A team of SUAI scientists has improved the image recognition method. The objective of the invention is to develop a method that would increase the contrast of the generated vectors of image features without increasing their bit depth.
SUAI scientists examined various images with predefined reference frames of images of equal size. For the reference signal, as well as for the recognized image, a feature vector was formed, after which the recognized image was identified by comparing its feature vector with the feature vectors of each of the reference images. The decision was made based on the results of calculating the difference of feature vectors.
The technical result is achieved by using feature vectors averaged over the scale coordinate of the amplitude coefficients of a three-dimensional matrix as elements. The matrix is obtained by sequentially correlating the values of the histograms of the brightness distribution over the gradations of the image levels with Haar wavelets. From a practical point of view, the results can be used in the development of scientific and methodological apparatus that allows monitoring the consequences of various man-made disasters, for example, oil product spills in the waters of the seas and oceans and for other types of remote sensing.
