This module consists of three sub-modules, each performing different sensory tasks.
Optical character recognition (OCR): the main functionality of this sub-module is to recognize human-readable characters from images.
Object detection by chromatic discrimination: the main functionality of this sub-module is to detect objects on an image based on their color.
Optical line following: the main functionality of this sub-module is to implement movement algorithms along optically detectable tracks on the ground.
In case these sub-modules are in operation, i.e., the robot control software for the use case demonstration is operated by the end-user, then the end-user cannot perform any modification on these sub-modules because these sub-modules are parts of the robot control software. In case these sub-modules are included in another robot control software (developed by third-party SME), then a software developer can perform automated text recognition tasks, detect objects on an image based on their color and implement movement algorithms along optically detectable tracks on the ground, respectively.
The module is already available in source code and as a part of a standalone desktop application by contacting the authors of this description.
The module as a source code is available both for internal and external use.
The typical application scenarios for this module are the following:
• Automated text recognition,
• object detection,
• optical line following
This module is created entirely with LabVIEW™ software.
OCR is the process where the machine vision software recognizes text or characters in an image.
This sub-module is based on the OCR template created with LabVIEW™ software and modified to recognize only the text appropriate for the task at hand. The sub-module uses a so-called Character Set File, in which the character templates are stored.
The Object detection by chromatic discrimination sub-module doesn’t require any hardware.
This sub-module uses the HSL color space and calculates the position (X and Y coordinates) of the center of mass of the group of pixels represented with the desired color on a camera image.
The Optical line following sub-module requires the Festo Robotino® v2 mobile robot hardware equipped with 2 optical proximity switches.
The optical proximity switches should be mounted on the base plate of the Robotino®, facing downwards with relative distance between them matching the width of the track.
The sensors should be connected to the DI0 and DI1 inputs of the input and output port of the Robotino®. The sub-module can operate on a bright floor with dark track and on a dark floor with a bright track also.
Inputs and outputs
• Inputs: grayscale image in JPG format, Character Set File, text to be recognized.
• Outputs: Recognized text in ASCII string format.
Object detection by chromatic discrimination
• Inputs: image, HSL parameters of the desired color, minimum number of pixels.
• Outputs: the desired color is found, the position of the center of mass, segmented image highlighting the group of pixels represented with the desired color.
Optical line following
• Inputs: State of the optical proximity switches.
• Output: movement speeds for the Robotino®.
Formats and standards
The JPEG image format, ASCII string format, HSL color space representation is used in this module.
Budapest University of Technology and Economics
To learn more about the solution, click on the link below to access the training on the Moodle platform