Modification and Acceleration of the Detection Objects Algorithm by Interactively Set Color and Texture Attributes | IEEE Conference Publication | IEEE Xplore

Modification and Acceleration of the Detection Objects Algorithm by Interactively Set Color and Texture Attributes


Abstract:

We discuss a detection objects algorithm by the color and texture attributes to apply the recognition of the road signs. The basis of this algorithm is the technique of f...Show More

Abstract:

We discuss a detection objects algorithm by the color and texture attributes to apply the recognition of the road signs. The basis of this algorithm is the technique of forming image features vectors, which contain all information about color and texture characteristics of the specially selected road signs. Feature vectors are used to train a neural network, which subsequently classifies the outlines of the recognizable signs after color processing of the video stream. The proposed technique has a high potential for application in tasks of locating robot in space, constructing a terrain map and remote control of the radio engineering systems. But for the correct application of the proposed solution, it is necessary to develop a convenient interface for training the neural network by user of the program. It is also necessary to pre-process each image of the video stream to equalize the brightness, and parallelize all hard calculations in the detector by color and texture features.
Date of Conference: 14-17 September 2018
Date Added to IEEE Xplore: 08 November 2018
ISBN Information:

ISSN Information:

Conference Location: Kazan, Russia

References

References is not available for this document.