Artificial intelligence technologies based on deep convolutional neural networks and large language models have made significant breakthroughs in many tasks, such as image recognition, target detection, semantic segmentation, and natural language processing, but also face a conflict between the high computational capacity of the algorithms and limited deployment resources. Quantization, which converts floating-point neural networks into low-bit-width integer networks, is an important and essential technique for efficient deployment and cost reduction in edge computing. This paper analyzes various existing quantization methods, showcases the deployment accuracy of advanced techniques, and discusses the future challenges and trends in this domain.

Semantic-level land-use scene classification is a challenging problem, in which deep learning methods, e.g., convolutional neural networks (CNNs), have shown remarkable capacity. However, a lack of sufficient labeled images has proved a hindrance to increasing the land-use scene classification accuracy of CNNs. Aiming at this problem, this paper proposes a CNN pre-training method under the guidance of a human visual attention mechanism. Specifically, a computational visual attention model is used to automatically extract salient regions in unlabeled images. Then, sparse filters are adopted to learn features from these salient regions, with the learnt parameters used to initialize the convolutional layers of the CNN. Finally, the CNN is further fine-tuned on labeled images. Experiments are performed on the UCMerced and AID datasets, which show that when combined with a demonstrative CNN, our method can achieve 2.24% higher accuracy than a plain CNN and can obtain an overall accuracy of 92.43% when combined with AlexNet. The results indicate that the proposed method can effectively improve CNN performance using easy-to-access unlabeled images and thus will enhance the performance of land-use scene classification especially when a large-scale labeled dataset is unavailable.

When a satellite performs complex tasks such as discarding a payload or capturing a non-cooperative target, it will encounter sudden changes in the attitude and mass parameters, causing unstable flying and rolling of the satellite. In such circumstances, the change of the movement and mass characteristics are unpredictable. Thus, the traditional attitude control methods are unable to stabilize the satellite since they are dependent on the mass parameters of the controlled object. In this paper, we proposed a reinforcement learning method to re-stabilize the attitude of a satellite under such circumstances. Specifically, we discretize the continuous control torque, and build a neural network model that can output the discretized control torque to control the satellite. A dynamics simulation environment of the satellite is built, and the deep Q Network algorithm is then performed to train the neural network in this simulation environment. The reward of the training is the stabilization of the satellite. Simulation experiments illustrate that, with the iteration of training progresses, the neural network model gradually learned to re-stabilize the attitude of a satellite after unknown disturbance. As a contrast, the traditional PD (Proportion Differential) controller was unable to re-stabilize the satellite due to its dependence on the mass parameters. The proposed method adopts self-learning to control satellite attitudes, shows considerable intelligence and certain universality, and has a strong application potential for future intelligent control of satellites performing complex space tasks.

Artificial intelligence technologies based on deep convolutional neural networks and large lan-guage models have made significant breakthroughs in many tasks such as image recognition, target detection, semantic segmentation and natural language processing, but also face the problem of contradiction between the high computational capacity of the algorithms and the limited deployment resources. Quantization, which converts floating-point neural networks into low-bit-width integer networks, is an important and essential technique for efficient deploy-ment and cost reduction in edge computing. This paper analyze various existing quantization methods, showcase the deployment accuracy of advanced techniques, and discuss the future challenges and trends in this domain.

Local features extraction is a crucial technology for image matching navigation of an unmanned aerial vehicle (UAV), where it aims to accurately and robustly match a real-time image and a geo-referenced image to obtain the position update information of the UAV. However, it is a challenging task due to the inconsistent image capture conditions, which will lead to extreme appearance changes, especially the different imaging principle between an infrared image and RGB image. In addition, the sparsity and labeling complexity of existing public datasets hinder the development of learning-based methods in this research area. This paper proposes a novel learning local features extraction method, which uses local features extracted by deep neural network to find the correspondence features on the satellite RGB reference image and real-time infrared image. First, we propose a single convolution neural network that simultaneously extracts dense local features and their corresponding descriptors. This network combines the advantages of a high repeatability local feature detector and high reliability local feature descriptors to match the reference image and real-time image with extreme appearance changes. Second, to make full use of the sparse dataset, an iterative training scheme is proposed to automatically generate the high-quality corresponding features for algorithm training. During the scheme, the dense correspondences are automatically extracted, and the geometric constraints are added to continuously improve the quality of them. With these improvements, the proposed method achieves state-of-the-art performance for infrared aerial (UAV captured) image and satellite reference image, which shows 4–6% performance improvements in precision, recall, and F1-score, compared to the other methods. Moreover, the applied experiment results show its potential and effectiveness on localization for UAVs navigation and trajectory reconstruction application.