In the field of image enhancement, using deep learning methods to enhance low-light images is currently mainstream. However, the methods often have complex network structures of a large number of parameters, and their training often uses paired data-sets, which are difficult to obtain in actual practice. To solve these problems, this paper proposes a simple generative adversarial network and Retinex model, dubbed RetinexGAN, that is completely trained using unpaired data-sets. It contains a decomposition network and two discriminator networks. To reduce the parameters of the network, only two convolution layers are used in the decomposition network. We show more challenging testing data where some parts of the image are underexposed and others are normal light. Both quantitative and visual results show that RetinexGAN is largely superior to state-of-the-art methods.

How to generate the path planning of mobile robots quickly is a problem in the field of robotics. The Q-learning(QL) algorithm has recently become increasingly used in the field of mobile robot path planning. However, its selection policy is blind in most cases in the early search process, which slows down the convergence of optimal solutions, especially in a complex environment. Therefore, in this paper, we propose a continuous local search Q-Learning (CLSQL) algorithm to solve these problems and ensure the quality of the planned path. First, the global environment is gradually divided into independent local environments. Then, the intermediate points are searched in each local environment with prior knowledge. After that, the search between each intermediate point is realized to reach the destination point. At last, by comparing other RL-based algorithms, the proposed method improves the convergence speed and computation time while ensuring the optimal path.

The orthodontic treatment often relies on the experience of doctors in traditional methods, and there are often fewer doctors with good experience, which is not conducive to improving the efficiency of patient consultation. Therefore, it has become the mainstream research direction in recent years to assist doctors in improving the efficiency of diagnosis by simulating the dental orthodontic process through computers. The orthodontic process is a multiobjective and high-dimensional path planning problem. To optimize the movement path of multiobjective orthodontics and compensate for the movement efficiency of invisible appliances, a preovercorrection orthodontic motion path scheme based on an improved multi-PSO algorithm was proposed to reduce the dimension disaster and the movement cost and improve the success rate of orthodontic surgery. Firstly, the solution set of the multiobjective particle swarm optimization (MOPSO) algorithm is introduced into the multi-PSO path planning algorithm to obtain the orthodontic movement path. Secondly, by analyzing the movement efficiency of the invisible appliance, tooth displacement compensation is designed and evaluated, and the final orthodontic scheme was generated for patients through the overcorrection method. Finally, the scheme is visualized by VTK visualization. The experimental results show that compared with the multi-PSO algorithm, the improved algorithm can reduce the length of the motion path by 10%, and the rotation angle is reduced by 4%. Meanwhile, the preovercorrection scheme designed can provide movement allowance for the orthodontic process, which guarantees that the optimal orthodontic path obtained by the scheme conforms to the clinical experiment results, ensuring that the tooth can move according to the expected path in the clinical experiment and assuring the success rate of orthodontic treatment.

Most safety accidents are caused by human factor in underground resource mining industry. This is because the non-uniform lighted and noisy and dangerous environment easily evokes the negative mental state and results in nonstandard production operation. This paper proposes an edge computing mental state framework of the internet of things in the underground mining industry. Moreover, a filtering algorithm using a defined threshold function is developed. Furthermore, a complemented capsule network model is constructed by using two residual modules. In addition, a two-stage mental state fusion algorithm is proposed with EEG signals and facial expression. At last, the mental state variation characteristics are explored with the illuminating and coloring. Experiments show that the mental state detection accuracy is increased by 2.6%. The higher mental arousal is at the illumination between 320 Lx and 330 Lx.

Alertness mechanism of unmanned monitoring vehicles to environment is important. Especially, the vigilance modeling of underground security robots has a particularly significance because the underground is a dangerous environment. However, there is no a mature methodology for the alertness computation. In this work, four parts of the alertness estimation are focused. First, an autonomous robot alertness mechanism framework is proposed by using the deep reinforcement learning model of the human alertness mechanism. Second, a fast K-T filtering algorithm is developed to eliminate the multiple noises of the electroencephalograph (EEG) signals by the blind source separation and the adjustable Q factor wavelet transform. Third, the description problem of the directed interaction stability of the cortical EEG signals is solved by the ensemble empirical mode decomposition and the directional transfer function. Fourth, the human alertness estimation is explored by using the support vector regression of the dynamically spatial-temporal brain network connection parameters. Experiments show that, the mean square error and the determination coefficient of the explored alertness estimation are respectively 0.115 and 0.8337. Compared with the scalp EEG alertness estimation, it has a better performance because the mean square error is decreased by 0.0684, and the determination coefficient is increased by 0.023.

To address the widespread challenges of significant multi-category dental morphological variations and interference from overlapping anatomical structures in panoramic dental X-ray images, this paper proposes a dual-stream dental segmentation model based on Transformer heterogeneous feature complementarity. Firstly, we construct a parallel architecture comprising a Transformer semantic parsing branch and a Convolutional Neural Network (CNN) detail capturing pathway, achieving collaborative optimization of global context modeling and local feature extraction. Furthermore, a Pooling-Cooperative Convolutional Module was designed, which enhances the model’s capability in detail extraction and boundary localization through weighted centroid features of dental structures and a latent edge extraction module. Finally, a Semantic Transformation Module and Interactive Fusion Module are constructed. The Semantic Transformation Module converts geometric detail features extracted from the CNN branch into high-order semantic representations compatible with Transformer sequential processing paradigms, while the Interactive Fusion Module applies attention mechanisms to progressively fuse dual-stream features, thereby enhancing the model’s capability in holistic dental feature extraction. Experimental results demonstrate that the proposed method achieves an IoU of 91.49% and a Dice coefficient of 94.54%, outperforming current segmentation methods across multiple evaluation metrics.