FRC Kickoff - Release Candidate 1

Our first release candidate for the FRC Kickoff on Saturday.
Please report any bugs/crashes/defects you find. These are incredibly important and will help us make GRIP stable for the 2016 season.

---

Changes

• Publish values to NetworkTables
  • Will not work on NI roboRIO until you have the 2016 NI image
  • Should work on your local machine, please test this out with outline viewer.
• Deploy GRIP to NI roboRIO with Stop/Start functionality
  • You should also get all console logs printed back to GRIP for debugging
• Saves will now automatically append the .grip extension
• IP Camera protocol will automatically be added if you don't provide it when creating a IP Camera source
• Cleaner Socket Hint Labels
• Various bug fixes

FRC Beta 3

The third preview release of GRIP for FRC beta teams.
Expect more updates later this week!
We really need your feedback at this point. Please report bugs that you encounter to us.
This will be released to all teams very soon and we want a solid start.
We will be fixing bugs for the entirety of the next week.

---

Changes

• Logging to file (useful in the case of a crash)
• View problems with steps using the new exception view button

FRC Beta 2 - Patch 1

Changes

Fixes

• Divide by zero error from camera sources

Features

• Previews now follow the steps when moved in the pipeline.

---

Please post defects, suggestions, and other discussions to the issue tracker.

FRC Beta 2

The second preview release of GRIP for FRC beta teams.

---

Changes

• Saving and opening projects
• Finding contours
• Webcam support for Windows (#142)
• Multi Image Sources
  • When opening images, shift click to select multiple files. You can switch between these images to test your algorithm on different samples.
• Fuzzy operation searching
• More in-application operation documentation
• Support for starting and pausing webcams
• Many stability fixes and user interface enhancements

Please post defects, suggestions, and other discussions to the issue tracker or to the FRC 2016 Beta TeamForge! We need feedback on what features to add and what bugs to fix.

FRC Beta

The first preview release of GRIP for FRC beta teams.

---

GRIP is a tool for developing computer vision algorithms interactively rather than through trial and error coding. With GRIP, you choose vision operations to create a graphical pipeline. You can use GRIP with a webcam, an IP camera like the Axis camera, or sample images, so you don't have to rely on having physical access to a robot to test your algorithm.

GRIP is based on OpenCV, one of the most popular computer vision software libraries used for research, robotics, and vision algorithm implementations. The low-level operations that are available in GRIP are almost a one-to-one match with the operations available in OpenCV, but with instant feedback on what they do. In addition, some high-level operations are provided for easy learning and convenience. Let us know in the FRC TeamForge page if there are more features you'd like to see.

Tested operating systems:

• Windows 7
• Windows 10
• Fedora 23
• Ubuntu 15.04
• Mac OS X

Documentation

See the the wiki for documentation (in progress).

There is also a quick getting started guide on the FRC ScreenSteps.

Here's an incomplete list of things you can do in GRIP right now:

• Add sources from images on your computer, webcams, and IP cameras
• Add OpenCV and other high-level computer vision operations
• Connect outputs and inputs of operations together to form an image processing pipeline
• Manipulate the parameters to any step using graphical controls such as sliders and checkboxes
• Preview any output at any point in the pipeline

This isn't quite enough to use for a competition yet, but it should be functional enough to get a feel for how GRIP works.

Notes

This is an early release, and some features are currently missing. So you don't waste your time trying to find a feature that we haven't actually implemented yet, here are some of the major ones that haven't been finished.

• You cannot save or load a project
• Your cannot publish results to NetworkTables (or any other protocol)
• Many higher-level feature detection functions don't exist

Please post defects, suggestions, and other discussions to the issue tracker or to the FRC 2016 Beta TeamForge! We need feedback on what features to add and what bugs to fix.