The problem of automatic image recognition based on the minimum information discrimination principle is formulated and solved. Discrimination calculation in the Kullback–Leibler information metric based on colour histograms comparison is proposed. It’s combined with a method of directed enumeration of the set of alternatives as opposed to the method of complete enumeration of competing hypotheses. Results of an experimental study of the discrimination in the problem of face images recognition are presented. It is shown that the proposed algorithm is characterized by increased accuracy and reliability of automatic image recognition.
Diagnostics and measurement are necessary for the effective management decision making. The social and economic processes are hardly to measure due to its nature. This article explores the approaches to the diagnostics of self-organizing and self-developing systems on the enterprise level. While the authors take into account the world measurement practice of management systems. The relationships between elements of self-organizing and self-developing systems and their key characteristics were revealed in purpose of diagnostic. The paper presents four steps of the diagnostics of the self-organization and self-development maturity level: the preparation phase, express-analysis of external experts, enterprise self-diagnostics with the help of questionnaires, precise diagnostics by external experts. The data obtained from questionnaires are analyzed with the help of special program. The results of the diagnostics are pictured with the Radar method.
Studied is a possibility of increasing the accuracy of diagnostics by examining a number of diagnostic rules as a set of expert assessments, which allows one to combine them («mix of expert opinions»). Proposed is to use of the principle of minimum-information-mismatch in Kullback - Leibler metric to highlight the rule most appropriate for classification of a particular object. Program and results of experimental study are presented in the problem of automatic recognition of gray-scale images. It is shown that the developed approach can significantly improve the quality of diagnostics.