Dot pattern feature extraction, selection and matching using LBP, Genetic Algorithm and Euclidean distance

Analysis of Dots in a pattern is useful for many patterns and image analysis problems. This paper states some of the methods like Gabor Wavelet, Fourier descriptor, Local binary pattern that can be used for extracting the features from a Dot pattern image. The Dot pattern image is first pre-processed (Re-constructed, Rotated, Enhanced). The paper then analyzes the dot pattern image by finding the irregularities (missing pattern) in the regular dot pattern image. local binary pattern (LBP) is then applied for extracting the Dot pattern image features. Genetic Algorithm is stated for retaining only the more discriminated features by discarding the less discriminated features. The optimized features thus obtained can be used for matching the two dot patterns for similarity using Euclidean Distance.