Intelligent evaluation method and system for creation process of art works of primary and secondary school students

CN122929544APending Publication Date: 2026-10-16NANJING UNIV OF INFORMATION SCI & TECH
0 Cites 0 Cited by

Patent Information

Application Number
CN202611081327.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-21
Publication Date
2026-10-16

AI Technical Summary

Technical Problem

现有评价手段通常仅能记录绘画笔迹的坐标序列或整体耗时，难以分辨画面中出现的反复涂抹、颜色覆盖与修正操作究竟是学生自主摸索的真实行为痕迹，还是为掩饰非独立创作而进行的程序化模仿

Benefits of technology

[0018] 1. In this invention, by dividing the canvas image into spatial grids and extracting the color distribution feature vector sequence, a dynamic visual state transition chain is constructed, realizing the refined restoration and quantitative recording of the entire art creation process. This solves the problem that traditional evaluation only focuses on the result and ignores the conceptual framework. Furthermore, by identifying visual state reproduction events and associating them with brushstroke behavior units, the distribution density and frequency of exploration and correction units are calculated. This can distinguish between the genuine traces of students' independent conception and non-independent creative behaviors such as ghostwriting and mechanical copying, thereby improving the authenticity of process evaluation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application relates to the field of computer vision and discloses an intelligent evaluation method and system for the creation process of art works of primary and secondary school students, which comprises the following steps: obtaining canvas image data, pixel block areas corresponding to the canvas image data and color distribution feature vectors, generating visual state nodes and transition chains by analyzing the pixel block areas, clustering nodes with time continuity and spatial adjacency into stroke behavior units, identifying the similarity between a current node and a previous node by traversing the transition chains, marking visual state reproduction events and associating the visual state reproduction events with the stroke behavior units, classifying the stroke behavior units into single fixed units or exploration and correction units according to the number of nodes and the reproduction event conditions, counting the number and distribution of the stroke behavior units, and outputting a creation process authenticity score based on the distribution density of the exploration and correction units and the frequency of the reproduction events. The application realizes deep analysis of the creation behavior logic and improves the accuracy and reliability of the authenticity evaluation.
Need to check novelty before this filing date? Find Prior Art