Early and accurate detection of retinal pathologies is critical for preventing vision loss and enabling timely clinical intervention. Traditional computer vision techniques, such as thresholding, edge detection, morphological filtering, and Hough transforms, have long been used to extract features from retinal fundus images, yet their performance is often constrained by image variability and complex pathological presentations. This study presents a hybrid deep learning architecture that integrates Convolutional Neural Networks (CNNs) for image-based classification with Recurrent Neural Networks (RNNs), specifically Long Short-Term Memory (LSTM) units, to model geometric and anatomical features derived from classical methods. This architecture allows for the fusion of pixel-level deep features with clinically interpretable descriptors, including optic disc-fovea distance, lesion spatial distribution, and vessel curvature sequences. Comparative analysis demonstrates that the proposed hybrid model achieves superior diagnostic accuracy, reaching 97%, significantly outperforming both conventional image processing approaches and CNN-only baselines. The results indicate that incorporating structured domain knowledge into neural models improves both performance and interpretability, offering a robust framework for real-world retinal disease screening applications.
19 July 2024 in Optoelectronic Information-Power Technologies
Based on machine learning algorithms, a method for predicting risks in the European energy markets has been proposed. The work is aimed at developing intelligent risk management systems that utilize advanced artificial intelligence technologies for assessing and minimizing potential threats. Utilizing historical data and current market trends, a comprehensive approach to identifying price volatility and risk zones in the energy markets is presented. The study demonstrates how artificial intelligence can enhance the effectiveness of decisions made by managers in the energy markets and ensure more sustainable resource management in conditions of increasing uncertainty. The results show that the use of complex machine learning algorithms and data analysis can significantly improve the accuracy of risk prediction and contribute to the adoption of well-founded managerial decisions.
6 October 2025 in Engineering, Technology & Applied Science Research
Optimizing growth conditions for tomato crops is essential due to their sensitivity to environmental factors. Pre-sowing seed treatments, particularly irradiation with specific wavelengths, can significantly influence germination and subsequent plant development. This study investigates how five laser wavelengths (red 630 nm, green 530 nm, blue 470 nm, ultraviolet 390 nm, infrared 780 nm) and four exposure durations (15, 30, 45, 60 minutes), along with a control sample, affected two tomato cultivars (Moneymaker and Bull's Heart). A total of 42 treatment combinations were tested using tabular experimental data (numeric input features: cultivar, wavelength, exposure time), with growth outcomes, such as plant height and fruit yield, recorded. A feedforward neural network with two hidden layers was trained to predict the final height of the plant from the seed treatment parameters. The model achieved a strong predictive accuracy (R2≈0.92 and MSE≈9.5 cm2) using an 80:20 train-test data split. This study demonstrates that deep learning can effectively model plant growth responses to physical seed priming and can be used to optimize treatment protocols for improved agricultural outcomes.
19 November 2024 in Optoelectronic Information-Power Technologies
T The automatic selection method for rectangle bounding of image objrcts exposed on the film is described, to optimize the image's size and increase the decoding efficiency. The image processing algorithm has been developed to improve the efficiency of grouping individual elements into a single structure. The simulation was carried out to prove the effectiveness of the proposed method. The main tool that has been used for developing the application is the open-source library of functions and algorithms of computer vision, image processing, and numerical algorithms of general-purpose OpenCV.