Method for identifying digits in blurred images based on blur classifier
A fuzzy image and fuzzy classification technology, applied in image data processing, character and pattern recognition, image enhancement, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0043] For the convenience of description, we assume the following application example: There are 500 fuzzy grayscale images each containing 1 number, from which 8 images with more positive and normal numbers are selected for each number as training samples, and the rest are used as test samples. A fuzzy classifier is constructed by using the training samples, and digit recognition is performed on the test samples.
[0044] The specific embodiment of the present invention is:
[0045] Step 1. Construct fuzzy classifier, comprise feature selection and determine the required parameter of classifier, concrete steps are as follows:
[0046] Step (1) Firstly select a number of relatively standardized grayscale images as training samples for numbers 0 to 9, and calculate the features of each known number image according to the following steps:
[0047] Step (11) Adaptive binarization of grayscale images: find the two grayscale values c1 and c2 with the highest frequency of occurr...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 
