Revision history

  • 09/06/2006
    1. Initialize button to set keywords and options in the Search MUTCD codebook window to its initial status.
    2. Now a user can move bewteen options in the Search MUTCD codebook using the TAB key to move down and Shift-TAB to move back.
    3. It is possible now to embed the database within the program. This means that we do not need to install MySQL database separately. I embedded the MySQL source codes into the MUTCD Traffic Sign Coder program. However, I will check more its stability and it is necessary to develop the data query interface. So it is not enabled yet.
  • 08/17/2006 The first release of MUTCD Traffic Sign Coder. Its interface is more enhanced and the keyboard shortcut is supported now. Screen shots are also updated. User traffic sign tagging database and its interface implemented. The image zoom in/out function is now embedded. A user can select a region using a number of region selection tools including rectangle, rounded rectangle, and polygon.
  • 08/14/2006 Preliminary MUTCD code database search interfaces are implemented. Please see below screen shots
  • 08/12/2006 Update the MUTCD code database completely manually (now 624 codes with 720 images). I also extracted the additional context of each traffic sign including its shape, background color, legend color, width, height, etc. Image files are directly stored into the database in various sizes and formats (available size: 200x200, 100x100, 50x50, formats: BMP and PNG). Please see Construct the traffic sign database for details.
  • 08/08/2006 Found the current MUTCD code database is missing many codes (currently 328 codes only)
  • 08/07/2006 Begin the preliminary user interface design: File folder, image file filter, image view display algorithms are implemented.
  • 08/06/2006 Order CodeJock Xtreme Toolkit Pro library

Screen shots

Most recent screen shots

MUTCD Traffic Sign Coder (09/07/2006)

Screen shot history

Code database

Please see Construct the traffic sign database.

To-do list

  1. Import existing polygon detection algorithms.
  2. Implement the polygon detection result display algorithm.
  3. Mark the background of the sign region from the selected area.
  4. Compute the image features including its colors, shapes, and texts.
  5. Compute the best traffic sign candidates using ANN.
  6. Implement the user interface for the above all.

Things to discuss

  1. We may import all image files into the database and process all images within the database. No more tons of separate files. Is this necessary?

Download

Please see Download page.