Computer Vision Sandbox 1.2.7 release
December 23, 2017
As the end of the year approaches, it is time to summarize the ongoing development and announce the new 1.2.7 release of the Computer Vision Sandbox. The new version is mostly about new features, some of which are quite interesting and may serve well for many different applications.
The first feature to mention is detection of square binary glyphs. Being simple shapes with well-defined structure, those are easy to locate in images and can be used in a number of applications. For example, glyphs can be used to give commands to a robot or to provide directions for navigation in some environment. Another common application is augmented reality, where glyph from the source image can be replaced with something artificial. We are not quite there to do 3D augmented reality yet, but a simple 2D can be done.
Another feature to highlight is detection of 1D bar codes. This one represents only the first step of some bigger future development, where the detection part is planned to be completed with recognition part, so the encoded string could be extracted from the found bar code.
The blobs processing module also gets extended and now provides plug-ins to detect circles and quadrilaterals in images. This can be really useful in applications where only blobs of certain shapes must be processed. And of course, these plug-ins report size/position of detected objects, so some custom analysis can be implemented using scripting plug-in.
On top of the above, some more useful plug-ins are provided as well as bug fixes being implemented. Overall, it looks like a really nice new version to try out.