<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Data Augmentation of Railway Images for Track Inspection · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1802.01286/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/1802.01286/feed.xml" rel="self" type="application/rss+xml" /><item><title>Data Augmentation of Railway Images for Track Inspection — mentioned by davidthewatson</title><link>https://news.ycombinator.com/item?id=32121034</link><guid isPermaLink="false">urn:hackerpapers:mention:1802.01286:32121034</guid><pubDate>Sat, 16 Jul 2022 19:17:15 GMT</pubDate><description>&lt;p&gt;Computer vision can be used to classify sun kinks at speed from a train-mounted sensor. The GPS-triangulated sun kink data can then be used to dispatch repair teams to the appropriate location preventing derailments in the process. This has been studied academically and commercially in the US for a few years, at least. I&amp;#x27;m not aware of a commercial deployment, but it may be in progress given the published R&amp;amp;D. For instance: https://arxiv.org/pdf/1802.01286.pdf From the paper: &amp;quot;Sun kink classifier could classify professionally simulated sun kink videos with a precision of 97.5%.&amp;quot;&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1802.01286/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item></channel></rss>
