Image Neural Network Classifier by Detail Attributes
S.V. Sai, Viacheslav Nikonov · Pattern Recognition and Image Analysis · 2023
Abstract The paper presents the features of the development of a neural network analyzer of image detail. The neural network analyzer is a binary classifier based on the algorithm of search and identification of small image structures with different levels of detail (FDL). The developed neural network has a small number of neurons, requires a small amount of data and can be applied in embedded devices. The neural network classifier is considered in the context of neural network applications for computation optimization. Examples of using the classifier in image analysis and digital processing tasks are given.