Detection of Texture Objects on Multichannel Images

Abstract

A method for detecting extended texture objects on multi-channel images is proposed. The method is based on the representation of multi-digit digital images by a set of bit images and their approximation by three-dimensional Markov chains with two states. It is proposed to use transition probabilities for three-dimensional Markov chains and brightness characteristics averaged within a sliding window as textural features. To reduce computational resources, determination of signs is carried out by the bit planes of the senior, most informative digits of the digital image. The detection of texture objects is based on the analysis of the histogram of the texture feature and the brightness of the pixels. The simulation results confirm the effectiveness of the proposed method.

Keywords

• Markov multidimensional chains
• Detecting texture objects
• Image segmentation