Multi-Strategy Enhanced Harris Hawks Optimization for Global Optimization and Deep Learning-Based Channel Estimation Problems

Mathematics, Год журнала: 2023, Номер 11(2), С. 390 - 390

Опубликована: Янв. 11, 2023

Harris Hawks Optimization (HHO) simulates the cooperative hunting behavior of hawks and it has advantages fewer control parameters, simple principles, excellent exploitation ability. However, HHO also disadvantages slow convergence easy falling into local optimality. Aiming at above shortcomings, this paper proposes a Multi-strategy Enhanced (MEHHO). Firstly, map-compass operator Cauchy mutation strategy are used to increase population diversity improve ability algorithm jump out optimal. Secondly, spiral motion is introduced exploration phase enhance search efficiency. Finally, speed accuracy improved by greedy selection fully retain dominant individuals. The global capability proposed MEHHO verified 28 benchmark test functions, then parameters deep learning network for channel estimation optimized using verify practicability MEHHO. Experimental results show that more in solving optimization problems improving method based on learning.

Язык: Английский

---

Automatic diagnostic system for segmentation of 3D/2D brain MRI images based on a hardware architecture

Microprocessors and Microsystems, Год журнала: 2023, Номер 98, С. 104814 - 104814

Опубликована: Фев. 28, 2023

Язык: Английский

---

Novel Air2water Model Variant for Lake Surface Temperature Modeling With Detailed Analysis of Calibration Methods

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Год журнала: 2022, Номер 16, С. 553 - 569

Опубликована: Дек. 5, 2022

The air2water model is a simple and efficient tool for modeling surface water temperature in lakes based solely on the air temperature. In this article, we propose to modify such way that different parameters would be associated with lake stratification of cold waters than warm waters. situation mix both also considered. tested 22 lowland Polish against two classical variants. As new variant slightly more complicated basic versions, focus importance choice calibration method. Each calibrated eight optimization methods, which are compared various benchmark problems. We show proposed superior models about 90% lakes, but only if approach properly selected, confirms links between appropriate procedures. performs well exception large shallow ones, probably due weak lakes.

Язык: Английский

---

Giza pyramids construction algorithm with gradient contour approach for multilevel thresholding color image segmentation

Applied Intelligence, Год журнала: 2023, Номер 53(18), С. 21248 - 21267

Опубликована: Май 25, 2023

Язык: Английский

---

Improved Northern Goshawk Optimization Algorithm for Medical Image Segmentation

Lecture notes in electrical engineering, Год журнала: 2024, Номер unknown, С. 344 - 354

Опубликована: Янв. 1, 2024

Язык: Английский

---