This study introduces a novel population-based metaheuristic algorithm called secretary bird optimization algorithm (SBOA), inspired by the survival behavior of secretary birds in their natural environment. Survival for secretary birds involves continuous hunting for prey and evading pursuit from predators. This information is crucial for proposing a new metaheuristic algorithm that utilizes the survival abilities of secretary birds to address real-world optimization problems. The algorithm's exploration phase simulates secretary birds hunting snakes, while the exploitation phase models their escape from predators. During this phase, secretary birds observe the environment and choose the most suitable way to reach a secure refuge. These two phases are iteratively repeated, subject to termination criteria, to find the optimal solution to the optimization problem. To validate the performance of SBOA, experiments were conducted to assess convergence speed, convergence behavior, and other relevant aspects. Furthermore, we compared SBOA with 15 advanced algorithms using the CEC-2017 and CEC-2022 benchmark suites. All test results consistently demonstrated the outstanding performance of SBOA in terms of solution quality, convergence speed, and stability. Lastly, SBOA was employed to tackle 12 constrained engineering design problems and perform three-dimensional path planning for Unmanned Aerial Vehicles. The results demonstrate that, compared to contrasted optimizers, the proposed SBOA can find better solutions at a faster pace, showcasing its significant potential in addressing real-world optimization problems.

Metaheuristic algorithms, widely applied across various domains due to their simplicity and strong optimization capabilities, play a crucial role in problem-solving. While the Aquila Optimizer is recognized for its effectiveness, it often exhibits slow convergence rates and susceptibility to local optima in certain scenarios. To address these concerns, this paper introduces an enhanced version, termed Tent-enhanced Aquila Optimizer (TEAO). TEAO incorporates the Tent chaotic map to initialize the Aquila population, promoting a more uniform distribution within the solution space. To balance exploration and exploitation, novel formulas are proposed, accelerating convergence while ensuring precision. The effectiveness of the TEAO algorithm is validated through a comprehensive comparison with 14 state-of-the-art algorithms using 23 classical benchmark test functions. Additionally, to assess the practical feasibility of the approach, TEAO is applied to six constrained engineering problems and benchmarked against the performance of the same 14 algorithms. All experimental results consistently demonstrate that TEAO outperforms other advanced algorithms in terms of solution quality and stability, establishing it as a more competitive choice for optimization tasks.

The accurate reconstruction of a defective part of the mandible is a time-consuming task in maxillofacial surgery. In order to design accurate 3D implants quickly, a method for generating a mandibular defect implant model based on deep learning was proposed. First, an algorithm for generating a defective mandible 3D model randomly from a complete mandible 3D model was proposed due to the insufficiency of 3D models. Then a mandible 3D model dataset that consists of defective mandible 3D models and a complete mandible 3D model was constructed. An improved 3D Unet network that combines residual structure and dilated convolution was designed to generate a repaired mandibular model automatically. Finally, a mandibular defect implant model was generated using the reconstruction–subtraction strategy and was validated on the constructed dataset. Compared with the other three networks (3D Unet, 3D RUnet, and 3D DUnet), the proposed method obtained the best results. The Dice, IoU, PPV, and Recall for mandible repair reached 0.9873, 0.9750, 0.9850, and 0.9897, respectively, while those for implants reached 0.8018, 0.6731, 0.7782, and 0.8330. Statistical analysis was carried out on the experimental results. Compared with other methods, the P value of the method proposed in this paper was less than 0.05 for most indicators, which is a significant improvement.