is within a predetermined range.
7. A non-transitory computer readable medium storing a program causing a computer to execute a process for image processing, the process comprising:
for each character of a plurality of characters recognized from an image,
obtaining,
(a) position of the character in the image,
(b) size of the character, and
(c) confidence level of a character recognition result of the character; and
determining whether to regard the character as a noise based on a distance between the character and its nearest character, the size of the character, and the confidence level of the character recognition result of the character.