Evaluation form identification method based on image identification and K neighbor

Abstract

The present invention discloses an evaluation form identification method based on image identification and K neighbor. The method comprises the following steps: 1, determining an evaluation form; 2, printing the evaluation form and performing site evaluation; 3, scanning the evaluation form and performing storage of the evaluation form; 4, performing preprocessing of the images; 5, extracting bar codes and fill-out form area, and determining the concrete position of the bar codes and the fill-out form area and performing information extraction after the image inclination is correct; 6, identifying the bar codes; and 7, identifying the fill-out form. The present invention provides a random option arrangement and image identification examination evaluation form identification method. when the evaluation form is created, the examination object, the examination indexes and the examination context are subjected to random arrangement, the content between the evaluation forms is the same, and the display sequences of the evaluation forms are different to prevent adjacent participating persons from mutual influence; and the horizontal crossing features and the longitudinal crossing features of the symbol images are extracted, the K neighbor method is provided to perform symbol identification, and the identification accuracy is high.

Description

technical field [0001] The invention belongs to the field of image processing and data mining, in particular to an evaluation table recognition method based on image recognition and K-nearest neighbors. Background technique [0002] The Fourth Plenary Session of the 16th Central Committee of the Party put forward the requirement of "grasp the development of cadre performance evaluation standards that reflect the requirements of the scientific development concept and the correct performance concept". In order to do a good job in the evaluation and evaluation of leading cadres, each unit needs to set the evaluation content and indicators according to the actual situation, scientifically set the way and method of realization, and correctly use the evaluation results, so as to produce different types of evaluation forms and statistical analysis methods. Therefore, there is an urgent need for a complete and efficient assessment method to identify various assessment contents and c...

AI Technical Summary

Problems solved by technology

This method is efficient, but the main problem is that it is impossible to uniquely distinguish different participants and the anonymity of the participants; at the same time, due to the lack of ticket stubs, it is impossible to publicly check the tickets, its credibility is questioned, and its safety and reliability issues has not been well resolved

There are two types of identification of symbols in the form filling area. One is to only identify whether the content is filled in. This method directly calculates the proportion of black pixels in the form filling area. If it is higher than a certain threshold, the content is considered to be filled in, but the content cannot be recognized. The specific form of , such as √, × or ○, etc.

Images