A method for evaluating the clarity of certificate pictures

Abstract

The invention discloses a method for evaluating the clarity of a license picture, comprising the steps of: obtaining an original license picture; performing bilateral filtering on the original license picture to obtain a grayscale image of the license picture; performing gradient-based edge detection on the grayscale image to obtain a gradient Fig. Obtain the hidden gradient function according to the gray level change matrix around the measurement pixel, and analyze the continuous change of the gray level around each pixel in the gradient map based on the hidden gradient function; take the global average value of the analysis results as the evaluation value of the picture clarity; Standardize the evaluation value to obtain the picture clarity score. The present invention can effectively evaluate the clarity of the text area of ​​the license picture; it can directly and effectively judge the clarity of the license picture without extracting the text information from the fuzzy license picture, and it greatly improves the whole system especially in the stream processing environment. The efficiency; not only the evaluation effect is good, but also the calculation amount is low, saving a lot of time and operating costs.

Description

technical field [0001] The invention belongs to the technical field of picture processing, in particular to a method for evaluating the clarity of a license picture. Background technique [0002] The catering stores on the online catering platform not only have physical stores, but also franchise stores of online food delivery platforms such as Ele.me and Meituan, which are engaged in food delivery business. According to the relevant provisions of the State Administration for Industry and Commerce's "Online Transaction Management Measures" and the State Food and Drug Administration's "Online Catering Service Food Safety Supervision and Administration Measures", online catering service providers should publicize their business licenses, food Business license and other information. [0003] One of the supervisions for online platform merchants is whether the uploading of certificates is compliant, and there are potential risks for merchants who upload vague certificates. The...

AI Technical Summary

Problems solved by technology

The license picture contains a lot of text areas, and the existing methods cannot effectively evaluate the clarity of the text areas in the license picture, and the evaluation effect is poor

Putting all the license pictures into the text extraction system first and then judging the clarity will greatly increase the amount of calculation, and the time cost is high, so we need to judge the clarity of the license pictures before extracting the text information of the license pictures

Images