The article presents the development of an interactive system for recognizing and classifying human hand gestures based on machine learning technologies. A new approach to gesture representation is proposed, combining spatial and temporal characteristics of the location of key points of the hand, which ensures high accuracy, noise resistance, and adaptability of the system to various conditions of use.
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.
This study developed a real-time fish classification and counting system for six types of fish using the YOLOv5 machine learning model with high accuracy. The system achieved an F1-score of 0.87 and a precision confidence curve with an all-classes value of 1.00 at a confidence level of 0.920, demonstrating the model's reliability in object detection and classification. Real-time testing showed that the system could operate quickly and accurately under various environmental conditions with an average inference speed of 30 FPS. However, several challenges remain, such
In this paper, we address the challenge of visual geo-localization from low-quality UAV imagery captured in real world environments. We propose a two-stage architecture, which includes Super-Resolution and visual geo-localization. We introduced novel, non-learnable Ensemble Super-Resolution (ESR) module, which first refines upscaled aerial frames, then seamlessly feeds the enhanced imagery into a visual geo-localization pipeline.
This paper is dedicated to developing and evaluating the facial recognition system, focusing on its effectiveness and operational reliability under real-world conditions. The choice of the Raspberry Pi hardware platform for implementing the system has been justified by its capability to process video streams in real time, as well as its compatibility with the high-quality Raspberry Pi Camera V2, which enables the acquisition of images with sufficient resolution for the proper functioning of computer vision algorithms.
This paper presents a study on organizing requirements for automated meat quality control systems. It identifies key quality indicators–color, texture, marbling, and gloss–and analyzes the technical and functional parameters essential for practical assessment. The research highlights integrating computer vision, image processing, and machine learning algorithms to enhance objectivity, accuracy, and evaluation speed. The proposed approach aims to reduce human influence, enable real-time monitoring, and offer scalable solutions suitable for large-scale producers and small enterprises.
The article examines data augmentation methods in the task of image recognition, specifically introducing the exponential augmentation approach to enhance the performance of deep neural networks, particularly YOLO, in object detection tasks. The proposed methodology is based on the sequential and repeated application of various transformations, including horizontal and vertical flipping, 90° rotation, Gaussian Blur, brightness and contrast adjustment.
Object recognition systems often struggle to maintain accuracy in dynamic environments due to challenges such as lighting variations, occlusions, and limited training data. Traditional convolutional neural networks (CNNs) require extensive labeled datasets and lack adaptability when exposed to new conditions. This study aims to develop an adaptive object recognition framework that enhances model generalization and rapid adaptation in changing environments.
The article presents the software development for modeling and simulating the workspace of a collaborative robot taking into account the presence of people. This is an important step in creating safe and efficient robotic systems within Industry 5.0 concept. The problem is posed by the need to ensure safety during the interaction of the robot with the operator, which is relevant for modern production processes with high human participation.
The increasing focus on smart agriculture in the last decade can be attributed to various factors, including the adverse effects of climate change, frequent extreme weather events, increasing population, the necessity for food security, and the scarcity of natural resources. The government of Morocco adopts preventative measures to combat plant illnesses, specifically focusing on tomatoes. Tomatoes are widely acknowledged as one of the most important vegetable crops, but they are highly vulnerable to several diseases that significantly decrease their productivity. De