Computer vision is one of the most frontier technologies in computer science. It is used to build artificial systems to extract valuable information from images and has a broad range of applications in various areas such as agriculture, business, and healthcare. Convolutional neural networks represent the key algorithms in computer vision, and in recent years, they have attained notable advances in many real-world problems. The accuracy of the network for a particular task profoundly relies on the hyperparameters’ configuration. Obtaining the right set of hyperparameters is a time-consuming process and requires expertise. To approach this concern, we propose an automatic method for hyperparameters’ optimization and structure design by implementing enhanced metaheuristic algorithms. The aim of this paper is twofold. First, we propose enhanced versions of the tree growth and firefly algorithms that improve the original implementations. Second, we adopt the proposed enhanced algorithms for hyperparameters’ optimization. First, the modified metaheuristics are evaluated on standard unconstrained benchmark functions and compared to the original algorithms. Afterward, the improved algorithms are employed for the network design. The experiments are carried out on the famous image classification benchmark dataset, the MNIST dataset, and comparative analysis with other outstanding approaches that were tested on the same problem is conducted. The experimental results show that both proposed improved methods establish higher performance than the other existing techniques in terms of classification accuracy and the use of computational resources.

The cloud computing paradigm, as a novel computing resources delivery platform, has significantly impacted society with the concept of on-demand resource utilization through virtualization technology. Virtualization enables the usage of available physical resources in a way that multiple end-users can share the same underlying hardware infrastructure. In cloud computing, due to the expectations of clients, as well as on the providers side, many challenges exist. One of the most important nondeterministic polynomial time (NP) hard challenges in cloud computing is resource scheduling, due to its critical impact on the cloud system performance. Previously conducted research from this domain has shown that metaheuristics can substantially improve cloud system performance if they are used as scheduling algorithms. This paper introduces a hybridized whale optimization algorithm, that falls into the category of swarm intelligence metaheuristics, adapted for tackling the resource scheduling problem in cloud environments. To more precisely evaluate performance of the proposed approach, original whale optimization was also adapted for resource scheduling. Considering the two most important mechanisms of any swarm intelligence algorithm (exploitation and exploration), where the efficiency of a swarm algorithm depends heavily on their adjusted balance, the original whale optimization algorithm was enhanced by addressing its weaknesses of inappropriate exploitation–exploration trade-off adjustments and the premature convergence. The proposed hybrid algorithm was first tested on a standard set of bound-constrained benchmarks with the goal to more accurately evaluate its performance. After, simulations were performed using two different resource scheduling models in cloud computing with real, as well as with artificial data sets. Simulations were performed on the robust CloudSim platform. A hybrid whale optimization algorithm was compared with other state-of-the-art metaheurisitcs and heuristics, as well as with the original whale optimization for all conducted experiments. Achieved results in all simulations indicate that the proposed hybrid whale optimization algorithm, on average, outperforms the original version, as well as other heuristics and metaheuristics. By using the proposed algorithm, improvements in tackling the resource scheduling issue in cloud computing have been established, as well enhancements to the original whale optimization implementation.

Multilevel image thresholding is a very important image processing technique that is used as a basis for image segmentation and further higher level processing. However, the required computational time for exhaustive search grows exponentially with the number of desired thresholds. Swarm intelligence metaheuristics are well known as successful and efficient optimization methods for intractable problems. In this paper, we adjusted one of the latest swarm intelligence algorithms, the bat algorithm, for the multilevel image thresholding problem. The results of testing on standard benchmark images show that the bat algorithm is comparable with other state-of-the-art algorithms. We improved standard bat algorithm, where our modifications add some elements from the differential evolution and from the artificial bee colony algorithm. Our new proposed improved bat algorithm proved to be better than five other state-of-the-art algorithms, improving quality of results in all cases and significantly improving convergence speed.

Wireless sensor networks, as an emerging paradigm of networking and computing, have applications in diverse fields such as medicine, military, environmental control, climate forecasting, surveillance, etc. For successfully tackling the node localization problem, as one of the most significant challenges in this domain, many algorithms and metaheuristics have been proposed. By analyzing available modern literature sources, it can be seen that the swarm intelligence metaheuristics have obtained significant results in this domain. Research that is presented in this paper is aimed towards achieving further improvements in solving the wireless sensor networks localization problem by employing swarm intelligence. To accomplish this goal, we have improved basic versions of the tree growth algorithm and the elephant herding optimization swarm intelligence metaheuristics and applied them to solve the wireless sensor networks localization problem. In order to determine whether the improvements are accomplished, we have conducted empirical experiments on different sizes of sensor networks ranging from 25 to 150 target nodes, for which distance measurements are corrupted by Gaussian noise. Comparative analysis with other state-of-the-art swarm intelligence algorithms that have been already tested on the same problem instance, the butterfly optimization algorithm, the particle swarm optimization algorithm, and the firefly algorithm, is conducted. Simulation results indicate that our proposed algorithms can obtain more consistent and accurate locations of the unknown target nodes in wireless sensor networks topology than other approaches that have been proposed in the literature.