A high-precision method for predicting the linear dimensions of three-dimensional objects of arbitrary shape based on computed tomography data has been developed. The method is aimed at reducing measurement errors caused by partial-volume effects, shadow transitions, and arbitrary spatial orientation of the object.
This paper presents a hybrid approach to traffic sign recognition that combines classical preprocessing techniques (color segmentation, contour detection, Haar Cascade, and HOG) with a lightweight Convolutional Neural Network (CNN) for classification. The proposed method reduces the amount of processed image data by a factor of 10–20, as only preselected regions of interest are passed to the neural network.
Ensemble methods can be used for many tasks, some of the most popular being: classification, regression, and image segmentation. Image segmentation is a challenging task, where the use of ensemble machine learning methods provides an opportunity to improve the accuracy of neural network predictions.
Introducing neural network training process modification that uses combination of several datasets to optimize search of objects and obstacles using mobile robotic systems in a closed space. The study includes an analysis of papers and existing approaches aiming to solve the problem of object boundary detection and discovers the key features of several neural network architectures. During research, it was discovered that there is an insufficient amount of data about the effectiveness of using obstacle detection approaches by mobile robotics systems in a closed space.
It is acknowledged that each person's life, group of people and nation is formed depending on geographical, economic, political, cultural and religious conditions. Lifestyle is formed as a result of daily repetition and consists of the following factors: nutrition, exercise, the presence of bad habits, moral and spiritual development, and so on. In recent decades, lifestyle has been considered an integral part of well-being, leading to increased research. According to the scientist's study, more than half of health problems are related to diet.
This paper examines the main methods and principles of image formation, display of the sign language recognition algorithm using computer vision to improve communication between people with hearing and speech impairments. This algorithm allows to effectively recognize gestures and display information in the form of labels. A system that includes the main modules for implementing this algorithm has been designed.
We propose a computer vision system based on multi-processor architecture that uses multi-port memory with equal access to all special microprocessors.
We consider a microcontroller implementation of a histogram-based algorithm for finding object coordinates in monochrome video.
An efficient graph-based image segmentation algorithm (EGBIS) is considered. The aspects of an efficient algorithm implementation, in particular the use of “Disjoint sets union” data structure with its heuristics “path compression” and “union be rank”, are investigated. A suitability of the algorithm for the use in automated systems is reviewed. An applicability of cluster validity measures for image segmentation quality assessment is analyzed.