Thin sheet medium identification method and identification device

Abstract

The present invention relates to a thin sheet medium identification method and device, wherein the method includes: step 1, collecting image information; step 2, segmenting the image information of the object to be identified from the collected image information; step 3, reconstructing the characteristic area image data, and extract features from the reconstructed image data to form a feature vector; and step 4, send the feature vector to a classifier trained on standard data to obtain a recognition result. This method reconstructs the image according to the prior information of the feature area, improves the image quality, and helps to improve the accuracy of recognition. Since only the image area that needs to be extracted is reconstructed, the consumption of storage space is reduced, which saves a lot of storage space and time. Improve resource utilization, and effectively improve the accuracy of recognition, to achieve a better balance between resources and efficiency.

Description

Technical field [0001] The invention relates to a pattern recognition technology based on image information, in particular to a method and device for identifying sheet-like media based on image information. Background technique [0002] In the field of image-based pattern recognition, the first thing to do is to use the sensor to collect the recognized image. If the image resolution is insufficient, the accuracy of the recognition result will be greatly reduced, especially between the easily confused recognition objects, such as the character recognition. "O" and "D", etc., although computer technology is changing with each passing day, from cost considerations, high-end hardware platforms are not selected, but only to meet actual needs. In this case, in the actual production process, due to the hardware The differences between the two and the diversity of the recognition space will cause more confusable object recognition errors. [0003] There are generally two existing solution...

AI Technical Summary

Problems solved by technology

However, in the actual collected signals, the amount of useful signal data is much greater than the amount of noise data. At this time, to reconstruct all the collected signals, the amount of data processing is huge, and the processing speed of the algorithm is slow. Improve recognition accuracy, but not ideal in most cases

Images