This article explores the creation of recognition algorithms (RAs) utilizing one-dimensional threshold rules (OTRs) to address the challenge of classifying objects within the complex high-dimensional feature space (HDFS). A novel method is introduced, centered around the establishment of a collection of basic objects (BOs) and the subsequent development of an OTR for each. What sets this approach apart is the construction of OTRs derived from each chosen base object. The proposed RA is defined parametrically through a series of sequential steps designed to address specific subtasks. These key procedures include: determining the difference function (DF) in the subspace of representative features (SRF); identification of subsets of interrelated objects (SIOs); formation of a set of BOs; construction of tabular proximity functions (TPFs); calculating the proximity score between (PSB) the base and simple objects; calculating the PSB a class and an object. To evaluate how well the suggested RAs model works, we conducted experiments to address object classification, a model problem. The results of comparing the suggested RAs with existing ones are included.