This paper discusses the Viola-Jones algorithm for face detection and its implementation based on the STM32 microcontroller. The advantages of using embedded systems in implementing personal identification systems are given: low cost due to the reduction of the element base and low power consumption. The architecture of the hardware and software system for face detection based on a multi-core microcontroller is proposed. The following requirements are put forward for the implemented facial recognition system: processing frequency of not less than 1 frame per second, output in color format, display of faces in the form of rectangular frames on the frame, refusal to use external memory modules. Cascades and features used in the classical version of the Viola-Jones algorithm are described. MB-LBP is chosen as a feature due to the efficiency of calculation and storage within low-power embedded systems due to integer single-byte results. The structure of files of trained OpenCV classifiers is described and methods for their compression and conversion for use in 32-bit systems with limited RAM and the absence of a floating-point unit are proposed. A method for optimizing an integral image using overflow calculations is described. A multicriterial optimization problem for selecting optimal parameters of an integral image is formulated and solved using the gradient descent method. The application of SIMD instructions for parallelizing the calculation of an integral image on the STM32 is described. The results of measuring the operating time of the implemented system at different stages are presented, which confirm that the previously stated requirements are met.
Keywords: face detection, microcontroller, embedded systems, Viola-Jones algorithm, MB-LBP features, classifier optimization, integral image optimization, SIMD instructions
This paper analyzes the performance of solving the classification problem using various open-source artificial intelligence and machine learning libraries in the field of marketing and customer relationship management; based on the results of experiments, the best library is selected for the purpose of introducing artificial intelligence into domestic CRM systems based on numerical performance indicators.
Keywords: artificial intelligence, machine learning, big data, classification, marketing, customer relationship management, import substitution, open source
Is devoted to the actual problem of constructing classifiers objects given by a point in a multidimensional space of feature values. The principle of linear normal classification of objects in multi-dimensional space of attributes can be used to build a classifier in the case of many complex structures, in general, are inseparable one hyperplane. In such cases, proposed to use a set of hierarchically related normal separating hyperplanes, which is called the normal hierarchical classifier.
Keywords: recognition, classification, feature space, the geometric method
Is devoted to the actual problem of constructing classifiers objects given by a point in a multidimensional space of feature values. A version of the geometric separation of sets by hyperplanes normal to the center-distance data sets. This approach to separating planes reduces the computational operations performed. This author separability criterion allows a normal quite effective in terms of computational complexity the exact solution of the normal separation, which requires only a linear search of points separated sets. Proposed in the article the approach to classification of sets in the multidimensional space of values of their attributes can be used as a starting point for building effective in terms of computational complexity classification not only for normally separable sets, but also for more complex variations thereof. This is the most significant practical importance of materials submitted by the authors.
Keywords: recognition, classification, feature space, the geometric method