Scholarlink

<p>Line charts are one of the most effective ways to track real-time changes, and are widely used in web monitoring systems. In this study, a new method using computer vision technology is presented to automatically detect anomalies in line charts. To achieve this, a multi-modal analysis approach is proposed to identify red color increases, inspect bad words in the text contained in the graphs, and examine shape changes in the graphs. This allows the system to determine whether the situation is anomalous. These processes run automatically, and can be applied to multiple monitoring systems via remote web access, enhancing control center efficiency. Early identification and management of potential risks further improve overall system performance and safety.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automatic line chart analysis for intelligent operation monitoring

  • 期刊论文

  • 发表日期：

    2026年5月20日

查看全文
  • Gyubaek Kim,
  • Sanghyun Park

摘要

Line charts are one of the most effective ways to track real-time changes, and are widely used in web monitoring systems. In this study, a new method using computer vision technology is presented to automatically detect anomalies in line charts. To achieve this, a multi-modal analysis approach is proposed to identify red color increases, inspect bad words in the text contained in the graphs, and examine shape changes in the graphs. This allows the system to determine whether the situation is anomalous. These processes run automatically, and can be applied to multiple monitoring systems via remote web access, enhancing control center efficiency. Early identification and management of potential risks further improve overall system performance and safety.