Camouflaged Object Detection (COD) aims to detect objects hidden in complex environments. The existing COD algorithms ignore the influence of feature expression and fusion methods on detection performance when combining multi-level features. Therefore, a COD algorithm based on progressive feature enhancement aggregation was proposed. Firstly, multi-level features were extracted through the backbone network. Then, in order to improve the expression ability of features, an enhancement network composed of Feature Enhancement Module (FEM) was used to enhance the multi-level features. Finally, Adjacency Aggregation Module (AAM) was designed in the aggregation network to achieve information fusion between adjacent features to highlight the features of the camouflaged object area, and a new Progressive Aggregation Strategy (PAS) was proposed to aggregate adjacent features in a progressive way to achieve effective multi-level feature fusion while suppressing noise. Experimental results on 3 public datasets show that the proposed algorithm achieves the best performance on 4 objective evaluation indexes compared with 12 state-of-the-art algorithms, especially on COD10K dataset, the weighted F-measure and the Mean Absolute Error (MAE) of the proposed algorithm reach 0.809 and 0.037 respectively. It can be seen that the proposed algorithm achieves better performance on COD tasks.
A hybrid algorithm of Simulated Annealing (SA) and flower pollination algorithm was presented to overcome the problems of low-accuracy computation, slow-speed convergence and being easily relapsed into local extremum. The sudden jump strategy in SA was utilized to avoid falling into local optimum, and the global searching performance of SA was exploited to enhance the global searching ability of the hybrid algorithm. The hybrid algorithm was tested through six standard functions and compared to basic Flower Pollination Algorithm (FPA), Bat Algorithm (BA), Particle Swarm Optimization (PSO) algorithm and improved PSO algorithm. The simulation results show that the optimal value of 4 functions were found by the hybrid algorithm with better convergence precision, convergence rate and robustness. At the same time, the experimental results of solving nonlinear equation group verify the validity of the hybrid algorithm.
In order to solve the problems of Cuckoo Search (CS) algorithm including low optimizing accuracy and weak local search ability, an improved CS algorithm with differential evolution strategy was presented. The individual variation was completed in the algorithm before population with two weighted differences increased on its individuals entering the next iteration, then crossover operation and select operation were performed to obtain optimal individual, which making the CS algorithm lack of mutation mechanism have the variation mechanism, so as to increase the diversity of the CS algorithm, avoid individual species into local optimum and enhance the global optimization ability. The algorithm was put through several classical test functions and a typical application example. The simulation results show that the new algorithm has better global searching ability, and the convergence precision, convergence speed and optimization success rate are significantly better than those of the basic CS algorithm.