<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>OpenCV on David Hamp-Gonsalves</title><link>https://davidhampgonsalves.com/tags/opencv/</link><description>Recent content in OpenCV on David Hamp-Gonsalves</description><generator>Hugo -- gohugo.io</generator><language>en-US</language><lastBuildDate>Sun, 02 Aug 2015 00:00:00 +0000</lastBuildDate><atom:link href="https://davidhampgonsalves.com/tags/opencv/index.xml" rel="self" type="application/rss+xml"/><item><title>Failed Projects: Open Mosaic</title><link>https://davidhampgonsalves.com/failed-projects-open-mosaic/</link><pubDate>Sun, 02 Aug 2015 00:00:00 +0000</pubDate><guid>https://davidhampgonsalves.com/failed-projects-open-mosaic/</guid><description>The goal of this project was to recreate an image as a mosaic of other images. I wanted to achieve this with a low number of tiles and not simply use images as pixels based on their colors.
Mistakes The biggest mistake I made was to vastly underestimate the difficulty of the problem. I thought I could get good results by tuning my early histogram grid implementation and floundered. I have seen many other attempts at this project but yet to see any that produce good results.</description></item><item><title>OpenCV CompareHist Performance</title><link>https://davidhampgonsalves.com/opencv-comparehist-performance/</link><pubDate>Sat, 31 Jan 2015 00:00:00 +0000</pubDate><guid>https://davidhampgonsalves.com/opencv-comparehist-performance/</guid><description>I did some performance tests on the four histograms comparison methods that [compareHist](http://docs.opencv.org/java/org/opencv/imgproc/Imgproc.html#compareHist(org.opencv.core.Mat, org.opencv.core.Mat, int)) provides and initially found that in basic performance tests they were all about the same speed. More recently I profiled them in a real world application and found they were significantly different.
This is all using the Java OpenCV bindings.</description></item><item><title>OpenCV/Python Color Tracking</title><link>https://davidhampgonsalves.com/opencv/python-color-tracking/</link><pubDate>Mon, 30 May 2011 00:00:00 +0000</pubDate><guid>https://davidhampgonsalves.com/opencv/python-color-tracking/</guid><description>Preamble I needed some color based object tracking for a project I was hacking together last weekend. I choose to use the OpenCV Python bindings since I knew that I wouldn&amp;rsquo;t be doing anything fancy and I thought it would simplify the learning process. While the Python bindings are great I wasn&amp;rsquo;t able to find much documentation and what I thought would be an easy 10 minutes turned into a bit of an afternoon project.</description></item></channel></rss>