<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ocr on The 3TC</title><link>https://the-3tc.github.io/tags/ocr/</link><description>Recent content in Ocr on The 3TC</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Mon, 15 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://the-3tc.github.io/tags/ocr/index.xml" rel="self" type="application/rss+xml"/><item><title>The Character Error Vector: Measuring global vs local text extraction quality</title><link>https://the-3tc.github.io/blog/cev-score/</link><pubDate>Mon, 15 Jun 2026 00:00:00 +0000</pubDate><guid>https://the-3tc.github.io/blog/cev-score/</guid><description>&lt;link rel="stylesheet" href="https://the-3tc.github.io/css/cev.css"&gt;
&lt;h2 id="traditional-measure-of-ocr-error"&gt;Traditional measure of OCR error&lt;/h2&gt;
&lt;p&gt;Along with the Edit Distance, the Character Error Rate (CER) is one of the most widely used metrics when measuring OCR quality. It is calculated by comparing the observed string to the ground truth string by counting the number of insertions, deletions, and
substitutions needed to turn the predicted text into the ground truth,
normalised by the number of characters:&lt;/p&gt;
&lt;p&gt;$$\text{CER} = \frac{\text{Substitutions} + \text{Deletions} + \text{Insertions}}{\text{Total characters}}$$&lt;/p&gt;</description></item></channel></rss>